Violence Assessment and Management

BOOK OF THE MONTH TERMS OF USE
PsychiatryOnline.com currently offers access in PDF format to

a free book each month from the APPI Bookstore. This offer-
ing is intended as a promotional tool, to give subscribers a
chance to review books they might be interested in purchas-
ing separately for continued access. As such, the “Book of the
Month” is governed by the following terms of use:
The Book of the Month is supplied as PDF; no other formats will be

supplied. Subscribers have access to the Book of the Month for only
as long as the “download” link appears on the PsychiatryOnline.com
home page (usually one month). This feature may be discontinued
at any time without notice or reason to subscribers.
The User license does not cover storing the book in any format be-
yond the one-month period. Users may not copy, modify, reproduce,
publish, transmit, display, broadcast, rent, lend, sell, catalog, or
otherwise distribute the Book of the Month.
Click here to continue on to the Book of the Month.

Textbook of
Violence Assessment
and Management
This page intentionally left blank
Textbook of
Violence Assessment
and Management
Edited by
Robert I. Simon, M.D.
Kenneth Tardiff, M.D., M.P.H.
Washington, DC
London, England
Note: The authors have worked to ensure that all information in this book is accu-
rate at the time of publication and consistent with general psychiatric and medical
standards, and that information concerning drug dosages, schedules, and routes of
administration is accurate at the time of publication and consistent with standards
set by the U.S. Food and Drug Administration and the general medical community.
As medical research and practice continue to advance, however, therapeutic stan-
dards may change. Moreover, specific situations may require a specific therapeutic
response not included in this book. For these reasons and because human and me-
chanical errors sometimes occur, we recommend that readers follow the advice of
physicians directly involved in their care or the care of a member of their family.
Books published by American Psychiatric Publishing, Inc., represent the views and
opinions of the individual authors and do not necessarily represent the policies and
opinions of APPI or the American Psychiatric Association.
To purchase 25–99 copies of this or any other APPI title at a 20% discount, please
contact APPI Customer Service at [email protected] or 800–368–5777. To purchase 100
or more copies of the same title, please e-mail [email protected] for a price quote.
Copyright © 2008 American Psychiatric Publishing, Inc.
ALL RIGHTS RESERVED
Manufactured in the United States of America on acid-free paper
12 11 10 09 08 5 4 3 2 1
First Edition
Typeset in Adobe Palatino and The Mix
American Psychiatric Publishing, Inc.
1000 Wilson Boulevard
Arlington, VA 22209–3901
www.appi.org
Library of Congress Cataloging-in-Publication Data

Textbook of violence assessment and management / edited by Robert I. Simon, Ken-
neth Tardiff. -- 1st ed.
p. ; cm.
Includes bibliographical references and index.
ISBN 978-1-58562-314-3 (alk. paper)
1. Dangerously mentally ill. 2. Risk assessment. 3. Violence. I. Simon, Robert I. II.
Tardiff, Kenneth, 1944-
[DNLM: 1. Social Behavior Disorders--diagnosis. 2. Violence--prevention & con-
trol. 3. Mentally Ill Persons--psychology. 4. Professional-Patient Relations. 5. Risk
Assessment--methods. 6. Social Behavior Disorders--therapy. WM 600 T355 2008]
RC569.5.V55T47 2008
616.8582--dc22
2008004457
British Library Cataloguing in Publication Data

A CIP record is available from the British Library.
Dedicated to all who are committed to understanding,
treating, and preventing violence.
Contents
Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi
Foreword. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xvii
Paul S. Appelbaum, M.D.
Preface. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xxi
Acknowledgments. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .xxiii
I
ASSESSMENT PRINCIPLES
1 Clinical Risk Assessment of Violence. . . . . . . . . . . . . . . . . . . .3

2 Structured Risk Assessment of Violence . . . . . . . . . . . . . . 17

John Monahan, Ph.D.
3 Cultural Competence in Violence Risk Assessment . . . . 35

Russell F. Lim, M.D.
Carl C. Bell, M.D.
4 Psychological Testing in Violence Risk Assessment . . . . 59

Barry Rosenfeld, Ph.D., A.B.P.P.
Ekaterina Pivovarova, M.A.
II
MENTAL DISORDERS AND CONDITIONS
5 Mood Disorders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
Rif S. El-Mallakh, M.D.
R. Jeannie Roberts, M.D.
Peggy L. El-Mallakh, Ph.D.
6 Schizophrenia and Delusional Disorder . . . . . . . . . . . . . . 105

Martha L. Crowner, M.D.
7 Posttraumatic Stress Disorder . . . . . . . . . . . . . . . . . . . . . . .123

Thomas A. Grieger, M.D., D.F.A.P.A.
David M. Benedek, M.D., D.F.A.P.A.
Robert J. Ursano, M.D., D.F.A.P.A.
8 Substance Abuse Disorders. . . . . . . . . . . . . . . . . . . . . . . . . . 141

Rodney Burbach, M.D.
9 Personality Disorders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161

William H. Reid, M.D., M.P.H.
Stephen A. Thorne, Ph.D.
10 Neurological and Medical Disorders . . . . . . . . . . . . . . . . . 185

Karen E. Anderson, M.D.
Jonathan M. Silver, M.D.
11 Impulsivity and Aggression . . . . . . . . . . . . . . . . . . . . . . . . . 211

Sara T. Wakai, Ph.D.
Robert L. Trestman, Ph.D., M.D.
III
TREATM ENT S ET TINGS
12 Outpatient Settings. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237

James C. Beck, M.D., Ph.D.
13 Inpatient Settings. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .259

Cameron D. Quanbeck, M.D.
Barbara E. McDermott, Ph.D.
14 Emergency Services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 277
Jean-Pierre Lindenmayer, M.D.
Anzalee Khan, M.S.
IV
TREATM ENT AND MANAGEMENT
15 Psychopharmacology and Electroconvulsive

Therapy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 301
Leslie Citrome, M.D., M.P.H.
16 Psychotherapeutic Interventions . . . . . . . . . . . . . . . . . . . . 325

John R. Lion, M.D.
17 Seclusion and Restraint . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 339

John R. Lion, M.D.
V
SPECIAL POPULATIONS
18 Children and Adolescents . . . . . . . . . . . . . . . . . . . . . . . . . . . 359

Peter Ash, M.D.
19 The Elderly. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .381

Robert Weinstock, M.D.
Stephen Read, M.D.
Gregory B. Leong, M.D.
J. Arturo Silva, M.D.
VI
S P E C I A L TO P I C S
20 Forensic Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 409

Charles L. Scott, M.D.
21 Legal Issues of Prediction, Protection,

and Expertise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 429
Daniel W. Shuman, J.D.
Britt Darwin-Looney, J.D.
22 Sexual Violence and the Clinician. . . . . . . . . . . . . . . . . . . . 441
John M.W. Bradford, M.B.Ch.B., D.P.M.
Paul Fedoroff, M.D.
Philip Firestone, Ph.D.
23 Violence Toward Mental Health Professionals . . . . . . . . 461

William R. Dubin, M.D.
Autumn Ning, M.D.
24 Intimate Partner Violence and the Clinician. . . . . . . . . . 483

Susan Hatters Friedman, M.D.
Joy E. Stankowski, M.D.
Sana Loue, Ph.D., J.D., M.P.H.
25 Workplace Violence and the Clinician. . . . . . . . . . . . . . . . 501

Ronald Schouten, M.D., J.D.
26 Vehicular Crashes and the Role of

Mental Health Clinicians . . . . . . . . . . . . . . . . . . . . . . . . . . . . 521
Alan R. Felthous, M.D.
Thomas M. Meuser, Ph.D.
Thomas Ala, M.D.
27 School Violence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 537

Carl P. Malmquist, M.D., M.S.
28 Clinically-Based Risk Management of

Potentially Violent Patients. . . . . . . . . . . . . . . . . . . . . . . . . . 555
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .567
Contributors
Thomas Ala, M.D.

Associate Professor of Clinical Neurology, Center for Alzheimer’s Dis-
ease and Associated Disorders, Southern Illinois University School of
Medicine, Springfield, Illinois

Assistant Professor, Psychiatry and Neurology Movement Disorders,
University of Maryland School of Medicine, and Director, University of
Maryland Huntington’s Disease Clinic, Baltimore, Maryland

Elizabeth K. Dollard Professor of Psychiatry, Medicine and Law, and
Director, Division of Psychiatry, Law and Ethics, Department of Psychi-
atry, Columbia University College of Physicians and Surgeons, New
York, New York
Peter Ash, M.D.

Associate Professor, Department of Psychiatry and Behavioral Sciences,
Emory University, Atlanta, Georgia

Professor of Psychiatry, Harvard Medical School; Associate Director,
Law and Psychiatry Service, Department of Psychiatry, Massachusetts
General Hospital, Boston, Massachusetts
Carl C. Bell, M.D.

President/Chief Executive Officer, Community Mental Health Council
and Foundation, Inc.; Professor of Psychiatry and Public Health, Uni-
versity of Illinois at Chicago, Chicago, Illinois

Associate Professor, Department of Psychiatry, Uniformed Services
University, Bethesda, Maryland
xi
xii ❘ Textbook of Violence Assessment and Management

Professor and Head of Division of Forensic Psychiatry, University of
Ottawa; Associate Chief, Royal Ottawa Health Care Group, Ottawa,
Ontario, Canada

Medical Review Officer, U.S. Nuclear Regulatory Commission, Wash-
ington, D.C.; Chairman, Department of Family Practice, and former
Medical Director, Addiction Treatment Center, Suburban Hospital Be-
thesda, Maryland; Clinical Instructor, Georgetown University School of
Medicine, Washington, D.C.

Professor of Psychiatry, New York University School of Medicine, New
York, New York; Director, Clinical Research and Evaluation Facility,
Nathan S. Kline Institute for Psychiatric Research, Orangeburg, New York

Associate Clinical Professor of Psychiatry, Columbia College of Physi-
cians and Surgeons, New York, New York

Praesidium, Inc., Arlington, Texas

Professor and Vice Chair, Department of Psychiatry, Temple University
School of Medicine; Chief Medical Officer, Temple University Hospital-
Episcopal Campus, Philadelphia, Pennsylvania

Assistant Professor, Department of Nursing Education, University of
Louisville School of Nursing, Louisville, Kentucky

Associate Professor, Director, Mood Disorders Research Program, De-
partment of Psychiatry and Behavioral Sciences, University of Louis-
ville School of Medicine, Louisville, Kentucky
Paul Fedoroff, M.D.

Associate Professor, Division of Forensic Psychiatry, Department of
Psychiatry, University of Ottawa; Director of Forensic Research Unit,
University of Ottawa Institute of Mental Health Research, Ottawa,
Ontario, Canada
Contributors ❘ xiii

Professor and Director of Forensic Psychiatry Division, Department of
Neurology and Psychiatry, St. Louis University School of Medicine, St.
Louis, Missouri

Professor, Department of Psychology, Faculty of Social Sciences,
University of Ottawa, Ottawa, Ontario, Canada

Senior Instructor in Psychiatry, Case Western Reserve University School
of Medicine; forensic psychiatrist, Northcoast Behavioral Healthcare,
Cleveland, Ohio

Private practice in forensic psychiatry, Falls Church, Virginia
Anzalee Khan, M.S.

Ph.D. candidate, Department of Psychometrics, Fordham University
Rose Hill Campus, Bronx, New York

Clinical Professor of Psychiatry, Department of Psychiatry and Be-
havioral Sciences, University of Washington School of Medicine, Seat-
tle, Washington; psychiatrist, Western State Hospital, Tacoma, Wash-
ington

Associate Clinical Professor, Department of Psychiatry and Behavioral
Sciences, University of California Davis School of Medicine; Staff Psy-
chiatrist and Director of Diversity Education and Training, Adult Psy-
chiatric Support Services Clinic (APSSC) of Sacramento County,
Sacramento, California

Director, Psychopharmacology Research Unit, Manhattan Psychiatric
Center, Nathan S. Kline Institute for Psychiatric Research; Clinical Pro-
fessor, Department of Psychiatry, New York University, New York,
New York
John R. Lion, M.D.

Clinical Professor of Psychiatry, University of Maryland School of Med-
icine, Baltimore, Maryland
xiv ❘ Textbook of Violence Assessment and Management

Professor in Epidemiology and Biostatistics, Case Western Reserve Uni-
versity School of Medicine, Cleveland, Ohio

Professor of Social Psychiatry, University of Minnesota, Minneapolis,
Minnesota

Associate Professor of Clinical Psychiatry, Department of Psychiatry
and Behavioral Sciences, Division of Psychiatry and the Law, Univer-
sity of California, Davis, California

Director of Gerontology, Associate Professor of Social Work and Psy-
chology, University of Missouri, St. Louis, Missouri
John Monahan, Ph.D.

John S. Shannon Distinguished Professor of Law, Professor of Psychol-
ogy and Psychiatric Medicine, University of Virginia, Charlottesville,
Virginia
Autumn Ning, M.D.

Instructor and Assistant Training Director, Department of Psychiatry,
Temple University School of Medicine; Medical Director, Crisis Re-
sponse Center, Temple University Hospital-Episcopal Campus, Phila-
delphia, Pennsylvania

Doctoral candidate, Department of Psychology, Fordham University,
Bronx, New York

Assistant Clinical Professor, Department of Psychiatry and Behavioral
Sciences, Division of Psychiatry and the Law, University of California,
Davis, California
Stephen Read, M.D.

Clinical Professor of Psychiatry, Department of Psychiatry and Biobe-
havioral Sciences, University of California, Los Angeles, David Geffen
School of Medicine; psychiatrist, Greater Los Angeles Veterans Affairs
Health Care System, Los Angeles, California
Contributors ❘ xv

Clinical Professor of Psychiatry, University of Texas Health Science
Center, San Antonio, Texas

Mood Disorders Research Program; Instructor, Department of Psychia-
try and Behavioral Sciences, University of Louisville School of Medi-
cine, Louisville, Kentucky

Professor, Department of Psychology, Fordham University, Bronx, New
York

Associate Professor of Psychiatry, Harvard Medical School; Director,
Law and Psychiatry Service, Massachusetts General Hospital, Boston,
Massachusetts

Chief, Division of Psychiatry and the Law, Clinical Professor of Psychi-
atry, and Director, Forensic Psychiatry Fellowship, University of Cali-
fornia Davis Medical Center, Sacramento, California

M.D. Anderson Foundation Endowed Professor of Health Law, Ded-
man School of Law, Southern Methodist University, Dallas, Texas

Clinical Professor of Psychiatry and Director, Program in Psychiatry
and Law, Georgetown University School of Medicine, Washington, D.C.

Private practice of psychiatry, San Jose, California

Clinical Professor of Psychiatry, New York University School of Medi-
cine, New York, New York; Fellow, American Neuropsychiatric Associ-
ation; Diplomate, Behavioral Neurology and Psychiatry

Senior Instructor in Psychiatry, Case Western Reserve University School
of Medicine; Chief Clinical Officer, Northcoast Behavioral Healthcare,
Cleveland, Ohio
xvi ❘ Textbook of Violence Assessment and Management

Professor of Psychiatry and Public Health, Department of Psychiatry,
Weill Cornell Medical College, New York, New York

Adjunct faculty, St. Edwards University, Austin, Texas

Director, Connecticut Health; Director, Center for Correctional Mental
Health Services Research and Professor of Medicine and Psychiatry,
University of Connecticut Health Center, Farmington, Connecticut

Professor and Chairman, Department of Psychiatry, Uniformed Ser-
vices University, Bethesda, Maryland

Assistant Professor, Center for Correctional Mental Health Services Re-
search, Department of Medicine, University of Connecticut Health Cen-
ter, Farmington, Connecticut

Clinical Professor of Psychiatry, Department of Psychiatry and Biobe-
havioral Sciences, University of California, Los Angeles, David Geffen
School of Medicine; psychiatrist, Greater Los Angeles Veterans Affairs
Health Care System, Los Angeles, California
Foreword
How is it that clinicians in the mental health system have come to have
responsibility for the assessment and management of violence? The an-
swer lies deep in the history of our field. People with serious mental dis-
orders have long been feared for the oddness of their behavior, and in
some cases for the occasional acts of violence that may punctuate their
lives. Indeed, funding for the first hospital in the colonial United States,
the Pennsylvania Hospital in Philadelphia, which opened in 1751, was
obtained in part by the promise of having a place to contain the mentally
ill, thus diminishing the perceived threat they posed to the populace.
From its inception, then, the mental health system has been intimately
linked with the prediction and prevention of violence.
One cannot proceed further without noting that the relationship be-
tween mental illnesses and violence has been much exaggerated over
the years. Epidemiologic and cohort studies taken as a whole suggest
some increased risk of violence in persons with major mental disorders.
But not all studies support this conclusion, and in any event the contri-
bution of serious mental disorders is dwarfed by the effects of substance
abuse (especially alcohol abuse) and personality traits such as psycho-
pathy. The best available estimate suggests that only 3%–5% of the risk
for violence in the United States is attributable to mental illnesses, and
it has long been clear that people with serious mental illnesses are much
more likely to be the victims of violence than its perpetrators.
Nonetheless, mental illness is sometimes causally linked with vio-
lence, as when patients act on command hallucinations to harm other
people or strike out in fear against imagined persecutors. Manic pa-
tients in their irritability or depressed patients in their hopelessness
may also cause harm to others. All of these conditions and others are ex-
acerbated by the simultaneous use of alcohol and other disinhibiting
substances, the abuse of which is more common among people with
mental disorders.
As a result, nearly every phase of the mental health evaluation and
treatment process may involve the assessment of violence risk and de-
cisions about its management. This includes outpatient screening and
xvii
xviii ❘ Textbook of Violence Assessment and Management
intake, inpatient admissions and discharges, and emergency assess-

ments. In addition, clinicians are often called upon to assess violence risk
of people with mental illnesses in the criminal justice system, including
in court clinics, forensic hospitals, and jails and prisons. Increasingly
common diversion programs, such as mental health courts, may also call
for clinical assessments of violence risk. Such importance does society
confer on these tasks—and such confidence does it have in the ability of
clinicians to perform them—that failure to conform to accepted stan-
dards of assessment and management may leave clinicians open to the
imposition of liability and to no small amount of public opprobrium.
How well do clinicians perform their assigned predictive tasks?
Most studies have not been encouraging for long-term predictions of
patient violence, and even over shorter time periods, only modest pre-
dictive accuracy has been found. (It should be noted, though, that all
research on clinical prediction of violence is complicated by the need to
intervene to prevent violence when its occurrence is thought likely.) The
less than impressive data on clinical prediction have led to the develop-
ment of actuarial prediction instruments, with the hope that they would
show better results. These models are based on the standardized collec-
tion of information about key variables, to which a predetermined
algorithm is then applied. Indeed, studies of the application of these
predictive models generally yield accuracies superior to that of unaided
clinical judgment. Moreover, this work has stimulated an important re-
conceptualization of the predictive process, from making dichotomous
judgments of the dangerousness of given individual patients to estimat-
ing the violence risk category to which a patient belongs.
Have we then reached the promised land of violence risk assess-
ment, where reliance on the new actuarial approaches will supplant the
admittedly imperfect clinical predictions that have dominated practice?
Not quite yet. Whatever virtues the actuarial approach may have—and
they clearly include systematizing the assessment and ensuring that im-
portant variables are not overlooked—existing instruments, too, are far
from perfect. Their accuracy is less than ideal and falls sharply as the
base rate of violence in the population being examined drops. In their
rigidity, they do not permit consideration of contingent factors that may
affect risk (e.g., the patient has a broken leg and will not be walking for
the next six weeks). Many models assume that a single set of predictors
will apply to all patients, an assumption implausible on its face. Per-
haps most important of all, they rely heavily on “static” risk variables,
invariant over time (e.g., past history of violence, exposure to abuse as
a child). Thus, once labeled “high risk” by one of the actuarial ap-
proaches, a patient has only a faint chance to lose that designation.
Foreword ❘ xix
If the promised land is not yet in sight, there is still no question that
this is an exciting time for the science and practice of violence risk as-
sessment. Clinicians and researchers are experimenting with ways of
combining actuarial and clinical approaches, in the hope of achieving
levels of accuracy unattainable with either alone. Behavioral geneticists
have begun to identify gene variants that may be implicated in violence
risk and have begun to develop initial models of the interactions of these
variants with environmental variables. Functional magnetic resonance
imaging, positron emission tomography, and other brain scanning tech-
niques are being used to identify neural circuits that may inhibit or
facilitate violent acts. With the knowledge being gained, it seems likely
that violence risk prediction in the next generation will look very differ-
ent than it does today.
The same can be said with regard to the present relatively primitive
level of management and treatment of violence risk. When violence ap-
pears to be causally linked to the symptoms of mental illnesses, our
contemporary interventions typically focus on symptom control. Rela-
tively few interventions—anger management programs are one—target
the propensity to violence per se. This situation may also improve in the
future as advances in the science of violence causation promote the de-
velopment of more specific interventions. Indeed, we may some day
legitimately be able to speak of prevention of violent propensities,
rather than management of risk, as clinicians’ primary task.
However promising the future, clinicians must address the needs of
patients and the demands of society today. Thus the importance of this
book, which summarizes in an accessible format the state of the art of vi-
olence assessment and management for mental health clinicians. Here,
clinical and actuarial approaches to risk assessment are reviewed, the
association of violence with specific psychiatric disorders is addressed,
and management of violence is considered in depth. The important set-
tings of the home, workplace, school, and healthcare facility are consid-
ered, as are the legal and risk management issues of which all clinicians
should be aware. As a guide to dealing with violence today this volume
is unparalleled, and I commend it to your thoughtful attention.

Elizabeth K. Dollard Professor of Psychiatry, Medicine and Law
Director, Division of Psychiatry, Law and Ethics
Department of Psychiatry
Columbia University College of Physicians and Surgeons
Preface
The assessment and management of violent ideations and behaviors is

a core competency that clinicians must possess or acquire. It is the rare
clinician who does not assess and treat patients at risk for violence. Ac-
cordingly, this book leads off with “Assessment Principles.” In this
important first section, Drs. Tardiff and Monahan present tutorials on
the clinical and actuarial risk assessment of violence. Drs. Lim and Bell
underscore the importance of clinicians’ competence in evaluating indi-
viduals for violence who are from different cultures and racial groups.
Dr. Rosenfeld and Ms. Pivovarova provide a much-needed review of
adjunctive psychological testing in violence risk assessment.
The textbook’s 28 chapters address the diversity of clinical settings
and situations where psychiatrists and other mental health profession-
als evaluate, treat, and manage people with violent ideations and
behaviors. The different sections of the book address assessment prin-
ciples, mental disorders and conditions, treatment settings, treatment
and management, special populations, and various special topics.
Chapter authors combine evidence-based medicine with expert opin-
ion. The chapters include clinical case examples that are integrated into
an in-depth discussion. Each chapter ends with a list of key points that
underscore the main learning objectives.
We are fortunate to have enlisted distinguished academicians and
clinicians to contribute their experiences to this textbook. Some authors
have collaborated with junior colleagues to craft up-to-date, cutting-
edge chapters. Each chapter was carefully reviewed by the editors.
Much latitude was given to the different authors’ writing styles and
clinical perspectives.
Overlap among chapters is inevitable but useful. Very few people
read a textbook from cover to cover. Instead, readers will select chapters
of special importance to review in regard to pressing clinical situations
or teaching needs.
Violence is endemic and epidemic. Its roots and causes are many
and complex. Violence is not a diagnosis, although it can be associated
with psychiatric conditions. People arrested for violent acts come under
xxi
xxii ❘ Textbook of Violence Assessment and Management
the jurisdiction of law enforcement agencies and the judicial system.

Many have mental disorders. Few are seen by psychiatrists. Some of-
fenders are diverted to mental health courts and forensic treatment
facilities. Although psychiatrists and other mental health professionals
consult with these agencies, their encounters with individuals who be-
have violently or have violent thoughts usually occur in the treatment
of outpatients and inpatients.
For clinicians in the trenches, evaluating and treating patients with
violent ideations and behaviors can be anxiety-provoking, frustrating,
sometimes dangerous, and occasionally legally fraught. However, the
clinician does not have to worry alone. The Textbook of Violence Assess-
ment and Management is on call 24 hours a day for expert consultation,
just an arm’s reach away.

Kenneth Tardiff, M.D.
Acknowledgments
T his textbook could not have been published without the committed
efforts of many individuals. We want to express our gratitude to all the
authors for the time and effort they devoted to researching and writing
chapters of such high quality. It is their book as much as ours.
We also want to thank Robert E. Hales, M.D., M.B.A., Editor-in-
Chief of American Psychiatric Publishing, Inc., and John McDuffie, Ed-
itorial Director, for their vision and support for a textbook on violence
assessment and management.
Many thanks go to Ms. Tina Coltri-Marshall for her outstanding
work in the difficult task of coordinating the entire textbook project
among the editors, numerous authors, and American Psychiatric Pub-
lishing staff. Special appreciation goes to Ms. Carol A. Westrick for her
competence and undaunted work ethic.
This work was funded, in part, by grant DA06534 from the National
Institute on Drug Abuse.
—R.I.S., K.T.
xxiii
xxiv ❘ Textbook of Violence Assessment and Management
Disclosure of Interests
The contributors have declared all forms of support received within the 12 months prior
to manuscript submittal that may represent a competing interest in relation to their
work published in this volume, as follows:
Carl C. Bell, M.D.: Consultant, AstraZeneca
John M. Bradford, M.D.: Grant support, Canadian Institute Health Research,
Janssen-Ortho; Speakers’ bureau, Janssen-Ortho, Pfizer
Leslie Citrome, M.D., M.P.H.: Consultancy, honoraria, or clinical research support,
Abbott Laboratories, AstraZeneca, Barr Laboratories, Bristol-Myers Squibb,
Eli Lilly and Company, GlaxoSmithKline, Janssen Pharmaceutica, Jazz Phar-
maceuticals, Pfizer
William R. Dubin, M.D.: Speakers’ bureau, AstraZeneca, Pfizer
The following contributors stated that they had no competing interests during the year
preceding manuscript submittal:
Thomas Ala, M.D.; Peter Ash, M.D.; James C. Beck, M.D.; Martha Crowner, M.D.;
Peggy L. El-Mallakh, Ph.D., R.N.; Rif S. El-Mallakh, M.D.; Paul Fedoroff, M.D.;
Alan R. Felthouse, M.D.; Philip Firestone, Ph.D.; Susan Hatters Friedman, M.D.;
Thomas A. Grieger, M.D., D.F.A.P.A.; Anzalee Khan, M.S.; Jean-Pierre Linden-
mayer, M.D.; John R. Lion, M.D.; Barbara E. McDermott, Ph.D.; John Monahan,
Ph.D.; Autumn Ning, M.D.; Cameron D. Quanbeck, M.D.; William H. Reid, M.D.,
M.P.H.; R. Jeannie Roberts, M.D.; Charles L. Scott, M.D.; Daniel W. Shuman, J.D.;
Jonathan M. Silver, M.D.; Robert I. Simon, M.D.; Kenneth Tardiff, M.D.; Robert L.
Trestman, Ph.D., M.D.; Sara T. Wakai, Ph.D.
P A R T I
Assessment Principles
C H A P T E R 1
Clinical Risk Assessment

of Violence
T his chapter presents a clinical model for the short-term risk assess-
ment of violence. Assessment of the risk of violence by patients is ex-
pected of all clinicians who have a relationship with a patient for eval-
uation or treatment: psychiatrists, other physicians, psychologists,
nurses, social workers, activity therapists, and all other staff members
who have responsibilities for patients. The responsibility to assess the
risk of violence exists when clinicians evaluate a patient in the emer-
gency department and decide whether to discharge or admit; see a pa-
tient in an office setting for the first time and between outpatient visits;
admit a patient to a hospital and order a level of observation; and pro-
vide other immediate treatment as the patient enters the hospital. It ex-
ists during in-hospital diagnosis and treatment, including monitoring
the patient and deciding whether seclusion or restraint should be used.
It exists in the decision to discharge the patient and in the planning and
implementation of care after discharge.
The focus of this chapter is to describe clinical methods to evaluate
the risk of violence in the short term (within days or a week), because
an increased risk of violence should result, as soon as possible, in pre-
ventive clinical actions such as change in medication, monitoring, and
admitting to or discharge from a hospital. Unlike clinical methods of
assessing the risk of violence, actuarial methods of assessing the risk of
violence use structured instruments with items that are selected to
3
4 ❘ Textbook of Violence Assessment and Management
measure areas thought to be related to the overall risk of violence in an

individual. Usually these items are scored and used to predict the long-
term risk of violence after discharge from prison or forensic psychiatric
treatment facilities. The actuarial method has been applied to psychiat-
ric patients in the long-term prediction of violence; however, this does
not assist the clinician in the treatment of a potentially violent patient,
because the clinician managing such a patient is primarily interested in
the risk of violence in the next few days to a week. A number of re-
searchers have reviewed many instruments that assess the risk of vio-
lence by using actuarial methods. They found that greater integration
of clinical, dynamic data more relevant to general adult psychiatry is
needed in the assessment of the short-term risk of violence (Harris et al.
2004; Kumar and Simpson 2005; Kroner et al. 2005; Mills 2005).
Principles in the Assessment of the Risk of Violence

A well-trained psychiatrist or other mental health professional should
be able to assess a patient’s short-term violence potential with assess-
ment techniques analogous to those used in the short-term prediction
of suicide potential. The time frame for both is several days to a week.
Beyond that time, many factors may intervene after the initial decision
is made about risk, as in the case of the stabilized schizophrenic patient
who stops his or her medication or the abstinent spouse abuser who re-
sumes drinking. As in the assessment of suicide risk, the evaluator fo-
cuses on the clinical aspects of the evaluation—namely, psychopathol-
ogy—but also must take into consideration demographic, historical,
and environmental factors that may be related to an increased risk of vi-
olence or suicide (Tardiff et al. 2000).
The evaluation of violence potential is analogous to that of suicide
potential. Even if the patient does not express thoughts of violence, the
clinician should routinely ask the subtle question, “Have you ever lost
your temper?” in much the same way as one would check for suicide
potential with the question “Have you ever felt that life was not worth
living?” If the answer is “yes” in either case, the evaluator should pro-
ceed with the evaluation in terms of how, when, and so on, with refer-
ence to violence as well as suicide potential.
When making a decision about violence potential, the clinician also
should interview family members, police, and other persons with infor-
mation about the patient and about violent incidents to ensure that the
patient is not minimizing his or her dangerousness. It is also important
to contact or attempt to contact the patient’s current and past therapists
Clinical Risk Assessment of Violence ❘ 5
and review old charts for previous episodes of violence, police and ar-
rest reports, and other available records such as judicial proceedings.
Factors in the Assessment of the Risk of Violence

The model presented in this section describes at least 10 factors that
must be evaluated in determining whether a patient poses a short-term
risk of violence (Table 1–1). These factors are not scored to produce a
global numerical indication of risk, such as 6 out of 10 indicating risk
but 4 out of 10 not indicating risk of short-term violence. Rather, infor-
mation obtained in each area should be synthesized and weighed by the
evaluator to form a clinical decision about short-term risk of violence.
The clinician must keep in mind that some factors may be more impor-
tant than others for the individual patient, such as a history of violence
with alcohol use or accompanying noncompliance with medication or
other aspects of treatment. Even after making a decision about the pa-
tient’s risk of violence, the clinician must keep in mind that unexpected
events can still occur that may provoke violence, such as resumption of
alcohol or drug use or a patient’s spouse asking for a divorce.
This model represents a consensus among experts that has not been
empirically tested but has been used as a standard by myself and other
psychiatrists, both in testifying as expert witnesses in a number of mal-
practice suits and in daily practice.
Appearance of the Patient

The appearance of the patient may prompt further scrutiny of the
potential for violence. This applies both to the loud, agitated, angry-
TABLE 1–1. Factors that must be evaluated in the assessment of

short-term risk of violence
1. Appearance of the patient
2. Presence of violent ideation and degree of formulation and/or planning
3. Intent to be violent
4. Available means to harm and access to the potential victim
5. Past history of violence and other impulsive behaviors
6. Alcohol or drug use
7. Presence of psychosis
8. Presence of certain personality disorders
9. History of noncompliance with treatment
10. Demographic and socioeconomic characteristics
appearing patient who is impatient and refuses to comply with the

usual intake procedures in the emergency department or clinic and to
the quiet, guarded patient to whom one must carefully listen to detect
subtle violent ideation. Dysarthria, unsteady gait, dilated pupils, trem-
ors, and other signs of acute drug or alcohol intoxication dictate caution
and serious consideration of the potential for violence, even though
threats of violence may not have been expressed.
Presence of Violent Ideation and Degree of

Planning/Formulation
The clinician should begin by assessing whether the patient has
thoughts of violence toward other persons. As in the evaluation of sui-
cide ideation, evaluation of violent ideation includes assessment of how
well planned the ideation or threat is—that is, the degree of formu-
lation. Relatively speaking, vague threats of killing someone, such as
“I’m going to get even with her” or “She’ll be sorry to see me,” are not
as serious as the patient’s saying, “I’m going to kill my wife with a gun
because she had an affair.”
Intent
If a patient has thoughts of harming someone, it is important to explore
whether he or she really intends to do something or is just having
thoughts of violence. This disclosure may arise during an outpatient
treatment session, as an offhand comment on the inpatient unit, or dur-
ing any other contact with the patient. The patient’s mere thought of vi-
olence may not be sufficient for the clinician to take actions such as
warning someone, changing medication, or hospitalizing the patient.
For some patients these thoughts of violence may seem intrusive, alien,
and disturbing, and they will say that they do not intend to do anything
to carry them out.
Available Means
The availability of a means of inflicting injury or death is important in
the assessment of violence potential. If the patient is thinking about get-
ting a gun or already has one, the clinician should obviously take a
threat of violence more seriously. The clinician always should ask a po-
tentially violent patient if he or she has or has ready access to a gun. Vig-
orous efforts should be made to have the patient get rid of the gun or to
have it taken from the patient by family members or others. Removal of
the gun must be verified by a callback by staff. When guns are removed,
the potential for homicide is reduced; however, that does not necessar-
ily preclude the patient’s attacking the victim in other, less lethal ways.
“Available means” also applies to the physical availability of the po-
tential victim. How easily accessible is he or she to the patient? Does the
potential victim live in a secluded place or in a city building without a
doorman? Geography is another aspect of availability. A schizophrenic
patient who threatens his or her father may be more of an immediate
threat if actually living with the father as opposed to living in a different
city or state at a distance from him.
Past History of Violence or Impulsive Behaviors

A past history of violence or other impulsive behaviors is often related
to future violence. Clinicians should ask about injuries to other persons,
destruction of property, suicide attempts, reckless driving, reckless
spending, criminal offenses, sexual acting out, and other impulsive be-
haviors. Past violence increases the risk of future violence by a patient.
Episodes of past violence should be “dissected” in a detailed, concrete
manner by the clinician. This includes obtaining details as to the time
and place of past violence; who was present; who said what to whom;
what the patient saw; what the patient remembers; what family mem-
bers, friends, or staff members remember about the violent episode; why
the patient was violent (e.g., because of psychosis); and what could have
been done to avoid the violence. Often there is a pattern of escalation of
violence, whether it involves the dynamics of a couple in a domestic vi-
olence situation or the increasing agitation of a schizophrenic inpatient
for whom interactions with other patients have become too intense.
The past history of violence should be treated as any other medical
symptom. This includes noting the date of onset, frequency, place, and
severity of violence. Severity is measured by degree of injury to the vic-
tim, from pushing to punching, causing injuries such as bruises, and on-
ward to more serious injuries such as broken bones, lacerations, internal
injuries, or even death. Severity, target, and frequency of violence can be
measured by a written instrument such as the Overt Aggression Scale
(Yudofsky et al. 1986 ; see Figure 1–1). Information that should be ob-
tained and recorded about past history of violence includes prior psy-
chological testing, imaging, laboratory testing, and other evaluations, as
well as past treatment, hospitalizations, and response to treatments.
Alcohol and Drug Use

Alcohol and drug use can exacerbate the psychopathology in other
psychiatric disorders and can cause violence in persons with no other
Name of patient ________________ Name of rater ________________

Sex of patient ______ Date ___________ Shift _________________
Aggressive behavior (check all that apply)
Verbal aggression
____ Makes loud noises, shouts angrily
____ Yells mild personal insults (e.g., “You’re stupid”)
____ Curses viciously, uses foul language in anger, makes moderate threats to
others or self
____ Makes clear threats of violence toward other or self (e.g., “I’m going to
kill you”) or requests help to control self.
Physical aggression against objects
____ Slams door, scatters clothing, makes a mess
____ Throws objects down, kicks furniture without breaking it, marks the wall
____ Breaks objects, smashes windows
____ Sets fires, throws objects dangerously
Physical aggression against self
____ Picks or scratches skin, hits self, pulls hair (with no or minor head injury
only)
____ Bangs head, hits fist into objects, throws self onto floor or into objects
(hurts self without serious injury)
____ Small cuts or bruises, minor burns
____ Mutilates self, makes deep cuts, bites that bleed, internal injury, fracture,
loss of consciousness, loss of teeth
Physical aggression against other people
____ Makes threatening gestures, swings at people, grabs at clothes
____ Strikes, kicks, pushes, pulls hair (without injury to them)
____ Attacks others, causing mild or moderate physical injury (bruises, sprain,
welts)
____ Attacks others, causing severe physical injury (broken bones, deep
lacerations, internal injury)
Time incident began_________________ Duration ______________________

Intervention: _______________________________________________________
__________________________________________________________________
FIGURE 1–1. The Overt Aggression Scale.

Adapted from Yudofsky et al. 1986.
psychiatric disorder. It is important to recognize that alcohol and many

drugs can produce violence through intoxication as well as withdrawal.
Heavy use of alcohol and drugs can cause changes in the brain that may
lead to chronic impairment and psychiatric symptoms related to violent
behavior (Tardiff et al. 2005; Volavka and Tardiff 1999). The ingestion
of alcohol can be associated with aggression and violence as a result of
disinhibition, particularly in the initial phase of intoxication. Intoxica-
tion is accompanied by emotional lability and impaired judgment. In
some cases, alcohol withdrawal may lead to delirium, and violence may
result from gross disorganization of behavior or as a response to threat-
ening auditory hallucinations or delusional thinking (Bushman 1997;
Langevin et al. 1987; McCormick and Smith 1995).
Cocaine, particularly when absorbed through the nasal route, ini-
tially produces a feeling of well-being and euphoria. With continued
use, particularly when the cocaine is taken intravenously or smoked in
the form of crack, the euphoria turns to grandiosity, psychomotor agi-
tation, suspiciousness, and, frequently, violence. Suspiciousness be-
comes first paranoid ideation and then paranoid delusional thinking.
Thus, violence results from delusional thinking as well as from the stim-
ulation effect of cocaine (Denison et al. 1997; Linaker 1994).
Violence may occur during intoxication with a number of hallucino-
gens, but less commonly than it occurs in phencyclidine (PCP) intoxica-
tion. Within 1 hour of oral use (5 minutes if the drug is smoked or taken
intravenously), PCP often produces marked violence, impulsivity, un-
predictability, and grossly impaired judgment. There also may be delu-
sional thinking or delirium (Convit et al. 1988).
With intense or prolonged amphetamine use, a feeling of well-being
and confidence turns to confusion, rambling, incoherence, paranoid
ideation, and delusional thinking, which are accompanied by agitation,
fighting, and other forms of aggression (Miczek and Tidey 1989).
Inhalants are substances containing hydrocarbons, such as gasoline,
glue, paint, and paint thinners, that are used by young children and early
adolescents to produce intoxication. Inhalant intoxication may be charac-
terized by belligerence and violence as well as impaired judgment.
Anabolic steroids are used by young men to enhance muscle growth
and performance in athletics. Reports and systematic studies have
found that after several months of self-administering these drugs, these
men become irritable, combative, and violent (Choi et al. 1989; Pope
and Katz 1994).
Psychosis
Psychosis is not a diagnosis, but it is a symptom that can be found in a
number of disorders, including schizophrenia, delusional disorder,
neurological and medical disorders, substance abuse disorders, and
mood disorders, especially with mania. These disorders are discussed
elsewhere in this book. When psychosis is present, regardless of the dis-
order, it increases the risk of violence (Anderson and Silver 1999).
Schizophrenic patients can be delusional and can have ideas of per-
secution. Patients may believe that people are trying to harm them; that
the police, the FBI, or other organizations are spying on them; that some
unknown mechanism is controlling their minds; or that the therapist is
harming them (e.g., through medication). Patients with paranoid delu-
sions in schizophrenia may react to these persecutory delusions by re-
taliating against the presumed source of the persecution. Patients with
other types of schizophrenia may attempt to kill other persons because
of some form of psychotic identification with the victim. Hallucinations
associated with schizophrenia have been known to result in violent be-
havior and homicide (Andreasen et al. 1995; Dixon et al. 1991; Modestin
and Ammann 1996).
Other aspects of schizophrenia, apart from psychotic processes, can
also result in violence. Sudden, unpredictable changes in affect may be
associated with anger, aggression, and violent behavior. Some schizo-
phrenic patients are violent because of generalized disorganization of
thought and a lack of impulse control accompanied by purposeless ex-
cited psychomotor activity (akathisia), or they may inadvertently come
into physical contact with other patients, which may lead to fights.
Schizophrenic patients also may use violence to attain what they want,
to express anger, or to deliberately hurt others.
The psychotic paranoid patient, regardless of diagnosis, poses a
problem because his or her delusions may not be obvious or the patient
may attempt to hide them. Therefore, the evaluator must listen for sub-
tle clues and should follow up regarding the assessment of violence
toward others but must be careful not to confront the patient with insis-
tent questioning about the presence of paranoid delusions (Taylor and
Felthous 2006).
A manic patient may become violent as a result of delusional think-
ing in which the patient believes he or she is being persecuted because
of some special attribute. Manic patients usually put all their impulses,
including violent ones, into action. A typical situation in which manic
patients erupt with violence is when they feel contained and not free to
do what they want to do (McElroy et al. 1992).
Personality Disorders
Violence by persons with antisocial personality disorder is often vicious
and persistent. They will continue punching, or hitting with objects, be-
yond what is needed to subdue the other person and win the fight.
These patients have no remorse for their actions, and the victim is per-
ceived as deserving the beating. The person with borderline personality
disorder can be violent and make suicidal gestures when rejected or
feeling rejected by others. The violence and suicide attempts are part of
a broader picture of impulsivity and instability of interpersonal rela-
tionships. Persons with narcissistic personality can be violent occasion-
ally when angry, such as when they are not given something they think
they deserve. The person with paranoid personality rarely attacks those
seen as persecutors, but when violence does occur it can be severe, even
taking the form of mass murder. The person with intermittent explosive
disorder is violent during circumscribed episodes, often with little ap-
parent precipitating cause or out of proportion to any identifiable cause
(Bernstein et al. 1993; Gunderson et al. 1991; Hare et al. 1991; Herpertz
et al. 1997; Kemperman et al. 1997).
Noncompliance With Treatment

A history of noncompliance with treatment should alert the clinician
that the patient is at an increased risk of violent behavior. Noncompli-
ance may be indicated by a history of irregular attendance at scheduled
appointments or laboratory and other clinical workups, or by the pa-
tient’s refusing to take certain medications for a psychiatric or medical
disorder or deliberately missing doses of medication. Measuring the
patient’s blood levels of medication may assist the clinician in monitor-
ing the patient’s compliance with the medication. Contact with the
patient’s family—with the consent of the patient—also may assist in
determining whether the patient is taking medication as prescribed.
Depot medication, particularly antipsychotic medications for schizo-
phrenia and other psychotic disorders, also can be used to ensure com-
pliance by these patients.
Demographic Characteristics
Demographic characteristics of patients should be considered in the as-
sessment of violence potential. Young persons and men have been
found to be at increased risk of violence, as are persons from environ-
ments of poverty, familial disruption, or decreased social control in
which violence is considered an acceptable means of attaining a goal in
the absence of other legitimate means or adequate education. The envi-

ronment from which the patient comes is thus an important consider-
ation in the determination of violence potential.
Case Examples
Case Example 1
The patient was a 25-year-old single white man with a history of para-
noid schizophrenia who was discharged from the military in the 1970s
because he developed delusions that he was a spy for the government
and had killed people. He was sent to a Veterans Administration (VA)
hospital for treatment and was discharged to the clinic for outpatient
treatment.
After discharge from the hospital, he still voiced paranoid ideation
and the delusion that he was a spy. He was not certain that he had killed
people and stated that there was no history of violence. He denied any
suicidal or violent ideation when first seen in the clinic. An in-depth,
comprehensive assessment of his short-term violence potential was per-
formed at that time. He was given oral haloperidol and continued to
attend the clinic on a monthly basis. He was compliant with the medi-
cation for years and was seen by a psychiatric resident who updated the
initial evaluation of violence and suicide potential monthly under the
supervision of an attending psychiatrist.
The patient continued to be delusional about being a spy but denied
any suicidal or violent ideation or intent. He lived alone and worked
part-time installing carpet for a small business. One day, he slashed and
killed his employer with a carpet knife. A lawsuit resulting from this act
was filed against the VA.
It was determined that the murder was not predictable and that the
patient had been monitored consistently, had received adequate treat-
ment, and had been compliant with medication; the judge consequently
ruled in favor of the VA. Although it was not admissible in the court,
I read the patient’s diary, and he had made no mention of thoughts or
intent to kill anyone, including his employer.
Case Example 2
The patient was a 36-year-old single, biracial man with a history of
schizoaffective disorder and polysubstance abuse, including cocaine
and alcohol. He was brought into the psychiatric emergency depart-
ment by the police after a physical altercation with a staff member at his
residence. He had become verbally aggressive for several weeks and
had stopped taking his valproic acid and olanzapine. On the day of the
violent episode, he had run into the street, was brought back to the res-
idence, and then threw a chair at a staff member. He had a history of two
suicide attempts and had been psychotic on a number of occasions. In
the emergency department he de-escalated considerably but made fre-
quent references to people trying to kill him. He was given haloperidol

and lorazepam intramuscularly after refusing oral medication and was
subsequently able to speak calmly with the psychiatric resident and at-
tending psychiatrist. He was placed in the locked seclusion room and al-
lowed to remain on the gurney. Twenty minutes later, he swung a metal
intravenous pole that he had detached from the gurney and broke the
camera with which the staff had been monitoring him. The police were
called, and three officers in riot gear subdued him. He was put in four-
point restraints in seclusion and placed on constant observation with
additional intramuscular haloperidol and lorazepam.
This case illustrates a psychotic patient who stopped medications

and probably used cocaine and alcohol. He became violent but ap-
peared to calm down with medication in the emergency department.
However, once in seclusion with no restraints or close observation, he
tore a metal pole from the gurney. He had become a very seriously vio-
lent person. Adequate police were called to subdue him as he vigor-
ously swung the pole. Eventually he was properly restrained and put
under adequate supervision with additional medication.
Conclusion
This chapter has described violence by patients with a number of psy-
chiatric disorders. Violence differs among psychiatric disorders in
terms of frequency, the manner in which it is expressed, and the psycho-
pathology and dynamics that produce it. Violence is seen more fre-
quently by clinicians and is more problematic in antisocial personality,
borderline personality, and intermittent explosive disorders; schizo-
phrenia and other psychotic disorders; mania; and alcohol and drug
abuse.
A model has been presented here for assessing the potential for vio-
lence among patients in the short term (days or a week). This time
period is of great relevance in clinical decisions regarding a patient’s
admission to a hospital, monitoring and treatment in the hospital, and
discharge from the hospital; the development of an aftercare plan; and
outpatient monitoring of the patient’s potential for violence between of-
fice visits. The clinician assessing violence potential must rely on as
many sources of data as possible, including interviews with the patient,
friends and family, police, current and former treaters, and past clinical
and other types of records.
Key Points
■ Ten factors involved in evaluating a patient’s potential for violence
are outlined in Table 1–1. In brief, they include appearance, idea-
tion and planning, intent, means and access, past history, alcohol
and drug use, presence of psychosis, presence of personality disor-
der, history of treatment noncompliance, and demographic and
socioeconomic characteristics.
■ All ten of the factors are weighed by the clinician in the final
assessment of whether the patient poses a short-term risk of
violence to others.
■ If the patient poses a short-term risk of violence, some action is
necessary on the part of the evaluator. Action may include chang-
ing the treatment plan, hospitalizing the patient, warning the
intended victim and/or the police, and other creative maneuvers
to prevent the imminent violence by the patient.
■ All of the data used to determine whether a patient is at risk for
violence must be documented in writing; the thinking process by
which the decision was made also must be evident in the written
documentation.
■ Reassessment of violence potential should be made at short inter-
vals (e.g., from visit to visit or every few days) if the patient is to
continue to be treated outside the hospital or other institution.
■ In the hospital and emergency department, safety and monitoring
of a potentially violent patient are essential, and treatment after
discharge must be detailed in writing and arranged in a timely
manner.
References
Anderson KE, Silver JM: Neurological and medical diseases and violence, in
Medical Management of the Violent Patient: Clinical Assessment and Ther-
apy. Edited by Tardiff K. New York, Marcel Dekker, 1999, pp 87–124
Andreasen NC, Arndt S, Alliger R, et al: Symptoms of schizophrenia: methods,
meanings, and mechanisms. Arch Gen Psychiatry 52:341–351,1995
Bernstein DP, Useda D, Siever LJ: Paranoid personality disorder: a review of its
current status. J Personal Disord 7:53–62, 1993
Bushman BJ: Effects of alcohol on human aggression: validity of proposed ex-
planation. Recent Dev Alcohol 13:227–304, 1997
Choi PYL, Parrott AC, Cowan D: High dose anabolic steroids in strength ath-
letes: effects upon hostility and aggression. J Psychopharmacol 3:102–113,
1989
Convit A, Nemes ZC, Volavka J: History of phencyclidine use and repeated as-
saults in newly admitted young schizophrenic men. Am J Psychiatry
154:1176–1183, 1988
Denison ME, Paredes A, Booth JB: Alcohol and cocaine interactions and aggres-
sive behaviors. Recent Dev Alcohol 13:283–291, 1997
Dixon L, Haas G, Weiden PH, et al: Drug abuse in schizophrenic patients: clin-
ical correlates and reasons for use. Am J Psychiatry 148:224–230, 1991
Gunderson JG, Ronningstam E, Smith LE: Narcissistic personality disorder: a
review of data on DSM-III-R descriptions. J Personal Disord 5:167–177,
1991
Hare RD, Hart SD, Harper TJ: Psychopathy and the DSM-IV criteria for antiso-
cial personality disorder. J Abnorm Psychol 100:391–398, 1991
Harris GT, Rice ME, Camilleri JA: Applying a forensic actuarial assessment (the
Violence Risk Appraisal Guide) to nonforensic patients. J Interpers Vio-
lence 19:1063–1074, 2004
Herpertz S, Gretzer EM, Steinmeyer V, et al: Affective instability and impulsiv-
ity in personality disorder. J Affect Disord 44:31–37, 1997
Kemperman I, Russ MJ, Shearin E: Self-injurious behavior and mood regulation
in borderline patients. J Personal Disord 11:146–157, 1997
Kroner DG, Mills JF, Reddon JR: A coffee can, factor analysis and prediction of
antisocial behavior: the structure of criminal risk. Int J Law Psychiatry
28:360–374, 2005
Kumar S, Simpson AI: Application of risk assessment for violence methods to
general adult psychiatry: a selective review of the literature. Aust NZ J Psy-
chiatry 39:328–335, 2005
Langevin R, Ben-Aron G, Wortzman R, et al: Brain damage, diagnosis, and sub-
stance abuse among violent offenders. Behav Sci Law 5:77–86, 1987
Linaker OM: Assaultiveness among institutionalized adults with mental retar-
dation. Br J Psychiatry 164:62–78, 1994
McCormick RA, Smith M: Aggression and hostility in the substance abuser: the
relationship to abuse patterns, coping style, and relapse trigger. Addict Be-
hav 20:555–564, 1995
McElroy SL, Keck PE, Pope HG, et al: Clinical and research implications of the
diagnosis of dysphoric or mixed mania or hypomania. Am J Psychiatry
149:1633–1644, 1992
Miczek KA, Tidey JW: Amphetamines: aggressive and social behavior. NIDA
Research Monographs 94:68–79, 1989
Mills JF: Advances in the assessment and prediction of interpersonal violence.
J Interpers Violence 20:236–241, 2005
Modestin T, Ammann R: Mental disorders and criminality: male schizophrenia.
Schizophr Bull 22:69–82, 1996
Pope HG, Katz DL: Psychiatric and medical effects of anabolic-androgenic ste-
roid use: a controlled study of 160 athletes. Arch Gen Psychiatry 51:375–
386, 1994
Tardiff K, Leone AC, Marzuk PM: Suicide risk measures, in Handbook of Psy-
chiatric Measures. Edited by Rush AJ. Washington, DC, American Psychi-
atric Press, 2000, pp 261–270
Tardiff K, Wallace Z, Tracy M, et al: Drug and alcohol use as determinants of
New York City homicide trends from 1990–1998. J Forensic Sci 50:1–5, 2005
Taylor PJ, Felthous AR: Introduction to this issue: international perspectives on

delusional disorders and the law. Behav Sci Law 24:235–240, 2006
Volavka J, Tardiff K: Substance abuse and violence, in Medical Management of
the Violent Patient: Clinical Assessment of Therapy. Edited by Tardiff K.
New York, Marcel Dekker, 1999, pp 153–177
Yudofsky SC, Silver JM, Jackson W: The Overt Aggression Scale for the objective
rating of verbal and physical aggression. Am J Psychiatry 143:35–39, 1986
C H A P T E R 2
Structured Risk Assessment

of Violence
John Monahan, Ph.D.
F or more than 50 years it has been commonplace in the behavioral sci-

ences to distinguish “clinical” from “actuarial” methods of risk assess-
ment and to conclude that the advantage in predictive validity lies with
the actuarial (Meehl 1954). For example, William Grove and Paul Meehl
(1996) located 136 empirical studies comparing clinical and actuarial
prediction and found them overwhelmingly to support the superiority
of the latter over the former. Their conclusion: “We know of no social
science controversy for which the empirical studies are so numerous,
varied, and consistent as this one” (p. 318; see also Grove et al. 2000). A
more recent comprehensive review disaggregated studies in terms of
the type of behavior being predicted and found that “one area in which
the statistical method is most clearly superior to the clinical approach is
the prediction of violence” (Aegisdottir et al. 2006, p. 368; see also Swets
et al. 2000).
There is less consensus, however, on exactly what is meant by “ac-
tuarial” risk assessment and how it differs from the “clinical” kind. I ar-
gue here that the dichotomous organization of the field of risk assess-
ment may have outlived its usefulness. Risk assessment may be better
Case examples reprinted from Monahan J, Steadman HJ, Appelbaum PS, et al.:
Classification of Violence Risk Professional Manual. Lutz, FL, Psychological Assess-
ment Resources, 2005, pp. 12 and 18. Used with permission.
17
seen to exist on a continuum of structure, with completely unstructured

(corresponding to “clinical”) risk assessment occupying one end of the
continuum, completely structured risk assessment (corresponding to
“actuarial”) occupying the other, and additional forms of more-than-
unstructured-but-less-than-fully-structured risk assessment lying be-
tween these poles.
I first consider unstructured violence risk assessment and then de-
scribe and illustrate three types of increasingly structured approaches
to violence risk assessment. At the outset, however, it is necessary to
make three clarifying points. First, it is sometimes incorrectly believed
that actuarial prediction eschews all reliance on clinical judgment. For
example, Sreenivasan et al. (2000, p. 439) stated, “The actuarial method
requires no clinical input, just a translation of the relevant material from
the records to calculate the risk score. Indeed, there is no compelling
reason for a clinician to be involved.” Although the risk factors on a
given actuarial tool may and often do include those obtained from
records, there is no reason why this needs to be so in order for the risk
assessment to be properly called “actuarial.” Indeed, “actuarial tables
can be constructed that rely entirely on data that must be obtained
through clinical judgment (e.g., ‘add ego strength score to impulse con-
trol score and subtract maternal deprivation score,’ etc.)” (Monahan
1981, p. 64). As Hilton et al. (2006, p. 402)—among the strongest propo-
nents of actuarial methods—noted, “because some of the best indica-
tors [of risk] require clinical skill to measure, accurately appraising vio-
lence risk is likely to remain a task for the clinician, but the place for
clinical judgment is within rather than outside actuarial tools” (emphasis
added; see also Litwack 2001; Westen and Weinberger 2004).
Second, it is sometimes inaccurately claimed that one factor that
clearly distinguishes clinical from actuarial prediction is the time frame
over which the prediction is valid, with clinical prediction aiming to as-
sess violence risk in the short term (e.g., over days or weeks) and actu-
arial prediction designed to assess long-term violence risk (e.g., over
months or years). It is certainly true that actuarial instruments are
sometimes validated over long periods of time (e.g., the Violence Risk
Appraisal Guide [VRAG, discussed later in this chapter] was developed
using a 7-year follow-up). However, in the seven major empirical stud-
ies that form the evidence base for estimating the validity of clinical pre-
diction of violence in the community, the follow-up periods varied from
6 months to 5 years, with a median of 3 years (Monahan 2006b). It
should also be noted that the most recently developed structured vio-
lence risk assessment instrument, the Classification of Violence Risk
(COVR, discussed later) was validated over a period of only 20 weeks,
Structured Risk Assessment of Violence ❘ 19
which is substantially shorter than the validation period used in any of

the studies of clinical prediction.
The final clarifying point is that clinical and actuarial prediction are
not neatly distinguished by the former’s exclusive reliance on “dy-
namic” risk factors and the latter’s exclusive reliance on “static” ones.
Many appear to believe that clinical prediction, because it emphasizes
changeable risk factors, has obvious implications for treatment whereas
actuarial prediction, because it uses unchangeable risk factors, has no
implications for clinical intervention. It is no doubt true that discussions
of changeable risk factors are more often to be found in the literature on
clinical than on actuarial prediction. The point is easily overstated,
however. Guides to clinical prediction (e.g., Monahan 1981) often stress
“static” unchangeable risk factors (e.g., a history of violence, gender),
and structured violence risk assessment tools can and do include “dy-
namic” changeable risk factors that are amenable to clinical treatment
(e.g., lack of insight and impulsivity on the HCR-20 [discussed later]
and anger control and substance abuse on the COVR) (Monahan and
Appelbaum 2000).
Unstructured Violence Risk Assessment

Unstructured risk assessment relies on the subjective judgment of pro-
fessionally educated people who are experienced at making predictive
judgments; in the case of violence, these typically include psychiatrists,
psychologists, and social workers. In unstructured assessment, risk fac-
tors are selected and measured based on the mental health profes-
sional’s theoretical orientation and prior clinical experience. What these
risk factors are, or how they are measured, might vary from case to case
depending on which seem most relevant to the professional doing the
assessment. At the conclusion of the assessment, risk factors are com-
bined in an intuitive or holistic manner to generate an overall profes-
sional opinion about a given individual’s level of violence risk.
Research has not been kind to unstructured violence risk assess-
ment. One early review of studies challenging the predictive accuracy
of unstructured risk assessments of violence concluded that “Of those
predicted to be dangerous, between 54% and 99% are false positives—
people who will not, in fact, commit a violent act” (Monahan 1981, p. 21).
Little has transpired in recent decades to increase confidence in the abil-
ity of mental health professionals, using their unstructured clinical
judgment, to accurately assess risk of violence in the community (Mon-
ahan 2007). In the most methodologically sophisticated study on this
topic, for example, researchers took as their subjects male and female
patients being examined in the acute psychiatric emergency depart-

ment of a large civil hospital (Lidz et al. 1993). Psychiatrists and nurses
were asked to assess the risk of patient violence to others over the next
6-month period. Patients who elicited professional concern regarding
future violence were moderately more likely to be violent after dis-
charge (53%) than were patients who had not elicited such concern
(36%). In other words, of the patients predicted to be violent by the
clinicians, one out of two later committed a violent act, whereas of the
patients predicted to be safe by the clinicians, one out of three later com-
mitted a violent act. The accuracy of clinical predictions of violence was
statistically significant for male patients but not for female ones.
Taken as a whole, as Douglas Mossman (1994, p. 790) has stated, re-
search supports the conclusion that “clinicians are able to distinguish
violent from nonviolent patients with a modest, better-than-chance
level of accuracy.” In recent years, however, the lack of strong empirical
support for the validity of unstructured violence risk assessment has
motivated clinical researchers to explore alternative forms of risk as-
sessment, ones that disaggregate the process of risk assessment into its
component parts and then proceed to structure some or all of those com-
ponents.
Varieties of Structured Violence Risk Assessment

Violence risk assessment might usefully be seen as having three compo-
nents: selecting and measuring risk factors, combining them, and gen-
erating a final risk estimate (Monahan 2006b).
In the first component, selecting and measuring risk factors, the mental
health professional performing the assessment decides which risk fac-
tors to measure and how to measure them. In unstructured risk assess-
ment, as described earlier, risk factors are selected and measured on the
basis of the mental health professional’s theoretical orientation and
prior clinical experience and may vary from case to case as theory or ex-
perience dictate. In contrast, in all forms of structured risk assessment,
decisions about which risk factors to measure and how to measure them
are made in advance, before the risk assessment begins. Explicit rules
specify a risk factor ’s operational definition and quantification. In
structured risk assessment, the mental health professional performing
the assessment has no discretion regarding the selection or measure-
ment of risk factors: these decisions are “structured” for him or her in
advance by the appearance-specified variables, with instructions on
how these variables are to be scored on a formal risk assessment instru-
ment.
The second component, combining risk factors, involves taking the

person’s individually measured risk factors (i.e., “scores” on each of the
risk factors) and assembling them into a single overarching estimate of
violence risk. In unstructured risk assessment, risk factors are assem-
bled in an intuitive or holistic manner to generate a clinical opinion
about violence risk. In some forms of structured risk assessment, risk
factors are assembled into an estimate of risk by means of a mathemat-
ical process specified in advance. That process is usually as simple as
adding the scores of the individual risk factors together to yield a total
score, but it can involve more complex statistical procedures as well
(Banks et al. 2004).
In the third component, generating a final risk estimate, the mental
health professional responsible for the risk estimate reviews the likeli-
hood of violence produced by the first two components of the risk assess-
ment process. In unstructured risk assessment, because the risk factors
are already combined in an intuitive or holistic manner to generate a
clinical opinion about violence risk, there is nothing for the clinician to
“review.” His or her clinical opinion is the clinician’s final estimate of vi-
olence risk. In structured risk assessment, however, the final risk estimate
offered by the clinician may differ from the risk estimate produced by the
first two (structured) components of the assessment process, on the basis
of additional (unstructured) information the clinician has gathered from
interviews, significant others, or available records—information not in-
cluded on the structured risk assessment instrument. As we will see,
some forms of structured risk assessment allow for a final clinical review,
whereas others preclude it by structuring even the final risk estimate.
To illustrate these three increasingly structured types of risk assess-
ment, I briefly describe recently available instruments that structure
one, two, or all three components of the risk assessment process.
The HCR-20
The “HCR-20,” first published in 1995 and revised in 1997, consists of a
series of 20 ratings addressing Historical, Clinical, and Risk manage-
ment factors (Webster et al. 1997). In one study, the HCR-20 was com-
pleted for civilly committed patients who were followed for approxi-
mately two years after discharge into the community. When the scores
were divided into five categories, 11% of the patients scoring in the low-
est category were found to have committed or threatened a physically
violent act, compared with 40% of the patients in the middle category
and 75% of the patients in the highest category (Douglas et al. 1999; see
also Douglas et al. 2005).
Selecting and Measuring Risk Factors

The 20 factors on this structured risk assessment tool were not derived
from a specific empirical research project. Rather, they represent the au-
thors’ judgment of which risk factors have emerged most strongly
across many empirical studies of violence risk. The 10 “historical” items
on the HCR-20 are 1) previous violence, 2) young age at first violent in-
cident, 3) relationship instability, 4) employment problems, 5) substance
use problems, 6) major mental illness, 7) psychopathy, 8) early malad-
justment, 9) personality disorder, and 10) prior supervision failure. The
five “clinical” items are 11) lack of insight, 12) negative attitudes, 13) ac-
tive symptoms of major mental illness, 14) impulsivity, and 15) unre-
sponsiveness to treatment. The five “risk management” items are 16)
plans lack feasibility, 17) exposure to destabilizers, 18) lack of personal
support, 19) noncompliance with remediation attempts, and 20) stress.
Each of the 20 items is measured on a three-point scale according
to the certainty that the risk factor is present: A score of 0 equals “no”—
the item is definitely absent or does not apply; a score of 1 equals
“maybe”—the item is possibly present, or is present only to a limited
extent; and a score of 2 equals “yes”—the item definitely is present.
Combining Risk Factors

The HCR-20 structures the process of selecting and measuring risk fac-
tors, but for clinical purposes, it does not structure the process of com-
bining risk factors to reach an overall estimate of risk. As stated in the
professional manual for the HCR-20 (Webster et al. 1997, pp. 21–23):
For clinical purposes, it makes little sense to sum the number of risk fac-
tors present in a given case, and then use fixed, arbitrary cutoffs to clas-
sify the individual as low, moderate, or high risk… [It] is both possible
and reasonable for an assessor to conclude that an assessee is at high risk
for violence based on the presence of a single risk factor—if, for exam-
ple, that risk factor is “Active Symptoms of Major Mental Illness” and
reflects the assessee’s stated intent to commit a homicide…. In sum, at
present it may be neither possible nor desirable to develop cutoff scores
for the determination of summary or final risk judgments in clinical
settings.
Generating a Final Estimate of Risk

For clinical purposes the HCR-20 does not structure the process of com-
bining risk factors to reach an overall estimate of risk. Rather, it allows
the clinician to combine the 20 measured risk factors in an intuitive
manner to yield an overall estimate of risk. Thus there is no need for a
“clinical review” of a structured risk estimate, and so none is per-

formed.
The Classification of Violence Risk

The first violence risk assessment software, called the Classification of
Violence Risk (COVR), was published in 2005. COVR is an interactive
software program designed to estimate the risk that an acute psychiatric
patient will be violent toward others over the next several months. The
program can measure 40 risk factors. Used on a laptop or desktop com-
puter, COVR guides the evaluator through a brief chart review and a
10-minute interview with the patient and then generates a report that
places the patient’s violence risk into one of five categories—with a 1%
likelihood of violence in the first category, a 26% likelihood of violence
in the middle category, and a 76% likelihood of violence in the highest
category—including the confidence interval for the given risk estimate.1

The COVR software was constructed from data generated in the Mac-
Arthur Violence Risk Assessment Study (Monahan et al. 2001). In this
research, more than 1,000 patients in acute civil psychiatric facilities
were assessed on 134 potential risk factors for violent behavior. Patients
were followed for 20 weeks in the community after discharge from the
hospital, and their violence toward others was assessed. The software is
capable of assessing those 40 risk factors for violence that emerged as
most predictive of violence in the original research, but in any given
case assesses only those risk factors necessary to classify the patient’s
violence risk. Among the risk factors assessed most frequently by the
COVR are the seriousness and frequency of prior arrests, young age,
male gender, being unemployed, the seriousness and frequency of hav-
ing been abused as a child, a diagnosis of antisocial personality dis-
order, the lack of a diagnosis of schizophrenia, whether the individual’s
father used drugs or left the home before the individual was 15 years
old, substance abuse, impaired anger control, and violent fantasies.

To combine risk factors into a preliminary estimate of risk, the COVR re-
lies on “classification tree” methodology. This approach allows many
1
Note that the author of this chapter is one of the owners of COVR.
different combinations of risk factors to classify a person as high or low

risk. On the basis of a sequence established by the classification tree, a
first question is asked of all persons being assessed. Contingent on the
answer to that question, one or another second question is posed, and
so on. The classification tree process is repeated until each person is
classified into a final risk category. This “interaction” model contrasts
with the more typical “main effects” approach to structured risk assess-
ment, such as used by the HCR-20 and the VRAG (discussed later), in
which a common set of questions is asked of everyone being assessed.

In the view of its authors, the COVR software is useful in informing, but
not replacing, clinical decision making regarding risk assessment. The
authors recommend a two-phased violence risk assessment procedure
in which a patient is first administered the COVR, and then the prelim-
inary risk estimate generated by the COVR is reviewed by the clinician
ultimately responsible for making the risk assessment, in the context of
additional information believed to be relevant and gathered from clini-
cal interviews, significant others, and/or available records. The authors
of the COVR believed it essential to allow for such a review, for two rea-
sons. The first has to do with possible limits on the generalizability of
the validity of the software. For example, is the predictive validity of the
COVR generalizable to forensic patients, to people outside the United
States, to people who are younger than 18 years old, or to the emer-
gency-department assessments of persons who have not recently been
hospitalized? The predictive validity of this instrument may well gen-
eralize widely. Yet there comes a point at which the sample to which a
structured risk assessment instrument is applied differs so much from
the sample on which the instrument was constructed and validated
(Monahan et al. 2005b) that one would be hard pressed to castigate the
evaluator who took the structured risk estimate as advisory rather than
conclusive.
The second reason for allowing clinicians the option to review struc-
tured risk estimates is that the clinician may note the presence of rare
risk or protective factors in a given case, factors that—precisely because
they are rare—will not have been taken into account in the construction
of the structured instrument (Appelbaum et al. 2000). In the context of
structured instruments for assessing violence risk, the most frequently
mentioned rare risk factor is a direct threat—that is, an apparently seri-
ous statement of intention to do violence to a named victim (as with the
HCR-20).
Case Examples of the Use of the COVR

The COVR manual (Monahan et al. 2005a) gives three case examples of
the use of the COVR in clinical practice:
Case Example 1: High Risk

Mr. Smith is a 27-year-old salesman who has been hospitalized for the
eighth time with a diagnosis of bipolar disorder. After 5 days in the hospi-
tal, he is being considered for discharge. Mr. Smith had been hospitalized
for making an aggressive act toward his wife while manic and intoxicated.
Because of this, the clinician responsible for the discharge decision re-
quests that the COVR be administered. The next day, a COVR report is
given to the responsible clinician that concludes “The likelihood that Mr.
Smith will commit a violent act toward another person in the next several
months is estimated to be between 65% and 86%, with a best estimate of
76%.” The report also lists the risk factors used to produce this estimate.
The clinician, after reviewing the COVR report and all the informa-
tion in Mr. Smith’s hospital chart, interviews Mr. Smith. The interview
fails to uncover any unusual protective factors that would call into ques-
tion the estimate of violence risk that the COVR had produced. More-
over, it is clear that his manic state has not fully resolved. The clinician
decides not to discharge Mr. Smith at the current time but rather to con-
tinue a course of medication and anger management groups, designed
to lower his violence risk, and to recommend that Mr. Smith continue
with anger management and intensive substance abuse treatment in the
community after discharge. With Mr. Smith’s consent, his wife is coun-
seled about her risk if his symptoms recur and he starts drinking again.
Case Example 2: Low Risk

Ms. Jones is a 42-year-old female accountant who has been hospitalized
for the first time for several days with a diagnosis of major depression.
She is being considered for discharge. An ambiguous threat Ms. Jones
made about a coworker had been noted in her hospital chart by a nurse,
so the clinician responsible for the discharge decision requests that the
COVR be administered. The next day, a COVR report is given to the re-
sponsible clinician that concludes “The likelihood that Ms. Jones will
commit a violent act toward another person in the next several months
is estimated to be between 0% and 2%, with a best estimate of 1%.” The
report also lists the risk factors used to produce this estimate.
tion in Ms. Jones’ hospital chart, interviews Ms. Jones. The interview
fails to uncover any unusual risk factors that would call into question
the estimate of violence risk that the COVR had produced, and Ms. Jones
explains the ambiguous comment, which turns out not actually to have
been a threat, to the clinician’s satisfaction. Because she seems less de-
pressed and is not suicidal, the clinician decides to discharge Ms. Jones
at the current time and to follow up with routine care in the community.
Case Example 3: Moderate Risk

Mr. Brown is a 21-year-old security guard who has been hospitalized for
several days with a diagnosis of borderline personality disorder with
comorbid substance dependence, after getting into a shouting match
with his girlfriend and cutting his arms. He is being considered for dis-
charge. Because the chart indicates that Mr. Brown had been involun-
tarily committed on two prior occasions as “dangerous to others,” the
clinician responsible for the discharge decision requests that the COVR
be administered. The next day, a COVR report is given to the responsible
clinician that concludes “The likelihood that Mr. Brown will commit a
violent act toward another person in the next several months is esti-
mated to be between 20% and 32%, with a best estimate of 26%.” The re-
port also lists the risk factors used to produce this estimate.
tion in Mr. Brown’s hospital chart, interviews Mr. Brown. During the in-
terview, Mr. Brown states his apparently serious intention to “teach a
lesson she’ll never forget” to his girlfriend, who has told him that he
cannot come back to live with her. He also responds affirmatively to a
question about whether he has a firearm in the house. The clinician be-
lieves that this clinical information is indicative of a high risk of immi-
nent violence. The clinician decides not to discharge Mr. Brown at the
current time but rather to continue a course of medication and psycho-
therapy designed to lower his violence risk. The clinician also decides to
inform Mr. Brown’s former girlfriend of the threat.
The Violence Risk Appraisal Guide

The Violence Risk Appraisal Guide (VRAG), first published in 1993,
measures 12 risk factors designed to predict violence in mentally ill of-
fenders. In a more recent prospective study with 467 male forensic pa-
tients, the VRAG showed impressive predictive validity. Patients were
placed into one of nine categories of violence risk: 11% of the patients
who scored in the lowest category on the VRAG were later found to
commit a new violent act, compared with 42% of the patients in the
middle category and 100% of the patients in the highest category (Har-
ris et al. 2002).

The VRAG was developed from a sample of more than 600 men from a
maximum-security hospital in Canada. All had been charged with seri-
ous criminal offenses. Approximately 50 predictor variables were
coded from institutional files. The criteria used to develop the instru-
ment were any new criminal charge for a violent offense or return to the
institution for an act that otherwise would have resulted in a criminal
charge for a violent offense. The average time at risk in the community
was approximately 7 years after discharge. A series of analyses identi-

fied 12 variables for inclusion in the instrument: 1) score on the Hare
Psychopathy Checklist–Revised, 2) separation from parents at under
age 16, 3) victim injury in index offense, 4) diagnosis of schizophrenia,
5) never married, 6) elementary school maladjustment, 7) female victim
in index offense, 8) failure on prior conditional release, 9) property of-
fense history, 10) age at index offense, 11) alcohol abuse history, and
12) diagnosis of personality disorder. For all variables except numbers
3, 4, 7, and 10, the nature of the relationship to subsequent violence was
positive (i.e., subjects who injured a victim in the index offense, had re-
ceived a diagnosis of schizophrenia, chose a female victim for the index
offense, or were older were significantly less likely to be violent recidi-
vists than other subjects).

Each of the 12 risk factors measured by the VRAG is statistically
weighted, and the weighted scores are summed together to yield an
overall estimate of violence risk.

Importantly, the authors of the VRAG (Quinsey et al. 2006) do not allow
for any clinical review of the structured risk estimate that this instru-
ment produces:
What we are advising is not the addition of actuarial methods to exist-

ing practice, but rather the replacement of existing practice with actuar-
ial methods. This is a different view than we expressed a decade ago,
when we advised the practice of adjusting actuarial estimates of risk by
up to 10% when there were compelling circumstances to do so…. We no
longer think this practice is justifiable: Actuarial methods are too good
and clinical judgment is too poor to risk contaminating the former with
the latter. (p. 197)
A comparison of unstructured violence risk assessment and the

various forms of increasingly structured violence risk assessment is
presented in Table 2–1.
Use of Structured Violence Risk Assessment

in Clinical Practice
The literature on the incorporation of any form of structured risk assess-
ment into the clinical practice of predicting violence is thin, but all of it
TABLE 2–1. Unstructured and structured methods of violence risk

assessment
Structured
Number of selection and Structured Structured
structured measurement combination final risk
Method components of risk factors? of risk factors? estimate?
Unstructured 0 No No No
HCR-20 1 Yes No No
COVR 2 Yes Yes No
VRAG 3 Yes Yes Yes
Note. COVR=Classification of Violence Risk; HCR=Historical, Clinical, and Risk
Management; VRAG=Violence Risk Appraisal Guide.
suggests that only a minority of mental health professionals routinely

employ some form of structured risk assessment.
Elbogen et al. (2002) surveyed 134 clinicians in Nebraska and asked
about the relevance to violence risk assessment of a large number of risk
factors. Some of the risk factors were those found on structured risk as-
sessment instruments such as the VRAG, the HCR-20, and the COVR.
Others were not research based but rather were variables obtained from
interviews with clinicians regarding their beliefs about what predicted
violence (e.g., “impulsive behavior while in care”): “results show that
nearly every clinician perceived dynamic, behavioral variables to be
significantly more relevant than research-based factors…. Behavioral
risk factors were perceived as more relevant than research risk factors
from the HCR-20 and the VRAG, and from three of the four domains of
the MacArthur Risk Assessment Study [i.e., the COVR]” (p. 43).
Tolman and Mullendore (2003) surveyed 93 practitioners of general
clinical psychology and 71 diplomates of the American Board of Foren-
sic Psychology from Michigan regarding instruments used in conduct-
ing violence risk assessments. They found that the VRAG was used in
making violence risk assessments by 27% of the diplomates and by 9%
of the general practitioners, and the HCR-20 was used by 31% of the
diplomates and 2% of the general practitioners. For diplomates, the
VRAG and the HCR-20 were among the top five instruments used to
assess violence risk, whereas for the general practitioners, these struc-
tured risk assessment instruments were not among the top five instru-
ments used. Rather, general practitioners tended to rely on all-purpose
instruments whose relationship to violence risk is either unsubstanti-
ated, such as the Minnesota Multiphasic Personality Inventory–2 (see
Melton et al. 1997), or proven to be invalid, such as the Rorschach (Lili-

enfeld et al. 2000).
Finally, Lally (2003) surveyed a national sample of 64 diplomates
from the American Board of Forensic Psychology regarding the use of
various procedures for assessing violence risk. The VRAG was rated as
“Acceptable”—but not as “Recommended”—by more than half of the
respondents. The HCR-20 was rated by the majority as somewhere be-
tween “Acceptable” and “No Opinion.”
Although it would appear from these three surveys that most men-
tal health professionals have yet to incorporate structured violence risk
assessment tools into their routine clinical or forensic practice, when
mental health professionals do predicate their risk assessments on the
use of a structured instrument, courts uniformly find these risk assess-
ments to be admissible scientific evidence (see Monahan 2006a for a
compilation of recent federal and state cases). Not only courts but also
legislatures are increasingly coming to look favorably on structured vi-
olence risk assessment. In 2003, Virginia became the first jurisdiction to
require that a named structured violence risk assessment tool, with a
cutoff score specified by law, be used to assess violence risk. This was
done for the purpose of evaluating candidates for civil commitment as
a sexually violent predator. The relevant statute directs the Department
of Corrections to identify all prisoners incarcerated for sexually violent
offenses “who receive a score of four or more on the Rapid Risk Assess-
ment for Sexual Offender Recidivism or a like score on a comparable,
scientifically validated instrument as designated by the Commissioner
[of the Department of Mental Health, Mental Retardation, and Sub-
stance Abuse Services]” (Va. Code. Ann. § 37.2–903[c] 2005). The Rapid
Risk Assessment of Sexual Offender Recidivism (RRASOR) is a com-
pletely structured (i.e., actuarial) instrument consisting of four items:
1) number of prior sex offense convictions or charges (from 1 to 6 or
more); 2) age at release (25 years or older versus younger than 25);
3) victim gender (only females versus any males); and 4) relationship to
victim (only related versus any nonrelated). The latter parenthetical
items receive a higher score than the former items. A total score of 4 on
the RRASOR corresponds to a 10-year recidivism rate of 48.6%, whereas
a score of 5 corresponds to a 73.1% recidivism rate (Hanson 2004). In
2006, this statute was amended by the Virginia legislature to replace the
RRASOR with another actuarial tool of somewhat higher predictive va-
lidity, the Static-99 (Hanson and Thornton 2000). Only prisoners who
score above the specified cutoff score on the structured risk assessment
instrument are sent on for a subsequent clinical evaluation of violence
risk and “mental abnormality.”
Conclusion
In unstructured risk assessment, neither the selection nor the measure-
ment of the risk factors used in the assessment is specified in advance.
Therefore, there are no risk factor “scores” that can be combined to yield
a quantitative estimate of risk and no need for a clinical review of such
an estimate. The three forms of structured risk assessment described
here all specify in advance at least which risk factors are to be addressed
and how those risk factors are to be measured. The HCR-20 structures
only the choice and measurement of risk factors. The COVR goes on to
also structure the manner in which the risk factors are combined to yield
an estimate of risk, but the COVR allows the clinician to review this es-
timate in the context of other (unstructured) available information be-
fore issuing a final risk estimate. The VRAG, in contrast, is a completely
structured (i.e., actuarial) risk assessment tool. No clinical review is al-
lowed: the structured risk estimate that is produced when the risk fac-
tors are combined is the final product of the risk assessment process.
Although the three specific structured risk assessment tools consid-
ered here are the most frequently discussed in the literature, it should be
emphasized that they are merely illustrative of a larger group of instru-
ments that—like the HCR-20—structure only the choice and measure-
ment of risk factors (e.g., Kropp and Hart 2000), or—like the COVR—
also structure the manner in which the risk factors are combined to yield
an overall estimate of risk but allow the clinician to review this struc-
tured estimate in the context of other available information (e.g., Han-
son 1997), or—like the VRAG—stipulate that the structured risk esti-
mate that is available when the risk factors are combined is the final
product of the risk assessment process (Harris et al. 2003).
If structured violence risk assessment is superior to unstructured vi-
olence risk assessment, which specific form of structured risk assess-
ment has the highest predictive validity? Should the clinician structure
only one component of the risk assessment process (as the HCR-20
does), two components (as the COVR does), or all three components (as
the VRAG does)? On this issue, perhaps because some of these instru-
ments are so new, there are many strong opinions but no widespread
acceptance of a single view among either researchers or practitioners.
Finally, although structured violence risk assessment—of whatever
form—appears to be demonstrably superior to unstructured violence
risk assessment, and despite the increasing receptivity of courts and
legislatures to the use of structured violence risk assessment, only a mi-
nority—perhaps only a small minority—of practicing mental health
professionals in the United States routinely employ any form of struc-
tured violence risk assessment at the latter end of the first decade of the
twenty-first century.
Key Points
■ To improve the predictive validity of violence risk assessment, the
assessment process can fruitfully be disaggregated into its three
components: 1) selecting and measuring risk factors, 2) combin-
ing risk factors, and 3) generating a final estimate of risk.
■ Violence risk assessment instruments recently have been created
that structure one, two, or all three of these component parts of
the risk assessment process.
■ All forms of structured violence risk assessment appear to have
greater predictive validity than unstructured (“clinical”) violence
risk assessment.
■ Consensus has not yet been achieved as to which form of struc-
tured violence risk assessment has the greatest predictive validity.
■ Courts and legislatures are increasingly open to the use of struc-
tured violence risk assessment.
■ At the present time, relatively few practicing mental health profes-
sionals employ any form of structured violence risk assessment.
References
Aegisdottir S, White M, Spengler P, et al: The Meta-Analysis of Clinical Judg-
ment Project: fifty-six years of accumulated research on clinical versus sta-
tistical prediction. Couns Psychol 34:341–382, 2006
Appelbaum P, Robbins P, Monahan J: Violence and delusions: data from the
MacArthur Violence Risk Assessment Study. Am J Psychiatry 157:566–572,
2000
Banks S, Robbins P, Silver E, et al: A multiple-models approach to violence risk
assessment among people with mental disorder. Crim Justice Behav
31:324–340, 2004
Douglas K, Ogloff J, Nicholls T, et al: Assessing risk for violence among psychi-
atric patients: the HCR-20 violence risk assessment scheme and the Psy-
chopathy Checklist: Screening Version. J Consult Clin Psychol 67:917–930,
1999
Douglas K, Yeomans M, Boer D: Comparative validity analysis of multiple mea-
sures of violence risk in a sample of criminal offenders. Crim Justice Behav
32:479–510, 2005
Elbogen E, Mercado C, Scalora M, et al: Perceived relevance of factors for vio-
lence risk assessment: a survey of clinicians. International Journal of Foren-
sic Mental Health 1:37–47, 2002
Grove W, Meehl P: Comparative efficiency of informal (subjective, impression-

istic) and formal (mechanical, algorithmic) prediction procedures: the clin-
ical–statistical controversy. Psychol Public Policy Law 2:293–323, 1996
Grove W, Zald D, Lebow B, et al: Clinical versus mechanical prediction: a meta-
analysis. Psychol Assess 12:19–30, 2000
Hanson R: The Development of a Brief Actuarial Scale for Sexual Offense Recid-
ivism. Ottawa, ON, Canada, Department of the Solicitor General of Can-
ada, 2004
Hanson R, Thornton D: Improving risk assessments for sex offenders: a com-
parison of three actuarial scales. Law Hum Behav 24:119–136, 2000
Harris G, Rice M, Cormier C: Prospective replication of the Violence Risk Ap-
praisal Guide in predicting violent recidivism among forensic patients.
Law Hum Behav 26:377–394, 2002
Harris G, Rice M, Quinsey V, et al: A multi-site comparison of actuarial risk in-
struments for sex offenders. Psychol Assess 15:413–425, 2003
Hilton N, Harris G, Rice M: Sixty-six years of research on the clinical versus ac-
tuarial prediction of violence. Couns Psychol 34:400–409, 2006
Kropp P, Hart S: The Spousal Assault Risk Assessment (SARA) Guide: reliabil-
ity and validity in adult male offenders. Law Hum Behav 24:101–118, 2000
Lally S: What tests are acceptable for use in forensic evaluations? A survey of
experts. Prof Psychol Res Pr 34:491–498, 2003
Lidz C, Mulvey E, Gardner W: The accuracy of predictions of violence to others.
JAMA 269:1007–1011, 1993
Lilienfeld S, Wood S, Garb H: The scientific status of projective techniques. Psy-
chological Science in the Public Interest 2: 27–66, 2000
Litwack T: Actuarial versus clinical assessments of dangerousness. Psychol
Public Policy Law 7:409–443, 2001
Meehl P: Clinical Versus Statistical Prediction: A Theoretical Analysis and a Re-
view of the Evidence. Minneapolis, University of Minnesota, 1954
Melton G, Petrila J, Poythress N, et al: Psychological Evaluations for the Courts:
A Handbook for Mental Health Professionals and Lawyers, 2nd Edition.
New York, Guilford, 1997
Monahan J: The Clinical Prediction of Violent Behavior. Washington, DC, Gov-
ernment Printing Office, 1981
Monahan J: A jurisprudence of risk assessment: forecasting harm among pris-
oners, predators, and patients. Va Law Rev 92:391–435, 2006a
Monahan J: Tarasoff at thirty: How developments in science and policy shape the
common law. Univ Cincinnati Law Rev 75:497–521, 2006b
Monahan J: The scientific status of research on clinical and actuarial predictions
of violence, in Modern Scientific Evidence: The Law and Science of Expert
Testimony, Vol 1. Edited by Faigman D, Kaye D, Saks M, et al. St. Paul, MN,
West Publishing Company, 2007, pp 120–147
Monahan J, Appelbaum P: Reducing violence risk: diagnostically based clues
from the MacArthur Violence Risk Assessment Study, in Effective Preven-
tion of Crime and Violence Among the Mentally Ill. Edited by Hodgins S.
Dordrecht, The Netherlands, Kluwer Academic Publishers, 2000, pp 19–34
Monahan J, Steadman H, Silver E, et al: Rethinking Risk Assessment: The Mac-
Arthur Study of Mental Disorder and Violence. New York, Oxford Univer-
sity Press, 2001
Monahan J, Steadman H, Appelbaum P, et al: The Classification of Violence

Risk. Lutz, FL, Psychological Assessment Resources, 2005a
Monahan J, Steadman H, Robbins P, et al: An actuarial model of violence risk
assessment for persons with mental disorders. Psychiatr Serv 56:810–815,
2005b
Mossman D: Assessing predictions of violence: being accurate about accuracy.
J Consult Clin Psychol 62:783–792, 1994
Quinsey V, Harris G, Rice M, et al: Violent Offenders: Appraising and Managing
Risk, 2nd Edition. Washington, DC, American Psychological Association,
2006
Sreenivasan S, Kirkish P, Garrick T, et al: Actuarial risk assessment models:
a review of critical issues related to violence and sex-offender recidivism
assessments. J Am Acad Psychiatry Law 28:438–448, 2000
Swets J, Dawes R, Monahan J: Psychological science can improve diagnostic de-
cisions. Psychological Science in the Public Interest 1:1–26, 2000
Tolman A, Mullendore K: Risk evaluations for the courts: is service quality a
function of specialization? Prof Psychol Res Pr 34: 225–232, 2003
Va. Code. Ann. § 37.2–903(c) 2005
Webster C, Douglas K, Eaves D, et al: HCR-20: Assessing Risk for Violence (Ver-
sion 2). Vancouver, BC, Canada, Simon Fraser University, 1997
Westen D, Weinberger J: When clinical description becomes statistical predic-
tion. Am Psychol 59:595–613, 2004
C H A P T E R 3
Cultural Competence in
Violence Risk Assessment
Carl C. Bell, M.D.
T he impact of cultural factors on the assessment of the risk of violence

is complex and multifactorial. In earlier chapters of this textbook, the
assessment of violence potential is dependent on historical information:
psychiatric, medical, developmental, social, substance use, family his-
tory, and socioeconomic status. Almost all risk factors are affected by
culture, including age, gender, mental illness diagnoses, substance
abuse, and age at first event of violence. Cultural beliefs about violence
and religion vary from group to group. Some groups may feel that
wives are supposed to be subservient (e.g., Muslims, Chinese, Vietnam-
ese, Cambodians, Christians) and use that as justification for violence,
whereas others strongly adhere to nonviolence. Psychiatric diagnosis is
particularly vulnerable to influence by ethnicity; for example, we know
that African Americans and Hispanics are overdiagnosed with psy-
chotic disorders and underdiagnosed with bipolar disorder (Adebimpe
1981; Bell and Mehta 1980). Age at first episode of violence also can be
inaccurate; for Asian Americans, for example, admitting to a history of
violence would be shameful, and the information thus would be sup-
pressed (Yick 1999). Substance abuse is more prevalent in some ethnic-
ities, such as in Native Americans (Substance Abuse and Mental Health
Services Administration 2003), but that does not necessarily mean that
people of these ethnicities are more prone to violence.
35
The supplement to the Surgeon General’s report on mental health

(U.S. Department of Health and Human Services 2001a) “Mental Health:
Culture, Race and Ethnicity” acknowledges that because very little pub-
lished mental health research exists on cultural, racial, and ethnic issues
in general—and even less on the issue of violence risk assessment in dif-
ferent cultural, racial, and ethnic groups in varying contexts—discussing
cultural sensitivity/competence in violence risk assessment is a chal-
lenge.
Recognizing this reality, we must first understand why there is so lit-
tle creditable information about different cultural, racial, and ethnic
groups in mental health literature. For years, the United States has tried
to be a “color-blind” melting pot, and the consideration of the dynamics
of culture, race, and ethnicity have been selectively ignored by science.
Accordingly, when the topic of cultural sensitivity or competence sur-
faces, individuals become anxious, defensive, and very rigid in their
perspectives. “Race constitutes a stubbornly resistant malady in the
United States because of ‘the color line’—a visible (and invisible) bar-
rier that separates whites from nonwhites” (Pulera 2002, p. 3). The
American Psychiatric Association recognized the existence of structural
(institutional) racism by asserting that racist policies occur at an organi-
zational or group level and that these policies are embedded in the op-
erating contexts of particular organizations or institutions in such a way
that racist assumptions may be difficult to recognize (American Psychi-
atric Association 2006).
One such assumption is that “one size fits all,” and that the standard
is European or European American. Of course this is a form of “mono-
cultural ethnocentrism” (Sue and Sue 1999) and is as difficult to rec-
ognize for the majority culture in the United States as it is for a fish to
recognize water. Recent studies have implicated racism and racial dis-
crimination, both individual and structural, as factors leading to dispar-
ities in overall healthcare and mental healthcare, including diagnosis
and treatment (U.S. Department of Health and Human Services 2001a).
Racially biased attitudes may implicitly affect provider decision mak-
ing, leading to denial of services for some populations or to inappropri-
ate assessment and diagnosis that in turn leads to ineffective treatment
(Miranda et al. 2002; West et al. 2006). In the case of violence risk assess-
ment, this may lead to an overdetermination of the risk of violence.
Thus, considering the nation’s overall lack of success at appropriately
responding to cultural, racial, and ethnic differences, the issue of cul-
tural sensitivity or competence in violence risk assessment and man-
agement is extremely sensitive.
Cultural Competence in Violence Risk Assessment ❘ 37
Problems in Assessing Violence Risk

in People of Color
There are a great many myths and stereotypes related to violence and
other symptoms in people of color—that is, people from racial and cul-
tural groups other than European American (Pierce et al. 1999). For ex-
ample, concerning youth, Dr. Satcher’s Surgeon General’s Report on
Youth Violence noted one myth about youth violence, namely that
“African American and Hispanic youths are more likely to become in-
volved in violence than other racial or ethnic groups” (U.S. Department
of Health and Human Services 2001b, p. 5). However, the fact is that race
and ethnicity have little bearing on the overall proportions of racial
and ethnic groups that engage in nonfatal violent behavior (U.S. De-
partment of Health and Human Services 2001b). Furthermore, although
African American and Hispanic youths have higher homicide rates than
European American youths, these differences drop out when variables
of socioeconomic status are controlled (Hollinger et al. 1994; Sampson
et al. 1997).
Dr. Satcher’s report on youth violence also noted that risk factors can
predict the likelihood of future violence. Accordingly, these factors are
useful in identifying vulnerable populations that may benefit from inter-
vention efforts. However, risk markers such as race and ethnicity are fre-
quently confused with risk factors. The distinction is that risk markers
have no causal relation to violence (U.S. Department of Health and Hu-
man Services 2001b). Finally, Dr. Satcher’s report noted that “no single
risk factor or combination of factors can predict violence with unerring
accuracy” (U.S. Department of Health and Human Services 2001b, p. 77).
The reasons why this is true can be summed up in the maxim “risk factors
are not predictive factors due to protective factors” (Bell 2007, p. 14).
In studies of violence risk assessment, the findings in adults parallel
the findings in youths with regard to issues of culture, race, and ethnic-
ity. The ethnic groups at greatest risk for interpersonal violence are
American Indian/Alaskan Native women and men, African American
women, and Hispanic women, at 25% risk (Tjaden and Thoennes 2000).
Statistics from 2002 show that the death rate due to homicide per 100,000
was 4.1 for Caucasians, 39.6 for African Americans, 11.1 for American
Indians and Native Alaskans, 5.2 for Asians and Pacific Islanders, and
17.5 for Hispanic Americans (Keppel et al. 2002). In 2001, arrests for ag-
gravated assault show 99.6 per 100,000 for Caucasians, 320 for African
Americans, 144.6 for Americans Indians and Native Alaskans, and 37.5
for Asians and Pacific Islanders. Hispanics were not listed as a separate
ethnic group (Tseng et al. 2004). Suicide, anger/revenge, and mass mur-
der stemming from domestic or romantic conflict are the chief forms of
violence in the European-American community (Petee et al. 1997). Serial
killing has been stereotypically linked to European Americans; how-
ever, a recent article by Walsh (2005) showed that this type of homicide
by African Americans has been underreported. Thus, being a minor-
ity—with the exception of being Asian American—was associated with
being a victim of a violent attack. The distinction must be made that be-
ing associated with something does not necessarily predict it, hence the
distinction of a risk marker not being a risk factor because there has been
no causal link proven. Furthermore, although it is true that there are ra-
cial and ethnic differences in homicide rates, these differences drop out
when variables of socioeconomic status are controlled (Bell 2002). Socio-
economic inequality, not race, facilitates higher rates of violence among
ethnic minority groups (Johnson 2000).
Demographic factors associated with increased risk for violence in
adults are race and ethnicity; however, because these risk factors “tend
to dissipate when other factors are taken into account statistically”
(Hucker 2004), it is likely that in adults, race and ethnicity are also actu-
ally risk markers and not risk factors, as was seen in adolescents. This
finding is supported by the MacArthur Violence Risk Assessment
Study (Steadman et al. 1998), which found that although there was an
overall association between race and violence, African Americans and
European Americans who lived in comparably disadvantaged neigh-
borhoods had the same rates of violence.
A common stereotype is that people of color (e.g., African Ameri-
cans) are more dangerous than people without color (i.e., perceived Eu-
ropean Americans). Another variation of this effect is the finding that mi-
nority children are more frequently evaluated and reported as victims of
child abuse, according to Lane et al. (2002), which shows a reporting bias
and does not necessarily indicate that minorities are more likely to be
abusive toward their children. Clinician biases are also seen in increased
rates of seclusion and restraint of African Americans compared with
whites in the psychiatric inpatient ward (Flaherty and Meagher 1980)
and in the more and larger doses of oral and injectable antipsychotic
medications given to African American patients than to similar white
patients by psychiatric clinicians in psychiatric emergency and inpatient
services (Primm 2006). Flaherty and Meagher (1980) speculated that “the
stereotype of the black male made the staff feel and act as if blacks were
more dangerous, prompting more restrictive measures” (p. 681). Fla-
herty and Meagher’s study also found that black patients were less likely
to be referred to recreational and occupational therapy, whereas these
services were routinely ordered for whites on the unit. The physicians on
the unit reported that they ordered these therapies routinely unless a pa-
tient was too dangerous or psychotic to participate.
The reasons for the overdetermination of dangerousness in people
of color have been poorly studied, and the few studies examining this
issue tend to be dated. Flaherty and Meagher (1980) found that all-
white inpatient treatment teams spent less time discussing black patient
issues compared with white patient issues. Adebimpe (1981) and others
(Gross et al. 1969) have observed similar low allocations of time on
black patient issues and attribute part of this problem to the social and
cultural distance between the patient and clinician. Adebimpe (1981)
cited differences in vocabulary, modes of communication, value sys-
tems, and expression of distress and a breakdown in rapport as factors
that increase diagnostic errors. Jones and Gray (1985) stated that “white
psychiatrists seem to have more difficulty relating to black male pa-
tients than to [black] female patients” (p. 25) and hypothesized that this
may be because white psychiatrists expect black men to be threatening
(i.e., they adhere to a common societal negative stereotype of black men).
Although these important studies are classic, they are from a time when
experimental design and statistical methodology were less sophisti-
cated, and modern studies should be performed to expand upon and
update these findings.
The following two cases illustrate how an individual’s or clinician’s
cultural experiences and assumptions can influence violent behavior
and the risk assessment of violent behavior.
Case Example 1:
Racism and Oppression in a Hawaiian Man
A 35-year-old biracial Caucasian and Hawaiian man was imprisoned af-
ter being convicted of killing his ex-wife’s father, a Portuguese Ameri-
can man who had disapproved of his daughter’s marrying someone
who was not Portuguese. The patient had murdered the victim by
punching him to death. The patient was an amateur boxer and at the
time of the killing was using cocaine and complaining that he was in
danger from others. He was eventually released after the completion of
his prison sentence and placed in a board and care for persons with
mental disorders and substance abuse disorders. He lived there with
other men belonging to various ethnic groups and would get into fights
when a roommate would call him by a racial slur.
According to the HCR-20 (Historical, Clinical, and Risk Manage-

ment) Survey, this patient had many risk factors, including substance
abuse, a mental disorder (substance-induced delusional disorder), and
a previous history of violence. Because he belonged to a group that had
been taken advantage of by previous explorers of Hawaii, decimated by

imported diseases, and overthrown from power in 1893, he had a sense
of disempowerment. Likewise, his cultural institutions were disman-
tled by missionaries and foreign businessmen. He felt a sense of shame
about his Hawaiian identity because he felt powerless and unable to be
an effective part of society (Schultz-Ross 1997).
In response to the resident's violence, the staff of the board and care
facility started a group that emphasized the understanding of Hawaiian
culture. His individual work focused on his owning his nonviolence, as
opposed to his seeing it as obedience to orders that he resented. The
intervention was intended to increase his sense of self-respect and thus
reduce his levels of shame. Another intervention would have been to
target improvement of interpersonal relationships; for example, hoopo-
nopono, an indigenous mental health intervention, might have been
more effective (Rezentes 1996).
Case Example 2:
Racism, Transference, and Countertransference
A male Caucasian patient was to be evaluated at a prison for a psychi-
atric disorder. He was 6 feet tall, heavy, and muscular. His voice was
loud, and he spoke with a heavy Southern accent. He glared angrily at
anyone who came near. His psychiatrist, a thin, 5-foot-tall African
American wearing glasses, began to question the patient, but he refused
to answer, saying, “Not to you!” The patient was a devout Christian and
read the Bible daily. He had been convicted of murdering an elderly
Caucasian woman by tackling her, lifting her up in the air, and dropping
her on the sidewalk. He was convinced that she meant him harm and
felt no remorse for his actions. He also believed that the correctional of-
ficers meant him harm.
Eventually, the patient agreed to speak with the psychiatrist. Ini-

tially the patient distrusted him because of his race, but the larger rea-
son was because the physician was employed by the prison. Although
the patient’s racism could have been the reason he did not want to
speak to his doctor, his reluctance was eventually found to be driven by
paranoia (Schultz-Ross 1997).
Given this background, how confident can we be as psychiatrists in
assessing the risk of violence in psychiatric patients? Tardiff (1998)
stated that psychiatrists are fairly reliable assessors, but the influence of
the ethnicity of the patient and the evaluator was not examined. Our
position is that knowing some cultural information about patients—
such as any differing cultural norms, religious beliefs, and experiences
with racism—and knowing our own stereotypic biases will improve
psychiatric diagnosis and improve the therapeutic alliance so that the

potential for violence is much reduced.
Interpersonal Violence
Interpersonal violence is defined as a pattern of assaultive and coercive
behaviors, including physical, sexual, and psychological attacks as well
as economic coercion. The assessment of IPV is somewhat different
from risk assessments done in psychiatric hospitals, clinics, and prisons.
The person being evaluated is usually the victim and may or may not
volunteer that he or she is being abused. Culture and religion also affect
the assessment and management of this potentially violent situation.
Intimate partner violence (IPV) is a form of interpersonal violence be-
tween two people who are intimate, including spouses, couples, and
partners. The term was developed to replace wife abuse or spousal abuse
in order to include gay partners. Within the women’s movement, IPV
usually refers to violence toward women; the term is also used in refer-
ence to gay female relationships.
IPV is pervasive, with one in four women in the United States expe-
riencing abuse during their lifetime (Tjaden and Thoennes 2000). In
2002, the World Health Organization identified domestic violence as a
serious public health problem, with victims experiencing more opera-
tive procedures, visits to doctors, and hospital stays than nonvictims.
Domestic violence not only causes acute injuries but also has been
linked to serious health consequences such as chronic pain, abdominal
complaints, sexually transmitted infections, unwanted pregnancies, de-
pression, posttraumatic stress disorder, miscarriages, and premature la-
bor (Krug et al. 2002). Unfortunately, many victims suffer in silence and
receive no assistance for their abusive situation. Healthcare and mental
healthcare professionals play a crucial role in identifying victims be-
cause these professionals have regular opportunities to ask their pa-
tients about domestic violence, regardless of the reason for the medical
visit. Patients should be routinely screened for domestic violence by di-
rectly asking about domestic violence, regardless of symptoms, injuries,
or reason for the visit (Mayor’s Office to Combat Domestic Violence
2003). In the latter part of this chapter, we discuss some culturally spe-
cific information useful in the assessment and prevention of IPV.
Case Example 3:
Domestic Violence in a Puerto Rican Woman
Ms. A, a 30-year-old Puerto Rican woman from New York, married an
Argentinean man she met in college. She was a second-generation
Puerto Rican American woman born to immigrant parents in the lower

middle class, whereas her husband was wealthy and belonged to the ar-
istocracy. They had moved to Los Angeles to be near the husband’s
older brother. Ms. A’s husband began to drink and would beat her when
he was intoxicated. She asked her husband to stop his behavior, and he
would apologize and agree not to hit her anymore, but it continued to
happen. She eventually asked her brother-in-law for advice, but he said
that she must be doing something wrong. Her feeling after this conver-
sation was that the brother-in-law did not approve of her. She became
pregnant, and after the birth of their first child, the beatings intensified.
Ms. A filed for divorce and moved back to New York to live with her
family. Her family was not supportive of her because they felt that di-
vorce was wrong. After several years, she remarried to a Puerto Rican
man. He also began to beat her several years after they married. She re-
mained silent about the abuse for 3 years, but she eventually went to a
counseling center. Separation was suggested, but she adamantly re-
fused. Her husband left her for several months, and she allowed him to
move back in.
This case demonstrates the powerful influence of family support.

Ms. A would rather risk being physically hurt than bear the brunt of her
family’s disapproval and suffer their withdrawal of support (Schultz-
Ross 1997).
Critical Concepts:
Culturally Appropriate Assessment
Culture can be defined as a set of meanings, norms, beliefs, and values
shared by a group of people. These beliefs and values are taught, rein-
forced, and reproduced to the next generation. Culture refers to a sys-
tem of meanings in which words, behaviors, events, and symbols have
attached meanings that are agreed upon by the members within the cul-
tural group. Thus, an individual’s culture shapes how he or she makes
sense of the social and natural world. Finally, culture includes both the
subjective components of human behavior (the shared ideas and mean-
ings that exist within the minds of individuals within a group) as well
as the objective components (the observable behaviors and interactions
of these individuals).
One’s culture shapes what symptoms one expresses and how they
are expressed (Mezzich et al. 2000), and it influences the meaning that
one attributes to symptoms and how one interacts with the healthcare
system. Culture also influences what a society regards as appropriate or
inappropriate behavior, and it thus exerts a powerful influence on an in-
dividual’s potentially violent behavior. War, with its sociocultural
upheaval and association with posttraumatic stress disorder, is one ex-

ample of how cultural-historical events can cause or contribute to psy-
chopathology (Du and Lu 1997; Kirmayer 2001). Likewise, culture can
also exert a protective influence on mental health. Traditional healing
approaches and spiritual/religious interventions may also provide
meaningful benefits to patients (Ton and Lim 2006), as may interven-
tions that are culturally syntonic.
Culturally competent care involves combining general culture-
specific knowledge with specific information from the patient and be-
ing aware of biases that either the clinician or the patient may bring to
the evaluation. Because individuals can belong to more than one cul-
tural group, they may not comply with the norms of their stated cul-
tural group and may in fact emphasize some cultural values of that
group and deemphasize others.
The DSM-IV-TR Outline for Cultural Formulation

The publication of DSM-IV and its text revision, DSM-IV-TR (American
Psychiatric Association 1994, 2000) represented a turning point in the ap-
plication of cultural psychiatry principles with the introduction of the
Outline for Cultural Formulation (OCF; DSM-IV-TR Appendix I). The
OCF gives clinicians a framework for assessing the impact of culture on
the diagnosis and treatment of psychiatric illness (Table 3–1). Because
culture plays such a crucial role in all aspects of mental health and illness,
it is important to incorporate cultural assessment as part of any interven-
tion. The culturally competent clinician seeks to acquire knowledge
about the cultural groups of his or her patients. Although such knowl-
edge is essential, a framework to organize and make sense of the infor-
mation is extremely helpful. However, the clinician inevitably will en-
counter many patients who are affiliated with cultural groups of which
the clinician has inadequate knowledge. Furthermore, patients may not
fully engage in all the beliefs and practices of a given cultural group.
The first part of the DSM-IV-TR OCF describes how the individual sees
him- or herself and relationships to others. Clinicians should be aware that
cultural identity is multidimensional and can have many different aspects,
such as country of origin, language spoken, religious beliefs, identified
ethnicity, sexual orientation, marital status, and so on. The patient’s cul-
tural identity is constructed during the interview but can be completed in
greater detail during the social and developmental history.
Part two of the formulation has to do with the patient’s beliefs about
his or her illness. The clinician asks what the patient thinks is causing
the problem and what the patient would do to solve it. In the case of
TABLE 3–1. DSM-IV-TR outline for cultural formulation

A. Cultural identity of the individual
B. Cultural explanations of the individual’s illness
C. Cultural factors related to psychosocial environment and levels of
functioning
D. Cultural elements of the relationship between the individual and the
clinician
E. Overall cultural assessment for diagnosis and care
Source. American Psychiatric Association 2000.
violent behavior, this question would relate to the reasons behind and
justification for such behavior. Part three of the formulation concerns
stressors and supports and includes an extended family assessment re-
garding the family’s influence on the patient as well as the role that re-
ligion plays in the patient’s life. Part four is an examination done by the
clinician to assess what role the clinician’s and patient’s ethnicities are
playing in the interaction. (As in Case 2 above, there could be transfer-
ence on the patient’s part and countertransference from the evaluator’s
perspective, leading to fear on the evaluator’s part and to aggression
from the patient.) Part five assembles the previous four parts to make a
formulation that informs treatment, such as the cultural history in Case
1 above, leading to a culturally appropriate treatment plan.
DSM-IV-TR also includes culture-bound syndromes, such as amok,
boufée delirante, pibloktoq, and zar. Although these can provide an explana-
tion of violent behavior, they are seen in cultures not often encountered
in daily clinical practice. These syndromes are included in Table 3–2 for
reference. Using the DSM-IV-TR OCF in concert with a cultural consult-
ant or someone familiar with a particular culture’s beliefs, values, and
norms to help understand individuals’ beliefs about their behavior, their
family relationships, and their religious beliefs will yield much more use-
ful information for the assessment and management of violence.
Culturally Appropriate Assessment in the

Psychiatric Setting
In clinical settings, a potentially violent situation occurs when an indi-
vidual who has a history of violence, but is not currently threatening to
become violent, is being interviewed. In this situation there is a lot of
time to assess the patient’s risk for violence and to plan how to respond
to the violence if it does become urgent or emergent. All patients, re-
gardless of cultural, racial, or ethnic origin, should be assessed for the
TABLE 3–2. DSM-IV-TR culture-bound syndromes that involve

violence or aggression
Amok: A dissociative episode characterized by a period of brooding followed
by an outburst of violent, aggressive, or homicidal behavior directed at
people and objects. The episode tends to be precipitated by a perceived slight
or insult and seems to be prevalent only among males. The episode is often
accompanied by persecutory ideas, automatism, amnesia, exhaustion, and a
return to premorbid state following the episode. Some instances of amok
may occur during a brief psychotic episode or constitute the onset or an
exacerbation of a chronic psychotic process. The original reports that used
this term were from Malaysia. A similar behavior pattern is found in Laos,
Philippines, Polynesia (cafard or cathard), Papua New Guinea, and Puerto
Rico (mal de pelea), and among the Navajo (iich’aa).
Boufée delirante: A syndrome observed in West Africa and Haiti. This French
term refers to a sudden outburst of agitated and aggressive behavior, marked
confusion, and psychomotor excitement. It may sometimes be accompanied
by visual and auditory hallucinations or paranoid ideation. These episodes
may resemble an episode of brief psychotic disorder.
Pibloktoq: An abrupt dissociative episode accompanied by extreme
excitement of up to 30 minutes’ duration and frequently followed by
convulsive seizures and coma lasting up to 12 hours. This is observed
primarily in arctic and subarctic Eskimo communities, although regional
variations in name exist. The individual may be withdrawn or mildly
irritable for a period of hours or days before the attack and will typically
report complete amnesia for the attack. During the attack, the individual may
tear off his or her clothing, break furniture, shout obscenities, eat feces, flee
from protective shelters, or perform other irrational or dangerous acts.
Zar: A general term applied in Ethiopia, Somalia, Egypt, Sudan, Iran, and
other North African and Middle Eastern societies to the experience of spirits
possessing an individual. Persons possessed by a spirit may experience
dissociative episodes that may include shouting, laughing, hitting the head
against a wall, singing, or weeping. Individuals may show apathy and
withdrawal, refusing to eat or carry out daily tasks, or may develop a long-
term relationship with the possessing spirit. Such behavior is not considered
pathological locally.
Source. American Psychiatric Association 2000.
risk for violence. Assessing for potential violence is a serious undertak-

ing and, if possible, should not be rushed. Furthermore, repeated risk
assessments for violence should occur in the same way that repeated
risk assessments for suicide occur. Time permitting, these assessments
should involve a prolonged clinical assessment and a thorough review
of the individual’s documented history from every possible source,
including legal records and discussions with family, friends, witnesses,
attorneys, and victims (Johnson 2000). Occasional use of screening
instruments such as the HCR-20 may be helpful in the clinical assess-

ment. The single strongest predictive factor for the assessment of violence
risk is a history of violence. It is vital that the evaluator assess whether
the patient has had singular or repetitive episodes of violence and
whether these episodes have been planned or impulsive. Assessment of
the outcome of the patient’s violent episodes (whether they lead to
harm to others, harm to property, harm to self, or no physical or emo-
tional harm) and comparison of these episodes with the norms of the
patient’s cultural group is also important.
When assessing an individual for future violence, one must distin-
guish between a clinically oriented assessment and a statistically based
assessment. A clinical assessment is based on an evaluator’s skill, expe-
rience, and knowledge. A statistical assessment is an actuarial predic-
tion based on statistical models and the use of risk factor instruments.
In any setting, an actuarial assessment is more accurate than a clinical
assessment. Ideally, psychiatrists performing violence risk assessment
should be aware of their cultural, racial, and ethnic stereotypes and
prejudices, and ethically, they should take these biases into account dur-
ing their assessment. Here is where the OCF of DSM-IV-TR could be
helpful, but in our personal experience we find it is underused.
When assessing patients of different cultural, racial, and ethnic
backgrounds who may be potentially violent, it is important not to
micro-insult or micro-aggress against the patient in the process of the
evaluation (Bell et al. 2006). For example, for a younger white male psy-
chiatrist to call an elderly black male patient by his first name may be
perceived by the patient as a subtle insult. Equally insulting is asking a
person of color for his or her Medicaid card instead of asking how he or
she plans to pay for the service (Bell et al. 2006). It is also important to
be culturally sensitive to how some people of color perceive and re-
spond to dominance or authority. For example, because of African
Americans’ concerns about racism, asking an African American to sub-
mit to a search for weapons upon entering a clinical setting such as an
emergency department or inpatient unit may be mistaken for discrimi-
nation instead of standard operating procedure. This may also be true
of visible security presence in a clinical setting. Probably one of the most
culturally insensitive acts of the majority culture against people of color
is the denial of the existence of racism, which frequently happens when
a person of color raises the issue of racism or is being victimized by a
racial stereotype—a common experience for people of color. One exam-
ple of stereotyping is the presumption of guilt and criminal intent when
people of color are singled out by police for questioning (e.g., the traffic
stop for “driving while Black or Hispanic”). This is their reality.
Culturally Appropriate Assessment in

Interpersonal Violence
We now focus on specific cultural knowledge that will help in the as-
sessment of patients who may be subject to IPV but who come to the
psychiatrist’s office for another reason. There is much more in the liter-
ature about the assessment of IPV in ethnic minorities than there is
about the risk assessment of ethnic minorities for violent behavior. The
following case illustrates how a culturally competent clinician may use
cultural knowledge and norms to properly assess a patient for IPV.
Case Example 4:
A Pregnant Pakistani Adolescent
Sheryl comes in for her first visit to a psychiatrist. She is 17 years old and
has been married to her 19-year-old husband for 3 months. He answers
most of the questions for her and states that she is depressed and does
nothing around the house. The husband refuses to leave the room when
asked to do so. Nevertheless, the psychiatrist insists, and the husband
leaves the room. Sheryl never makes eye contact with the physician.
When the psychiatrist assesses her for IPV, she denies it. When asked if
her husband controls what she does, she states, “Of course he does. He
is my husband.”
Knowing nothing further about the patient than the stated case, we
would be alarmed by the overly controlling partner and the patient’s
lack of eye contact. We know, however, that she emigrated from Paki-
stan and is Muslim. The culturally specific information that we would
find helpful in this situation is that generally, Muslim women expect to
be married and expect that the marriage will be arranged. A study by
Hassouneh-Phillips (2001) showed that American Muslim women view
marriage as a means to achieving personal and spiritual happiness.
They also believe that good wives are obedient, because the Qur’an
states that men have more strength so their duty is to protect and sup-
port their women (Mayor’s Office to Combat Domestic Violence 2003).
As mentioned earlier, culture influences how people view and per-
ceive abuse; whether they seek help, how they communicate their expe-
riences; and from whom they are likely to seek assistance. Cultural fac-
tors may serve as barriers to treatment, such as an extended family
structure in which a family elder supports the abuse, or a church leader
who advises the woman to go back to her husband. The clinician should
communicate with each patient as an individual, without expecting
generalized reactions from their respective cultural groups.
IPV is more common in cultures in which women are considered to

be inferior. African Americans tend to use violence as a resource of con-
trol to compensate for a lack of other resources such as money, respect,
power, prestige, or knowledge (Weil and Lee 2004). Before going on to
more specific cultural information, we would like to discuss an ap-
proach to using such information. Table 3–3, presents a “culturally com-
petent assessment A–E” for incorporating cultural knowledge with the
patient evaluation. A is for Assumptions, and cultural assessment in-
volves looking at our own cultural assumptions. B is for the Beliefs of
the group being evaluated, as in knowing and understanding those be-
liefs. C signifies that effective Communication can be a bridge between
the belief systems of the evaluator and the patient. D, for Diversity, al-
lows us to understand how patients’ individual experiences make them
different from others in their cultural group. E is for Education—what a
clinician needs in order to understand how other groups differ from our
own—and for Ethics and how they are changed by differing cultural be-
liefs (Thompson 2005). With this A–E mnemonic as a backdrop, we
present some culture-specific information about cultural groups found
in the literature to see how culture influences their experience of IPV.
Somalis
Pan et al. (2006) conducted interviews in San Diego, California, with
members of the Somali, Latino, and Vietnamese communities. They
found that Somali community members felt that physical violence was
an unacceptable means of conflict resolution. However, IPV is viewed as
an acceptable means of maintaining the patriarchal structure of the So-
mali family. The major sources of conflict within the family were changes
in gender roles and responsibilities since resettling in the United States.
The power dynamic in the family was reversed when families came to the
United States because government aid checks were issued to the Somali
women, not the men. Thus, Somali men reported feelings of helplessness
and uselessness because they have lost their role as the breadwinner for
the family, and many try to regain control through violence.
Somali women are responsible for maintaining harmony within the
family by supporting their husband, obeying his wishes, and not upset-
ting or angering him. IPV perpetrated by the husband can be justified in
situations in which the wife defied the husband’s wishes. Somali men
view this as the husband’s right to “teach his wife a lesson.” Interestingly
enough, there is no term for IPV in the Somali language. Both genders
report that it is a commonly held belief among women that if a husband
does not beat his wife, it means he does not love her. Most Somalis are
TABLE 3–3. Culturally competent assessment A–E

A Assumptions The act of taking for granted or supposing that a
concept or idea is true.
B Beliefs Shared concepts about how a group operates.
C Communication Two-way sharing of information that results in an
understanding between the sender and the
receiver.
D Diversity The way in which people actually differ
(regardless of other people’s assumptions or
beliefs) and the effect those differences have on
their response to healthcare and to the
practitioner.
E Education, Ethics Gaining knowledge about a diverse group and
understanding that ethical issues may be viewed
differently by different groups.
Source. Adapted from Thompson 2005.
Muslims, and Islamic traditions are thought to reduce tension in families

and thus reduce the incidence of domestic violence. Hassouneh-Phillips
(2001) reported that renegotiated Islamic marriage contracts did not stop
IPV, nor did consulting with the Imam. However, Potter (2007) reported
that in one case, consulting with religious leaders resulted in the removal
of the husband from the home. As in many non-Western cultures, family
members and community elders are frequently called upon as a resource
for resolving conflict between spouses. Traditionally, the wife consults
with the men in her family, who then talk to the husband. However, So-
mali women in the United States do not have access to this type of family
support because they are refugees fleeing a lengthy civil war. During the
resettlement process, refugees receive information about American laws
related to IPV, and women are told to call the police to report violence in
the home. However, this can be counterproductive to their marriage be-
cause involving outsiders in family matters is deemed “Americanized”
and is an appropriate reason to seek a divorce. Somali community mem-
bers report that men can divorce their wives if they become “too Amer-
icanized.” Thus, many Somali women are trapped in violent situations
with no culturally viable means of resolving the problem.
Asian Americans
In assessing any Asian culture, it may be useful to know that Asians
value the importance of the family over the individual. They believe in
conflict avoidance and that personal problems such as marital issues
should remain private so as not to shame and dishonor the family and
cause a loss of face. Before marriage, a woman follows and obeys her fa-
ther; after marriage, she follows and obeys the husband; and after the
death of her husband, she follows and obeys the son (Xu et al. 2001).
Yoshioka and Dang (2000) did a survey of Asian American families
about their attitudes toward family violence. The men had the highest
score in support of male privilege. They believed that a man has the
right to discipline his wife, that he should be able to have sex whenever
he wants it, that he is the ruler of his home, and that some wives deserve
beatings. It is a sign of weakness to ask for help, and family members
would discourage the disclosure of problems and would make excuses
for the abuser. Asian Americans’ respect for their elders would result in
pressure not to report (see Table 3–4 for a list of barriers to reporting
IPV). Violence was justified if the wife had an extramarital affair, lost
emotional control, or made a financial decision without consulting the
husband. Older respondents and men were more tolerant toward the
use of force to resolve family conflict (Yick and Agbayani-Siewart 1997).
Vietnamese
The Vietnamese community in San Diego (Pan et al. 2006), as do many
non-Western ethnic communities, sees domestic violence as a family
matter. Sharing information about the family with outsiders is viewed
as inappropriate. Violence, ignoring problems, and seeking outside as-
sistance (court, counselors, police) are cited as unacceptable ways of re-
solving conflict, which creates a trap, because the victim cannot involve
outsiders to resolve his or her problems with violence. Shame is a major
barrier to accessing services. The responsibility for maintaining peace
and family harmony falls on the woman, who accomplishes this task by
obeying her husband’s wishes and attending to the needs of her hus-
band, much like what was seen in Somali families. Strong family ties
and respect for family members are cited as ways of promoting har-
mony in the family. Acceptable strategies for conflict resolution include
soliciting and listening to the advice of parents and elders or discussing
problems in a peaceful manner. The primary stressors for IPV were eco-
nomic. Bui and Morash (1999) stated that “for Vietnamese Americans,
women’s economic contributions could not reduce husbands’ domi-
nant positions and violence, but economic hardship could prevent
abused women from leaving an abusive relationship ” (p. 790). Viet-
namese participants reported that sending money to family members in
Vietnam or sponsoring family members to the United States are major
sources of tension. In addition, they repeatedly identified excessive
gambling as a cause of tension within the family. Both men and women
TABLE 3–4. Barriers to culturally appropriate interpersonal violence

assessment in Asian American patients
Institutional
Monolingual worker
Immigration status
Welfare policy
Refugee resettlement
Racism and homophobia
Lack of health insurance, training, child care, and affordable housing
Nonadaptive systems
Cultural
Values/Beliefs
Isolation
Shame
Other kinds of relationships (i.e., same gender or interracial)
Community
Religion and spirituality
No support from community or family
Individual
Values around shame
Low self-esteem/self-confidence
Inability to speak English
Lack of cultural fluency
Age
Lack of marketable skills
Status
Socialization patterns
Not knowing resources and law
Source. This material was reprinted and/or adapted from the Family Violence Preven-
tion Fund’s publication entitled “(Un)heard Voices: Domestic Violence in the Asian
American Community” (2007). The report was authored by Sujata Warrier, Ph.D. Pro-
duction was made possible by a grant from the Violence Against Women Office, Office
of Justice Programs, U.S. Department of Justice.
gamble, and many spend their families’ income either in the casinos or
in underground gambling rings.
Cambodians
Weil and Lee (2004) described cultural factors that increase the risk of
IPV for Cambodian women. The cultural expectations of the wife are
that she will obey and respect the husband, not be sexually promiscu-
ous before or after marriage, and accept the problems of the marriage.
Women are blamed for problems regardless of fault, and in fact, the cul-
tural belief is that the woman must have done something wrong to de-
serve such punishment. A contributing factor is that most women were
physically abused by their parents before their marriages. Women do
not have the right to divorce or leave a husband who is hitting them and
cannot have their husbands arrested for violent acts against the family.
Latinos
In the Latino community, family harmony is supported by following
family traditions and celebrations and helping each other. Physical vio-
lence and verbal aggression are considered unacceptable ways of re-
solving conflict. Open communication between family members and
leaving potentially volatile situations are viewed as acceptable ways of
resolving conflict. Gender roles in this community appear to be slowly
changing as families adjust to living in the United States; men are start-
ing to recognize that women may have more to contribute to the family
than the domestic tasks of cooking, house cleaning, and childcare.
Latina women are asking for more equitable distribution of labor
and decision making in the household; however, they are still responsi-
ble for the vast majority of housework and childcare. Adolescent girls
reported frustration about the amount of responsibility they held in the
household compared with their brothers. The frequently mentioned
causes of tension in the family were economics, immigration status, and
substance use. Women from the Latino community reported that the
threat of deportation due to their undocumented status is often used as
a means of controlling them and ensuring that they do not leave abu-
sive situations. For example, women reported that men often say that if
their wives call the police, they will be deported. When compared with
the Somali and Vietnamese communities, the Latino community ap-
pears to be more aware of the availability of domestic violence interven-
tion services but has a limited understanding of how to use the services
and how to work with service providers. Significant barriers to access
are language and cultural differences, fear of deportation, and the in-
ability to effectively use identified services (Pan et al. 2006).
African Americans
African American families have a legacy of racism and stereotypes that
works against both members of the married couple. Black men, like
most men, may experience entitlement dysfunction when they see that
they are being fitted into a gender role stereotype. They are often seen
as menacing, so they have been legitimized by this stereotype to be in-
timidating and controlling. African American women have an image of
having much sexual, social, physical, and economic power. They can
also be seen as invulnerable, insensitive, stoic, and in need of domesti-
cation and control. Victims of IPV feel shame about their inability to
have a perfect family. Those who have darker complexions, tall phy-
siques, are overweight, fight back, or have a mental illness are thought
to deserve abuse (Bell and Mattis 2000). Some African American adoles-
cent women are coerced into intimate relationships with older African
American males and may be labeled as morally suspect or hypersexual.
Financial pressures may make leaving an abusive relationship seem-
ingly impossible, because the single mother and her children would
end up homeless (Bell and Mattis 2000). Finally, Richie (1996) noted that
many African American women who are being battered by their part-
ners or sons are reluctant to report violence out of fear of contributing
to the victimization of African American men.
Potter (2007) found that many African American women seek sup-
port from religious leaders when trying to deal with IPV, and many are
not supported for reporting the violence or for wanting to leave their
husbands. The victim is often sent back to the perpetrator with the man-
date that they “should work things out” because Ephesians 5:21–33
states “submit yourselves unto your own husbands, as unto the Lord.
For the husband is the head of the wife, even as Christ is the head of the
church…. Therefore as the church is subject unto Christ, so let the wives
be to their own husbands in everything.”
Key Points
■ The culturally appropriate assessment of the risk of violence is
vital in developing effective interventions that are culturally con-
gruent.
■ As the United States becomes more ethnically diverse every year,
the likelihood increases that our patients will be from cultures
with which we are unfamiliar.
■ A violence risk assessment of culturally diverse patients requires
the clinician to become familiar with basic norms about violence,
coping strategies and behaviors, gender roles, and the roles of spir-
ituality and religion in the patient’s culture.
■ Frameworks such as the DSM-IV-TR OCF or the “culturally appro-

priate assessment A–E” mnemonic outlined in this chapter may
prove helpful in organizing the assessment of violence risk in cul-
turally diverse patients.
■ Ultimately, the clinician has the responsibility not to stereotype
patients either diagnostically or as “more prone to violence” when
the data do not support that linkage.
■ More research needs to be done so that clinicians may further
understand the links between ethnicity, socioeconomic status,
and the risk of violent behavior.
References
Adebimpe VR: Overview: white norms and psychiatric diagnosis of black pa-
tients. Am J Psychiatry 138:279–285, 1981
American Psychiatric Association: Diagnostic and Statistical Manual of Mental
Disorders, 4th Edition. Washington, DC, American Psychiatric Association,
1994
Disorders, 4th Edition, Text Revision. Washington, DC, American Psychiat-
ric Association, 2000
American Psychiatric Association: Resolution Against Racism and Racial Dis-
crimination and Their Adverse Impacts on Mental Health: Position State-
ment, July 2006. Available at http://www.psych.org/edu/other_res/
lib_archives/archives/200603.pdf. Accessed February 9, 2007
Bell CC: Violence prevention 101: implications for policy development, in Per-
spectives on Crime and Justice 2000–2001 Lecture Series, Vol V. Washing-
ton, DC, National Institute of Justice, March 2002, pp 65–93. Available at
http://www.ncjrs.gov/pdffiles1/nij/187100.pdf. Accessed February 10,
2007
Bell CC: Prevention is the future. Clinical Psychiatry News 35:14, 2007
Bell CC, Mattis J: The importance of cultural competence in ministering to Afri-
can American victims of domestic violence. Violence Against Women
6:515–532, 2000
Bell CC, Mehta H: The misdiagnosis of black patients with manic depressive ill-
ness. J Natl Med Assoc 72:141–145, 1980
Bell CC, Dove HW, Williamson JL: Challenges and obstacles in treating men-
tally ill black patients. Psychiatric Times 23:48-49, 2006. Available at http://
giftfromwithin.org/html/challenge.html. Accessed December 16, 2007
Bui HN, Morash M: Domestic violence in the Vietnamese immigrant commu-
nity: an exploratory study. Violence Against Women 5:769–795, 1999
Du N, Lu F: Assessment and treatment of posttraumatic stress disorder among
Asians, in Working With Asian Americans. Edited by Lee E. New York,
Guilford, 1997, pp 275–294
Flaherty JA, Meagher R: Measuring racial bias in inpatient treatment. Am J Psy-
chiatry 137:679–682, 1980
Gross H, Herbert MR, Kantterud GL, et al: The effects of race and sex on the
variation of diagnosis and disposition in a psychiatric emergency room.
J Nerv Ment Dis 148:638–642, 1969
Hassouneh-Phillips D: Marriage is half faith and the rest is fear of Allah. Vio-
lence Against Women 7:927–946, 2001
Hollinger PC, Offer D, Barter JT, et al: Suicide and Homicide Among Adoles-
cents. New York, Guilford, 1994
Hucker S: Psychiatric Aspects of Risk Assessment, 2004. Available at http://
www.violence-risk.com/risk/assessment.htm. Accessed February 10, 2007
Johnson B: Assessing the risk for violence, in Psychiatric Aspects on Violence:
Understanding Causes and Issues in Prevention and Treatment. Edited by
Bell CC. San Francisco, CA, Jossey-Bass, 2000, pp 33–36
Jones BE, Gray BA: Black and white psychiatrists: therapy with blacks. J Natl
Med Assoc 77:19–25, 1985
Keppel KG, Pearcy JN, Wagener DK: Trends in Racial and Ethnic-Specific Rates
for the Health Status Indicators: United States, 1990–98. (Healthy People
2000 Statistical Notes No 23, pp 1–12). Rockville, MD, U.S. Department of
Health and Human Services, 2002
Kirmayer LJ: Cultural variations in the clinical presentation of depression and
anxiety: implications for diagnosis and treatment. J Clin Psychiatry 62
(suppl):22–28, 2001
Krug EG, Dahlberg LL, Mercy JA, et al: World Report on Violence and Health.
Geneva, Switzerland, World Health Organization, 2002
Lane WG, Rubin DM, Monteith R, et al: Racial differences in the evaluation of
pediatric fractures for physical abuse. JAMA 288:13:1603–1609, 2002
Mayor’s Office to Combat Domestic Violence: Medical Providers’ Guide to Man-
aging the Care of Domestic Violence Patients Within a Cultural Context, Au-
gust 2003. Available at http://www.nyc.gov/html/ocdv/downloads/pdf/
providers_dv_guide.pdf. Accessed March 25, 2007
Mezzich JE, Otero AA, Lee S: International psychiatric diagnosis, in Compre-
hensive Textbook of Psychiatry, 7th Edition. Edited by Kaplan HI, Sadock
BJ. Baltimore, MD, Williams & Wilkins, 2000, pp 839–853
Miranda J, Lawson W, Escobar J: Ethnic minorities. Ment Health Serv Res 4:231–
237, 2002
Pan A, Daley S, Rivera LM, et al: Understanding the role of culture in domestic
violence: the Ahimsa Project for Safe Families. J Immigr Minor Health 8:35–
43, 2006
Petee TA, Padgett KG, York TS: Debunking the stereotype: an examination of
mass murder in public places. Homicide Studies 1:317–337, 1997
Pierce CM, Earls FJ, Kleinman A: Race and culture in psychiatry, in The Harvard
Guide to Psychiatry. Edited by Nicholi AR. Cambridge, MA, Belknap Press,
1999, pp 735–743
Potter H: Battered black women’s use of religious services and spirituality for
assistance in leaving abusive relationships. Violence Against Women
13:262–284, 2007
Primm AB: The assessment of African American patients, in Clinical Manual of
Cultural Psychiatry. Edited by Lim RF. Washington, DC, American Psychi-
atric Publishing, 2006, pp 35–68
Pulera DJ: Visible Differences: Why Race Will Matter to Americans in the
Twenty-First Century. New York, Continuum International Publishing,
2002
Rezentes WC: Ka Lama Kukui Native Hawaiian Psychology: An Introduction.
Honolulu, HI, A’Alii Books, 1996
Richie B: Compelled to Crime: The Gender Entrapment of Black Battered
Women. New York, Routledge, 1996
Sampson RJ, Raudenbush SW, Earls F: Neighborhoods and violent crime: a
multilevel study of collective efficacy. Science 277:918–924, 1997
Schultz-Ross RA: Violent behavior, in Culture and Psychopathology. Edited by
Tseng WS, Strelzer J. New York, Brunner/Mazel, 1997, pp 173–189
Steadman H, Mulvey E, Monahan J, et al: Violence by people discharged from
acute psychiatric inpatient facilities and by others in the same neighbor-
hoods. Arch Gen Psychiatry 55:393–401, 1998
Substance Abuse and Mental Health Services Administration: Results from the
2002 National Survey on Drug Use and Health: National Findings, 2003.
Available at http://www.oas.samhsa.gov/nhsda/2k2nsduh/Results/
2k2Results.htm. Accessed April 15, 2007
Sue DW, Sue D: Counseling the Culturally Different: Theory and Practice, 3rd
Edition. New York, Wiley, 1999
Tardiff K: Prediction of violence in patients. Journal of Practical Psychology and
Behavioral Health 4:12–19, 1998
Thompson R: Intimate partner violence: a culturally sensitive approach. Adv
Nurse Pract 13:57–59, 2005
Tjaden P, Thoennes N: Full Report of the Prevalence, Incidence, and Conse-
quences of Violence Against Women: Findings from the National Violence
Against Women Survey (Report No NCJ 183781). Washington, DC, U.S. De-
partment of Justice, 2000
Ton H, Lim, RF: The assessment of culturally diverse individuals, in Clinical
Manual of Cultural Psychiatry. Edited by Lim RF. Washington, DC, Amer-
ican Psychiatric Publishing, 2006, pp 3–31
Tseng WS, Matthews D, Elwyn TS: Cultural Competence in Forensic Mental
Health. New York, Brunner-Rutledge, 2004
U.S. Department of Health and Human Services: Culture, Race, and Ethnicity:
A Supplement to Mental Health: A Report of the Surgeon General. Rock-
ville, MD, U.S. Department of Health and Human Services, Public Health
Service, Office of the Surgeon General, 2001a
U.S. Department of Health and Human Services: Youth Violence: A Report of
the Surgeon General. Rockville, MD, U.S. Department of Health and Hu-
man Services, 2001b
Walsh A: African Americans and serial killing in the media: the myth and the
reality. Homicide Studies 9:271–291, 2005
Warrier S: Unheard voices: domestic violence in the Asian American com-
munity. Family Violence Prevention Fund, 2007. Available at http://www
.endabuse.org/programs/immigrant/files/UnheardVoices.pdf. Accessed
March 3, 2007
Weil JM, Lee HH: Cultural Considerations in understanding family violence
among Asian American Pacific Islander families. J Comm Health Nurs
21:217–227, 2004
West JC, Herbeck DM, Bell CC, et al: Race/Ethnicity among psychiatric pa-
tients: variations in diagnostic and clinical characteristics reported by prac-
ticing clinicians. Focus 4:48, 2006
Xu X, Campbell JC, Zhu F: Intimate partner violence against Chinese women:
the past, present, and future. Trauma Violence Abuse 2:296–315, 2001
Yick AG: Domestic violence in the Chinese American community: cultural ta-
boos and barriers. Family Violence and Sexual Assault Bulletin 15:16–23,
1999
Yick AG, Agbayani-Siewart P: Perception of domestic violence in a Chinese
American Community. J Interpers Violence 12:832–897, 1997
Yoshioka MR, Dang Q: Asian family violence: a study of the Cambodian, Chi-
nese, Korean, South Asian, and Vietnamese communities in Massachusetts.
Boston, MA, Asian Task Force Against Domestic Violence, 2000
C H A P T E R 4
Psychological Testing in
In recent years, the research and clinical literature on risk assessment

has focused almost exclusively on the development and validation of
structured assessment techniques. Yet despite the increased attention to
actuarial risk assessment (discussed in Chapter 2 by Monahan), most
practicing clinicians rely on unstructured or structured professional
judgment, essentially conducting a clinical evaluation and reaching a
conclusion about violence risk. A recent survey of board certified foren-
sic psychologists indicated that very few of those surveyed routinely
use formal risk assessment measures (Archer et al. 2006). Instead, foren-
sic evaluators continue to incorporate specialized psychological testing
to address important elements of an individual’s violence risk.
Even among ardent supporters of actuarial risk assessment, psycho-
logical testing and measurement can often play an invaluable role (e.g.,
in quantifying psychopathy or identifying excessive defensiveness). Al-
though the specific tests and techniques that are used necessarily vary
depending on patient history, evaluation setting, and the nature of the
violence in question (e.g., sexual, domestic), there are a number of in-
struments that are sufficiently common to be critical for any thorough
risk assessment. However, these measures should not be considered a
replacement for systematic approaches to risk assessment (i.e., “struc-
59
tured professional judgment” instruments) but rather an adjunctive

method to inform the clinical evaluation process.
A comprehensive review of all psychological tests that may be used
or useful in violence risk assessment is beyond the scope of this chapter.
Instead, we present a brief review of clinical factors and commonly
used measures that should be considered when evaluating risk of vio-
lence. In addition, we address some of the variations and specialized in-
struments that may be useful with special populations such as sexual
offenders, although it is not possible to address adequately all settings
and populations (for example, death penalty evaluations and suicide
risk assessments are not discussed). Finally, in this chapter we focus
chiefly on risk assessment, but it should be noted that the identification
of risk management strategies is an equally important aspect of violence
risk assessment. Many of the measures described here might be used to
help monitor violence risk or response to treatment as part of an ongo-
ing risk management approach. The following case example highlights
many of the assessment issues discussed in this chapter.
Case Example
Assessing Violence Risk in a Criminal Defendant
For the past several months, Shawn, a 23-year-old medical student, had
been dating one of his medical school classmates, Veronica. As the rela-
tionship evolved, Shawn became increasingly jealous, accusing Veron-
ica of flirting with one of their professors and eventually demanding
that she confess to her infidelity. Veronica ended the relationship shortly
after this confrontation, but Shawn continued to initiate contact with
Veronica after class despite her repeated insistence that they had noth-
ing more to discuss. One day Shawn arrived at Veronica’s apartment un-
announced, demanding she let him in to talk about their relationship.
Veronica refused to allow him into the apartment and told Shawn she
would call the police if he continued to harass her. Soon afterward
Veronica began receiving telephone calls in the night; the caller typically
would hang up without speaking, but occasionally she heard a man’s
voice on the phone calling her insulting names. Several days later
Veronica thought she saw Shawn sitting in a car parked across the street
from her apartment. She telephoned the police, who confronted Shawn
and, on searching his car, discovered a camera, a stun gun, a large knife,
and a pair of handcuffs. However, because Shawn had not actually as-
saulted Veronica, the charges of harassment and stalking carried a rela-
tively modest penalty. Nevertheless, the court referred Shawn for a
mental health evaluation, requesting an evaluation of his mental state
and violence risk as well as recommendations for what treatment, if any,
was necessary.
Psychological Testing in Violence Risk Assessment ❘ 61
Evaluating Violence Risk

As is evident from the preceding case example, not all violence risk as-
sessments occur in “typical” mental health or criminal justice settings.
In fact, some of the most challenging evaluations concern atypical situ-
ations or individuals. Because certain measures may have been shown
to be more adequate for particular populations, context plays a critical
role in determining what psychological instruments may be useful in
violence risk assessment.
Clinicians engage in violence risk assessments both in civil and
criminal settings. Risk assessment in a civil context generally focuses on
whether an individual should be placed in a psychiatric institution or
forcibly medicated over his or her objection. Questions about violence
risk in criminal settings, on the other hand, typically arise in the context
of parole and sentencing decisions, where risk of future violence and
criminal behavior are important considerations. In general, violence
risk assessment in a criminal context is concerned with longer-term
risks rather than the acute risk of violence necessary for involuntary
psychiatric hospitalization, but this dichotomy is not always clear-cut
(e.g., sex offender commitment hearings, although civil in nature, arise
after a criminal conviction and involve long-term violence risk assess-
ments). In short, a critical issue in evaluating violence risk is determin-
ing the time frame of concern (short-term versus long-term predictions).
Once the context and nature of the risk assessment question are
clear, decisions can be made as to which procedures and instruments
might be useful. Some of these instruments are useful because they in-
clude elements of a structured or actuarial risk assessment (e.g., psycho-
pathy), whereas others may directly address critical elements of the
individual’s clinical or risk profile (e.g., psychosis or substance abuse).
Psychopathy
Psychopathy is probably the most well established and widely known
risk factor for future violence, cutting across most contexts and popula-
tions. The term psychopathy has been used to describe the subset of indi-
viduals who engage in violent and criminal behavior, show no remorse
for their actions or empathy for their victims, and yet maintain a super-
ficial veneer of sociability and poise. A thorough review of the psych-
opathy literature could easily fill several volumes. For the purpose of
violence risk assessment, it is probably sufficient to note that this con-
struct comprises a central element of both clinical and actuarial (as well
as structured professional judgment) approaches to risk assessment.
The central “measure” of psychopathy for the past two decades has
been Hare’s Psychopathy Checklist (PCL) and its subsequent revision
(PCL-R; Hare 1991). The PCL was developed to measure the core con-
cepts of psychopathy and has been extensively validated in forensic
mental health research. Not only is the PCL-R considered by many to be
a critical element of any violence risk assessment, it is also incorporated
into many of the empirically supported actuarial risk assessment mea-
sures (e.g., the HCR-20 [Historical, Clinical, and Risk Management] and
Violence Risk Appraisal Guide). However, the PCL-R is not a psycho-
logical “test,” per se, but rather a clinical rating scale. The measure con-
sists of 20 items that are evaluated by a trained clinician on the basis of
clinical interview, official records, and third-party information. Each
item is scored on a scale of 0–2, with 0 indicating that the item does not
apply, 1 indicating that the item applies “somewhat,” and 2 indicating
that the item clearly applies to the individual. Scores are typically eval-
uated against published normative data, either by identifying the indi-
vidual’s level of psychopathy (e.g., percentile relative to typical crimi-
nal offenders or forensic psychiatric patients) or by identifying those
individuals who score above the threshold used to identify “psycho-
pathy” (i.e., greater than 30). Although considerable research has fo-
cused on elucidating a factor structure for this measure, most clinicians
rely on the total score for evaluating violence risk.
However, despite its importance in violence risk assessment, the
PCL-R has a number of limitations that are rarely acknowledged by
mental health evaluators. For example, although research studies typi-
cally demonstrate a high degree of interrater reliability across different
clinicians, individual evaluators may not have received comparable
training. Thus, the accuracy of PCL-R ratings will vary considerably
across clinicians, and contradictory PCL estimates are not uncommon.
In addition, mental health evaluators may be confused about the
different versions of the PCL that exist, at times using measures incor-
rectly. For example, Hare and Hart developed a screening version of the
PCL for use in research settings where identification of “probable” psy-
chopaths may be useful (Hart et al. 1995). Yet many clinicians utilize
this briefer instrument in clinical evaluations, seemingly unaware of the
different purpose and validation data that pertain to the screening ver-
sion (e.g., a higher rate of “false positives”: incorrect classifications of
nonpsychopathic offenders as psychopathic). Likewise, the PCL has
been adapted for adolescents (the PCL-YV, or Youth Version; Forth et al.
2003), although the validity of this measure for violence risk assessment
is far less convincing than for the PCL-R.
The utility of the PCL for evaluating violence risk in women is also
far less clear than its utility in men, because the construct of psycho-
pathy has been less often explored in women offenders. The few studies
that have examined the use of the PCL-R in women have failed to clar-
ify whether this instrument is equally useful in women. Finally, cross-
cultural research on the construct and measurement of psychopathy is
still in a relatively nascent stage. Although a growing number of studies
have supported the utility of the PCL in other countries and settings,
and considerable research has analyzed the cross-racial validity of the
PCL in North American offenders (e.g., comparing Caucasian and Afri-
can American samples), this research has largely been restricted to
Western, highly developed countries (e.g., Sweden, the Netherlands,
the United Kingdom). In fact, little evidence exists to support (or con-
tradict) the utility of the PCL as an aid for evaluating violence risk in
non-Western cultures. Hence, considerable caution is warranted when
utilizing this measure in violence risk assessment with individuals who
fall outside the primary validation sample (North American males in
prison). In the vignette described earlier, for example, the PCL-R would
likely lead to an unrealistically low estimate of violence risk, largely
because of Shawn’s high-achieving background and lack of childhood
behavior problems or other serious antisocial behaviors.
General Personality Assessment

Although psychopathy may be one of the most powerful violence risk
factors, the assessment of an individual’s violence risk often requires a
broader, more comprehensive approach to personality assessment. The
most common approach to personality assessment in forensic settings
is the use of broad-based self-report inventories such as the Minnesota
Multiphasic Personality Inventory–2 (MMPI-2; Butcher et al. 1989), the
Personality Assessment Inventory (PAI; Morey 1991), and the Millon
Clinical Multiaxial Inventory–3 (MCMI-III; Millon 1994). As is dis-
cussed later, not only can these measures provide important informa-
tion along a number of different personality dimensions but also, and
perhaps more importantly, they can inform the evaluator about the in-
dividual’s approach to testing (i.e., the presence of defensiveness or
symptom exaggeration).
The latter element of objective personality measures—the assess-
ment of test-taking style—is a critical component of psychological eval-
uations of violence risk. Because many individuals are motivated to
mask the true extent of their psychological difficulties, the mental health
evaluator engaged in violence risk assessment must consider the possi-
bility that the individual being evaluated has not been honest in report-
ing his or her history, symptoms, behaviors, or thoughts. Hence, a thor-
ough assessment of test-taking style is typically critical to any violence
risk assessment, particularly when the individual’s self-report will make
up a significant component of the evaluation (which is often, but not
necessarily, the case). Although several instruments (e.g., MMPI-2,
MCMI-III, PAI) include scales that assess defensiveness, there is little
dispute that the MMPI-2 is the best-validated measure for evaluating
test-taking style. Of course, the presence of defensiveness does not cor-
respond to an elevated risk of violence, but it does suggest that the indi-
vidual’s denial, lack of insight or awareness, or deliberate minimization
of symptoms likely results in inaccurate (unrealistic) test results and self-
report. It is worth noting that although these instruments also include
multiple scales and indices to evaluate symptom exaggeration (malin-
gering), this pattern is rarely observed in the evaluation of violence risk
because little motive exists for exaggerating the severity of one’s symp-
toms. Furthermore, for the sake of brevity, the discussion that follows fo-
cuses primarily on the MMPI-2, with the acknowledgement that similar
scales and interpretations may be available from other measures.
Assuming that the individual has responded honestly to the MMPI-
2, interpretation of the Clinical and Supplementary scales of the instru-
ment can be helpful in identifying general aggressive tendencies. It
should, of course, be noted that the presence of a “profile” that is linked
to an elevated risk of violence is by no means sufficient to conclude that
an individual is at high risk for violence. Nevertheless, a personality
style associated with an elevated risk of violence may provide one indi-
cator of possible violence risk. Likewise, evidence of a psychotic disor-
der, which in itself may not correspond to an elevated risk of violence,
could provide important data in cases where the individual’s history or
clinical presentation suggests delusional beliefs of a threatening or
paranoid nature, as in the vignette previously described.
The scenario of paranoid delusions notwithstanding, the MMPI-2
profile most commonly associated with violence is the 3–4/4–3 profile.
This profile has been described as indicative of poorly controlled anger
and hostility, although more recent research on these scales and this
code type has been equivocal. In addition, the Overcontrolled Hostility
Scale, which was developed to identify individuals who engaged in se-
vere acts of violence that seemed disproportionate to the provocation,
has also been used to explain seemingly extreme violent incidents by in-
dividuals who had little prior history of violence, although most au-
thors caution against using this scale as a predictive measure because it
was developed using retrospective analyses (Greene 2000).
A number of other MMPI-2 subscales exist that are both less contro-
versial and less widely studied than the Overcontrolled Hostility Scale.
For example, a number of subscales have been developed specifically
for the Psychopathic Deviance Scale (Scale 4) that break down eleva-
tions on this clinical scale into smaller, content-based subdivisions. The
five subscales developed by Harris and Lingoes are Familial Discord,
Authority Conflict, Social Imperturbability, Social Alienation, and Self
Alienation (Harris R, Lingoes J: Subscales for the Minnesota Multipha-
sic Personality Inventory. San Francisco, CA, The Langley Porter Clinic,
unpublished manuscript, 1955). Although some of these subscales (e.g.,
Social and Self Alienation) are less intuitively relevant to violence risk,
Osberg and Poland (2001) found modest correlations (r= 0.36) between
Authority Conflict and Self Alienation and future criminal behavior
among a sample of inmates who were eventually released. In fact, these
subscales provided considerable incremental validity beyond the larger
scale. Thus, although relatively little research has applied these sub-
scales to violence risk assessment, particularly in prospective studies,
they may provide a useful element of MMPI-2 interpretation in risk as-
sessment settings.
The PAI, another widely used, multi-scale, objective personality in-
ventory, has also been used in violence risk assessment. The PAI not
only includes several scales intended to measure aggressive tendencies
(Aggressive Attitude, Verbal Aggression, and Physical Aggression), but
also includes a summary index intended to quantify violence potential.
Unfortunately, despite this scale’s potential, relatively little research has
supported its use as a risk assessment tool (Morey and Quigley 2002).
Likewise, the PAI Correctional Report includes an “Institutional Risk
Circumplex” intended to help evaluate the offender’s likelihood of en-
gaging in violent behavior within a prison setting, although at present
this index too has little empirical support for its utility (Edens and Ruiz
2005).
Finally, the MCMI-III includes subscales for Antisocial Personality
Disorder along with scales tapping Sadistic and Negativistic (opposi-
tional) personality traits (Millon 1994). Although elevations on these
scales, either in isolation or in combination with other scales, are often
interpreted as indicating aggressive tendencies, relatively little research
has addressed—let alone supported—these interpretations. In short, al-
though multi-scale inventories such as the MMPI-2, PAI, and MCMI-III
have considerable potential and are frequently utilized in violence risk
assessment, empirical support for many of the indices and interpreta-
tions is extremely limited, and the need for caution in making any con-
clusions regarding violence risk cannot be overemphasized.
What About the Rorschach?

Despite growing questions about the reliability and validity of the Ror-
schach, even using the widely taught Exner scoring system (Lillienfeld
et al. 2000), many mental health clinicians continue to use or consider
using the Rorschach test in violence risk assessment (Archer et al. 2006).
Although it might be tempting to declare unequivocally that this prac-
tice is unwarranted, there may be particular settings and individuals for
whom this measure provides useful information. As noted earlier, de-
fensiveness and denial of psychological difficulties are particularly
common phenomena in violence risk assessment, and thus psychologi-
cal tests that might be less susceptible to that defensiveness are clearly
desirable. However, the potential and reality are often far apart when
analyzing the utility of projective tests such as the Rorschach.
A number of researchers have attempted to identify scores or deter-
minants on projective testing that might indicate an elevated risk of vi-
olence (Gacano and Meloy 1994; Greco and Cornell 1992), although this
limited literature has yet to result in any reliable and valid indices that
consistently correspond to a heightened risk of violence. For example,
Gacano and Meloy (1994) described a series of “supplemental aggres-
sion ratings” for the Rorschach that have been used in several studies,
with occasional significant associations with a history of violence or sa-
distic personality traits. However, support for these indices has been ex-
tremely limited and largely inconsistent. Researchers have occasionally
reported significant associations with one or more of these variables,
but research has yet to demonstrate a consistent pattern of findings sug-
gestive of clear concurrent—let alone predictive—validity.
However, the failure to identify Rorschach indices that correspond
to heightened risk of violence does not necessarily render the measure
clinically useless. In fact, even ardent critics of the Rorschach acknowl-
edge the potential utility of this measure for identifying psychotic
thinking. Particularly in cases in which the risk for future violence ap-
pears intertwined with psychotic symptoms (e.g., delusional beliefs of
a threatening nature), the Rorschach may help the evaluator identify the
presence of psychosis even in the context of an individual’s defensive-
ness or denial of symptoms. Of course, identifying indicators of psycho-
sis does not necessarily—or even usually—correspond to heightened
risk of violence; however, it may provide useful data to inform the eval-
uating clinician. Thus, in these limited circumstances, the Rorschach
test may provide incremental utility over other, more frequently uti-
lized assessment techniques. In fact, in the vignette described earlier,
the Rorschach might represent a useful technique for evaluating the ex-
tent to which Shawn’s jealousy and seemingly irrational behavior rep-

resent the emergence of a psychotic disorder rather than more chronic
aggressive tendencies (i.e., psychopathy).
Measures That Address Specific Personality

Characteristics
In addition to broad constructs (such as psychopathy) and broad mea-
sures of personality functioning (such as the MMPI-2 and PAI), a num-
ber of specific personality characteristics have potential relevance in
violence risk assessment, including anger, hostility, impulsivity, and
aggression. Although multiple measures exist for each of these char-
acteristics (some of which are briefly described later), their utility is
clearly contingent on honest responses, because self-report measures
are particularly vulnerable to distorted responses. Although some eval-
uators may rely on the measures described, considerable caution must
be exercised unless convincing evidence exists to support the accuracy
and honesty of the respondent. Hence, these measures are much more
frequently utilized in risk assessment research than in clinical practice.
Moreover, the research literature addressing predictors of violence has
generated conflicting findings regarding the salience of personality
traits that may predispose an individual to violence.
One of the personality characteristics that has received considerable
attention as a potential risk factor for violence is anger. Several measures
of anger exist, including the Novaco Anger Scale (Novaco 1994), the
State/Trait Anger Expression Inventory (Spielberger 1988), and the An-
ger subscale of the Aggression Questionnaire (Buss and Perry 1992).
Most of these measures conceptualize anger as a personality trait, with
the assumption that individuals who have either excessive anger or dif-
ficulty modulating the expression of their anger will be more prone to
violent behaviors. For example, the Novaco Anger Scale was designed
to evaluate responses to situations that are intended to provoke anger.
Initial research demonstrated strong reliability—both internal consis-
tency and test-retest (critical to the conceptualization of anger as a char-
acter trait)—as well as strong associations with violent behavior across
a number of different populations (e.g., incarcerated felons, domestic
violence offenders, and psychiatric patients). In fact, the MacArthur
Risk Assessment Study of violence committed by psychiatric patients
during and after release into the community found strong support for
the Novaco scale as a predictor of future violence (Monahan et al. 2001).
Other measures of anger, however, have had far less empirical support,
typically being used only in correlational or retrospective studies with
little evidence to support their predictive validity. Nevertheless, the

findings from the MacArthur study provide some support for the hy-
pothesis that elevated levels of anger are associated with an increased
risk of violence.
A closely related construct, hostility, has received far less support. In
fact, measures of hostility (e.g., the Cook-Medley Hostility Inventory,
the Hostility subscale of the Aggression Questionnaire) have rarely
been used in prospective research (Buss and Perry 1992; Cook and
Medley 1954). Indeed, the distinction between hostility and anger is
not altogether clear, and considerable overlap exists between the two
constructs. Some theorists conceptualize anger as a largely cognitive
process (and therefore one that can be directed at inanimate objects as
well as humans), whereas they see hostility as more interpersonal in na-
ture, but correlations between measures of anger and hostility are usu-
ally quite high (e.g., greater than 0.7). Thus, the distinction between
these constructs for the purposes of violence risk assessment may be
largely semantic.
Distinguishing the construct of aggression from anger and hostility
is equally complex and is further complicated by the lack of a singular
accepted definition of aggression. In fact, many theorists have de-
scribed aggression in purely behavioral terms, essentially relegating
this term to a milder version of violence. For example, Leonard Berkow-
itz (1993), an eminent social psychologist, described aggression as “any
form of behavior that is intended to injure someone physically or psy-
chologically” (p. 3). Thus, measures of aggression such as the Aggres-
sion Questionnaire typically elicit information regarding the frequency
of verbal and physical aggression (e.g., “If somebody hits me, I hit
back”). Although elevations on this scale are often associated, for obvi-
ous reasons, with a history of violence, the utility of this scale for iden-
tifying violence potential among individuals without a clear history is
less apparent.
A final personality characteristic that has received growing em-
pirical support as a risk factor for future violence is impulsivity. Al-
though frequently studied by neurologists and biologists, impulsivity
has only occasionally been studied as a risk factor for human violence
(and it is one of the items on the PCL-R). In fact, Barratt (1994) proposed
that impulsive aggression may often be due to an underlying biological
predisposition he termed behavioral disinhibition. Barratt conceptualized
impulsivity as comprising three factors—motor, cognitive, and non-
planning—and he developed the Barratt Impulsivity Scale (now in its
eleventh revision) to measure these three dimensions of impulsivity. Al-
though the Barratt scale was not intended to aid in violence risk assess-
ment, the MacArthur Risk Assessment Study found strong evidence for
the predictive validity of the measure in identifying psychiatric patients
at elevated risk of future violence. Thus, although still relatively under-
studied, this measure of impulsivity may have some utility in violence
risk assessment.
Other Measures of Potential Relevance in

The importance of substance abuse in violence risk assessment, whether
in isolation or in conjunction with a major mental disorder, cannot be
overstated. Although considerable variability across drugs and drug us-
ers exists, there is little dispute that active substance abuse represents a
significant risk factor for future violence. Thus, measures of substance
abuse such as the Substance Abuse Subtle Screening Inventory (SASSI,
now in its third revision; Miller et al. 1988) have obvious appeal as a
supplement to any violence risk assessment. Not only does the SASSI
help identify individuals with both current and potential risk of sub-
stance abuse, but this measure also includes a scale intended to measure
defensive responding. Although research supporting the SASSI has not
addressed its utility in violence risk assessment, it nevertheless may
help address a potentially important risk factor. For example, in the case
vignette presented at the beginning of this chapter, the evaluation of
substance abuse as a potential contributing factor, perhaps in conjunc-
tion with the evidence of an emerging psychotic disorder, not only
might help improve the assessment of Shawn’s potential for future vio-
lence, but also may provide invaluable information for developing a
risk management strategy to help reduce the risk of future violence.
As noted earlier, defensiveness is a paramount concern in violence
risk assessment. Although scales to assess defensiveness are embedded
within a number of broader measures, clinicians who are not inclined to
use these measures, whether due to time constraints or other consider-
ations, may find it helpful to use a measure specifically intended to
evaluate defensiveness. The most widely used and well validated mea-
sure of defensiveness is the Paulhus Deception Scales (PDS, formerly
called the Balanced Inventory of Desirable Responding) (Paulhus 1998).
The PDS is a 40-item self-report measure designed to assess two types
of socially desirable responding: impression management and self-
deception. Numerous studies have demonstrated strong reliability and
concurrent validity of this measure, with strong correlations with other
measures of defensiveness. Thus, high scores on this scale may help
identify individuals for whom self-report may be unreliable and may
even facilitate distinguishing between individuals who are deliberately

attempting to distort their presentation (the Impression Management
subscale) and those who lack insight and rely excessively on denial (the
Self-Deceptive Enhancement subscale). However, little research has uti-
lized the PDS prospectively in violence risk assessment, and thus ques-
tions arise as to the incremental validity of this scale in the assessment
process.
Assessing the Risk of Sexual Violence

Sexual offenders are a unique population of offenders that warrant ad-
ditional consideration during the risk assessment process. Not only are
most sexual offenses violent in nature, but the risk factors associated
with sexual offending (or re-offending) are often quite different from
those associated with nonsexual violence (Hanson and Bussiere 1998).
Particularly given the emergence of Sexually Violent Predator (SVP)
commitment statutes, the need for sexual violence risk assessments has
consistently grown in recent decades. Although the emergence of SVP
statutes has resulted in the development of numerous actuarial risk as-
sessment instruments specifically designed for sex offender evalua-
tions, structured professional judgment instruments are also widely
used. However, both approaches rely heavily on static risk factors (e.g.,
nature of prior sexual offenses, age of victim), leaving clinicians with lit-
tle ability to differentiate among large groups of offenders or, more im-
portantly, to monitor change or improvement. Dynamic risk factors,
which are better represented in structured professional judgment in-
struments, include the presence of deviant sexual urges, attitudes toler-
ant of sexual assault, and cognitive distortions (Craig et al. 2005). Iden-
tifying methods for evaluating these risk factors represents a critical
step in the clinician’s risk assessment process.
One particularly controversial technique to aid in evaluating sex of-
fenders is the penile plethysmograph (PPG), a method of quantifying
physiological arousal to deviant sexual imagery (Murphy et al. 1991).
Despite its invasive nature and the discomfort most mental health eval-
uators experience at the thought of utilizing this technique, the PPG is
considered to be the most well established laboratory method of assess-
ing sexual deviance and represents one of the strongest predictors of
sexual offense recidivism, typically surpassing actuarial risk assess-
ment techniques (Hanson and Bussiere 1998; Laws et al. 2000). In gen-
eral, the PPG is used with either visual or auditory stimuli of various
types of sexual content (e.g., violent, consensual, or nonconsensual sex),
which are presented to the individual in order to determine relative
arousal patterns. However, despite its frequent use in sex offender treat-
ment settings, this technique has a number of significant limitations, in-
cluding the relatively modest research literature supporting its predic-
tive validity. Questions also exist regarding the potential impact of
anxiety, sexual abstinence, and conscious repression of arousal on the
relative arousal ratio on the PPG. Furthermore, clinicians are likely to
encounter privacy and ethical issues in using this highly intrusive in-
strument, particularly when the risk assessment has been mandated
and true consent is not obtained.
In response to the limitations of the PPG, Gene Abel, one of the lead-
ing researchers on treatment of sexual offenders, developed an alterna-
tive approach to evaluating deviant sexual arousal, the Abel Assess-
ment of Sexual Interest (AASI; Abel et al. 1998). The AASI, designed to
be a less invasive measure than the PPG, is composed of three parts: the
individual’s subjective rating of sexual interest, a measure of Visual Re-
action Time (VRT) in response to sexual stimuli, and a questionnaire de-
signed to elicit attitudes and cognitions thought to correspond to sexual
offending. However, the key component of the AASI is the measure of
VRT, which records the amount of time the individual examines visual
stimuli presented on the computer screen. This measure hinges on the
assumption that sexual offenders who spend relatively more time view-
ing images that depict sexually deviant material are likely to have
greater arousal to these stimuli and are therefore more likely to re-
offend than individuals who spend relatively little time viewing sexu-
ally deviant images. In addition, by comparing viewing patterns for
multiple possible types of paraphilic interest, the AASI may help differ-
entiate the specific nature of an individual’s deviant arousal, helping
guide treatment and risk management strategies.
However, relatively little research has examined either the AASI in
general or the VRT in particular. Abel et al. (2004) found some support
for the utility of the VRT in identifying child molesters, but virtually no
research has addressed the validity of this measure in other populations
of sex offenders. Moreover, the effectiveness of the VRT in differentiat-
ing different subgroups of sex offenders (e.g., pedophiles versus adult
rapists) or identifying offenders with a high likelihood of re-offending
is not clear. As discussed earlier, the self-report sections of the AASI are
also clearly vulnerable to the biased responding that occurs so often in
violence risk assessment (and even more so among sex offender evalu-
ations, given the negative ramifications of admitting deviant sexual
arousal). However, despite the limited research on its validity, the AASI
holds some promise as a potential aid to violence risk assessment for
sexual offenders. That said, at present it is clear that the AASI does not
yet have the requisite empirical support necessary to be considered a

measure of risk of future sexual violence.
Conclusion
The importance of accurate violence risk assessment cannot be over-
stated, because the potential ramifications of errors can be devastating.
Fortunately, a growing number of psychological tests and techniques
can assist the clinician performing such evaluations. In the case example
presented earlier, psychological testing might help to illuminate many
potentially critical violence risk factors such as psychosis, impulsivity,
and anger management issues and substance abuse. For example, a Ror-
schach Inkblot test might be useful to evaluate the possibility of an un-
derlying psychosis that might account for Shawn’s jealousy. In addition,
personality tests such as the MMPI and SASSI could provide useful in-
formation regarding the extent of Shawn’s substance abuse that might
help explain his increasingly problematic behavior (which, given his
status as a medical student, likely reflects a significant change from his
baseline) and provide an alternative avenue for intervention. In addi-
tion, such measures would provide information regarding the validity
of Shawn’s self-report through an evaluation of his response style (i.e.,
the presence or absence of defensiveness). Finally, evaluating impulsiv-
ity and anger through either self-report (e.g., the Barratt or Novaco
scales) or clinical ratings (as elicited by the PCL) would help clarify the
extent to which Shawn’s recent behavior actually reflects a change in,
rather than a continuation of, long-standing personality characteristics.
A finding of changed behavior might support the hypothesis that
Shawn’s behavior is the result of an emerging mental disorder that is
potentially (likely) treatable, with specific treatment recommendations
being guided by the nature of the disorder. A determination that
Shawn’s behavior is a manifestation of long-standing personality char-
acteristics would highlight the need to more closely examine his past
behavior as an indication of his current violence risk (which may be less
relevant in the context of an emerging mental disorder).
This chapter outlines a number of different psychological assess-
ment techniques that may be useful supplements to a clinical evalua-
tion of violence risk. Whether the clinician uses an actuarial approach,
a structured professional judgment approach, or an unstructured clini-
cal evaluation, accurate determination of violence risk requires careful
assessment of the risk factors that have been identified in the research
literature. Although the scope and comprehensiveness of the measures
described in this chapter vary widely—as does their validity in risk as-
sessment settings—mental health evaluators have multiple options

available to them beyond the simple actuarial checklists often utilized.
Thorough risk assessment and risk management requires a comprehen-
sive assessment of a wide range of behaviors and cognitions in order to
minimize the risk of errors, both of omission or commission.
Key Points
■ Psychological testing can be a valuable supplement to any risk
assessment approach.
■ Evaluating defensiveness is critical, particularly when self-report
is relied on.
■ Assessing underlying psychosis, sexual deviance, or substance
abuse can help identify important risk-enhancing symptoms.
■ Formal assessment of psychopathy can bolster any risk assess-
ment method.
■ Limitations exist regarding validity of assessment in populations
other than Western, English-speaking adults.
References
Abel GG, Huffman J, Warberg B, et al: Visual reaction time and plethysmogra-
phy as measures of sexual interest in child molesters. Sex Abuse 10:81–95,
1998
Abel GG, Jordan A, Rouleau JL, et al: Use of visual reaction time to assess male
adolescents who molest children. Sex Abuse 16:255–265, 2004
Archer RP, Buffington-Vollum JK, Stredny RV, et al: A survey of psychological
test use patterns among forensic psychologists. J Personal Assess 87:84–94,
2006
Barratt ES: Impulsiveness and anger, in Violence and Mental Disorders: Devel-
opments in Risk Assessment. Edited by Monahan J, Steadman HJ. Chicago,
IL, University of Chicago Press, 1994
Berkowitz L: Aggression. New York, McGraw-Hill, 1993
Buss AH, Perry M: The Aggression Questionnaire. J Pers Soc Psychol 63:452–
459, 1992
Butcher JN, Dahlstrom WG, Graham JR, et al: Minnesota Multiphasic Personal-
ity Inventory-2: Manual for Administration and Scoring. Minneapolis, Uni-
versity of Minnesota Press, 1989
Cook SS, Medley DM: Proposed hostility and pharisaic-virtue scores for the
MMPI. J Appl Psychol 38:414–418, 1954
Craig LA, Browne KD, Stringer I, et al: Sexual recidivism: a review of static, dy-
namic and actuarial predictors. J Sex Aggress 11:65–84, 2005
Edens JF, Ruiz MA: Personality Assessment Inventory Interpretive Report for
Correctional Settings (PAI-CS). Odessa, FL, Psychological Assessment Re-
sources, 2005
Forth AE, Kosson DS, Hare RD: The Hare Psychopathy Checklist: Youth Ver-
sion. North Tonawanda, NY, Multi-Health Systems, 2003
Gacano CB, Meloy JR: The Rorschach Assessment of Aggressive and Psycho-
pathic Personalities. Mahwah, NJ, Lawrence Erlbaum, 1994
Greco C, Cornell D: Rorschach object relations of adolescents who committed
homicide. J Pers Assess 59:574–583, 1992
Greene RL: The MMPI-2: An Interpretive Manual, 2nd Edition. Boston, MA, Al-
lyn & Bacon, 2000
Hanson RK, Bussiere MT: Predicting relapse: a meta-analysis of sexual offender
recidivism studies. J Consult Clin Psychol 66:348–362, 1998
Hare RD: The Hare Psychopathy Checklist–Revised, 2nd Edition. North
Tonawanda, NY, Multi-Health Systems, 1991
Hart SD, Cox D, Hare RD: The Hare Psychopathy Checklist: Screening Version.
North Tonawanda, NY, Multi-Health Systems, 1995
Laws DR, Hanson RK, Osborn CA, et al: Classification of child molesters by ple-
thysmographic assessment of sexual arousal and a self-report measure of
sexual preference. J Interpers Violence 15:1297–1312, 2000
Lillienfeld SO, Wood JM, Garb HN: The scientific status of projective tech-
niques. Psychological Science in the Public Interest 1:27–66, 2000
Miller FG, Roberts J, Brooks MK, et al: SASSI-3: Substance Abuse Subtle Screen-
ing Inventory, 3rd Edition. Springville, IN, SASSI Institute, 1988
Millon T: The Millon Clinical Multiaxial Inventory–III Manual. Minneapolis,
MN, National Computer Systems, 1994
sity Press, 2001
Morey LC: Personality Assessment Inventory Professional Manual. Odessa, FL,
Psychological Assessment Resources, 1991
Morey LC, Quigley BD: The use of the Personality Assessment Inventory (PAI)
in assessing offenders. Int J Offender Ther Comp Criminol 46:333–349, 2002
Murphy WD, Haynes MR, Worley PJ: Assessment of adult sexual interest, in
Clinical Approaches to Sex Offenders and Their Victims. Edited by Hollin
CR, Howells H. Chichester, UK, Wiley, 1991
Novaco RW: Anger as a risk factor for violence among mentally disordered, in
Violence and Mental Disorders: Developments in Risk Assessment. Edited
by Monahan J, Steadman HJ. Chicago, IL, University of Chicago Press, 1994
Osberg TM, Poland DL: Validity of the MMPI-2 Basic and Harris-Lingoes sub-
scales in a forensic sample. J Clin Psychol 87:1369–1380, 2001
Paulhus DL: Paulhus Deception Scales. Toronto, ON, Canada, Multi-Health
Systems, 1998
Spielberger C: State-Trait Anger Expression Inventory Professional Manual.
Odessa, FL, Psychological Assessment Resources, 1988
P A R T I I
Mental Disorders and

Conditions
C H A P T E R 5
Mood Disorders
Mood disturbances and violence are intimately related in a variety of

ways. Violence may play a role in the development of subsequent mood
disorders and can exacerbate or perpetuate existing mood disorders.
Alternatively, violence may be a consequence or correlate of a mood
disturbance. In addition, several biological, social, and psychological
factors that are associated with violence and aggression frequently co-
exist with mood disturbances, so that co-occurring violence and mood
disturbances are frequently seen in individuals or systems. In these sit-
uations, violence can be directed toward self, others, or property.
Violence in the Genesis of Mood Disturbances

Exposure to violence is a major predictor of subsequent depressive symp-
toms. This has been repeatedly documented in a wide range of studies.
Passive Exposure
Witness to Domestic Violence
Early life experiences can affect both the biology and behavior of an in-
dividual. For example, abuse or neglect of young individuals will influ-
ence the development of mood disorders and problem behaviors later
77
in life. Miller (2005) reported that childhood abuse and exposure to

trauma may be linked to increased production and secretion of cortisol
and epinephrine, which have been linked to depression and anxiety. Re-
search also suggests that infants exposed to domestic violence between
their parents can exhibit signs of trauma and some behavioral prob-
lems, such as aggression (Bogat et al. 2006; Whitaker et al. 2006). How-
ever, Bogat et al. (2006) suggested that passive exposure to parental vi-
olence does not alter an infant’s temperament.
As children grow older, passive exposure to violence between their
parents can have more dramatic effects. As many as 10%–24% of children
may be exposed to intimate partner violence (IPV) between their parents
or to other family violence (Martin et al. 2006; Silverstein et al. 2006).
Martin et al. (2006) maintained that exposure to violence occurs prior to
age 11 in 80% of families with IPV. If community violence is included, the
rate of adolescents who have witnessed violence may be as high as 40%
(Hanson et al. 2006). Prevalence rates of depression and anxiety are in-
creased in adolescents and young adults (ages 13–21) who experience
passive exposure to IPV (Hindin and Gultiano 2006; Martin et al. 2006).
Young women may be at greater risk than young men (Hindin and Gul-
tiano 2006). In addition, Hazen et al. (2006) reported that problem behav-
iors increase in children ages 4–14 years who experience passive expo-
sure to IPV in the home. Behavioral problems span both internalizing
(e.g., depression, low self-esteem) and externalizing problems (e.g., ag-
gression, acting out) (Hazen et al. 2006). These effects are independent of
maternal depression (Hazen et al. 2006; Martin et al. 2006). This is an im-
portant observation, because maternal depression is associated with an
increase in adolescent depression and school dysfunction but not an in-
crease in problematic behavior (Peiponen et al. 2006; Silverstein et al.
2006). Sternberg et al. (2006) found that exposure to family violence in
older children, ages 10–16 years, also increased subsequent depression
and behavioral problems; this effect was greater for girls than boys.
Parental Substance Abuse

Parental substance abuse can result in both direct and indirect problems
for children. At the very least, children of substance-abusing parents are
neglected. However, more frequently, substance abuse is associated
with a variety of factors that independently or in combination can be
quite harmful. These include domestic violence and several forms of
abuse, including verbal, emotional, physical, and sexual abuse. Parental
substance abuse is associated with an increased risk of depression, ag-
gression, behavioral problems, and substance abuse in the children (Ed-
Mood Disorders ❘ 79
wards et al. 2006; Hanson et al. 2006; Peiponen et al. 2006; Sher et al.
2005; Whitaker et al. 2006).
Direct Abuse or Neglect

Childhood abuse and neglect are clearly associated with a substantial
increase in the risk for subsequent depression and maladaptive behav-
iors (Cukor and McGinn 2006; Reigstad et al. 2006; Widom et al. 2007).
This is true in all cultures in which it has been studied (Afifi 2006).
Verbal and Emotional Abuse

The experience of verbal abuse during childhood (e.g., “You are stu-
pid”) increases depression, anger, and hostility in young adults (Sachs-
Ericsson et al. 2006; Teicher et al. 2006). Verbal and emotional abuse
influence the development of self-concept and lead to a self-critical style
of cognitive processing that contributes to low self-esteem (Cukor and
McGinn 2006; Sachs-Ericsson et al. 2006). This impaired self-image may
be one of the underlying phenomena that increase the risk of subse-
quent sexual victimization as a young adult (Rich et al. 2005).
Physical Abuse
Physical abuse may be a major contributing factor in the development
of violence in later life (Huizinga et al. 2006). Physical abuse is also piv-
otal in the development of depression in youth and on into adulthood
(Cukor and McGinn 2006; Reigstad et al. 2006; Widom et al. 2007;
Wright et al. 2004). Physical abuse may occur in either the home envi-
ronment or in school. Bullying is a form of verbal and physical violence
that can have major impact on development. The odds of experiencing
social problems and depression with suicidal ideation and attempts are
3.9 times higher among victims of bullying compared with nonvictims
(Brunstein Klomek et al. 2007; Kim et al. 2006). Furthermore, bullying
behaviors have been linked to mood disturbances. The odds of bullies
developing social problems, depression, and suicidality are 1.8 times
higher compared with people who are not bullies, and bullies who are
also targets of other bullies are 4.9 times as likely to develop social prob-
lems (Brunstein Klomek et al. 2007; Kim et al. 2006). High-profile school
shooters such as those at Columbine High School or Virginia Tech had
been bullied by classmates.
Sexual Abuse
Sexual abuse of children is associated with a wide variety of physical
and psychological sequelae, many of which are lifelong. Early sexual
abuse is associated with a significant increase in depression in both

males and females (Conway et al. 2004; Gladstone et al. 2004; Martin et
al. 2004; Peleikis et al. 2005). The risk of subsequent suicide attempts is
15 times higher in boys who experience early sexual abuse compared
with nonabused boys (Martin et al. 2004); among women, suicide ide-
ation is 4.5 times higher (Masho et al. 2005). The consequences of child-
hood sexual abuse include greater severity of depressive illness in adult
patients over age 50 (Gamble et al. 2006; McGuigan and Middlemiss
2005). Adult women who have experienced childhood sexual abuse are
more likely to be victims of violence (Gladstone et al. 2004) and other
forms of trauma, including sexual assault (Banyard et al. 2002; Rich et
al. 2005). Sexual abuse perpetrated by adult women can be just as harm-
ful as sexual abuse perpetrated by men (Denov 2004).
Adult Assault
After personality development is complete, adult assaults (sexual or
physical) can increase the likelihood of the development of mood dis-
turbances (Johansen et al. 2006). Consequences of being a victim of as-
sault include depression, anxiety disorders, and substance abuse; these
can persist for decades (Acierno et al. 2007).
Intimate Partner Violence

Intimate partner violence, perhaps the most common type of violence in
our society, is pervasive throughout several socioeconomic classes and
ethnic groups. Thirty percent of African American women seeking
medical care in a large public hospital reported severe IPV (Paranjape
et al. 2007), and 54% of women attending a rural family practice clinic
reported IPV (Coker et al. 2005). Researchers have estimated that 10%–
24% of representative samples of children may be exposed to IPV (Mar-
tin et al. 2006; Silverstein et al. 2006). In addition, 13% of middle-class
women also have experienced IPV (Anderson et al. 2002). Exposure to
IPV is associated with a significant increase in the risk for both depres-
sion and posttraumatic stress disorder (PTSD; Avdibegovic and Si-
nanovic 2006; Bonomi et al. 2006; Houry et al. 2006; Lipsky et al. 2005;
Paranjape et al. 2007; Varma et al. 2006) as well as more medical prob-
lems, reduced functioning, and increased medical disability (Bonomi et
al. 2006; Coker et al. 2005). Depression risk is almost 6 times higher in
women who are victims of IPV compared with those who are not, and
PTSD is 9.4 times higher (Houry et al. 2006). Sexual IPV is specifically
associated with an increase in depression and suicide ideation (Pico-
Alfonso et al. 2006); Houry et al. (2006) have observed that suicidal idea-
tion in women who have experienced IPV is 17.5 times higher com-
pared with women who have not experienced IPV. However, depres-
sion frequently predates the episodes of IPV, and the presence of
depression in young women actually increases the likelihood of dating-
violence (Foshee et al. 2004; Rivera-Rivera et al. 2006). African Ameri-
can women may be at particular risk for mood disturbances due to high
rates of IPV; 18% also abuse alcohol, which can worsen prognosis
(Paranjape et al. 2007). Women in abusive relationships have a great
need for emotional support (Theran et al. 2006), and African American
women appear to obtain much support through spirituality and affilia-
tion with religious institutions (Mitchell et al. 2006; Watlington and
Murphy 2006).
Community Violence, War, and Terrorism

Although violence at a personal level is a major factor in the develop-
ment of mood disturbances, violence at the community level also
contributes to subsequent depression, suicidal ideation, and suicide at-
tempts. For example, community violence can increase the risk of
depressive symptoms in adolescents, particularly girls (Goldstein et al.
2007; Hammack et al. 2004).
Terrorists count on the psychological impact of indirect violence to
achieve their aims. After the September 11, 2001, attacks in New York
and Washington, D.C., and the March 11, 2004, attack in Madrid, Spain,
there was an increase in the prevalence rates of major depression (9.4%
prevalence in New York City and 8% in Madrid, compared with 6.4% in
population-based surveys [Kessler et al. 2006]) and, to a lesser degree,
PTSD (Miguel-Tobal et al. 2006; Person et al. 2006). This increase was also
associated with a 49% increase in suicide attempts along the East coast of
the United States after the September 11 attacks (Starkman 2006).
Gaylord (2006) has estimated that 10%–17% of combat veterans will
experience psychiatric problems, including PTSD and depression.
These disorders may last for long periods of time after the end of hos-
tilities (Fiedler et al. 2006). However, among civilians who are trapped
in war zones or are direct targets of attacks or abuse, rates of PTSD have
been estimated at almost 33%, and rates of depression are approxi-
mately 41% (Hashemian et al. 2006; Loncar et al. 2006).
Treatment Approaches
Pharmacological approaches to treatment are geared toward treating
the depression and PTSD that may be associated with past or current
exposure to violence. In addition, researchers have investigated the
effectiveness of various forms of psychotherapy in the treatment of

depression and PTSD resulting from exposure to violence; these include
supportive therapy, cognitive-behavioral therapy, and forgiveness ther-
apy (Deblinger et al. 2006; Reed and Enright 2006). Focused therapies
such as cognitive-behavioral or forgiveness therapy appear to be more
effective than unfocused supportive therapy (Deblinger et al. 2006;
Reed and Enright 2006). Forgiveness therapy has been central to na-
tional attempts at healing past abuse, such as the South African Truth
and Reconciliation Commission (Potter 2006). Adult treatment for
childhood abuse is effective in reducing symptoms and dysfunction
(Martsolf and Draucker 2005). The approach for women involved in
IPV depends on the timing of the abuse. Women in a current abusive re-
lationship benefit more from emotional support, whereas women with
past abuse require practical support (Theran et al. 2006).
Prevention is a critically important focus for those at risk for devel-
oping violence-related mood disturbances. Identification of children at
greatest risk due to violence or substance abuse in their families, and
provision of appropriate support to prevent depression, aggression,
substance abuse, and future victimization, would be the ideal approach
(Sternberg et al. 2006). Past abuse predicts future abuse; policymakers
can use this knowledge to direct appropriate resources toward preven-
tion of future abuse among those at risk.
Case Example 1
Ms. A is a 36-year-old woman who presented to the emergency psychi-
atric service with a complaint of worsening depression and suicidal ide-
ation. She had a previous psychiatric history of recurrent major depres-
sion since adolescence, PTSD, and prescription benzodiazepine abuse.
She was not currently in treatment but had been treated in the past with
psychotherapy and a variety of antidepressant, anxiolytic, and antipsy-
chotic medications. She had had three previous hospitalizations, all as-
sociated with suicide attempts. The current episode began when her
boyfriend, with whom she lived, had become more abusive. He had at-
tempted to strangle her and had then raped her after an argument about
his alcohol abuse. She reported a remote history of childhood emotional
abuse by an alcoholic father prior to her parents’ divorce and sexual mo-
lestation by her mother’s boyfriend when she was 13 years old.
Ms. A was admitted to a crisis stabilization unit for 10 days, started on
an antidepressant (sertraline, 100 mg/day), and engaged in supportive
and insight-oriented psychotherapy. After discharge she engaged in out-
patient psychotherapy and pharmacotherapy. She continued to complain
of anxiety, and due to her history of benzodiazepine abuse, gabapentin
was started and increased to 1,200 mg three times daily. She reconciled
with her abusive boyfriend after he stopped using alcohol. However, the
relationship remained tumultuous, and he relapsed into alcohol abuse

again soon after the reconciliation. As the chaos in the relationship
increased, the patient dropped out of treatment and was lost to follow-up.
This case illustrates many of the relationships among violence, vic-

timization, and mood disturbance. Early verbal and emotional abuse
led to Ms. A’s low self-esteem. Early sexual abuse contributed to her
role as a victim. These early experiences contributed to early-onset de-
pression and worsening of these symptoms in adulthood. Similarly,
they allowed Ms. A to be tolerant of her role as victim and allow herself
to return to, and remain within, an emotionally, physically, and sexually
abusive relationship. The severity of the ongoing depressive and anxi-
ety symptoms had previously led the patient to prescription benzodiaz-
epine abuse. Despite participation in both psychotherapy and pharma-
cotherapy, she returned to her previous maladaptive behaviors.
Summary
Aggression and depression are intimately related. Early exposure to ag-
gression, either as a witness or as a victim, increases the likelihood of
depression, the chronicity of depression, and the likelihood of becom-
ing a perpetrator of future aggression. Similarly, early abuse increases
the risk of subsequent victimization and chronic depression.
Mood Disorders in the Genesis of Violence

As noted earlier, violence and aggression are associated with subse-
quent onset of mood and anxiety symptoms as well as full depressive
and posttraumatic stress disorders. However, mood disorders have also
been identified as a precursor to the onset of aggression. The presence
of a mood disorder increases the likelihood of an individual’s being a
victim of violence (Brunstein Klomek et al. 2007; Lehrer et al. 2006) and
a perpetuator of violence (Brunstein Klomek et al. 2007).
Depression in adolescents is one of the major predictors of aggres-
sion, violence (a more extreme form of aggression) (Blitstein et al. 2005;
Teicher et al. 2006), or oppositional and delinquent behaviors (Rowe et
al. 2006). Major depressive disorder and bipolar disorder are both asso-
ciated with an increase in irritability, aggression, and potential violence
against others and self (Grunebaum et al. 2006; Knox et al. 2000; Najt et
al. 2007; North et al. 1994; Schuepbach et al. 2006). Bipolar illness, in par-
ticular, may be associated with aggression due to the nature of its core
symptoms of irritability, lability, grandiosity, and paranoia (Feldmann
2001; Swann 1999).
Patients with bipolar disorder who were admitted involuntarily to

an inpatient unit were more likely to have comorbid substance abuse
and up to three times more likely to be aggressive after admission (Bar-
low et al. 2000; Schuepbach et al. 2006). An analysis of 576 consecutive
admissions for mania suggested that acute mania may have four distinct
phenomenological subtypes: pure, aggressive, psychotic, and depres-
sive (mixed) mania (Sato et al. 2002). When a patient’s illness recurs, the
profile of symptoms, including aggression, remains relatively consistent
(Cassidy et al. 2002), supporting the clinical notion that there is a high
association between past and future violence. The increase in aggression
associated with mania is associated with an increase in legal problems.
Whereas patients with schizophrenia or schizoaffective illnesses are
more likely to be arrested (Grossman et al. 1995), patients with bipolar
disorder are more likely than those with unipolar depression to have le-
gal problems (Calabrese et al. 2006). At the time of their arrest, a large
number of bipolar subjects were manic (74.2% of the 66 subjects studied)
and/or psychotic (59%) (Quanbeck et al. 2004). Many of these patients
had already come to the attention of the healthcare system and had re-
cently been discharged from an inpatient unit but had not attended their
outpatient follow-up (Quanbeck et al. 2004). This may explain why bi-
polar subjects are overrepresented among sex offenders, with approxi-
mately 35% of sex offenders having a bipolar disorder (usually with
comorbid antisocial personality disorder or substance abuse) (Dunsieth
et al. 2004; McElroy et al. 1999).
However, aggression can also occur during depressive episodes. In
bipolar patients, aggression can be a relatively common presentation of
agitated depression (Maj et al. 2003). Aggression is also common in uni-
polar depression (Posternak and Zimmerman 2002). A syndrome of
high irritability and other hypomanic symptoms in unipolar depressed
patients has been labeled mixed depression (Sato et al. 2005) and may be
associated with significant aggression (Sato et al. 2005).
The effect of antidepressant medications in the treatment of aggres-
sion is unclear. Antidepressant treatment has been variously reported
to increase and decrease aggression (Bond 2005; Goedhard et al. 2006;
Healy et al. 2006; Mitchell 2005). If there is an anti-aggression effect of
antidepressants, it is weak (Goedhard et al. 2006). An increase in ag-
gression associated with antidepressant use may possibly occur exclu-
sively in individuals with bipolar disorder or occult bipolar disorder—
that is, those in whom an episode of mania has not yet occurred.
Antipsychotic medications or mood stabilizers are generally used to
treat aggression (Afaq et al. 2002; Barzman et al. 2006). Valproate is per-
haps one of the best studied agents and has been found to be superior
to other antiepileptics such as topiramate (Gobbi et al. 2006) or oxcarba-

zepine (MacMillan et al. 2006) A meta-analysis of controlled trials sug-
gests that the effect of these interventions is generally small (Goedhard
et al. 2006). Dopamine antagonist antipsychotic medications may be
minimally better than serotonin-dopamine antagonist medications
(Goedhard et al. 2006). Effective pharmacological treatment approaches
to reduce aggression and violence in those with mood disturbances are
greatly needed.
Case Example 2
Mr. B was a 34-year-old white man with a history of bipolar disorder. He
reported that he experienced frequent “mood swings.” He described
these as brief periods (minutes to hours) in which he quickly lost his
temper. During these periods, he could become aggressive toward
strangers or toward significant others to whom he was emotionally at-
tached. He reported frequent fights and previous arrests for assault. Mr.
B also had a history of significant alcohol and marijuana abuse, which
he minimized. He denied that his marijuana use was a problem and
stated that it helped him calm down. He also did not view his alcohol
abuse as a problem because he did not “drink every day” and main-
tained that he could quit “at any time.” On questioning, he reported ep-
isodes of reduced sleep, reduced need for sleep, increased irritability, in-
creased rapid thoughts and distractibility, and increased involvement in
multiple problematic behaviors. These periods lasted 4–5 days and oc-
curred three or four times annually. He also reported periods of depres-
sion that were generally brief, lasting only 2 weeks, during which he
also manifested irritability, loss of interest, loss of pleasure, low self-
esteem, increased frustration, and suicidal ideation. He reported that he
had used both alcohol and marijuana during these episodes. He had no
periods of significant sobriety. A tentative diagnosis of type II bipolar
disorder was made, along with intermittent explosive disorder and al-
cohol and marijuana abuse. Mr. B was offered treatment with dival-
proex and psychotherapy and was encouraged to attend treatment for
substance abuse. He never sought out substance abuse treatment and
never stopped using marijuana, but he had brief periods in which he
stopped using alcohol. These periods of abstinence from alcohol gener-
ally lasted less than 1 month. His compliance with divalproex and psy-
chotherapy was very poor. He continued to have periods of depression
and impulsive rage, and he continued to abuse alcohol and marijuana.
Two years after his initial presentation, Mr. B committed suicide with a
self-inflicted gunshot wound.
Mr. B’s case illustrates typical characteristics of a difficult patient

with comorbid mood disorder and substance abuse. Unlike Ms. A, Mr.
B had not experienced extreme adversity. His aggressive behavior was
common and usually directed at others. His depression was probably
made worse by his substance use. Marijuana users frequently believe

that marijuana use is calming and reduces aggression (Arendt et al.
2007); however, marijuana use is specifically associated with an increase
in violence (Maremmani et al. 2004). Ultimately, Mr. B directed the ag-
gression against himself.
Summary
Aggression is common in individuals who are experiencing an episode
of mania, and there may be a subtype of mania in which aggression is
the core feature. Aggression may also be common in unipolar major de-
pression. The effect of antidepressants on aggression is unclear. If there
is an anti-aggression effect of these agents, it is weak. Most commonly,
the mood-stabilizing antiepileptic medications and antipsychotic med-
ications are used to treat aggression, but their anti-aggression effect ap-
pears to be weak.
Mood Disorders and Violence Toward the Self

Deliberate Self-Harm
Violence can be directed toward the self as an act of nonsuicidal self-
harm. Tuisku et al. (2006) defined self-harm as “direct, socially unac-
cepted, repetitive behavior that causes minor to moderate physical in-
jury.” Deliberate self-harm (DSH) “appears to reflect an externalizing
response in an isolated individual who has commonly been exposed to
earlier deprivational experiences” (Parker et al. 2005). Risk factors for
DSH include early onset of mood symptoms, recent diagnosis, young
age, family history of suicide, and comorbid disorders, especially anxi-
ety disorders and substance abuse (particularly alcohol). It is important
to make the distinction between DSH and suicidal intent; whereas DSH
is self-directed injury without any suicidal intent and frequently with
the goal of reducing anxiety or dysphoria, suicide is a self-destructive
act with a specific intent to end one’s life. Despite this distinction, it re-
mains a fact that individuals with DSH also have a higher risk for sub-
sequent suicide (Groholt et al. 2000).
Deliberate Self-Harm and Depression

Although DSH is almost always associated with dysphoria, it is not
always associated with the syndrome of depression. DSH can occur in
the setting of depression associated with both bipolar disorder and uni-
polar major depressive disorder. Haw et al. (2001) looked at a cohort of
106 patients who presented to a hospital following an episode of DSH

and found that 92% of these patients had a psychiatric diagnosis and
that the most common diagnosis was affective disorder (72% using
ICD-10 criteria).
As stated earlier (see “Violence in the Genesis of Mood Distur-
bances”), early adverse life events have a major impact on subsequent
mood states. Similarly, early adversity is a major correlate of subse-
quent DSH behaviors (Gladstone et al. 2004; Parker et al. 2005). Glad-
stone et al. (2004) examined DSH behaviors, personality characteris-
tics, and childhood variables, including parental styles and childhood
sexual/physical abuse, among 125 women with depressive disorders.
Findings indicated that participants who were victims of childhood
sexual assaults were more likely to engage in DSH as adults (Gladstone
et al. 2004). In addition, respondents who were victims of childhood
sexual abuse became depressed earlier in life than nonabused control
subjects (Gladstone et al. 2004).
Adolescents with DSH generally have less severe depressive symp-
toms than individuals with suicidal ideation but more severe symp-
toms than those without any history of self-injurious ideation. In a com-
munity sample, adolescents who have a history of self-harm reported
more depressive symptoms than those without a self-harm history
(Muehlenkamp and Gutierrez 2004). In a study of 218 adolescents, ages
13–19, who were receiving outpatient treatment for a depressive mood
disorder, adolescents who had DSH behavior had less severe depres-
sive symptoms than those with suicidal ideation or suicide attempts
(Tuisku et al. 2006). Similarly, among adults, the degree of seriousness
of a self-injurious act was associated with depression and intent. In a
study of 49 prisoners in Germany, measures of depression and hope-
lessness were both highly correlated with suicidal intent and lethality;
less lethal methods were not correlated with depression (Lohner and
Konrad 2006). Impulsive acts of self-harm were rarely associated with
depression (Lohner and Konrad 2006).
DSH behaviors are not fixed over the lifetime. For example, 70% of
132 adolescents who had deliberately poisoned themselves and who
were followed up for 6 years stopped the self-harm behaviors within
3 years of the index event (Harrington et al. 2006). DSH continued into
adulthood mainly among those with psychiatric disorders. Only 56% of
these study participants had a psychiatric disorder, and the most com-
mon psychiatric diagnosis was depression (Harrington et al. 2006).
DSH behaviors may appear de novo in the elderly. Lamprecht et al.
(2005) looked at older people presenting to an acute hospital with an ep-
isode of DSH. Sexual distribution among males and females was equal.
Only 37% had a major depressive illness at the time of the DSH assess-
ment, but 21% of the males had no psychiatric diagnoses at the time of
the DSH (Lamprecht et al. 2005).
In young adults, the lack of depression in subjects with DSH has also
been noted. Among 1,986 high-functioning military recruits (62%
male), only 10% of those with a history of DSH reported depressive
symptoms on the Beck Depression Inventory (Klonsky et al. 2003).
Peers viewed self-harmers as having strange and intense emotions and
a heightened sensitivity to interpersonal rejection (Klonsky et al. 2003).
Given that DSH may not necessarily be associated with depression,
why does it occur? Tzemou and Birchwood (2007) examined dysfunc-
tional thinking patterns and intrusive memories in patients diagnosed
with both unipolar depressive and bipolar mood disorders. They re-
cruited 49 participants diagnosed with major depression or manic or
hypomanic episodes. Twenty healthy control subjects were also re-
cruited from the same areas in Central England. Compared with the
healthy controls, dysfunctional attitudes were abnormal in the mood-
disordered groups when ill (Tzemou and Birchwood 2007). Interest-
ingly, whereas dysfunctional attitudes resolved in bipolar subjects as
they became euthymic, they persisted into euthymia for those diag-
nosed with unipolar major depression (Tzemou and Birchwood 2007).
Deliberate Self-Harm and Bipolar Illness

Intentional self-harm in mania is rare and is probably related to the de-
pressed mood that can occur during manic episodes (Ostacher and Ei-
delman 2006). However, DSH is more common during bipolar depres-
sions than it is in unipolar depressive illness (Parker et al. 2005). Parker
et al. (2005) reported that across samples of depressed individuals, more
individuals with bipolar disorder tended to report DSH behaviors com-
pared with those with unipolar depression. Smith et al. (2005) examined
the prevalence rates of bipolar disorders and major depression among
87 young adults with recurrent depression; 83.9% of study respondents
met criteria for major depressive disorder, and 16.1% met criteria for a
DSM-IV-TR–defined (American Psychiatric Association 2000) bipolar
disorder. The authors reported that among the respondents diagnosed
with major depression, 45.7% had a history of DSH and 13.0% had a his-
tory of a previous suicide attempt. Of the 14 respondents diagnosed
with bipolar disorder, 71.4% had DSH and 28.6% had a history of delib-
erate self-harm.
One of the best-known occurrences of DSH was that performed by
the Dutch artist Vincent van Gogh (1853–1890), who had bipolar disor-
der (Jamison 1993). On Christmas Eve in 1888, van Gogh cut off his own
earlobe with a razor blade as he was apparently attempting to attack an
acquaintance. Following this episode of self-harm, van Gogh exhibited
alternating states of “madness and lucidity” and received treatment in
an asylum in Saint-Remy. Two months after his discharge from the asy-
lum, he committed suicide by shooting himself “for the good of all”
(Van Gogh Gallery 2007).
Mood Disorders and Suicide

Suicide, the act of ending one’s life, is the most dramatic form of self-
harm. Epidemiological research indicates that in 2004, 31,484 individu-
als in the United States died from suicide or self-inflicted injury (10.8
per 100,000 population; Centers for Disease Control and Prevention
2006). Extensive research has examined risk factors for suicide, and sev-
eral studies have identified a history of prior suicide attempts as a very
strong predictor of suicide risk (American Psychiatric Association 2003;
Borges et al. 2006; Gaynes et al. 2004). Certain sociodemographic char-
acteristics have also been associated with high suicide risk. These in-
clude male gender, European American ethnicity, and advanced age.
However, the National Comorbidity Survey Replication Study found
that low-income, “non-Hispanic Black” (p. 1750) ethnicity and age
younger than 45 years were significant correlates of suicide ideation
(Borges et al. 2006). Additional risk factors include the presence of a
psychiatric disorder (particularly depression), alcohol abuse, physical
and sexual abuse, and a family history of suicide (Gaynes et al. 2004).
Psychiatric disorders may be present in up to 90% of those who commit
suicide (American Psychiatric Association 2003). Divorced, separated,
or widowed individuals have a higher risk of suicide (American Psychi-
atric Association 2003). Conversely, high-conflict or violent marriages
may increase the risk for suicide among married individuals (American
Psychiatric Association 2003).
Unipolar Depression and Suicide

Numerous studies have identified depression as a significant risk factor
for suicide. This contributes to mortality rates associated with depres-
sion that are approximately 20 times higher than the general population
(American Psychiatric Association 2003). The fraction of people who
have committed suicide that were depressed at the time of their death
has been estimated to range from 15% (Rich et al. 1986) to 97.5% (Sinclair
et al. 2005). However, most studies, including those based on psycholog-
ical autopsies, estimate a rate of 30%–34% (Arato et al. 1988; Foster et al.
1999; Henriksson et al. 1993). The fraction of adolescent suicides that in-
volve depression may be slightly higher, at 43% (Brent et al. 1993).
Comorbid psychiatric conditions may additionally increase the risk
for suicide. Paramount among these is co-occurring substance use,
which accounts for some 45% of completed suicides (Rich et al. 1986).
Additionally, aggression (Dervic et al. 2006; Keilp et al. 2006) and Clus-
ter B personality disorders (Dervic et al. 2006) are associated with sui-
cide attempts in depressed individuals with a history of childhood sex-
ual abuse.
A decline in depression and hopelessness was associated with a de-
cline in suicidal ideation in 198 people diagnosed with major depres-
sion (Sokero et al. 2006). There is a close correlation between the in-
creased use of antidepressants and an observed decline in overall
suicide rate (Gibbons et al. 2006; Korkeila et al. 2007), but this trend may
have begun prior to the introduction of antidepressants (Safer and Zito
2007). Antidepressants may have no effect on suicide ideation (Ham-
mad et al. 2006) or may actually increase the risk of suicide attempts
among depressed adults (Tiihonen et al. 2006) and suicide ideation
among adolescents (Bridge et al. 2007; Dubicka et al. 2006), but they
may reduce completed suicides (Tiihonen et al. 2006). The U.S. Food
and Drug Administration has placed a warning on all antidepressants
that they may increase suicidal ideation in adolescents (Kuehn 2007).
Although lithium is rarely used in major depressive disorder, it appears
to have an antisuicide effect similar to that seen in bipolar illness (Guz-
zetta et al. 2007).
Bipolar Disorder and Suicide

Lifetime prevalence of all bipolar disorders is approximately 2%; bipo-
lar type I disorder has an incidence rate of 0.8% compared with 1.2% for
bipolar II disorder. Suicide risk is high in bipolar disorder. Angst and
Preisig (1995) followed up 406 patients for 36 years; their findings indi-
cated that 11% committed suicide regardless of whether they were di-
agnosed with type I or type II disorder. Other estimates approach 19%
(Ostacher and Eidelman 2006). This risk appears higher than in unipo-
lar major depression. Chen and Dilsaver (1996) examined data from the
Epidemiologic Catchment Area study to estimate lifetime rates of sui-
cide attempts in mood disorders; findings indicated that 29.2% of re-
spondents with bipolar disorder attempted suicide compared with
15.9% of those with unipolar depressive disorder.
Additionally, when subjects with bipolar disorder attempt suicide,
the lethality of that attempt may be greater. Among 2,395 hospital ad-
missions of patients with unipolar depression and bipolar disorder,

subjects with bipolar disorder had a higher incidence of more lethal sui-
cide attempts (Raja and Azzoni 2004). The odds of completed suicide in
those with bipolar disorder is 2.0 times higher compared with those
with unipolar depression (Raja and Azzoni 2004). However, prevalence
rates of suicide may be inflated, because researchers typically focus on
hospitalized patients and those who have received treatment from a
mental health provider. This self-selected population may be more ill
compared with those who receive treatment from primary care provid-
ers or those who do not receive any psychiatric treatment.
Risk for suicide is highest during a depressive episode of bipolar
disorder. Isometsa et al. (1994) found that among patients diagnosed
with bipolar disorder, 80% of completed suicides occurred during a de-
pressive episode. Mortality from suicide in persons with bipolar de-
pression may be 30 times that of normal control subjects (Ostacher and
Eidelman 2006). However, suicidal ideation and suicide completions
may occur during the mixed (Dilsaver et al. 1994) or even manic phase
(Cassidy and Carroll 2001). Rapid cycling also carries a higher likeli-
hood for more serious suicide attempts but not an increase in completed
suicides compared with other types of episodes (42% vs. 27%; MacKin-
non et al. 2003).
Suicide risk is highest in newly diagnosed bipolar patients. Fagiolini
et al. (2004) found that among 104 patients with bipolar disorder, 50%
attempted suicide within 7.5 years of the initial onset of the illness (ei-
ther mania or depression). In these young bipolar patients, suicide
rarely occurs during episodes of mania.
Lithium appears to have a clear effect on reducing completed suicide
in bipolar patients, with a fivefold reduction in relative risk (Baldes-
sarini et al. 2006; Tondo et al. 2001). More impressively, lithium reduces
nonsuicidal DSH and nonpsychiatric mortality in bipolar patients (Ci-
priani et al. 2005).
Case Example 3
Mr. C was a 66-year-old married white man with a lifelong history of al-
cohol abuse and depression. He had been treated intermittently with an-
tidepressants but never persisted in psychiatric treatment for more than
3 months. He had had a chaotic relationship with his wife of 44 years
and had been verbally and physically abusive. For unknown reasons, in
the setting of ongoing alcohol abuse, he shot and killed his wife and
then killed himself.
Suicide and homicide are ultimately two sides of the same coin.
Individuals who commit suicide are more likely to have a history of
violence. In enmeshed relationships, homicide-suicide is frequently

seen as the only solution.
Summary
DSH behaviors frequently occur independently of suicide and should
be considered separate phenomena. However, individuals who engage
in DSH are at a higher risk of subsequent suicide. Subjects who engage
in DSH behavior have generally experienced early abuse. Both major
depression and bipolar illness are significant risk factors for completed
suicides. Bipolar disorder carries a much higher risk for completed sui-
cide than does major depression. Comorbid disorders, particularly sub-
stance abuse, increase the suicide risk. Antidepressants may increase
suicide ideation but do not appear to increase completed suicides and
may reduce the severity of suicide attempts. Lithium has a clear antisui-
cide effect in both bipolar illness and unipolar major depression.
The Biology of Aggression in Mood Disorders

There are many biological associations between mood symptoms and
aggression or violence. These include increased aggression with in-
creased cytokine activity (Zalcman and Siegel 2006), catecholamine me-
tabolism (Volavka et al. 2004), testosterone (Pope et al. 2000), and hypo-
thalamic-pituitary-adrenal axis dysfunction (Malkesman et al. 2006;
Shea et al. 2005). However, the most consistent findings are associations
with the serotonergic system.
Among the many findings associated with serotonergic dysfunction
in aggression, platelet serotonin 2A receptor (5-HT2A) binding was in-
creased in subjects with trait aggression (Lauterbach et al. 2006). Pre-
frontal cortical 5-HT2A binding was also increased in aggressive sui-
cidal patients (Oquendo et al. 2006). Similarly, relative increases in
plasma tryptophan levels (a precursor to serotonin) are associated with
increased aggression and hostility (Lauterbach et al. 2006; Suarez and
Krishnan 2006). Lower cerebrospinal fluid 5-hydroxyindoleacetic acid
concentration was independently associated with severity of lifetime
aggressivity and a history of a higher-lethality suicide attempt and may
be part of the diathesis for these behaviors. The dopamine and norepi-
nephrine systems do not appear to be as significantly involved in sui-
cidal acts, aggression, or depression (Placidi et al. 2002). However, the
most compelling findings regarding the involvement of serotonin in
both mood disturbance and violence is found in the serotonin trans-
porter polymorphisms.
Several recent studies have investigated the role of polymorphisms

in the serotonin reuptake pump or the serotonin transporter gene
(5HTTLPR). A common polymorphism of this gene is a deletion in the
area of the gene that regulates its transcription into messenger RNA and
ultimate translation into expressed protein—the promoter region. Indi-
viduals with this deletion, called the short or “s” allele, express fewer
serotonin transporters. Individuals who are homozygous for the “s” al-
lele (ss) are more likely to develop depression (odds ratio, 1.5–179; Cer-
villa et al. 2006) and depression after a traumatic event (Caspi et al.
2003; Kaufman et al. 2004). Thus, the observed link between early life
adversity, or later life trauma, and subsequent depression, is related, at
least in part, to having the ss genotype (Caspi et al. 2003; Kaufman et al.
2004). Although stressful life events or extreme adversity are clearly as-
sociated with subsequent depression, adversity is quite potent in induc-
ing depression in subjects with the ss genotype, such that the dosage of
adversity required to produce depression is much lower in individuals
homozygous for the short form of 5HTTLPR (Cervilla et al. 2007). Sev-
eral studies have also found that the ss genotype is also associated with
subclinical depressive symptoms in individuals without depression
(Gonda and Bagdy 2006; Gonda et al. 2005, 2006).
The ss genotype of 5HTTLPR is also associated with aggression. In a
case control study of conduct disorder with or without aggression, it
was found that the ss genotype was strongly associated with aggression
but not conduct disorder without aggression (Sakai et al. 2006). A
positron emission tomography study of 5HTTLPR density found that
reduced transporter density is associated with impulsive aggression
(Frankle et al. 2005). Although this study did not examine the genotype
of the study subjects, it found that the phenotype that is expected with
the ss genotype is associated with aggression (Frankle et al. 2005).
Among schizophrenic patients who attempted suicide, the ss genotype
of 5HTTLPR was associated with violent suicide attempts but was not
associated with nonviolent suicide attempts or with nonattempters
(Bayle et al. 2003)
Key Points
■ Early exposure to aggression, either as a witness or as a victim,
increases the likelihood of future depression, the severity and
chronicity of future depressions, and the likelihood of perpetrat-
ing future aggression.
■ Early abuse increases the likelihood of subsequent victimization.

■ Early sexual abuse increases the likelihood of subsequent sexual
victimization.
■ Substance abuse increases the risk for perpetrating violence,
becoming a victim of violence, and depression related to exposure
to violence. Conversely, exposure to violence increases the risk of
subsequent substance abuse.
■ Aggression is common in individuals who are experiencing an epi-
sode of mania. There may be a subtype of mania in which aggres-
sion is the core feature.
■ Aggression may also be common in unipolar major depression.
■ The effect of antidepressants on aggression is unclear. If there is
an anti-aggression effect, it is weak.
■ Mood-stabilizing antiepileptic and antipsychotic medications are
used to treat aggression independent of diagnosis, but again, any
anti-aggression effect is weak.
■ There are many biological markers for associated aggression and
mood disturbance.
■ The serotonin system appears to be implicated in the interface of
aggression and mood disturbance.
■ A genetic polymorphism in the serotonin transporter is strongly
associated with depression after adversity and is also associated
with aggression in depressed and nondepressed subjects.
■ Deliberate self-harm (DSH) behaviors frequently occur indepen-
dently of suicide and should be considered separate phenomena.
However, acts of DSH increase risk for subsequent suicide.
■ DSH behavior is associated with early abuse.
■ DSH behavior is usually associated with dysphoria but not neces-
sarily with depression.
■ Both major depression and bipolar illness are significant risk fac-
tors for completed suicides. Bipolar disorder carries a much higher
risk for completed suicide than does major depression.
■ Comorbid disorders, such as substance use, increase suicide risk.
■ Antidepressants may increase suicide ideation but do not appear
to increase completed suicides and may reduce the severity of sui-
cide attempts.
■ Lithium has a clear antisuicide effect in both bipolar illness and
unipolar major depression.
References
Acierno R, Lawyer SR, Rheingold A, et al: Current psychopathology in previ-
ously assaulted older adults. J Interpers Violence 22:250–258, 2007
Afaq I, Riaz J, Sedky K, et al: Divalproex as a calmative adjunct for aggressive
schizophrenic patients. J Ky Med Assoc 100:17–22, 2002
Afifi M: Depression in adolescents: gender differences in Oman and Egypt. East
Mediterr Health J 12:61–71, 2006
American Psychiatric Association: Practice guideline for the assessment and
treatment of patients with suicidal behaviors. Am J Psychiatry 160(suppl):
1–60, 2003
Anderson C, Roux G, Pruitt A: Prenatal depression, violence, substance use,
and perception of support in pregnant middle-class women. J Perinat Educ
11:14–21, 2002
Angst J, Preisig M: Outcome of a clinical cohort of unipolar, bipolar and
schizoaffective patients: results of a prospective study from 1959 to 1985.
Schweiz Archiv Neurol Psychiatry 146:17–23, 1995
Arato M, Demeter E, Rihmer Z, et al: Retrospective psychiatric assessment of
200 suicides in Budapest. Acta Psychiatr Scand 77:454–456, 1988
Arendt M, Rosenberg R, Fjordback L, et al: Testing the self-medication hypoth-
esis of depression and aggression in cannabis-dependent subjects. Psychol
Med 37:935–945, 2007
Avdibegovic E, Sinanovic O: Consequences of domestic violence on women’s
mental health in Bosnia and Herzegovina. Croat Med J 47:730–741, 2006
Baldessarini RJ, Tondo L, Davis P, et al: Decreased risk of suicides and attempts
during long-term lithium treatment: a meta-analytic review. Bipolar Disord
8:625–639, 2006
Banyard VL, Williams LM, Siegel JA: Re-traumatization among adult women
sexually abuse in childhood: exploratory analyses in a prospective study.
J Child Sex Abuse 11:19–48, 2002
Barlow K, Grenyer B, Ilkiw-Lavalle O: Prevalence and precipitants of aggres-
sion in psychiatric inpatient units. Aust NZ J Psychiatry 34:967–974, 2000
Barzman DH, DelBello MP, Adler CM, et al: The efficacy and tolerability of que-
tiapine versus divalproex for the treatment of impulsivity and reactive ag-
gression in adolescents with co-occurring bipolar disorder and disruptive
behavior disorder(s). J Child Adolesc Psychopharmacol 16:666–670, 2006
Bayle FJ, Leroy S, Gourion D, et al: 5HTTLPR polymorphism in schizophrenic
patients: further support for association with violent suicide attempts. Am
J Med Genet B Neuropsychiatr Genet 119:12–17, 2003
Blitstein JL, Murray DM, Lytle LA, et al: Predictors of violent behavior in an
early adolescent cohort: similarities and differences across genders. Health
Educ Behav 32:175–194, 2005
Bogat GA, DeJonghe E, Levendosky AA, et al: Trauma symptoms among in-
fants exposed to intimate partner violence. Child Abuse Negl 30:109–125,
2006
Bond AJ: Antidepressant treatments and human aggression. Eur J Pharmacol

526:218–225, 2005
Bonomi AE, Thompson RS, Anderson M, et al: Intimate partner violence and
women’s physical, mental, and social functioning. Am J Prev Med 30:458–
466, 2006
Borges G, Angst J, Nock MK, et al: A risk index for 12-month suicide attempts
in the National Comorbidity Survey Replication (NCS-R). Psychol Med
36:1747–1757, 2006
Brent DA, Perper JA, Moritz G, et al: Psychiatric risk factors for adolescent sui-
cide: a case-control study. J Am Acad Child Adolesc Psychiatry 32:521–529,
1993
Bridge JA, Iyengar S, Salary CB, et al: Clinical response and risk of reported sui-
cidal ideation and suicide attempts in pediatric antidepressant treatment: a
meta-analysis of randomized controlled trials. JAMA 297:1683–1696, 2007
Brunstein Klomek A, Marrocco F, Kleinman M, et al: Bullying, depression, and
suicidality in adolescents. J Am Acad Child Adolesc Psychiatry 46:40–49,
2007
Calabrese JR, Muzina DJ, Kemp DE, et al: Predictors of bipolar disorder risk
among patients currently treated for major depression. MedGenMed 8:38,
2006
Caspi A, Sugden K, Moffitt TE, et al: Influence of life stress on depression: mod-
eration by a polymorphism in the 5-HTT gene. Science 301:291–293, 2003
Cassidy F, Carroll BJ: Frequencies of signs and symptoms in mixed and pure ep-
isodes of mania: implications for the study of manic episodes. Prog Neu-
ropsychopharmacol Biol Psychiatry 25:659–665, 2001
Cassidy F, Ahearn EP, Carroll BJ: Symptom profile consistency in recurrent
manic episodes. Compr Psychiatry 43:179–181, 2002
Centers for Disease Control and Prevention: Homicides and Suicides: National
Violent Death Reporting System, United States, 2003–2004. MMWR Morb
Mortal Wkly Rep 55:721–724, 2006 [Erratum in: MMWR Morb Mortal Wkly
Rep 55:1074–1075, 2006]
Cervilla JA, Rivera M, Molina E, et al: The 5-HTTTLPR s/s genotype at the se-
rotonin transporter gene (SLC6A4) increases the risk for depression in a
large cohort of primary care attendees: the PREDICT-gene study. Am J Med
Genet B Neuropsychiatr Genet 141:912–917, 2006
Cervilla JA, Molina E, Rivera M, et al: The risk for depression conferred by
stressful life events is modified by variation at the serotonin transporter
5HTTLPR genotype: evidence from the Spanish PREDICT-Gene cohort.
Mol Psychiatry 12:748–755, 2007
Chen YW, Dilsaver SC: Lifetime rates of suicide attempts among subjects with
bipolar and unipolar disorders relative to subjects with other Axis I disor-
ders. Biol Psychiatry 39:896–899, 1996
Cipriani A, Pretty H, Hawton K, et al: Lithium in the prevention of suicidal be-
havior and all cause mortality in patients with mood disorders: a system-
atic review of randomized trials. Am J Psychiatry 162:1805–1819, 2005
Coker AL, Smith PH, Fadden MK: Intimate partner violence and disabilities
among women attending family practice clinics. J Womens Health
(Larchmt) 14:829–838, 2005
Conway M, Mendelson M, Glannopoulos C, et al: Childhood and adult sexual

abuse, rumination on sadness, and dysphoria. Child Abuse Negl 28:393–
410, 2004
Cukor D, McGinn LK: History of child abuse and severity of adult depression:
the mediating role of cognitive schema. J Child Sex Abuse 15:19–34, 2006
Deblinger E, Mannarino AP, Cohen JA, et al: A follow-up study of a multisite,
randomized, controlled trial for children with sexual abuse-related PTSD
symptoms. J Am Acad Child Adolesc Psychiatry 45:1474–1484, 2006
Denov MS: The long-term effects of child sexual abuse by female perpetrators: a
qualitative study of male and female victims. J Interpers Violence 19:1137–
1156, 2004
Dervic K, Grunebaum MF, Burke AK, et al: Protective factors against suicidal
behavior in depressed adults reporting childhood abuse. J Nerv Ment Dis
194:971–974, 2006
Dilsaver SC, Chen YW, Swann AC, et al: Suicidality in patients with pure and
depressive mania. Am J Psychiatry 151:1312–1315, 1994
Dubicka B, Hadley S, Roberts C: Suicidal behaviour in youths with depression
treated with new generation antidepressants: meta analysis. Br J Psychiatry
189:393–398, 2006
Dunsieth NW Jr, Nelson EB, Brusman-Lovins LA, et al: Psychiatric and legal
features of 113 men convicted of sexual offenses. J Clin Psychiatry 65:293–
300, 2004
Edwards EP, Elden RD, Colder C, et al: The development of aggression in 18
and 48 month old children of alcoholic parents. J Abnorm Child Psychol
34:409–423, 2006
Fagiolini A, Kupfer DJ, Rucci P, et al: Suicide attempts and ideation in patients
with bipolar I disorder. J Clin Psychiatry 65:509–514, 2004
Feldmann TB: Bipolar disorder and violence. Psychiatr Q 72:119–129, 2001
Fiedler N, Ozakinci G, Hallman W, et al: Military deployment to the Gulf War
as a risk factor for psychiatric illness among U.S. troops. Br J Psychiatry
188:453–459, 2006
Foshee VA, Benefield TS, Ennett ST, et al: Longitudinal predictors of serious
physical and sexual violence victimization during adolescence. Prev Med
39:1007–1016, 2004
Foster T, Gillespie K, McClelland R, et al: Risk factors for suicide independent
of DSM-III-R Axis I disorder. Br J Psychiatry 175:175–179, 1999
Frankle WG, Lombardo I, New AS, et al: Brain serotonin transporter distribu-
tion in subjects with impulsive aggressivity: a positron emission study with
[11C]McN 5652. Am J Psychiatry 162:915–923, 2005
Gamble SA, Talbot NL, Duberstein PR, et al: Childhood sexual abuse and depres-
sive symptom severity: the role of neuroticism. J Nerv Ment Dis 194:382–385,
2006
Gaylord KM: The psychosocial effects of combat: the frequently unseen injury.
Crit Care Nurs Clin North Am 18:349–357, 2006
Gaynes BN, West SL, Ford CA, et al: Screening for suicide risk in adults: a sum-
mary of the evidence for the U.S. Preventive Services Task Force. Ann In-
tern Med 140:822–835, 2004
Gibbons RD, Hur K, Bhaumik DK, et al: The relationship between antidepres-
sant prescription rates and rate of early adolescent suicide. Am J Psychiatry
163:1898–1904, 2006
Gladstone GL, Parker GB, Mitchell PB, et al: Implications of childhood trauma for
depressed women: an analysis of pathways from childhood sexual abuse to
deliberate self-harm and revictimization. Am J Psychiatry 161:1417–1425,
2004
Gobbi G, Gaudreau PO, Leblanc N: Efficacy of topiramate, valproate, and their
combination on aggression/agitation behavior in patients with psychosis.
J Clin Psychopharmacol 26:467–473, 2006
Goedhard LE, Stolker JJ, Heerdink ER, et al: Pharmacotherapy for the treatment
of aggressive behavior in general adult psychiatry: a systematic review.
J Clin Psychiatry 67:1013–1024, 2006
Goldstein AL, Walton MA, Cunningham RM, et al: Violence and substance use
as risk factors for depressive symptoms among adolescents in an urban
emergency department. J Adolesc Health 40:276–279, 2007
Gonda X, Bagdy G: Relationship between serotonin transporter gene 5HTTLPR
polymorphism and the symptoms of neuroticism in healthy population.
Psychiatr Hung 21:379–385, 2006
Gonda X, Juhasz G, Laszik A, et al: Subthreshold depression is linked to the
functional polymorphism of the 5HT transporter gene. J Affect Disord
87:291–297, 2005
Gonda X, Rihmer Z, Zsombok T, et al: The 5HTTLPR polymorphism of the se-
rotonin transporter gene is associated with affective temperaments as mea-
sured by TEMPS-A. J Affect Disord 91:125–131, 2006
Groholt B, Ekeberg O, Haldorsen T: Adolescents hospitalised with deliberate
self-harm: the significance of an intention to die. Eur Child Adolesc Psychi-
atry 9:244–254, 2000
Grossman LS, Haywood TW, Cavanaugh JL, et al: State psychiatric hospital pa-
tients with past arrests for violent crimes. Psychiatr Serv 46:790–795, 1995
Grunebaum MF, Galfalvy HC, Nichols CM, et al: Aggression and substance
abuse in bipolar disorder. Bipolar Disord 8:496–502, 2006
Guzzetta F, Tondo L, Centorrino F, et al: Lithium treatment reduces suicide risk
in recurrent major depressive disorder. J Clin Psychiatry 68:380–383, 2007
Hammack PL, Richards MH, Lug Z, et al: Social support factors as moderators
of community violence exposure among inner-city African American
young adolescents. J Clin Child Adolesc Psychol 33:450–462, 2004
Hammad TA, Laughren TP, Racoosin JA: Suicide rates in short-term random-
ized controlled trials of newer antidepressants. J Clin Psychopharmacol
26:203–207, 2006
Hanson RF, Self-Brown S, Fricker-Ethal A, et al: Relations among parental sub-
stance use, violence exposure and mental health: the national survey of ad-
olescents. Addict Behav 31:1988–2001, 2006
Harrington R, Pickles A, Aglan A, et al: Early adult outcomes of adolescents
who deliberately poisoned themselves. J Am Acad Child Adolesc Psychia-
try 45:337–345, 2006
Hashemian F, Khoshnood K, Desai MM, et al: Anxiety, depression, and post-
traumatic stress in Iranian survivors of chemical warfare. JAMA 296:560–
566, 2006
Haw C, Hawton K, Houston K, et al: Psychiatric and personality disorders in

deliberate self-harm patients. Br J Psychiatry 178:48–54, 2001
Hazen AL, Connelly CD, Kelleher KJ, et al: Female caregivers’ experiences with
intimate partner violence and behavioral problems in children investigated
as victims of maltreatment. Pediatrics 117:99–109, 2006
Healy D, Herxheimer A, Menkes DB: Antidepressants and violence: problems
at the interface of medicine and law. PLoS Med 3:e372, 2006
Henriksson MM, Aro HM, Marttunen MJ, et al: Mental disorders and comor-
bidity in suicide. Am J Psychiatry 150:935–940, 1993
Hindin MJ, Gultiano S: Associations between witnessing parental domestic vi-
olence and experiencing depressive symptoms in Filipino adolescents. Am
J Public Health 96:660–663, 2006
Houry D, Kemball R, Rhodes KV, et al: Intimate partner violence and mental
health symptoms in African American female ED patients. Am J Emerg
Med 24:444–450, 2006
Huizinga D, Haberstick BC, Smolen A, et al: Childhood maltreatment, subse-
quent antisocial behavior, and the role of monoamine oxidase A genotypes.
Biol Psychiatry 60:677–683, 2006
Isometsa ET, Henriksson MM, Aro HM, et al: Suicide in bipolar disorder in Fin-
land. Am J Psychiatry 151:1020–1024, 1994
Jamison KR: Touched with Fire: Manic-Depressive Illness and the Artistic Tem-
perament. New York, Free Press, 1993
Johansen VA, Wahl AK, Eilertsen DE, et al: Acute psychological reactions in as-
sault victims of non-domestic violence peritraumatic dissociation, post-
traumatic stress disorder, anxiety and depression. Nord J Psychiatry
60:452–462, 2006
Kaufman J, Yang BZ, Douglas-Palumberi H, et al: Social supports and serotonin
transporter gene moderate depression in maltreated children. Proc Natl
Acad Sci USA 101:17316–17321, 2004
Keilp JG, Gorlyn M, Oquendo MA, et al: Aggressiveness, not impulsiveness or
hostility, distinguishes suicide attempters with major depression. Psychol
Med 36:1779–1788, 2006
Kessler RC, Akiskal HS, Ames M, et al: Prevalence and effects of mood disor-
ders on work performance in a nationally representative sample of U.S.
workers. Am J Psychiatry 163:1561–1568, 2006
Kim YS, Leventhal BL, Koh YJ, et al: School bullying and youth violence: causes
or consequences of psychopathologic behavior? Arch Gen Psychiatry
63:1035–1041, 2006
Klonsky ED, Oltmanns TF, Turkheimer E: Deliberate self-harm in a nonclinical
population: prevalence and psychological correlates. Am J Psychiatry
160:1501–1508, 2003
Knox M, King C, Hanna GL, et al: Aggressive behavior in clinically depressed
adolescents. J Am Acad Child Adolesc Psychiatry 39:611–618, 2000
Korkeila J, Salminen JK, Hiekkanen H, et al: Use of antidepressants and suicide
rate in Finland: an ecological study. J Clin Psychiatry 68:505–511, 2007
Kuehn BM: FDA panel seeks to balance risks in warnings for antidepressants.
JAMA 297:573–574, 2007
Lamprecht HC, Pakrasi S, Gash A, et al: Deliberate self-harm in older people re-
visited. Int J Geriatr Psychiatry 20:1090–1096, 2005
Lauterbach E, Brunner J, Hawelleck B, et al: Platelet 5-HT2A receptor binding

and tryptophan availability are not associated with recent history of sui-
cide attempts but with personality traits characteristic for suicidal behav-
ior. J Affect Disord 91:57–62, 2006
Lehrer JA, Buka S, Gortmaker S, et al: Depressive symptomatology as a predic-
tor of exposure to intimate partner violence among U.S. female adolescents
and young adults. Arch Pediatr Adolesc Med 160:270–276, 2006
Lipsky S, Field CA, Caetano R, et al: Posttraumatic stress disorder symptoma-
tology and comorbid depressive symptoms among abused women re-
ferred from emergency department care. Violence Vict 20:645–659, 2005
Lohner J, Konrad N: Deliberate self-harm and suicide attempt in custody: dis-
tinguishing features in male inmates’ self-injurious behavior. Int J Law Psy-
chiatry 29:370–385, 2006
Loncar M, Medved V, Jovanovic N, et al: Psychological consequences of rape on
women in 1991–1995 war in Croatia and Bosnia and Herzegovina. Croat
Med J 47:67–75, 2006
MacKinnon DF, Zandi PP, Gershon E, et al: Rapid switching of mood in families
with multiple cases of bipolar disorder. Arch Gen Psychiatry 60:921–928,
2003
MacMillan CM, Korndorfer SR, Rao S, et al: A comparison of divalproex and ox-
carbazepine in aggressive youth with bipolar disorder. J Psychiatr Pract
12:214–222, 2006
Maj M, Pirozzi R, Magliano L, et al: Agitated depression in bipolar I disorder:
prevalence, phenomenology, and outcome. Am J Psychiatry 160:2134–2140,
2003
Malkesman O, Maayan R, Weizman A, et al: Aggressive behavior and HPA axis
hormones after social isolation in adult rats of two different genetic animal
models for depression. Behav Brain Res 175:408–414, 2006
Maremmani I, Lazzeri A, Pacini M, et al: Diagnostic and symptomatological
features in chronic psychotic patients according to cannabis use status.
J Psychoactive Drugs 36:235–241, 2004
Martin G, Bergen HA, Richardson AS, et al: Sexual abuse and suicidality: gen-
der differences in a large community sample of adolescents. Child Abuse
Negl 28:491–503, 2004
Martin J, Langley J, Millchamp J: Domestic violence as witnessed by New
Zealand children. N Z Med J 119:U1817, 2006
Martsolf DS, Draucker CB: Psychotherapy approaches for adult survivors of
childhood sexual abuse: an integrative review of outcomes research. Issues
Ment Health Nurs 26:801–825, 2005
Masho SW, Odor RK, Adera T: Sexual assault in Virginia: a population-based
study. Womens Health Issues 15:157–166, 2005
McElroy SL, Soutullo CA, Taylor P Jr, et al: Psychiatric features of 36 men con-
victed of sexual offenses. J Clin Psychiatry 60:414–420, 1999
McGuigan WM, Middlemiss W: Sexual abuse in childhood and interpersonal
violence in adulthood: a cumulative impact on depressive symptoms in
women. J Interpers Violence 20:1271–1287, 2005
Miguel-Tobal JJ, Cano-Vindel A, Gonzalez-Ordi H, et al: PTSD and depression
after the Madrid March 11 train bombings. J Trauma Stress 19:69–80, 2006
Miller MC: The biology of child maltreatment. Harv Ment Health Lett 21:1–3,
2005
Mitchell MD, Hargrove GL, Collins MH, et al: Coping variables that mediate the
relations between intimate partner violence and mental health outcomes
among low-income, African American women. J Clin Psychol 62:1503–
1520, 2006
Mitchell PJ: Antidepressant treatment and rodent aggressive behavior. Eur J
Pharmacol 526:147–162, 2005
Muehlenkamp JJ, Gutierrez PM: An investigation of differences between self-
injurious behavior and suicide attempts in a sample of adolescents. Suicide
Life Threat Behav 34:12–23, 2004
Najt P, Perez J, Sanches M, et al: Impulsivity and bipolar disorder. Eur Neuro-
psychopharmacol 17:313–320, 2007
North CS, Smith EM, Spitznagel EL: Violence and the homeless: an epidemio-
logic study of victimization and aggression. J Trauma Stress 7:95–110, 1994
Oquendo MA, Russo SA, Underwood MD, et al: Higher postmortem prefrontal
5-HT2A receptor binding correlates with lifetime aggression in suicide.
Biol Psychiatry 59:235–243, 2006
Ostacher MJ, Eidelman P: Suicide in bipolar depression, in Bipolar Depression:
A Comprehensive Guide. Edited by El-Mallakh RS, Ghaemi SN. Washing-
ton, DC, American Psychiatric Publishing, 2006, pp 117–144
Paranjape A, Heron S, Thompson M, et al: Are alcohol problems linked with an
increase in depressive symptoms in abused, inner-city African American
women? Womens Health Issues 17:37–43, 2007
Parker G, Malhi G, Mitchell P, et al: Self-harming in depressed patients: pattern
analysis. Aust NZ J Psychiatry 39:899–906, 2005
Peiponen S, Laukkanen E, Korhonen V, et al: The association of parental alcohol
abuse and depression with severe emotional and behavioral problems in
adolescents: a clinical study. Int J Soc Psychiatry 52:395–407, 2006
Peleikis DE, Mykletun A, Dahl AA: Long-term social status and intimate rela-
tionship in women with childhood sexual abuse who got outpatient psy-
chotherapy for anxiety disorder and depression. Nord J Psychiatry 59:31–
38, 2005
Person C, Tracy M, Galea S: Risk factors for depression after a disaster. J Nerv
Ment Dis 194:659–666, 2006
Pico-Alfonso MA, Garcia-Linares MI, Celda-Navarro N, et al: The impact of
physical, psychological and sexual intimate male partner violence on
women’s mental health: depressive symptoms, posttraumatic stress disor-
der, state anxiety, and suicide. J Womens Health (Larchmt) 15:599–611, 2006
Placidi GP, Oquendo MA, Malone KM, et al: Aggressivity, suicide attempts, and
depression: relationship to cerebrospinal fluid monoamine metabolite lev-
els. Biol Psychiatry 52:375–376, 2002
Pope HG Jr, Kouri EM, Hudson JI: Effects of supraphysiologic doses of tes-
tosterone on mood and aggression in normal men: a randomized con-
trolled trial. Arch Gen Psychiatry 57:133–140, 2000
Posternak MA, Zimmerman M: Anger and aggression in psychiatric outpa-
tients. J Clin Psychiatry 63:665–672, 2002
Potter NN: Trauma, Truth and Reconciliation: Healing Damaged Relationships.
New York, Oxford University Press, 2006
Quanbeck CD, Stone DC, Scott CL, et al: Clinical and legal correlates of inmates
with bipolar disorder at time of criminal arrest. J Clin Psychiatry 65:198–
203, 2004
Raja M, Azzoni A: Suicide attempts: differences between unipolar and bipolar
patients and among groups with different lethality risk. J Affect Disord
82:437–442, 2004
Reed GL, Enright RD: The effects of forgiveness therapy on depression, anxiety,
and posttraumatic stress for women after spousal emotional abuse. J Con-
sult Clin Psychol 74:920–929, 2006
Reigstad B, Jorgensen K, Wichstrom L: Diagnosed and self-reported childhood
abuse in national and regional samples of child and adolescent psychiatric
patients: prevalences and correlates. Nord J Psychiatry 60:58–66, 2006
Rich CL, Young D, Fowler RC: San Diego suicide study: young vs. old subjects.
Arch Gen Psychiatry 43:577–582, 1986
Rich CL, Gidycz CA, Warkentin JB, et al: Child and adolescent abuse and sub-
sequent victimization: a prospective study. Child Abuse Negl 29:1373–
1394, 2005
Rivera-Rivera L, Allen B, Rodriguez-Ortega G, et al: Dating violence and asso-
ciations with depression and risk behaviors: female students in Morelos,
Mexico. Salud Publica Max 48(suppl):S288–S296, 2006
Rowe R, Maughan B, Eley TC: Links between antisocial behavior and depressed
mood: the role of life events and attributional style. J Abnorm Child Psy-
chol 34:293–302, 2006
Sachs-Ericsson N, Verona E, Joiner T, et al: Parental verbal abuse and the medi-
ating role of self-criticism in adult internalizing disorders. J Affect Disord
93:71–78, 2006
Safer DJ, Zito JM: Do antidepressants reduce suicide rates? Public Health
121:274–277, 2007
Sakai JT, Young SE, Stallings MC, et al: Case control and within-family tests for
an association between conduct disorder and 5HTTLPR. Am J Med Genet
B Neuropsychiatr Genet 141:825–832, 2006
Sato T, Bottlender R, Kleindienst N, et al: Syndromes and phenomenological
subtypes underlying acute mania: a factor analytic study of 576 manic pa-
tients. Am J Psychiatry 159:968–974, 2002
Sato T, Bottlender R, Kleindienst N, et al: Irritable psychomotor elation in de-
pressed inpatients: a factor validation of mixed depression. J Affect Disord
84:187–196, 2005
Schuepbach D, Goetz I, Boeker H, et al: Voluntary vs. involuntary hospital ad-
mission in acute mania of bipolar disorder: results from the Swiss sample
of the EMBLEM study. J Affect Disord 90:57–61, 2006
Shea A, Walsh C, Macmillan H, et al: Child maltreatment and HPA axis dysreg-
ulation: relationship to major depressive disorder and post traumatic stress
disorder in females. Psychoneuroendocrinology 30:162–178, 2005
Sher L, Oquendo MA, Conason AH, et al: Clinical features of depressed patients
with or without a family history of alcoholism. Acta Psychiatr Scand
112:266–271, 2005
Silverstein M, Augustyn M, Cabral H, et al: Maternal depression and violence
exposure: double jeopardy for child school functioning. Pediatrics 118:792–
800, 2006
Sinclair JMA, Harriss L, Baldwin DS, et al: Suicide in depressive disorders: a ret-
rospective case-control study of 127 suicides. J Affect Disord 87:107–113,
2005
Smith DJ, Harrison N, Muir W, et al: The high prevalence of bipolar spectrum
disorders in young adults with recurrent depression: toward an innovative
diagnostic framework. J Affect Disord 84:167–178, 2005
Sokero P, Eerola M, Rytsälä H, et al: Decline in suicidal ideation among patients
with MDD is preceded by decline in depression and hopelessness. J Affect
Disord 95:95–102, 2006
Starkman MN: The terrorist attacks of September 11, 2001, as psychological
toxin: increase in suicide attempts. J Nerv Ment Dis 194:547–550, 2006
Sternberg KJ, Lamb ME, Guterman E, et al: Effects of early and later family vio-
lence on children’s behavior problems and depression: a longitudinal,
multi-informant perspective. Child Abuse Negl 30:283–306, 2006
Suarez EC, Krishnan KR: The relation of free plasma tryptophan to anger, hos-
tility, and aggression in a nonpatient sample of adult men and women. Ann
Behav Med 31:254–260, 2006
Swann AC: Treatment of aggression in patients with bipolar disorder. J Clin
Psychiatry 60(suppl):25–28, 1999
Teicher MH, Samson JA, Polcari A, et al: Sticks, stones, and hurtful words: rela-
tive effects of various forms of childhood maltreatment. Am J Psychiatry
163:993–1000, 2006
Theran SA, Sullivan CM, Bogat GA, et al: Abusive partners and ex-partners: un-
derstanding the effects of relationship to the abuser on women’s well-
being. Violence Against Women 12:950–969, 2006
Tiihonen J, Lonnqvist J, Wahlbeck K, et al: Antidepressants and the risk of sui-
cide, attempted suicide, and overall mortality in a nationwide cohort. Arch
Gen Psychiatry 63:1358–1367, 2006
Tondo L, Hennen J, Baldessarini RJ: Lower suicide risk with long-term lithium
treatment in major affective illness: a meta-analysis. Acta Psychiatr Scand
104:163–172, 2001
Tuisku V, Pelkonen M, Karlsson L, et al: Suicidal ideation, deliberate self-harm
behavior and suicide attempts among adolescent outpatients with depres-
sive mood disorders and comorbid Axis I disorders. Eur Child Adolesc
Psychiatry 15:199–206, 2006
Tzemou E, Birchwood M: A prospective study of dysfunctional thinking and
the regulation of negative intrusive memories in bipolar 1 disorder: impli-
cations for affect regulation theory. Psychol Med 37:689–698, 2007
Van Gogh Gallery: Vincent van Gogh: Biography. Available at http://www
.vangoghgallery.com. Accessed March 2007
Varma D, Chandra PS, Thomas T, et al: Intimate partner violence and sexual co-
ercion among pregnant women in India: relationship with depression and
post-traumatic stress disorder. J Affect Disord 102:227–235, 2006
Volavka J, Bilder R, Nolan K: Catecholamines and aggression: the role of COMT
and MAO polymorphisms. Ann NY Acad Sci 1036:393–398, 2004
Watlington CG, Murphy CM: The roles of religion and spirituality among Afri-
can American survivors of domestic violence. J Clin Psychol 62:837–857,
2006
Whitaker RC, Orzol SE, Kahn RS: Maternal mental health, substance use, and
domestic violence in the year after delivery and subsequent behavior prob-
lems in children at age 3 years. Arch Gen Psychiatry 63:551–560, 2006
Widom CS, DuMont K, Czaja SJ: A prospective investigation of major depres-
sive disorder and comorbidity in abused and neglected children grown up.
Wright J, Friedrich W, Cinq-Mars C, et al: Self-destructive and delinquent be-
haviors of adolescent female victims of child sexual abuse: rates and cova-
riates in clinical and nonclinical samples. Violence Vict 19:627–643, 2004
Zalcman SS, Siegel A: The neurobiology of aggression and rage: role of cyto-
kines. Brain Behav Immun 20:507–514, 2006
C H A P T E R 6
Schizophrenia and
Delusional Disorder
In this chapter I discuss assessment and management of acutely violent

patients with primary psychotic disorders. Most studies of violent pa-
tients are of diagnostically mixed populations that include a large pro-
portion diagnosed with schizophrenia, reflecting clinical populations.
I highlight studies of groups of adults with schizophrenia but also dis-
cuss studies of mixed groups. Some aspects of assessment and much of
management can apply to all severely mentally ill adults. Diagnoses are
often imprecise, especially in large community surveys, in which pa-
tients are often identified by their predominant symptoms rather than
diagnosis. There is little literature to guide discussions of violence in pa-
tients with delusional disorder, but these patients are similar to those
with chronic schizophrenia, with prominent positive symptoms and
low levels of negative symptoms.
My discussion of management of violence primarily covers environ-
mental and interpersonal strategies to prevent violence. Pharmacologi-
cal management is discussed at length in Chapter 15.
Violence Assessment
The first step in assessing a violent incident is to find out exactly what
happened. Ask who did what to whom. The term violent, like aggression,
agitation, and assaultiveness, is often used quite loosely and imprecisely,
105
sometimes when a patient is simply loud and belligerent. If there was

physical contact, learn whether the incident was serious and if so, how
serious. Were any weapons involved? Was anyone injured? If so, how
badly?
After it is clear that all are safe, interview the victim, assailant, and
others present at the time of the incident. Attempt to learn what hap-
pened between assailant and victim before the assault. Attempt to iden-
tify and piece together the often complex interactions among assailant,
victim, external circumstances, and symptoms of illness. Many factors
may explain violence in people with psychosis besides psychosis and
other psychopathology: environmental factors; cultural, interpersonal,
and attitudinal factors; and other, unknown factors. Aggression is a uni-
versal human drive, the expression of which is shaped by social norms
and external contingencies. In some circumstances the expression of ag-
gression in assault can be adaptive.
When assessing violent patients, be especially alert to risk factors
that can be changed. These include certain types of psychopathology,
substance abuse, treatment nonadherence, and medical comorbidities.
Substance abuse is more common in outpatients, but clever and deter-
mined inpatients can obtain alcohol and substances of abuse. Medical
conditions are not commonly associated with violence but are poten-
tially life threatening and reversible. For example, a woman with diabe-
tes was referred to a unit for violent patients because she had been bel-
ligerent and was shoving others. Nurses on the unit saw her stumbling
and intermittently confused. Review of medication history revealed a
recent increase in insulin dosing, and blood sugar monitoring revealed
intermittent hypoglycemia.
Rates of Violence in Community Samples

With Psychotic Illness and With No Detected
Psychiatric Illness
Violence occurs at low rates in the general population and at moder-
ately higher rates in adults with psychotic illness. In a large survey of
about 10,000 adults in the United States, the Epidemiologic Catchment
Area survey (Swanson et al. 1990), subjects were classed as violent if
they admitted to one of five behaviors within the previous 12 months:
physical abuse of a child, physical fighting, physical fighting while
intoxicated, fighting with a spouse or partner, and fighting with a
weapon. Two percent of those with no disorder had been violent in the
previous 12 months. The rate for those with schizophrenia was 13%,
whereas for those with major depression it was 12%, for adults who
Schizophrenia and Delusional Disorder ❘ 107
abused cannabis, 19%, and for those who abused alcohol, 25%. Trained
nonclinicians administered the Diagnostic Interview Schedule, based
on DSM-III (American Psychiatric Association 1980), to arrive at Axis I
diagnoses. When a person met criteria for more than one diagnosis, he
or she was counted in both categories. Those with more diagnoses were
more likely to be violent. In logistic regression analysis, male gender,
youth, poverty, substance abuse, and major mental illness all predicted
violence, as did the interaction between substance abuse and major
mental illness.
Other investigators have had similar findings: a low base rate of vi-
olence and a moderate rate in adults with psychotic illness. A large Is-
raeli survey (Stueve and Link 1997) found a 7% rate of violence in the
previous 5 years in adults ages 24–33 without identified psychiatric ill-
ness and a 21% rate in adults with psychotic or bipolar disorders. They
also found higher rates in those with comorbid substance abuse (39%)
and very high rates (93%) in those with psychotic or bipolar disorders
comorbid with antisocial personality, with or without substance abuse.
The relationship between violence and psychotic and bipolar disorders
was significant after controlling for substance abuse, antisocial person-
ality disorder, and demographic characteristics. Investigators asked if
respondents had been in a physical fight within the previous 5 years or
had a weapon in hand during a fight.
Diagnoses of psychosis obtained in large community samples such as
these may agree poorly with clinical diagnoses, but these studies give a
general idea of comparative rates of violence across diagnoses and with
comorbidity. Community samples have the advantage of avoiding the
sampling bias that can occur in studies of hospitalized patients (i.e., that
many of these subjects are hospitalized because they have been violent).
Association of Symptoms of Mental Illness

and Violence
In a study of a diagnostically mixed population (Monahan et al. 2001),
and in surveys of large community samples (Arseneault et al. 2000; Link
et al. 1998), paranoia or suspiciousness has been associated with assault-
iveness. Link et al. (1998) identified a relationship between paranoia, de-
lusions of control and thought insertion (known as threat/control over-
ride), and past history of violence in both psychiatrically ill and control
subjects. Arseneault et al. (2000) concluded that violence in subjects with
schizophrenia was partly explained by excessive perceptions of threat.
A study of 1,410 patients in treatment for schizophrenia, a substudy
of the National Institute of Mental Health Clinical Antipsychotic Trials
of Intervention Effectiveness (the CATIE project; Swanson et al. 2006),

explored the association of violence and symptoms. Serious violence
was associated with hostility (odds ratio [OR] 1.65), suspiciousness and
persecutory delusions (OR 1.46), hallucinations (OR 1.43), grandiosity
(OR 1.31), and excitement (OR 1.30). Delusional thinking alone was not
associated with serious violence, but when it occurred with suspicious-
ness and persecutory delusions, it was highly associated (OR 2.9). Neg-
ative symptoms were associated with a decreased risk of serious vio-
lence. High levels of positive symptoms with high levels of negative
symptoms were not associated with increased risk of serious violence,
but high levels of positive symptoms with low levels of negative symp-
toms were (OR 3.05).
Serious violence was defined as assault with a lethal weapon or a
threat with a lethal weapon, assault resulting in injury, or sexual assault.
Minor violence was defined as simple assault without a weapon. This
study omitted patients in their first episode and patients with treat-
ment-resistant illness. Violence was detected by self-report and family
collateral history. Of 1,410 subjects, 15.5% reported minor violence dur-
ing the previous 6 months and 3.6% reported serious violence. Child-
hood conduct problems, younger age, and a history of arrests were as-
sociated with serious violence. Younger age, living with family, and not
feeling listened to by family were associated with minor violence. To-
gether, all variables considered in this study did not explain more than
about 18% of the variance in minor and serious violence, suggesting
they had little explanatory power. Other factors, tested in other inves-
tigations or not yet hypothesized, could be more powerfully linked to
violence.
Students and psychiatric residents are routinely taught to ask whether
patients experience command hallucinations to harm others. Command
hallucinations would seem to be an obvious risk, but experienced clini-
cians know of many patients who have never been known to comply
with hallucinated commands. Reported rates of compliance vary widely.
These differences may be partly due to differences in populations stud-
ied, because some investigators study only patients with schizophrenia
and others study diagnostically mixed groups. Investigators also con-
sider different time frames between command hallucinations and vio-
lence. For example, some ask patients if they have ever been violent in
response to commands, whereas others ask if patients have been violent
in the past year. Differences may be due to failure to consider response
sets in self-report data. For example, in some groups, especially forensic
populations, patients may be motivated to over-report commands. All
studies are retrospective and subject to recall bias.
Nevertheless, it is clear that some patients with schizophrenia act

aggressively in response to command hallucinations. British investiga-
tors have taken a different approach and describe factors associated
with compliance to commands. In their review, Braham et al. (2004)
noted that most patients with command hallucinations hear harmless
commands and that dangerous commands have been associated with
less compliance. Factors associated with compliance are perceived be-
nevolence, power, and familiarity of the voice. A delusion congruent
with a command hallucination can also make compliance more likely.
In a forensic setting, Taylor (1985) found that the crimes of a group
of men motivated by delusions were more serious than crimes of other
psychotic men. Investigators conducted lengthy interviews of psy-
chotic and nonpsychotic male prisoners remanded for violent and non-
violent crimes in order to deduce reasons for offending. In the psychotic
group, delusions were fairly common precipitants of crimes, but in non-
psychotic men, motives such as panic, self-defense, jealousy, and imme-
diate retaliation were more likely to be associated with serious violence.
Although half of the psychotic men in this sample claimed some of the
same nonpsychotic motives for their crimes, such as panic, self-defense,
immediate retaliation, material gain, and accident, the author wrote
that these explanations had psychotic underpinnings. Taylor suggested
that psychotic symptoms may frustrate men so they lose self-control
and tolerance of others; in this way she proposed an indirect mecha-
nism—that is, irritability or reactivity—through which psychotic symp-
toms can lead to assault.
Studies of inpatients differ from community samples. Inpatient
states associated with assault include paranoia, strong affect, confusion,
hyperarousal, and frustration. Hyperarousal and frustration can result
from tense, unpredictable, chaotic environments. Certain symptoms
may be more predictive of violence in recently hospitalized than in
chronically hospitalized patients. In newly admitted inpatients, posi-
tive psychotic symptoms and irritability are predictive of violence, but
in long-term, chronically psychotic patients, motor excitement, anger,
low frustration tolerance, difficulty in delaying gratification (Kay et al.
1988), hostility, suspiciousness, and irritability (Krakowski et al. 1999)
are more predictive.
Reactive Versus Instrumental Classification of Assaults

An important goal of interviews after an assault is to determine
whether the event can be characterized as reactive or instrumental. In
animals and unselected populations of humans, assault can be broadly
and crudely classed into these two groups. Reactive events are impul-
sive, affective, or explosive, whereas instrumental events are coercive,
predatory, or psychopathic (Eichelman 1990). This dichotomy is an
oversimplification, because motives are often mixed, but it can be use-
ful in treatment planning for mentally ill people. Investigators believe
that reactive aggression is more treatable than instrumental aggression
(Campbell et al. 1978, 1982, 1984, 1995; Sheard et al. 1976).
In reactive fights, assailants react out of strong emotion. For exam-
ple, in a study in which we asked assailants to explain their behavior
(Crowner et al. 1995), one said, “I couldn’t help it. She made me lose
face. She keeps insulting me in front of others. I just did it.” Another
said, “I was upset. She just ticked me off when she hit me with the ping-
pong racket. I’m going to kill her if she hits me again.” Another said, “I
was so angry with him, the way he plays games before he takes his
medication. We always have to suffer because he plays those games. We
can’t get our cigarettes on time. I pushed him to one side so I could get
my medication. I didn’t think he’d fall.”
Instrumental assaults have a concrete goal. For example, a patient-
assailant explained, “He didn’t want me to have any of his food or
candy. He just has to do what I say.” Another said, “He sat on my chair
[the chair where this man habitually sat] after I stood up to take my
medication. I asked him to get up from my chair but he pretended not
to hear me, so I lifted up my chair.”
Assailants express reasons that have both reactive and instrumental
aspects when they say they assaulted someone to make him or her stop
a noxious behavior. One said, “He started it. He was bothering me and
bothering me. I just have to hit him to make him stop. He instigated it.
It’s not my fault. I think I broke his nose, but it was not my fault.” An-
other said, “He’s always cursing and harassing me. I woke up one
morning and just couldn’t take it anymore. I just wanted to make him
stop, let him know I was in a bad mood that day.” A third patient said,
“I wanted him to stop bothering the patients. He was bothering another
patient. I just wanted to help. I am a fair guy, and I don’t like others be-
ing bullied.”
Research Attempts to Classify Inpatient Assaults

Are the assaults of patients with psychotic illness similar to those of
other humans in that they can be classed as coercive or reactive? Or do
their motivations usually result from delusions and hallucinations? No-
lan et al. (2003) classified inpatient assaults as due to psychosis, impul-
sivity, or psychopathy. They classed an assault as psychotic or possibly
psychotic according to assailant explanation and rater judgments. As-

saults due to psychopathy were characterized by planning, lack of re-
morse, and predatory gain. They characterized reactive assaults as
those with an immediate victim provocation, for example, an order to
do something or a denial of a request, and without advance planning or
predatory gain. Subjects were inpatients at a state hospital on a special
unit for violent patients. Investigators interviewed victims, assailants,
and witnesses, using a checklist to determine presence or absence of
planning, predatory gain, remorse, and victim provocation.
Seventy percent of assailants carried a chart diagnosis of schizo-
phrenia or schizoaffective illness. Of 55 assaults, 20% were judged to be
psychotic or possibly psychotic. Eighty percent were judged to be re-
lated to psychopathy, poor impulse control, or uncertain factors. Victim
provocation was the motive most frequently cited by assailants, but vic-
tims rarely agreed they had been provocative.
A more recent study (Quanbeck et al. 2007) made a similar attempt
among chronically assaultive patients in a long-term psychiatric hospi-
tal in California. Most patients had a primary psychotic disorder. Fifty-
seven percent of the patients had been committed under forensic laws.
Using record review, the authors classified 839 assaults made by 88 in-
dividuals as psychotic, impulsive, or organized. In psychotic assaults
there was no provocation or rational motive, and the assailant cited a
delusion or hallucination as a motive. Impulsive events were character-
ized by an immediate provocation. The assailant was pacing, angry,
yelling or threatening, could not be calmed, had no obvious secondary
gain, and expressed remorse after the incident. Organized assaults were
characterized by planning, little or no provocation, little warning, an ex-
ternal goal or social motive such as asserting dominance, and no agita-
tion before the incident. The investigators found that 17% of assaults
were psychotic, 54% were impulsive, and 29% were organized. Forensic
patients committed more organized assaults, although this result was a
nonsignificant trend only.
Videotape Recording of Inpatient Assaults

We installed a video camera system in the dayroom of an inpatient unit
for persistently violent patients (Crowner et al. 2005), a room where the
patients spent most of their waking hours, in order to study assaults and
their precipitants. This allowed detailed, replicable characterization of
events. We tested the hypothesis that assault does not come “out of the
blue,” as assault motivated by psychosis is often described, but can be
predicted by immediate antecedent behaviors in victim and assailant. In
victims these immediate antecedents could be seen as provocations. As-

sailants who respond to victim provocation could be seen as reactive.
We detected 155 assaults between 59 patients. Individual patients
were often involved in multiple events. Fifty-six additional patients
were present on the ward during the study period but were never in-
volved in assaults. Of the group of 59, 56% carried a diagnosis of schizo-
phrenia.
Many of the incidents did not seem serious. To better define this im-
pression, we classed events into seriousness categories based on target
of blows (head or remainder of body), staff intervention, and perceived
forcefulness of blows. The categories were “play,” “warn,” and “hurt”
(Crowner et al. 1994). Twenty-one of 155 events were classed as “play”
assaults. In 76 assaults, patients seemed to be trying to hurt each other.
Fifty-seven events had intermediate seriousness, in which patients did
not seem to be playful or rageful but to be annoyed and trying to com-
municate a warning to the victim. One assault could not be classified.
In the 5 minutes before assaults, we found certain threatening and
intrusive behaviors, or cues, in both victims and assailants (Crowner et
al. 2005). The threatening behaviors were fist shaking, pointing, yelling
and arguing, bumping, shoving, and pushing. The intrusive behaviors
were following, touching, or moving very close, to within approxi-
mately 6 inches. These behaviors could be directed toward the other
member of the victim-assailant dyad or toward other people in the area.
We found these behaviors before 60% of assaults. In contrast, cues were
present in 10% of control periods. When we counted cues, we found
more than 10 times as many before assaults compared with control pe-
riods. These results remained highly significant after play assaults were
dropped from analysis.
When we looked at all assaults, we found threatening and intrusive
cues were significantly more numerous in victims than in assailants. Be-
fore the events of intermediate severity (“warn” assaults), there were
significantly more assailant intrusive cues and more victim-threatening
and intrusive cues. These assaults were more likely than others to have
any victim cue or any victim intrusive cue.
Many of assaults we detected on camera have been little studied or
noted by other investigators. As described in this section, we discov-
ered a series of interactions between patients who were soon afterward
involved in assaults, and between them and others on the ward. More
serious assaults had fewer victim cues than assaults of intermediate
severity. In serious assaults, assailants seemed intensely emotional. In-
termediate assaults seemed to have a communicative function in which
assailants were telling victims to stop or back off.
Summary of Assessment of Violence

Assault can be understood in many ways. The reactive/instrumental/
psychotic classes can be useful but may overlap and blend into each
other. Assault can be motivated by reactivity and psychosis. Psychotic
patients can react in paranoid and irritable ways to external events
rather than solely to internal states. Assault can be motivated by reac-
tivity and a wish to coerce or by psychopathy when it is an acceptable
means to make someone else do something such as stop “bothering”
the person. Assault can be done in anger or as an attempt to communi-
cate. Motivations expressed by patients cannot always be taken at face
value. Reasons patients give can be the same as those given by nonpsy-
chotic people, but as Taylor (1985) concluded, they may have “psy-
chotic underpinnings.”
Violence Management
General Principles
Inpatient management of violent patients is truly a team effort, requir-
ing consensus and collaboration with all disciplines, particularly—but
not exclusively—nursing. Psychiatrists have a legal and professional re-
sponsibility to lead this team by managing group dynamics and by
guiding treatment decisions. Assaultive patients very often stir up in-
tensely emotional conflicts within groups of staff members and be-
tween patient and professionals. Management decisions should be
guided by data, an empathetic engagement with the patient, and calm.
Staff are obliged to make all possible efforts to preserve patient dignity
and protect patient safety. To balance our obligations to patients with
our obligations to the safety of staff demands astute judgment and un-
usual skill.
The treatment team should have established guidelines for identify-
ing potentially dangerous situations and for choosing interventions to
prevent escalation of threatening behaviors. Criteria for intramuscular
medication given without patient consent and for seclusion and re-
straint should be clear, concrete, and accepted by the entire ward staff.
Management strategies vary according to local custom; staff will
usually turn to what they have done before and what seems to have
worked. However, interventions should be the least restrictive neces-
sary to ensure safety and should be based on available data. Restraint
and seclusion should be interventions of last resort, because patients
find them traumatic and humiliating and because these interventions
often lead to injuries of patients and staff. Use of restraint and seclusion
is reviewed in this volume (Chapter 17) and also by Bernay and Elver-
son (2000).
Pharmacological management of violent adults with schizophrenia
is discussed at length elsewhere in this volume (see Chapter 15). I only
mention briefly a few points here. Medication adherence is often key.
The physician must attempt to establish an alliance with the patient,
however tenuous, and delicately balance adverse and beneficial effects.
Court-ordered treatment may be necessary. When it is clear that the pa-
tient is taking medication, and it is not effective after an adequate trial,
clozapine can be helpful. Many have documented clozapine’s benefits
for violent patients, but Krakowski et al. (2006) completed a study di-
rectly comparing haloperidol, olanzapine, and clozapine in the treat-
ment of assaultive patients with schizophrenia and schizoaffective dis-
order. They found that clozapine was more efficacious than olanzapine,
and olanzapine was more efficacious than haloperidol. The antiaggres-
sive effects seemed to be above and beyond the antipsychotic and sed-
ative effects of the medications.
Predicting Who Is Likely to Become Violent

The best management of violence is prevention. Use the predictors dis-
cussed earlier and past behavior to identify patients who are likely to
become violent. The best predictor of future behavior is past behavior.
Learn the circumstance of past assault—for example, if the patient as-
saulted staff members or fellow patients when denied discharge or
when being placed in seclusion close to the time of admission. This in-
formation can be difficult to obtain reliably from patients alone, so de-
tailed past records are helpful. Circumstances associated with assault
may be avoided in the future or at least anticipated with watchfulness
and caution.
Designing and Managing Physical Space to

Prevent Violence
Assault prevention is an effort of all clinical and even managerial staff.
Reactive assaults can be prevented by minimizing potential provoca-
tions. Coercive assaults can be prevented by minimizing factors that
lead to victimization.
Ward design and furnishing is part of violence prevention. Wards
should be designed to optimize patient observation. Staff should ob-
serve for heated arguments, threats, and intrusive behaviors and inter-
vene quickly. Eliminate potential weapons such as chairs that can be
easily thrown and blood pressure cuffs set in heavy metal posts on
wheels. Eliminate places where weapons can be hidden, such as
dropped ceilings made of fiberboard squares that can easily be lifted.
Safeguard individual patient privacy with barriers in bathrooms, show-
ers, and bedrooms so personal intrusions and reactive, paranoid as-
saults are less likely. Safeguard personal possessions to decrease thefts.
Patients should have secure lockers with functional locks. Valuables
such as cash and cigarettes may be best held by staff.
Physical closeness and threats can predict assault. This has practical
significance for management of space between patients, for example in
lines, elevators, dining rooms, and dayrooms. Avoid crowding, bump-
ing, shoving, and pushing. Do not require patients to wait in lines for
medications or meals. In a large psychiatric hospital, patients were
escorted daily to off-ward programming. So that escorts could make
fewer trips, patients were packed shoulder to shoulder and elbow to
belly into elevators; in this setting, angry shouts and shoving were com-
mon. Shouts and shoving could easily lead to fistfights in a small en-
closed space. This is an excellent example of what not to do.
Managing Interpersonal Interactions to

Prevent Violence
Staff should strive to provide a predictable, orderly, safe, and respectful
environment. When patients feel threatened or unsafe, they are more
likely to be assaultive. Hospital staff members should avoid behavior
that could be seen as threatening; always remain calm and nonconfron-
tational, and never yell. Coercion should be minimal and in the interests
of patient safety, not staff convenience. Expectations should be simple,
clear, and rational. Rules that are complicated or inconsistent can be
confusing and seem hostile. A punitive approach is also usually seen as
hostile or aggressive. Swanson et al. (2006) found that patients who felt
listened to most of the time were less likely to be violent than those who
did not feel listened to. This could certainly apply to patients who feel
listened to by hospital staff.
Hospital staff members bear a responsibility to protect patients from
victimization by other patients through bullying and theft. This can be
done through vigilant observation and means to secure patient prop-
erty. When staff cannot provide justice, patients provide their own ver-
sion, often through violence.
Because threatening and intrusive behaviors may often precede
assault, staff should observe patients for these and intervene. These
behaviors occur in both parties in assaults, so interventions can target
potential victims as well as assailants. Psychiatrists should note these

behaviors and consider whether they are manifestations of illness that
could be treated. For example, a white man with mania was often the
target of assaults by fellow patients who were African American. This
was a mystery until he was seen and heard edging in on one and mut-
tering racial epithets. When he was more aggressively treated with lith-
ium, he sat for extended periods and was not assaulted.
In the community, targets of assaults are likely to be family mem-
bers, rather than fellow patients or staff members, and are more likely
to be involved in relationships with the assailant marked by mutual
hostility and financial dependence (Estroff et al. 1998). Family members
should attempt to disentangle themselves from such relationships, de-
creasing hostility and patients’ financial dependence.
Case Examples
The following is an example of a very serious assault; this type of assault
is rare and is not part of the studies discussed earlier, except perhaps
Taylor’s (1985). It illustrates that psychotic factors, particularly delu-
sional thinking, can be forgotten, suppressed, or repressed while non-
psychotic motivations are expressed.
Case Example 1
Mr. A was 48 years old when he was released on parole and committed
to a psychiatric hospital after serving a 12-year sentence for killing his
common-law wife with a kitchen knife. He related that she started a
fight with him over a coat, which she said he did not hang up properly.
As she came at him with a large fork, he stabbed her. “She got like that,”
he said. He denied he was drinking or using illicit substances at the time
of the crime. After the crime, he turned himself in at the local precinct.
As he lay in bed in prison, he would occasionally see and hear his
wife and long for her. “Don’t cry over spilt milk? I still cry over spilt
milk,” he said. “That saying is wrong.”
Records of a psychiatric admission 4 years before the crime revealed
that he had gone to his wife’s workplace with a knife because he wanted
to break the spell he believed she had cast over him. He also believed
she was controlling his mind. Of this admission Mr. A had few memo-
ries. He could only recall, “They say I was hearing voices.” When re-
minded of the documented circumstances of his previous admission, he
became visibly disturbed and said at that time he was acting a fool and
was ill.
The following case illustrates the connections among paranoia, irri-

tability, and assaultiveness, as well as treatment recommendations.
Case Example 2
Mr. B was a man in his twenties with a diagnosis of chronic schizophre-
nia and mild mental retardation. He had been continuously hospitalized
for at least 5 years. On the ward he was constantly irritable and paranoid,
especially about the food he was served, believing it was deliberately
contaminated with crack cocaine or cyanide. He also worried that he was
losing weight and that his body was changing in various ways.
Mr. B was seen daily, and he frequently would present a complaint.
He asked to see an ophthalmologist for poor near vision and a podiatrist
for burning feet and dry skin. He asked to sleep in an open seclusion
room because his room was too hot. He refused breakfast, saying he had
a cold that he could feel in his chest and head. He asked for cough syrup
and to stay in bed. He said he did not get his oral antipsychotic medica-
tion and that he could not live without it. Later the same day he com-
plained his mind was exploding because of medication, asked for a de-
creased dose, and said he needed none.
This patient could be mollified or put off sometimes, but never con-
vinced or brought around. He wanted to explain that all the violent in-
cidents he had been involved in were not his fault. After a few weeks of
such complaints, he punched another patient in the head because, as he
related, he was upset that he had not received a kosher diet. Fifteen min-
utes later while passing in the hall, his victim hit Mr. B, and Mr. B be-
came enraged. Staff members tried at length to calm him down, but he
hit one in the face.
After it became clear that Mr. B was ingesting little of his medication
consistently, a court order for fluphenazine decanoate was obtained. As
a result he had increased attention span, better ability to express himself
verbally, fewer paranoid and somatic complaints, and much less as-
saultiveness.
The following case illustrates an assault apparently resulting from

physical closeness between two men and paranoia in an environment
that was threatening and disorderly and where the assailant did not feel
listened to. The patient told of another assault that was reactive, but his
reactivity was likely due to his psychotic symptoms. It might also be
called antisocial, because he believed he did the right thing in beating an-
other man. A psychiatrist wrote that Mr. C’s actions were driven by his
personality structure and that he was not agitated or paranoid. This inpa-
tient psychiatrist released him to a shelter the day after Mr. C had beaten
another patient with a chair and his fists. In so doing, the psychiatrist may
not have appreciated that treatment might prevent future assaults.
Case Example 3
Mr. C was sent to a hospital emergency department after he punched
another man at the shelter where he was living. The other man, who was
much larger, subsequently punched Mr. C in the eye, causing an orbital

fracture. Once he was admitted to the psychiatric unit, Mr. C. expressed
a desire to stay at least 6 months because it was a nice environment, un-
like the shelter, which he called a “dog-eat-dog world” where other res-
idents stole and counselors ignored him. He explained that he punched
the man because he thought he was playing around behind his back,
trying to harm him. He realized he was mistaken and, in his own lan-
guage, “paranoid,” but he acknowledged that he had felt nervous with
other people behind his back as long as he could recall. In the previous
2 years, he had heard voices telling him he was nothing and was going
to be nothing.
He recalled another assault during a previous hospitalization in
which he hit another patient with a chair. He said that in this case he was
not paranoid but in his “natural senses” and “did right.” The other man
had been yelling and cursing at him in a violent, hostile voice to get off
the phone. Mr. C. became angry and hit him because he felt disre-
spected, as though the man were telling him he was worthless.
On the psychiatric unit he was tapered off valproate and ziprasi-
done and started on gradually increasing doses of olanzapine. Perhaps
more importantly, he was assigned a private room and allowed to stay
there whenever he pleased, by himself, even at meals. Within 3 weeks
he started venturing out to take part in group activities. Soon he was
seen chatting with his fellow patients, and his grooming and dress im-
proved. He was quiet, watchful, never threatening, and never involved
in physical fights. Mr. C was usually abrupt with his psychiatrist but ac-
cepted all treatments offered. He was thankful when offered transfer to
a state hospital because he believed transfer would offer him a chance
to leave the shelter system and obtain Social Security.
Key Points
■ Large community surveys find rates of violent behavior in adults
with schizophrenia are somewhat higher than in adults with no
diagnosed psychiatric illness but are lower than in groups with
substance abuse disorders.
■ Serious violence in a large group of patients with schizophrenia
has been associated with hostility, suspiciousness, persecutory
delusions, hallucinations, grandiosity, and excitement.
■ Negative symptoms were associated with a decreased risk of
serious violence in this same large group. High levels of positive
symptoms with high levels of negative symptoms were not linked
to increased risk of serious violence, but high levels of positive
symptoms with low levels of negative symptoms were.
■ Psychopathology and historical factors explain only a small

amount of the differences in rates of violence between patients
with schizophrenia.
■ Assaults occur in an interpersonal context and rarely “out of the
blue.” They can be understood as reactive, coercive, or arising out
of psychosis—or, more likely, some combination of these.
■ Assaults are often preceded by identifiable behaviors in both the
patient who hits and the patient who gets hit: namely, threaten-
ing behaviors (e.g., arguing and fist shaking) and intrusive behav-
iors (e.g., getting very close).
■ Assault can be less likely if staff observe patients closely and main-
tain an orderly, predictable, and respectful milieu.
■ Inpatient management of violent patients is a team effort requir-
ing consensus and collaboration with all disciplines.
■ Management decisions should be guided by data and vigorous
efforts to preserve patient dignity and protect patient and staff
safety. Interventions should be the least restrictive necessary to
ensure safety.
■ Clozapine has been shown to be more efficacious than olanzapine,
and olanzapine to be more efficacious than haloperidol, in con-
trolling assaultiveness in inpatients with schizophrenia and
schizoaffective illness. The antiaggressive effects of these drugs
seem to be above and beyond the antipsychotic and sedative
effects.
■ The best management is prevention. Violence may be prevented
by identifying patients likely to become violent on the basis of
known predictors or past behavior. Violence may be prevented
by foresighted design of physical space and by providing a safe,
predictable milieu.
■ Physical closeness and threats can predict assault. Avoid situa-
tions that lead to patient crowding, bumping, shoving, and
pushing.
■ Family members should try to disentangle themselves from rela-
tionships with patients marked by mutual hostility and patients’
financial dependence.
References
Disorders, 3rd Edition. Washington, DC, American Psychiatric Associa-
tion, 1980
Arseneault L, Moffitt TE, Caspi A, et al: Mental disorders and violence in a total
birth cohort. Arch Gen Psychiatry 57:979–986, 2000
Bernay LJ, Elverson DJ: Managing acutely violent inpatients, in Understanding
and Treating Violent Psychiatric Patients. Edited by Crowner ML. Wash-
ington, DC, American Psychiatric Press, 2000, pp 49–68
Braham LG, Trower P, Birchwood M: Acting on command hallucinations and
dangerous behavior: a critique of the major findings in the last decade. Clin
Psychol Rev 24:513–528, 2004
Campbell M, Schulman D, Rapoport JL: The current status of lithium therapy in
child and adolescent psychiatry. J Am Acad Child Psychiatry 17:717–720,
1978
Campbell M, Cohen IL, Small AM: Drugs in aggressive behavior. J Am Acad
Child Psychiatry 21:107–117, 1982
Campbell M, Small AM, Green WH, et al: Behavioral efficacy of haloperidol and
lithium carbonate: a comparison in hospitalized aggressive children with
conduct disorder. Arch Gen Psychiatry 41:650–656, 1984
Campbell M, Adams PB, Small AM, et al: Lithium in hospitalized aggressive
children with conduct disorder: a double-blind and placebo-controlled
study. J Am Acad Child Adolesc Psychiatry 34:445–453, 1995
Crowner ML, Stepcic F, Peric G, et al: Typology of patient-patient assaults de-
tected by videocameras. Am J Psychiatry 151:1669–1672, 1994
Crowner M, Peric G, Stepcic F, et al: Psychiatric patients’ explanations for as-
saults. Psychiatr Serv 46:614–615, 1995
Crowner ML, Peric G, Stepcic F, et al: Assailant and victim behaviors immedi-
ately preceding inpatient assault. Psychiatr Q 76:243–256, 2005
Eichelman BS: Neurochemical and psychopharmacologic aspects of aggressive
behavior. Annu Rev Med 41:149–158, 1990
Estroff SE, Swanson JW, Lachiocotte WS, et al: Risk reconsidered: targets of vi-
olence in the social networks of people with serious psychiatric disorders.
Soc Psychiatry Psychiatr Epidemiol 33:S95–S101, 1998
Kay SR, Wolkenfeld F, Murrill LM: Profiles of aggression among psychiatric pa-
tients, II: covariates and predictors. J Nerv Ment Dis 176:547–555, 1988
Krakowski M, Czobor P, Chou JC: Course of violence in patients with schizo-
phrenia: relationship to clinical symptoms. Schizophr Bull 25:505–517, 1999
Krakowski M, Czobor P, Citrome L, et al: Atypical antipsychotic agents in the
treatment of violent patients with schizophrenia and schizoaffective disor-
der. Arch Gen Psychiatry 63:622–629, 2006
Link BG, Stueve A, Phelan J: Psychotic symptoms and violent behaviors: prob-
ing the components of “threat/control-override” symptoms. Soc Psychia-
try Psychiatr Epidemiol 33:S55–S60, 1998
arthur Study of Mental Disorder and Violence. New York, Oxford Univer-
sity Press, 2001
Nolan KA, Czobor, P, Biman R, et al: Characteristics of assaultive behavior
among psychiatric inpatients. Psychiatr Serv 54:1012–1016, 2003
Quanbeck CD, McDermott BE, Lam J, et al: Categorization of aggressive acts
committed by chronically assaultive state hospital patients. Psychiatr Serv
58:521–528, 2007
Sheard MH, Marini JL, Bridges CI, et al: The effect of lithium on impulsive ag-
gressive behavior in man. Am J Psychiatry 133:1409–1413, 1976
Stueve A, Link BG: Violence and psychiatric disorders: results from an epidemi-
ological study of young adults in Israel. Psychiatr Q 68:327–342, 1997
Swanson JW, Holzer CE, Ganju VK, et al: Violence and psychiatric disorder in
the community: evidence from the Epidemiologic Catchment Area sur-
veys. Hosp Community Psychiatry 41:761–770, 1990
Swanson JW, Swartz M, Van Dorn RA, et al: A national study of violent behav-
ior in persons with schizophrenia. Arch Gen Psychiatry 63:490–499, 2006
Taylor PJ: Motives for offending among violent and psychotic men. Br J Psychi-
atry 147:491–498, 1985
C H A P T E R 7
Posttraumatic Stress
Disorder
Posttraumatic stress disorder (PTSD) is classified as an anxiety disor-

der and is defined by exposure to a severe traumatic event and the pres-
ence of a cluster of symptoms (American Psychiatric Association 2000).
Individuals must have experienced, witnessed, or learned of an event
that involved death, injury, or threat to physical integrity and reacted to
that event with intense fear, helplessness, or horror. To meet diagnostic
criteria, traumatically exposed persons must also have symptoms of re-
experiencing the event, avoidance of reminders of the event or numb-
ing of responsiveness, and increased symptoms of arousal or vigilance.
The symptom pattern must be present for more than 1 month and result
in clinically significant distress or impairment of functioning.
PTSD was first formally recognized by the psychiatric community
in 1980 (American Psychiatric Association 1980). During and after the
war in Vietnam, clinicians observed that a substantial portion of sol-
diers were experiencing protracted problems with readjustment into ci-
vilian society due to symptoms related to their wartime experiences. A
core constellation of psychiatric symptoms became the basis for diag-
nostic criteria for PTSD, but other difficulties were also observed.
Among these were intermittent acts of aggression or violence. Other
123
symptoms associated with PTSD, but not required for the diagnosis, in-
clude impaired interpersonal relationships, impaired affect modula-
tion, self-destructive and impulsive behaviors, feelings of constant
threat, and changes in personality characteristics. The American Psy-
chiatric Association’s practice guidelines note that some individuals
with PTSD have an increased expectation of danger that results in an
“anticipatory bias” in their perception of their environment and in-
creased readiness for “flight, fight, or freeze” responses (Ursano et al.
2004). This increased readiness for aggression may take the form of a re-
duced ability to tolerate mild or moderate slights, resulting in acts that
are disproportionate to the degree of provocation. Other psychiatric
conditions commonly comorbid with PTSD include major depression,
substance abuse disorders, and other anxiety disorders (American Psy-
chiatric Association 2000).
Almost all studies of aggression and violence in patients with PTSD
have been conducted among combat veterans from the Vietnam era. Vi-
olence in patients who develop PTSD in response to sexual assaults,
physical assaults, motor vehicle crashes, acts of terrorism, or exposure
to natural and manmade disasters has not been well studied. In contrast
to other patients with PTSD (e.g., those whose PTSD results from a mo-
tor vehicle accident or an isolated sexual assault), war veterans during
combat will have experienced extended periods of heightened vigilance
and arousal lasting weeks to months and extreme and repeated inter-
personal violence. Exposures include being shot at by enemy forces,
killing enemy forces, and sometimes being responsible for the wound-
ing or death of noncombatants. Under certain wartime conditions,
some may also witness or participate in repeated non-warfare acts of
abusive violence or killing of prisoners or civilians (Laufer et al. 1984).
Some civilian law enforcement officers may also have similar, but less
repeated, exposures.
Veterans with PTSD demonstrate higher levels of anger, problems
with anger regulation, increased levels of criminality, increased levels of
violence, and greater potential for serious acts of violence when com-
pared with other patient populations. Domains of anger problems in-
clude inaccurate perception and processing of environmental cues,
heightened physiological and emotional activation, and behavioral in-
clinations to act in antagonistic or confrontational ways (Chemtob et al.
1997). Patients with regulatory deficits in all three domains display an-
ger and aggression that has been labeled a “ball of rage” (Chemtob et al.
1997).
Although the literature on violence and PTSD is extensive, findings
between studies have shown multiple possible explanations for this
Posttraumatic Stress Disorder ❘ 125
association and leave open many questions. Are those exposed to serious
personal threat more likely to have come from troubled backgrounds
prior to the trauma? Does exposure to violence lead to future acts of vi-
olence directly, or is it mediated through the development of PTSD?
What is the role of comorbid substance use and violence?
Perhaps due to the complex number of pathways to violence, mod-
els to predict future acts of violence among veterans with PTSD have
not shown useful predictive value. Among one group of veterans with
PTSD, demographic variables, exposure to atrocities, severity of PTSD
symptoms, severity of drug and alcohol problems, past violent behav-
iors, past suicidal behaviors, and prior treatment information were
used in an attempt to develop such a model (Hartl et al. 2005). Only
prior violence history was useful in predicting postdischarge violence;
PTSD and depression severity were both poor predictors of high- and
low-risk group membership.
Although no precise model exists for predicting violence among pa-
tients with PTSD, there are identified risk factors for future violence
that can be the focus for management and treatment. Research during
the past three decades has examined multiple risk factors in an effort to
determine which seem most strongly associated with the violence in pa-
tients with PTSD.
Risk Factors for Violence or Aggression

Childhood traumas, level of combat exposure, PTSD symptoms and se-
verity, number of combat roles, exposure to atrocities, and preservice
antisocial behaviors have all been examined in relationship to later an-
tisocial behavior and violence. The studies often used different mea-
sures, controlled for different potentially contributing variables, and
sometimes had conflicting findings. When examined together, preser-
vice antisocial behavior and level of combat exposure were associated
with postservice antisocial behavior, including incidents of violence,
other nonviolent illegal behaviors, occupational problems, and nonvio-
lent interpersonal problems (Resnick et al. 1989). Number of combat
roles, subjective stress in combat, number of specific stress exposures,
and total PTSD symptom severity have all been associated with postser-
vice assault and weapons charges (Wilson and Zigelbaum 1983).
Among participants in the National Vietnam Veterans Readjustment
Study (NVVRS; Kulka et al. 1990), male veterans with PTSD reported an
average of 13.3 acts of violence in the preceding year compared with 3.5
acts of violence in those without PTSD. They were also 1.5 times more
likely to have been arrested or jailed and 3 times as likely to have been
convicted of a felony crime. In another analysis of the NVVRS data, pre-

military behaviors and experiences and postservice PTSD were both as-
sociated with postservice antisocial behavior (Fontana and Rosenheck
2005).
Compared with other psychiatric inpatients, veterans hospitalized
with severe PTSD were seven times more likely to have engaged in one
of more acts of violence in the 4 months prior to hospitalization, six
times more likely to have destroyed property, six times more likely to
have threatened others without a weapon, four times more likely to
have engaged in physical fights, and three times more likely to have
made threats with a weapon (McFall et al. 1999). Severity of PTSD
symptoms was also associated with increased risk to make threats of vi-
olence without a weapon, engage in physical fights, and make threats
with a weapon. Among veteran psychiatric inpatients with any diagno-
sis, veterans with combat exposure were more likely to engage in as-
saults or assault-related behavior during hospital admission than veter-
ans without such experiences (Yesavage 1983).
Premilitary problems, exposure to war zone atrocities, and postwar
problems were common among veterans with PTSD (Hiley-Young et al.
1995). One-third of veterans reported childhood physical abuse and ap-
proximately one-half endorsed one or more significant adolescent be-
havioral problems. Eighty-six percent endorsed witnessing abusive
war zone violence, and 91% both witnessed and participated in abusive
violence (hurting, killing, or mutilating Vietnamese). Postmilitary prob-
lems included violence toward their spouse (58%), violence toward oth-
ers (71%), drug problems (62%), and alcohol problems (73%). Interest-
ingly, no association between premilitary factors and postmilitary
violence or criminal behavior was found. Participation in killing during
war was associated with postmilitary violence toward others and to-
ward spouses.
One of the few studies that examined the association between PTSD
and violence in individuals who developed PTSD as a consequence of
mostly non-wartime experiences was conducted in a population of
1,140 incarcerated male felons (Collins and Bailey 1990). Prison arrest
records indicated that 14% were currently incarcerated for acts of ex-
pressive violence (homicide, rape, or aggravated assault). Only 2.3% of
the sample met study criteria for presence of PTSD. Of those, 31% re-
ported combat trauma. Although most inmates did not meet criteria for
the disorder, 795 (70%) endorsed at least one of nine symptoms of
PTSD. When demographic variables, antisocial characteristics, and sub-
stance abuse were controlled for, those who met criteria for the diagno-
sis of PTSD were 4.58 times more likely to be currently incarcerated for
homicide, rape, or assault and were 6.75 times more likely to have been
arrested for violence within the past year. Among those who did not
meet full criteria for PTSD, the presence of each additional symptom of
PTSD increased risk of current incarceration for a violent crime (odds
ratio [OR] 1.22) and for arrest for violence in the past year (OR 1.26). Of
those arrested for a violent crime who endorsed at least one symptom
of PTSD (N=80), most reported the PTSD symptoms began 1 or more
years prior to the arrest. This suggested that the presence of the symp-
toms may have contributed to the commission of the crime.
Family Violence
Patients with PTSD may direct aggression toward intimate partners. On
the Standard Family Violence Index (throwing something at someone,
pushing, grabbing, shoving, slapping, kicking, biting, hitting, beating
up, threatening with a gun or knife, or using a gun or knife on some-
one), veterans with PTSD endorsed an average of 22 such acts in the
past year (Beckham et al. 1997). In contrast, combat veterans without
PTSD endorsed an average of 0.2 such acts in the past year. Socioeco-
nomic status, aggressive responding, and PTSD severity were associ-
ated with increased violence. Yet another study found that presence of
PTSD may mediate the effect of combat exposure on later intimate part-
ner violence (Orcutt et al. 2003). Among multiple studies of Vietnam-
era veterans, past-year partner violence rates range from 13% to 58%,
with higher rates generally seen among inpatients with substance de-
pendence, PTSD, or other psychiatric disorders (Marshall et al. 2005).
PTSD severity was also correlated with partner abuse severity. Partner
physical abuse has also been associated with interactions of alcohol
consumption (frequency and amounts) and severity of hyperarousal
symptoms (Savarese et al. 2001). Higher rates of depression and drug
abuse are seen in veterans who had engaged in partner violence (Taft et
al. 2005).
In a study of veterans with either PTSD or depression, but not both
conditions, those in each group endorsed similar rates of partner vio-
lence (roughly 80%) and severe partner violence (roughly 40%) during
the past year (Sherman et al. 2006). Compared with control couples in
which the veteran did not currently meet criteria for a serious psychiat-
ric illness, those with either depression or PTSD were twice as likely to
endorse any act of partner violence and four times as likely to endorse
an act of severe partner violence. The study did not include veterans
with comorbid depression and PTSD, so it did not assess the relation-
ship of comorbid illness and partner violence.
Firearm Ownership and Firearms Behaviors

Possession of firearms or presence of firearms within the household
may increase the risk of potential serious violence toward others or may
elevate the risk of a successful suicide act. Compared with veterans
with substance use problems, veterans with PTSD reported owning
more than four times as many total firearms (mean, 3.2 vs. 0.72), more
than five times as many handguns (mean, 1.6 vs. 0.28), and five times as
many rifles or shotguns (mean, 4.3 vs. 0.86). Interestingly, there was no
difference in overall gun ownership between the two groups prior to
military service (mean, 1.69 vs. 1.68) (Freeman and Roca 2001). Twenty-
two percent endorsed aiming a gun at a family member; 21% endorsed
firing a gun within their house; 39% endorsed firing a gun to protect
home, self, or family; and 54% endorsed holding a loaded gun with sui-
cide in mind. In a separate study, 33% of the PTSD group endorsed car-
rying a gun on their person at least some of the time, and 33% endorsed
killing or mutilating an animal “in a fit of rage” (not while hunting)
(Freeman et al. 2003). Both studies were conducted among clinical sam-
ples of veterans with chronic combat-related PTSD. The combination of
firearm-related aggressive acts and the presence of numerous firearms
in homes of veterans with PTSD suggest a strong potential for lethal vi-
olence against others or successful suicide.
Suicide
Patients with PTSD are also at increased risk of suicide or suicide at-
tempts. Comorbidity of PTSD and other psychiatric conditions is com-
mon, and a substantial portion of patients with PTSD are diagnosed with
three or more other conditions (Brady et al. 2000a). The most commonly
comorbid conditions are depressive disorders, substance use disorders,
and other anxiety disorders, all of which are associated with an increased
risk of suicide. In one study, patients with comorbid depression and
PTSD were at increased risk of suicide attempts compared with patients
with only depression (Oquendo et al. 2003). In a second study, the pres-
ence of Cluster B personality disorders (paranoid, narcissistic, border-
line, or antisocial personality) in addition to PTSD and depression fur-
ther increased the risk of suicide attempts (Oquendo et al. 2005). In both
of these studies, the majority of subjects were non-veteran women.
Subthreshold PTSD can also develop after exposure to traumatic
events. Individuals not meeting full diagnostic criteria for the disorder
experience comparable levels of impairment and suicidality when com-
pared with patients who meet full criteria for the disorder (Zlotnick et
al. 2002). In one large national screening study, roughly one in four sub-
jects reported at least one PTSD symptom of at least 1 month’s duration
(Marshall et al. 2001b). Functional impairment, number of comorbid
disorders, presence of a depressive disorder, and current suicidal ide-
ation increased linearly and statistically with each increasing additional
PTSD symptom. Individuals with subthreshold PTSD were at greater
risk of suicidal ideation even after controlling for the presence of a de-
pressive disorder. These studies highlight the importance of screening
all patients with a history of trauma for presence of PTSD symptoms
that may increase risk of suicide or suicide attempts.
Assessment and Management of

Posttraumatic Stress Disorder and Violence
There are numerous guidelines for the assessment and clinical manage-
ment of PTSD (National Center for PTSD 2004; Ursano et al. 2004; VA/
DOD Clinical Practice Guideline Working Group 2004). All guidelines
suggest that management should be prioritized according to the degree
to which each symptom or behavior is causing distress or loss of func-
tion or may affect future safety. A high percentage of patients with
PTSD experience comorbid conditions such as depression or substance
abuse. Such comorbid conditions must also be evaluated and may need
to be addressed first, because they may be the source of greatest risk for
morbidity or future dangerousness.
Knowing the nature of the events leading to the development of
PTSD is of key importance in assessing potential future dangerousness.
Combat exposures and direct interpersonal violence, such as physical
assault, appear much more likely to lead to PTSD-associated violence
than traumas such as motor vehicle crashes or natural disasters. Areas
to inquire about when assessing combat veterans or others who have
experienced extreme acts of interpersonal violence are outlined in Table
7–1. Patients should be asked to elaborate on the details, frequency, and
duration of each endorsed experience.
As with all psychiatric evaluations, patients with PTSD should be
questioned about present suicidal ideation and past suicidal behaviors.
High rates of comorbid depression, substance use disorders, and ten-
dency toward firearm ownership all increase the risk of suicide as well
as the risk of harm to others. Each of these areas should be carefully as-
sessed in both acute and chronic care settings. Because spouses are often
the most available target of violence, patients should be asked about
patterns of interaction and conflict resolution within relationships.
If their responses are guarded or inconsistent, it may be necessary to
TABLE 7–1. Violence risk factor assessment in the evaluation of

patients with posttraumatic stress disorder
Have you been the victim of a violent sexual or physical assault?
—How many times have you been assaulted?
—Did the assault(s) involve the use of a weapon?
Have you been in combat?

—Have you killed or wounded another in combat?
—Did you participate in or observe killing, mutilation, or torture of
civilians?
Are there specific settings or events that cause you to become irritable or
“on guard”?
Have you been involved in a physical altercation within the past 6 months?
Do you own a firearm?

—Do you keep it loaded?
—Do you carry a firearm on your person or keep one “at arm’s length”?
—Have you ever pointed a firearm at another person as a warning or threat?
contact family members for corroborating information. If spousal abuse

is active and severe, court protective orders or other protective actions
may be needed until other solutions can be developed.
If patients endorse angry or hostile attitudes or are the victims of vi-
olent interpersonal assault they should be asked about their own history
of violent acts. Frequency, severity, and time duration since most recent
episode should be obtained. Potential screening questions and follow-
up elaboration questions are provided in Table 7–2. For each past act of
violence, patients should be questioned about the specific events lead-
ing up to the incident; specific provocation by the target of their vio-
lence; whether alcohol or drugs were involved; and how their current
situation, condition, attitudes, and recent behaviors differ from those
present at the time of the prior act. If current conditions closely parallel
those present at the time of past acts of violence, specific behavioral
“trigger avoidance” or “emotional defusing” plans should be devel-
oped and rehearsed in the clinician’s office. Shortened intervals between
treatment sessions, warnings to individuals specifically at risk, and pos-
sibly hospitalization or other protective interventions should also be
considered. In all instances, the treatment record should reflect the com-
ponents of the risk assessment and management decision process.
Clinicians should always be vigilant for their own safety. In emer-
gency department settings, agitated, intoxicated patients with PTSD
TABLE 7–2. Acts of aggression inventory: “In the past year,

have you…”
Event inventory
1. Been involved in a physical or verbal altercation with a stranger?
2. Been involved in a physical or verbal altercation with an acquaintance?
3. Been involved in a physical or verbal altercation with a spouse or relative?
4. Hit, kicked, or otherwise harmed or killed an animal in anger?
5. Damaged property as a consequence of being angry?
6. Contemplated or attempted suicide?
For each positive response:
a. Were you under the influence of alcohol or drugs at the time?
b. Did you have in your possession a firearm, knife, or other weapon?
c. Did you use or consider using the weapon?
d. When was the last time such an incident occurred?
e. What were the specific circumstances that led up to the event?
f. What was the outcome of the event?
g. How did it end?
h. Did you feel your behavior was appropriate under the circumstances?
i. Would you likely respond the same way in a similar situation?
j. How commonly would you encounter similar situations?
may need to be relocated to quieter and less distracting settings. Per-

sonal belongings and clothing should be checked for firearms or other
weapons. All personnel should be trained in emergency response and
restraint techniques. Hospitalized patients with PTSD should be care-
fully assessed for potential violence prior to discharge. Follow-up visits
should be scheduled to occur shortly after discharge, preferably with a
provider known to the patient. Family members should be educated on
signs of pending violent behavior and given direction on methods for
obtaining an emergent reevaluation or engaging other safety plans
(such as leaving the home or calling police) if they perceive a threat of
violence.
Shoenfeld et al. (2004) provided an overview of pharmacological
treatments for PTSD. Most studies of treatment for PTSD have been in
non-veteran populations and have not examined the specific efficacy of
these agents on symptoms of aggression or irritability. The selective se-
rotonin reuptake inhibitors (SSRIs) have been shown to be effective,
well tolerated, and safe in treatment of non–combat-related PTSD. Ser-
traline was effective during the 12-week acute and 24-week continua-
tion stages of treatment, with improvements seen in intrusive symp-

toms, avoidance symptoms, and arousal symptoms (Brady et al. 2000b;
Davidson et al. 2001b; Londborg et al. 2001). The mean dosage at com-
pletion was roughly 150 mg/day. Further ongoing treatment was also
effective in preventing relapse of PTSD (Davidson et al. 2001a), and
study participants reported improvements in quality of life and func-
tional measures. Participants who discontinued the drug had a worsen-
ing of symptoms and a decline in quality of life (Rapaport et al. 2002).
Roughly 80% of the participants in these trials were women, and only
about 5% had PTSD as a consequence of combat experiences, so the de-
gree of benefit of sertraline in combat veteran populations is not known.
Similar response rates and improvements in symptoms and function
were seen in controlled studies of paroxetine versus placebo in the treat-
ment of PTSD (Marshall et al. 2001a). The efficacy of 20 mg/day was
comparable with that seen using 40 mg/day. The majority of partici-
pants were women, and only 5%–7% had PTSD as a consequence of
combat exposure. In one study examining the efficacy of fluoxetine in
treatment of PTSD, the majority of the participants were men (80%), and
more than half had PTSD from combat experience or other wartime
exposures (Martenyi et al. 2002). Dosages in the range of 60 mg/day re-
duced symptoms; however, the response was not as robust as in other
SSRI studies. The investigators did not attempt to analyze the effect of
trauma type on treatment response.
Recent studies have shown efficacy for prazosin (an α1 receptor–
blocking antihypertensive medication) in reducing nightmares, im-
proving sleep quality, reducing psychological responses to trauma cues,
and improving global clinical status (Daly et al. 2005; Raskind et al.
2006; Taylor et al. 2006). These results suggest that prazosin might also
be of benefit in reducing irritability and aggressive behavior. Many case
and case series reports also suggest the use of other antidepressants,
mood stabilizers, and atypical antipsychotic medications for augmenta-
tion treatment of refractory PTSD symptoms, including anger and irri-
tability, that may be tied to potential acts of violence (Friedman 2006;
Schoenfeld et al. 2004).
Among available psychotherapeutic choices, cognitive-behavioral
treatments have been shown to be most effective in treating patients
with PTSD (Ursano et al. 2004). Within this class of treatments, both pro-
longed exposure therapy (guided imagery of the events and in vivo ex-
periences) and cognitive therapy or cognitive processing therapy (cor-
rection of distorted perceptions or appraisal of events) have been
shown to have benefit in trauma survivors. As with most clinical trials
of PTSD treatments, the early studies have mostly involved women
with sexual assault histories or other single-event traumas rather than

PTSD arising from combat.
Case Example 1
A 39-year-old Drug Enforcement Agency (DEA) officer was medically
retired 2 years ago after he had been shot in the face at close range dur-
ing a drug raid. During his 15 years with the agency he had seen
multiple shootings and had observed fellow officers killed in the line of
duty. During covert assignment in South America he had seen drug
smugglers torture, kill, and mutilate the bodies of rival gangs. In the
past 5 years he had had frequent nightmares with themes of killing and
pervasive danger, from which he awakened sweating and shaking. He
became progressively withdrawn, ultimately divorced his wife of
10 years, and no longer visited or spoke with family members. He was
constantly vigilant of his environment and startled at the sound of loud
noises. He presently has frequent suicidal ideation but relates no history
of acts of self-harm. He has a concealed weapons permit and carries a
concealed handgun on his person whenever he leaves his apartment. He
experienced some improvement in his depressed mood, lack of plea-
sure, and poor sleep after being started on sertraline by his primary care
physician 4 months ago. He has nightmares nearly every night and has
gradually increased his alcohol use to a pint of vodka per night, con-
suming it between 4:00 and 11:00 P.M. Some mornings he has little
recollection of his activities the prior evening. He was referred to an out-
patient practice by his primary care physician.
Initial management should consist of a detailed history of trauma

events, current symptoms of PTSD, and presence of other comorbid
conditions including depression, history of alcohol use, and drug use.
The immediate focus is on safety for the patient and others. Past violent
acts and firearm-related behaviors should also be explored. Heavy alco-
hol use, presence of a firearm, loss of social supports, and suicidal ide-
ation all add to risk. Establishment of rapport and trust may be difficult
in an individual who would typically avoid mental health professionals
because of career concerns. He will also not likely agree to relinquish his
firearm. Because of the potentially depressive effects and disinhibition
caused by heavy alcohol use, this is the first area of treatment. He is at
risk for complicated withdrawal, so a careful withdrawal history must
be obtained and inpatient detoxification should be considered. If he is a
safe candidate for outpatient withdrawal, he should be monitored daily
for the first week after discontinuation and provided benzodiazepines
to ease the symptoms and prevent seizures. Benzodiazepines would
also assist with his sleep disturbance acutely, but they should not be
used for maintenance treatment. A trial of prazosin should be consid-
ered to reduce the frequency and severity of nightmares, and his sertra-
line dosage may need to be titrated to ensure optimal response. Cogni-
tive-behavioral therapy should be initiated to examine the accuracy of
his perception of threats, establish future goals and direction, identify
“triggers” for possible aggressive acts and develop alternative response
choices, and reestablish communication and social supports. To assist
with abstinence from alcohol and to establish lifestyle changes, referral
to a self-help group such as Alcoholics Anonymous may be useful.
Case Example 2
A 42-year-old career law enforcement officer was referred by his em-
ployee assistance program provider for an evaluation of possible PTSD
and medication treatment. During his career, the officer had seen multi-
ple partners wounded in the line of duty and had been shot at on three
occasions. One of these shootings resulted in a minor wound. He had re-
cently been reprimanded for excessive use of physical force during an
arrest, when he repeatedly struck a suspected drug dealer with his ba-
ton in response to verbally abusive statements. He has loud, verbally
abusive fights with his wife, but these have not escalated to physical vi-
olence. On his screening questionnaire, he reported “sleep problems”
and “anger control issues” as his primary concerns. He is on no medica-
tions and has no prior psychiatric treatment.
On the basis of his history of exposures, this patient may have at

least some symptoms of PTSD. A careful trauma history should be
gathered and a thorough review conducted for symptoms of PTSD, de-
pression, substance use, and violent acts toward self and others. If sub-
stance use is not a significant problem it would be best to focus on the
patient’s stated problems initially to help cement a therapeutic relation-
ship. To reduce the potential for disinhibition in this potentially aggres-
sive patient, non-γ-aminobutyric acid (GABA) medications such as
trazodone or ramelteon may be preferable to benzodiazepines or sleep
agents such as zolpidem. If significant symptoms of PTSD or depres-
sion are present, an SSRI medication may be helpful for those condi-
tions and may also resolve sleep problems. Anger control issues would
be best managed with cognitive-behavioral therapy focused on themes
and situations likely to cause anger and on developing alternative re-
sponse patterns for such situations. At least one appointment with the
patient’s spouse would also be beneficial to obtain collateral informa-
tion about her observations of the patient in comparison with his recol-
lections. The need for ongoing couples therapy could then also be as-
sessed. Informal peer counseling with another senior officer may also
be available through the department. This setting could assist the pa-
tient in further expanding his repertoire of response patterns in anger-

provoking situations.
Case Example 3
A 26-year-old married National Guard sergeant had completed two
tours in Iraq (20 months total) and had just been released from active
duty to resume his civilian employment. He was self-referred to an out-
patient clinic with symptoms of intrusion, avoidance, emotional numb-
ing, hypervigilance, and arousal. Like many of his friends from the war,
he carries a loaded pistol in his car and sometimes on his person, “be-
cause I feel naked and vulnerable without it.” He does not hold a con-
cealed weapons permit. The patient’s wife is concerned that whenever
the patient is around persons who appear to be of Western Asian origin
he becomes notably agitated and overly reactive to any movement on
their part. At times she has feared that he would draw his pistol and use
it. This behavior is worse on days following nightmares with themes
pertaining to the war. The patient drank heavily when he first returned
from deployment but cut back when his wife threatened to leave him.
The initial assessment should include a detailed trauma history with

information specific to each incident and the patient’s emotional re-
actions to each incident. All prior acts of violence by the patient should
also be reviewed with regard to the situation, persons involved, and out-
comes. Because the patient cannot legally carry a concealed weapon, this
behavior should be carefully explored with him in terms of his knowl-
edge of the law and likely consequences of breaking it, the risks and ben-
efits of being armed under such circumstances, the reality of perceived
threat conditions, and alternative means of self-protection. As part of
weekly cognitive-behavioral sessions, he should review incidents of en-
counters with persons of Western Asian heritage, the exact nature of the
situation, his observations, his appraisal of the situation, alternative ex-
planations of the situation, his behavioral responses, and the outcome of
any such interactions. The goal would be to develop a more realistic ap-
praisal of threat in relatively safe civilian settings. His wife should be en-
listed as a collaborator in the therapy process to assist in calming the pa-
tient in circumstances of perceived threat and to provide her observations
of his behavior to the therapist as treatment progresses. One of the SSRI
medications may be beneficial in reducing the symptoms of intrusion,
avoidance, numbing, and hypervigilance/arousal. Prazosin may also be
helpful in reducing the frequency and severity of nightmares.
Case Example 4
A 23-year-old active-duty Army sergeant presented to the clinic at the
urging of his wife. She has requested a divorce—and then recanted her
request—on three occasions since his return from a year-long tour in Af-
ghanistan 3 weeks ago. The sergeant notes that he believes his wife “fell
in with the wrong crowd” while he was deployed and “stayed out late
partying and messing with drugs.” She has acknowledged being un-
faithful on one occasion. She told the sergeant she quickly broke things
off with the man (whom the sergeant knows), but she continues to
makes excuses to spend time away from home. The sergeant believes
she is either continuing to see this man or is using drugs with her new
friends. She usually leaves the house after an argument about household
responsibilities. The sergeant reports he has been excessively irritable
and angry because she “isn’t keeping the house up like she did before I
left. She’s more concerned about her friends than about me.” He further
reports that he becomes filled with desperation when he thinks of his
wife leaving him. He acknowledges that he has punched walls and
kicked a door after her abrupt departures, but he has neither threatened
nor assaulted his wife. He explains that she is the “only girl that ever
loved me, so I could never hurt her,” but he acknowledges that when he
thinks of her with that other guy he gets so angry he sees “flashes of me
just choking him—or maybe her, and I can’t get those out of my head for
10 or 20 minutes until I turn the radio up loud and smoke a cigarette.”
He reports initial insomnia and restlessness, particularly when his
wife rejects his sexual advances. His sleep is further interrupted by
nightmares related to his experience in Afghanistan. He has become so-
cially withdrawn, noting that “I don’t want to go out with my wife’s
friends because they all know what she’s been up to while I’ve been
away—and they don’t know what I’ve been through anyway.” He de-
nies appetite, weight, energy, or concentration changes or any suicidal
thoughts. He notes “I was depressed [at age 11] when my parents di-
vorced. I talked to a counselor every week for 6 months back then, and
I took Prozac—but I don’t feel that way now.” He was raised by his
mother after she divorced his physically abusive father. He had very
few friends (no close friends and no girlfriends) and preferred to be
alone. There is no history of alcohol or illicit substance use. He joined the
military immediately after high school and had been successful in spe-
cial operations training. He married his wife in a courthouse ceremony
after a brief courtship, “mostly so that she could get away from her par-
ents and get benefits while I was deployed.” During deployment he
thought frequently about the life they would have together upon his re-
turn. Now he becomes “just so angry inside” when he recognizes that
these dreams may not be realized.
This case highlights the potential complications of current psycho-

social circumstances and chronic patterns of coping to the management
of a potentially violent patient. The soldier reports some symptoms of
PTSD, but the feelings of abandonment precipitated by his wife’s ac-
tion, more so than his PTSD symptoms, may prove to be the triggers of
interpersonal violence. Further quantification of current PTSD and
depressive symptoms is important. However, clarification of his wife’s
desires with regard to continuing the marriage, and efforts to help

the service member reframe the implications of his wife’s abrupt depar-
tures (whether they reflect that she merely needs time alone or that she
truly wants to end the marriage) may prove more useful in reducing
anger. A more detailed exploration of the extent of injurious ideation to-
ward his wife’s boyfriend is also warranted. Marital therapy may help
both the service member and his spouse clarify their present levels of
commitment to the relationship. A past history of response to support-
ive psychotherapy and SSRIs for depression suggests that medication
management and supportive therapy might assist the service member
in understanding the intensity of his feelings of abandonment, more ef-
fectively expressing his frustrations, and identifying alternative coping
mechanisms. Careful monitoring in therapy and the development of
rapport may facilitate more intensive care (e.g., day treatment or hospi-
talization) should the psychosocial situation deteriorate.
Key Points
■ In contrast to patients who developed PTSD after a single-event
trauma, PTSD patients who underwent repeated threats to life
from multiple sources and observed repeated acts of violence
toward others over long periods are more likely to show height-
ened vigilance and possibly aggressive or violent behavior in
future settings.
■ Further research is clearly necessary on therapeutic interventions
targeting combat-related PTSD and PTSD-related violence and
aggression. The recommended treatments for PTSD, though sup-
ported by reasonable evidence in general, have not been well
validated in combat veteran populations—those most likely to
feel aggression and display violence as a result of their experi-
ences and illness. Furthermore, no controlled studies exist on the
effects of treatments on reducing aggression or violence in
patients with PTSD.
■ For combat veterans with PTSD, present knowledge and clinical
experience suggest that assessment and management of aggres-
sion and violence should include the use of pharmacological and
psychotherapeutic treatments with demonstrated efficacy in
other (i.e., noncombat) PTSD patients.
■ Effective management also requires treatment of other comorbid
conditions and the development of a hierarchy of problems and
interventions.
References
tion, 1980
Beckham JC, Feldman ME, Kirby AC, et al: Interpersonal violence and its corre-
lates in Vietnam veterans with chronic posttraumatic stress disorder. J Clin
Psychol 53:859–869, 1997
Brady K, Killeen TK, Brewerton T, et al: Comorbidity of psychiatric disorders
and posttraumatic stress disorder. J Clin Psychiatry 61(suppl):22–32, 2000a
Brady K, Pearlstein T, Asnis GM, et al: Efficacy and safety of sertraline treatment
of posttraumatic stress disorder: a randomized controlled trial. JAMA
283:1837–1844, 2000b
Chemtob CM, Novaco RW, Hamada RS, et al: Anger regulation deficits in com-
bat-related posttraumatic stress disorder. J Trauma Stress 10:17–36, 1997
Collins JJ, Bailey SL: Traumatic stress disorder and violent behavior. J Trauma
Stress 3:203–220, 1990
Daly CM, Doyle ME, Radkind M, et al: Clinical case series: the use of prazosin
for combat-related recurrent nightmares among Operation Iraqi Freedom
combat veterans. Mil Med 170:513–515, 2005
Davidson J, Pearlstein T, Londborg P, et al: Efficacy of sertraline in preventing
relapse of posttraumatic stress disorder: results of a 28-week double-blind,
placebo-controlled study. Am J Psychiatry 158:1974–1981, 2001a
Davidson JR, Rothbaum BO, van der Kolk BA, et al: Multicenter, double-blind
comparison of sertraline and placebo in the treatment of posttraumatic
stress disorder. Arch Gen Psychiatry 58:485–492, 2001b
Fontana A, Rosenheck R: The role of war-zone trauma and PTSD in the etiology
of antisocial behavior. J Nerv Ment Dis 193:203–209, 2005
Freeman TW, Roca V: Gun use, attitudes toward violence, and aggression
among combat veterans with chronic posttraumatic stress disorder. J Nerv
Ment Dis 189:317–320, 2001
Freeman TW, Roca V, Kimbrell T: A survey of gun collection and use among
three groups of veteran patients admitted to veterans affairs hospital treat-
ment programs. South Med J 96:240–243, 2003
Friedman MJ: Posttraumatic stress disorder among military returnees from Af-
ghanistan and Iraq. Am J Psychiatry 163:586–593, 2006
Hartl TL, Rosen C, Drescher KD, et al: Predicting high-risk behaviors in veter-
ans with posttraumatic stress disorder. J Nerv Ment Dis 193:464–472, 2005
Hiley-Young B, Blake DD, Abueg FR, et al: Warzone violence in Vietnam: an ex-
amination of premilitary, military, and postmilitary factors in PTSD in-
patients. J Trauma Stress 8:125–141, 1995
Kulka RA, Schlenger WE, Fairbank J, et al: The prevalence of other postwar re-
adjustment problems, in Trauma and the Vietnam War Generation: Report
of the National Vietnam Veterans Readjustment Study. New York, Brunner-
Mazel, 1990, pp 139–188
Laufer RS, Gallops MS, Frey-Wouters E: War stress and trauma: the Vietnam
veteran experience. J Health Soc Behav 25:65–85, 1984
Londborg PD, Hegel MT, Goldstein S, et al: Sertraline treatment of posttrau-
matic stress disorder: results of 24 weeks of open-label continuation treat-
ment. J Clin Psychiatry 62:325–331, 2001
Marshall AD, Panuzio J, Taft CT: Intimate partner violence among military vet-
erans and active duty servicemen. Clin Psychol Rev 25:862–876, 2005
Marshall RD, Beebe KL, Oldham M, et al: Efficacy and safety of paroxetine treat-
ment for chronic PTSD: a fixed-dose, placebo-controlled study. Am J Psy-
chiatry 158:1982–1988, 2001a
Marshall RD, Olfson F, Hellman C, et al: Comorbidity, impairment, and suicid-
ality in subthreshold PTSD. Am J Psychiatry 158:1467–1473, 2001b
Martenyi F, Brown EB, Zhang H, et al: Fluoxetine versus placebo in posttrau-
matic stress disorder. J Clin Psychiatry 63:199–206, 2002
McFall M, Fontana A, Raskind M, et al: Analysis of violent behavior in Vietnam
combat veteran psychiatric inpatients with posttraumatic stress disorder.
J Trauma Stress 12:501–517, 1999
National Center for PTSD: The Iraq War Clinicians Guide, 2nd Edition. Wash-
ington, DC, Veterans Administration, 2004. Available at http://www
.ncptsd.va.gov/ncmain/ncdocs/manuals/iraq_clinician_guide_v2 .pdf.
Accessed January 16, 2007
Oquendo MA, Friend JM, Halberstam B, et al: Association of comorbid post-
traumatic stress disorder and major depression with greater risk for sui-
cidal behavior. Am J Psychiatry 160:580–582, 2003
Oquendo M, Brent DA, Birmaher B, et al: Posttraumatic stress disorder comor-
bid with major depression: factors mediating the association with suicidal
behavior. Am J Psychiatry 162:560–566, 2005
Orcutt HK, King LA, King DW: Male-perpetrated violence among Vietnam vet-
eran couples: relationships with veteran’s early life characteristics, trauma
history, and PTSD symptomatology. J Trauma Stress 16:381–390, 2003
Rapaport MH, Endicott J, Clary CM: Posttraumatic stress disorder and quality
of life: results across 64 weeks of sertraline treatment. J Clin Psychiatry
63:59–65, 2002
Raskind MA, Peskind ER, Hoff DJ, et al: A parallel group placebo controlled
study of prazosin for trauma nightmares and sleep disturbance in combat
veterans with post-traumatic stress disorder. Biol Psychiatry 61:928–934,
2006
Resnick HS, Foy DW, Donahoe CP, et al: Antisocial behavior and post-traumatic
stress disorder in Vietnam veterans. J Clin Psychol 45:860–866, 1989
Savarese VW, Suvak MK, King LA, et al: Relationships among alcohol use, hy-
perarousal, and marital abuse and violence in Vietnam veterans. J Trauma
Stress 14:717–732, 2001
Schoenfeld FB, Marmar CR, Neylan TC: Current concepts in pharmacotherapy
for posttraumatic stress disorder. Psychiatr Serv 55:519–531, 2004
Sherman MD, Sautter F, Jackson MH, et al: Domestic violence in veterans with
posttraumatic stress disorder who seek couples therapy. J Marital Fam Ther
32:479–490, 2006
Taft CT, Pless AP, Stalans LJ, et al: Risk factors for partner violence among a na-
tional sample of combat veterans. J Consult Clin Psychol 73:151–159, 2005
Taylor FB, Lowe K, Thompson C, et al: Daytime prazosin reduces psychological

distress to trauma specific cues in civilian trauma posttraumatic stress dis-
order. Biol Psychiatry 59:577–581, 2006
Ursano RJ, Bell C, Eth S, et al: Practice guideline for the treatment of patients
with acute stress disorder and posttraumatic stress disorder. Am J Psychi-
atry 161(suppl):3–31, 2004
VA/DOD Clinical Practice Guideline Working Group: Management of Post-
Traumatic Stress. Washington, DC, Department of Veterans Affairs and De-
partment of Defense, 2004. Available at http://www.oqp.med.va.gov/cpg/
PTSD/PTSD_base.htm. Accessed January 16, 2007
Wilson JP, Zigelbaum SD: The Vietnam veteran on trial: the relation of post-
traumatic stress disorder to criminal behavior. Behav Sci Law 1:69–83, 1983
Yesavage JA: Differential effects of Vietnam combat experiences vs. criminality
on dangerous behavior by Vietnam veterans with schizophrenia. J Nerv
Ment Dis 171:382–384, 1983
Zlotnick C, Franklin CL, Zimmerman M: Does “subthreshold” posttraumatic
stress disorder have any clinical relevance? Compr Psychiatry 43:413–419,
2002
C H A P T E R 8
Substance Abuse Disorders

And Noah began to be an husbandman, and he

planted a vineyard: And he drank of the wine,
and was drunken; and he was uncovered within
his tent.
Genesis 9:20–21, King James Version
For U.S. Troops at War, Liquor Is Spur to Crime

The New York Times, March 13, 2007
F rom Biblical times through the present, we have known that alcohol
is often associated with out-of-control behavior and violence. Alone
among abused chemicals, only alcohol directly and commonly in-
creases aggression (Roth 1994). With illegal drugs, in contrast, the asso-
ciated violence is more often due to drug commerce: conflicts between
distributors, arguments and robberies between buyers and sellers, or
stealing to raise drug money (Roth 1994). Alcohol sedates the frontal re-
gions of the brain, the regions necessary for more judicious, thoughtful
decisions. Almost one-third of American adults drink at levels that in-
crease their risk for physical, mental health, or social problems (Na-
tional Institute on Alcohol Abuse and Alcoholism [NIAAA] 2004).
Most instances of substance-related violence happen at times and
places with no physician in attendance (homes, the street, bars, clubs, dis-
cos) (Macdonald et al 1999; Steadman et al 1998). Yet we as clinicians can
141
reduce that type of violence by identifying and treating substance abuse.

As Volavka (2002) has noted, alcohol-dependent persons “are not at an
increased risk for offending as long as they are sober. Of course, such per-
sons are by definition at a very high risk for not staying sober” (p. 206).
Alcohol and Violence

Case Example 1
In recovery from alcoholism and tortured by guilt, Alex returned to the
scene of a bar fight that he had been involved in many months before.
He learned the day after the fight that the man he hit had died, and he
moved on. Now he returned, but no one remembered him or the man
who had died. They were both just passing through.
In a comprehensive study of homicide offenders in Northern Swe-

den, Lindqvist (1986) found that 66% of the offenders were intoxicated
at the time of the killing. A Scandinavian study showed that, in compar-
isons across countries and over time, a change of 1 liter in annual per
capita alcohol consumption was associated with a 2%–10% change in
criminal violence (Alcohol and Public Health Research Unit [New
Zealand], n.d.). Moderate intoxication can combine with feeling a loss
of control in personal relationships, being in a crowded space, and be-
ing ineptly refused service to bring on violence. Saying “I was drunk at
the time” is a familiar way of avoiding responsibility (Alcohol and Pub-
lic Health Research Unit [New Zealand], n.d.). Volavka (2002) reported
that in short-term experiments, low doses of alcohol “elicit or facilitate”
aggression but high doses reduce it (p. 197).
Although alcohol and drugs often play an essential role in violence,
many other factors can contribute to violent behavior. Various authors
have emphasized the importance of culture and context. In addition,
specific factors can include a history of alcoholism, psychological disor-
ders, sleep deprivation, and physical conditions such as temporal lobe
dysfunction and hypoglycemia (Benson et al. 2001). Lipsey et al. (1997)
described causality in terms of “an alcohol–person–situation interac-
tion,” arguing that “alcohol consumption increases the probability of
violent behavior only in some persons in some situations” (p. 247). One
such scenario is described below.
Case Example 2
Bert and his wife were arguing about their financial problems. Because of
his angry, intoxicated outburst, his wife called the police, who came and
left. His wife then left the house, taking their child with her. Bert drank
Substance Abuse Disorders ❘ 143
through that night. The next day he called his boss, saying he did not feel
well, and he continued to drink. That evening the police pounded on his
door to serve a warrant. He called 911, telling the operator that he did not
know what was going on, but that he had a gun and he was defending
his home. The police placed a cordon around the house.
Three mechanisms have been proposed to explain the link between

alcohol and violence: potentiation, inhibition, and disorganization (Pihl
and Lemarquand 1998). Thanks to the work of Anna Rose Childress,
Ph.D., and others, we have a much better understanding of the physio-
logical effect of alcohol and drugs that tilts the balance between the lim-
bic system (“GO!”) and the frontal lobes (“STOP!”) (“HBO: Addiction”
2007). The limbic circuits respond quickly and strongly to danger, food,
and sex. The frontal brain regions, also vital to the welfare of our spe-
cies, make us stop and think, bringing in memories and experiences
that should influence our behavior. Adolescents have not yet fully de-
veloped the circuits needed for the prefrontal cortex to regulate the lim-
bic system, and in addition, their amygdalas are more active.
The relations between rape and alcohol use have been summarized
as follows:
Conservative estimates of sexual assault prevalence suggest that 25% of

American women have experienced sexual assault, including rape. Ap-
proximately one-half of those cases involve alcohol consumption by the
perpetrator, victim, or both. Alcohol contributes to sexual assault
through multiple pathways, often exacerbating existing risk factors. Be-
liefs about alcohol’s effects on sexual and aggressive behavior, stereo-
types about drinking women, and alcohol’s effects on cognitive and mo-
tor skills contribute to alcohol-involved sexual assault. (Abbey et al.
2001, p. 43)
Date rape is an important kind of violence to which alcohol contrib-

utes. This type of assault in a college setting has been characterized in
this way by Koss (1988):
Among college students, a typical sexual assault occurs on a date, at ei-

ther the man’s or the woman’s home, and is preceded by consensual
kissing. The assault involves a single assailant who uses no weapon, but
twists the woman’s arm or holds her down. The woman believes she
has clearly emphasized her nonconsent, and tries to resist by reasoning
or by physically struggling. (pp. 242–250)
Presley et al. (1998) found that 1.2%–1.5% of college students had

tried to commit suicide during the past 12 months due to drinking or
drug use. Lower minimum-age drinking laws are associated with higher
youth suicide rates (National Institute of Mental Health, n.d.). Particular
risk factors include being American Indian or Alaskan Native (National

Institute of Mental Health, n.d.). People dependent on substances are
more likely to have financial and social problems, to be depressed, to be
impulsive, and to engage in high-risk behaviors that result in self-harm
(National Institute of Mental Health, n.d.). An impulsive personality is
associated with heavier drinking and with violence (Gelles 1985).
Maltreatment of children is another form of violence closely associ-
ated with alcohol use.
Case Example 3
Many years after the event, late in therapy, Carla shared her most pain-
ful, secret guilt. She had never told anyone this before. As a young
mother deserted by her husband, she felt overwhelmed by the needs of
their infant. While drinking and intoxicated, when bathing the child,
she had knocked a radio from the edge of the tub into the water, think-
ing it might electrocute the child. The child was not harmed.
The World Health Organization (WHO) has found a strong link be-
tween child maltreatment and alcohol abuse. Thirty-five percent of pa-
rental child abusers had consumed alcohol or drugs at the time of the
incident (U.S. Department of Justice 2001). Risk factors include being
young, poor, unemployed, and socially isolated (Krug et al. 2002).
WHO suggests a range of prevention strategies, including family sup-
port, parenting training, screening for child maltreatment, and services
for victims (World Health Organization, n.d.).
“Shush, be still, be still, no breath. Off just shut off. Find the ceiling’s cor-
ner and hide in it.”
“There’s no fucking safe place no matter what they say. Don’t matter
what ‘they’ say.”
“How long do I have to be 5 years old anyway.”
“Be still, be gone, hold onto the quiet numb solitude.”
—Notes written by Martha, an adult woman physician, early in her

recovery from alcoholism, dealing with the physical and sexual
abuse she experienced in childhood (quoted with her permission)
Abused children, when grown up, have an increased risk of alco-

holism. However, the evidence for the linkage of childhood abuse to
adult alcoholism is stronger for women than for men (Widom and
Hiller-Sturmhofel 2001).
“Thought I had killed my mother. I went outside. She was peeing in the
bushes. She came in. I hit her in the head, I was so angry. She was passed
out upstairs. I put a feather in front of her nose. She was still breathing.”
—A woman, in recovery from alcoholism, speaking of her early ad-

olescent relationship with her mother, who had severe alcoholism
Vulnerable adults also can be victims of alcohol-related abuse. In the

United States, 44% of male and 14% of female abusers of elders were de-
pendent on alcohol or drugs, as were 7% of the victims (Greenberg et al.
1990). Individuals with alcoholism may be financially dependent on
older relatives (Bradshaw and Spencer 1999) and may neglect their re-
sponsibilities to them (Department of Social Development [South Af-
rica] 2001).
Drinking alcohol exerts a major influence on intimate partner vio-
lence. In studying a group of 109 couples in which the women were par-
ticipating in a study on therapies for women with alcoholism, Drapkin
et al. (2005) found that 61% reported some violence and 27% reported
severe violence (kicking, biting, hitting). Generally, men are thought to
be more violent than women. But Drapkin and colleagues found that in
27% of the couples they studied, men and women contributed equally
to the violence. In couples with a disparity in violence, the more violent
person was more likely to be the woman than the man (23% vs. 11%).
Overall, the women in this study were more likely than the men to be
verbally aggressive and psychologically coercive. Among the women in
this study, more intensive drinking was associated with more severe
violence or verbal aggression (Drapkin et al. 2005). Following up on a
random sample of 1,635 U.S. couples, Ramisetty-Mikler and Caetano
(2005) found that female-on-male violence predicted marital separation
but that heavy drinking by women reduced the risk of separation.
A 1995 national survey found that 23% of black couples, 17% of His-
panic couples, and 11.5% of white couples reported an incident of male-
to-female violence in the 12 months preceding. The corresponding rates
of female-to-male violence were 30%, 21%, and 15%, respectively, in
each instance higher than male-to-female violence. At the time of vio-
lence against their partners, 30%–40% of the men and 27%–34% of the
women were drinking. The higher prevalence of intimate partner vio-
lence among ethnic minorities seems to be related to individual risk fac-
tors, the environment, and the type of relationship between the partners
(Caetano et al. 2000). The social-structural theory emphasizes poverty,
undereducation, high unemployment, and racial discrimination as con-
tributors to increased violence. Another explanation posits a subculture
of violence in which some groups in society accept violence as a means
of resolving conflicts (Gelles 1985).
Drinking may reduce fear in victims, making them more willing
to participate in a dangerous quarrel, and may make them less able to
respond appropriately to threats. The likelihood of aggression between

two people is “greatest when both are intoxicated, intermediate when
one person is intoxicated, and least probable when both are sober”
(Murdoch et al. 1990). The diaries of men attending a batterer interven-
tion program revealed that they were 20 times more likely to attack
their partner on heavy drinking days, as compared to non-drinking
days (Fals-Stewart 2003).
In gang life, alcohol and violence affirm masculinity and male to-
getherness. Being on the street is the natural social arena for many mi-
nority and working-class male adolescents. The entry to life on the
street is through a gang. A new member’s passing through initiation is
celebrated by drunkenness. Gang members may leave alcohol at the
gravesite of a dead member to symbolize their unity (Hunt and Laidler
2001).
Alcohol and Suicide

Hayward et al. (1992) found that alcohol was involved in 36% of sui-
cides. In 46% of suicide attempts, the alcohol had been ingested within
the previous 2 hours (Merrill et al. 1992).
Case Example 4
For 10 years, Dan was hypomanic and very successful. “People were
patting me on the back, telling me how amazing I was.” Sometimes he
felt an “electric buzz” that he would try to force down with alcohol.
Then he began to have financial losses, depression, and conflicts with
his wife. “All the brain power I had when I was manic turned against
me. I couldn't come up with a good reason not to kill myself.” He waited
until 11:00 P.M. so that there wouldn’t be any children on the road, then
got drunk at a local bar and drove his car into a tree at high speed. He
survived, with mild, persistent brain injury.
Case Example 5
Evan was a young man with a strong sense of right and wrong who had
intense feelings about justice and injustice. He bitterly criticized himself
for various deficiencies, including having failed several college courses.
He found a reason to live in a several-years’ relationship with a young
woman who was also troubled and struggling, but then she left and be-
gan sending him e-mails accusing him of sexual insensitivity in their re-
lationship. He purchased a .38 revolver, got drunk, and killed himself.
First he left money for his roommates for the month’s rent, and he left a
message on his therapist’s answering machine, thanking her for trying
to help him. Evan feared he didn’t have the courage to kill himself. With
alcoholic intoxication, he found the courage.
Cocaine, Methamphetamine, and Violence

The effects of stimulants in animals vary, depending on the animal
species and the primate’s social position in the group. Low doses may
occasion aggression, with higher doses having the opposite effect (Vo-
lavka 2002, p. 210)
Cocaine and methamphetamine are known to induce paranoia in
people with no history of psychotic illness.
Case Example 6
The sister of one of my patients phoned me, saying she was concerned
that her own husband “might have a little problem with cocaine.” As we
spoke, this woman’s husband was standing in their living room, shot-
gun at his side, peering through the blinds.
Marzuk et al. (1995) studied the relationship between cocaine use

and fatal injuries in New York City in 1990–1992. Of the 14,843 New
Yorkers who died from homicides, suicides, accidents, and drug over-
doses (the medical examiner’s laboratory analyzed blood or urine au-
topsy specimens for about 85% of cases), benzoylecgonine, a cocaine
metabolite with a serum half-life of up to 48 hours, was present in 27%
of the cases. Cocaine itself, with a serum half-life of 1.5 hours, was
present in 18% of cases. About one-third of these fatalities were from ac-
cidental overdose; the other two-thirds were from violence or trauma
(Marzuk et al. 1995).
Spunt et al. (1990) found that cocaine-related violent events in
women most often were a direct, pharmacologic effect of cocaine. In
contrast, in white males, cocaine-related violence was committed pri-
marily to obtain money to buy drugs, and in black males, cocaine-
related violence was predominantly associated with drug marketing
and sales.
Case Example 7
I saw a young man in the hospital who had a bullet in his right arm.
When trying to buy cocaine from a drug dealer, he had folded over a $10
bill, trying to fool the dealer into thinking he was paying two $10 bills.
Opioids and Violence

In experiments, ex-abusers given moderate doses of methadone became
friendly and “mellow” (Volavka et al. 1974). The irritability and dys-
phoria of opiate withdrawal may motivate opiate addicts to rob or steal
for money. Prostitutes in opiate withdrawal robbed their clients rather

then just providing sex (Goldstein 1985).
PCP and Violence

The disruption of sensory input by phencyclidine (PCP) can produce
unpredictable and exaggerated reactions to the environment (Zukin et
al. 2005). However, those who are violent under the influence of PCP
usually have a history of psychosis or antisocial behavior (Roth 1994).
Cannabis and Violence

In most animal experiments, cannabis extracts reduced aggression and
increased flight and submission (Miczek 1987). Adolescent delinquents
reported that cannabis made them calm and reduced assaultiveness
(Tinklenberg et al. 1976). Volavka et al. (1971) and others have con-
ducted numerous experiments studying the effects of cannabis use and
have not seen aggression or violence.
Mental Disorders, Substance Abuse, and Violence

Personality disorder, substance abuse, and neurologic impairment can
all contribute to violence.
Antisocial personality disorder is a risk factor for developing prob-
lem drinking and often develops a few years before the drinking (Bahl-
mann et al. 2002). Cloninger et al. (1981) describe a subtype of alcohol-
ism that develops early (usually before age 25), is strongly inherited
from father to son, has many features of antisocial behavior, and is as-
sociated with abnormal serotonin metabolism (Virkkunen and Linnoila
1990). Ondansetron (a selective 5-HT3 antagonist) reduces drinking in
patients with early-onset alcoholism (Johnson et al. 2000).
The general public has a great fear of violence due to mental illness.
For example, 81% of the public believes that children with major de-
pression are more likely to be dangerous to themselves or others (Pesco-
solido et al. 2007).
The National Institute of Mental Health’s Epidemiologic Catchment
Area (ECA) study found that patients with serious mental illness
(schizophrenia, major depression, bipolar disorder) reported a lifetime
prevalence of violence of 16%, as compared to 7% among people with-
out mental illness (Swanson 1994). Because serious mental illness is
relatively rare, however, it contributes only 3%–5% to society’s risk for
violence (Friedman 2006).
The increased lifetime prevalence of violence in people with serious

mental illness is strongly influenced by their propensity to abuse alco-
hol and drugs. Having a mental disorder doubles the risk of alcohol
abuse, and it increases the risk of drug abuse by four times (Regier et al.
1990). People with no mental disorder who abuse alcohol or drugs are
nearly seven times as likely to report violent behavior as are those who
do not abuse alcohol or drugs (Friedman 2006).
Steadman et al. (1998) followed 1,136 patients with various psychi-
atric disorders for 50 weeks after their discharge from inpatient care.
The control group was 519 people living in the same neighborhoods.
There was no increased risk of violence among the non–substance abus-
ing mentally ill persons. Substance abuse increased the risk among both
patients and those in the control group. Substance abuse in combination
with a personality disorder produced the highest risk of violence.
In adults with psychotic illness or major mood disorder, violence in-
dependently correlates with several risk factors: substance abuse, a his-
tory of having been a victim of violence, homelessness, and poor mental
health. The 1-year rate of violent behavior for persons with none or only
one of these risk factors was 2%, a prevalence close to the ECA’s esti-
mate for the general population. It appears, then, that in people with
serious mental illness, violence “probably results from multiple risk
factors in several domains” (Friedman 2006, p. 2066).
Delusions that someone is trying to harm one and delusions that
outside forces control one’s mind are both associated with violence. A
study of delusional, violent patients found that 83.5% had a history of
substance abuse (Beck 2004).
Brain damage (head injury) changes a person’s response to alcohol.
Less alcohol produces more effect (Finger and Stein 1982), and alcohol
is more likely to make the person feel paranoid and inferior (Langevin
et al. 1987), setting the stage for violence.
Addiction
Dackis and O’Brien (2005) have characterized addiction as “a disease of
brain reward centers that ensure the survival of organisms and spe-
cies.” They describe the mechanism as follows:
Given their function, reward centers have evolved the ability to grip at-
tention, dominate motivation and compel behavior directed toward sur-
vival goals, even in the presence of danger and despite our belief that we
are generally rational beings. By activating and dysregulating endoge-
nous reward centers, addictive drugs essentially hijack brain circuits
that exert considerable dominance over rational thought, leading to pro-
gressive loss of control over drug intake in the face of medical, interper-
sonal, occupational and legal hazards. There is even evidence that de-
nial, once thought to be purely “psychological,” may be associated with
drug-induced dysfunction of the prefrontal cortex. (Dackis and O’Brien
2005, p. 1431)
The case of a well-educated professional man illustrates these dys-

functional processes.
Case Example 8
Frank was a talented physician who could often get a year or two of re-
covery from his addiction to opiates. That was enough stability to per-
mit him to work and use his medical skills, although not in patient care.
In relapse, he returned to the place where he customarily bought drugs,
even though he had been robbed there with a knife at his throat only the
night before. Several months later, he died alone in a motel room in an-
other state, probably of an overdose.
Interestingly, a cynomolgus monkey that loses social rank under-

goes a reduction in dopamine D 2 receptors and is more likely to self-
administer cocaine. If that monkey is placed with a different group of
monkeys, among which he has higher social standing, he will have an
increase of D 2 receptors (Czoty et al. 2004). Could this physiology ex-
plain some of the drug abuse among underprivileged groups?
Modafinil may improve impulse control in addicted patients and
help them feel better (Dackis 2005), as illustrated in the following case
example.
Case Example 9
Geoff, always falling far short of the expectations of his successful fa-
ther, began using marijuana in early adolescence and later began using
cocaine. Geoff’s father understood these drug use–related failures as
willful misbehavior and tried to discipline him, sometimes physically.
Geoff cooperated superficially with the treatment program his parents
had tricked him into attending. “Bunch of crap, just sit down and talk.
Didn’t get anything out of it. Just listen to people’s problems.” He
planned to “get my life sorted out. Just become a social user. Not com-
pletely stop.” He continued to smoke marijuana while in the treatment
program.
After his completion of that treatment program, Geoff’s plans to be
a “social user” of cocaine quickly and disastrously crashed. He took his
second treatment program much more seriously, understanding that he
could not safely use marijuana, either. Modafinil seemed to free him of
craving for cocaine and probably also helped his attention-deficit/
hyperactivity disorder.
Identifying Addiction
For many years, literature and Hollywood have dramatized (some-
times accurately) the lives of alcoholics and drug addicts. The “Decade
of the Brain” has now flowered into a moving, scientifically based mass
media production, Home Box Office’s series Addiction (sponsored by
the Robert Wood Johnson Foundation, the National Institute on Alcohol
Abuse and Alcoholism, and the National Institute on Drug Abuse),
which began March 15, 2007. A superb collection of resources can be ac-
cessed, and a DVD of the series can be purchased, at the series Web site
(http://www.hbo.com/addiction).
Only 10% of patients with alcoholism receive assessment and refer-
ral to treatment from their primary care physicians (McGlynn et al.
2003). For physicians, the National Institute on Alcohol Abuse and Al-
coholism (NIAAA, 2005) has released an updated version of Helping Pa-
tients Who Drink Too Much: A Clinician’s Guide. “Too much” is defined as
five or more drinks in a day for a man, four or more drinks in a day for
a woman. (A standard drink is equivalent to 12 ounces of beer, 5 ounces
of wine, or 1.5 ounces of 80-proof spirits.)
The single question “How often in the past year have you had five
or more drinks [four for a woman] in a day?” can serve to screen for al-
cohol-related problems (Dawson et al. 2005). As an alternative to the
single question during the clinical interview, the guide suggests screen-
ing by the written self-report AUDIT—the Alcohol Use Disorders Iden-
tification Test.
For persons who screen positive for alcohol-related problems, the
next step is to assess the severity and extent of the problems, using a list
of symptoms derived from DSM-IV-TR (American Psychiatric Associa-
tion 2000). The NIAAA’s guide provides additional resources and treat-
ment templates.
About 30% of the U.S. population have what is called “at-risk drink-
ing” (National Institute on Alcohol Abuse and Alcoholism 2005),
“heavy drinking,” or “unhealthy alcohol use” (Saitz 2005). These levels
of use do not meet criteria for substance abuse (a substance-related fail-
ure to perform obligations at work, school, or home; use in hazardous
situations; or recurrent legal or social problems—often associated with
antisocial personality disorder). Nor do they meet criteria for substance
dependence (a loss of control of use of the substance; a life focused on
getting and using the substance).
Often, “at-risk” heavy drinkers who do not meet the criteria for
abuse or dependence can voluntarily reduce their alcohol consumption
and can benefit from learning that their consumption is greatly above
the norm. Habitually they associate with other people who drink
heavily, and they conclude, incorrectly, that such a level of consumption
is common. The NIAAA guide Helping Patients Who Drink Too Much
provides a useful comparison and a template for strategies for cutting
down.
Treating Addiction
For many years, 12-Step recovery groups have counseled separating
from “people, places, and things” associated with drug or alcohol use.
Science has now shown that even years into recovery, environmental
cues can powerfully induce relapse (Grusser et al. 2004).
“Evidence-Based” strategies to help patients reduce the risk of re-
lapse (Witkiewitz and Marlatt 2007) include the following:
1. Understanding relapse as a process and event, and learning how to

identify early warning signs
2. Identifying high-risk situations and developing coping responses
3. Enhancing communications skills, improving interpersonal rela-
tionships, and developing a constructive social network
4. Managing negative emotional states
5. Identifying and managing cravings and the “cues” that precede
cravings
6. Identifying and challenging cognitive distortions
Pharmacological treatments for addiction, an obvious corollary to

our understanding of addiction physiology, are demonstrably useful
and are underused. Helping Patients Who Drink Too Much succinctly re-
views medications approved by the U.S. Food and Drug Administration
for the treatment of alcoholism and makes suggestions for their use.
Naltrexone, an opioid-blocking medication, reduces alcohol crav-
ing. Volpicelli et al. (1992) and others (O’Malley et al. 1992) have found
that naltrexone most consistently helps people with alcoholism drink
less often and in less quantity, forestalling the worst consequences of a
relapse. Compared with normal subjects, patients with a genetic vul-
nerability for alcoholism have low β-endorphin levels, with increased
β-endorphin release and pleasure after drinking alcohol (Gianoulakis et
al. 1996). Naltrexone, an opioid-blocking agent, reduces that response,
especially in persons with the G allele form of the gene coding the mu-
opioid receptor, the OPRM1 gene (Ray and Hutchison 2007). However,
Krystal et al. (2001) found that naltrexone was not effective overall in
Veterans Affairs patients with chronic, severe alcoholism.
Acamprosate, the newest medication certified by the U.S. Food and

Drug Administration to treat alcoholism, alters γ-aminobutyric acid
and N-methyl D-aspartate systems (Rammes et al. 2001) and also re-
duces alcohol craving and relapse. It may help people who feel like
drinking because of withdrawal.
Case Example 10
Helen has 50 years of happy marriage, a graduate degree, and a close re-
lationship with her adult children. However, more than once, she was
intoxicated when she arrived to provide transportation for her grand-
children. Understandably fearful, Helen’s children would not let her
take the grandchildren, even when she was sober. Acamprosate seemed
to make her recovery more comfortable. (She was also taking anti-
depressant medication.) Although three times daily is the suggested
dosing schedule, Helen had loose stools if she took acamprosate more
than twice a day. With participation in 12-Step recovery, she stayed
sober and regained the trust of her children.
Although some well-done, randomly controlled trials have found

statistical efficacy for naltrexone and for acamprosate, other well-
conducted trials have not. Project Combine (Anton et al. 2006) ran-
domly assigned 1,383 patients to one of eight groups, to receive medical
management plus pills. Four groups received naltrexone, or acampro-
sate, or both naltrexone and acamprosate, or placebo. Four other groups
received one of these medication regimes plus a “combined behavioral
intervention” (CBI). A ninth group received CBI and no pills. Acampro-
sate showed no evidence of efficacy, with or without CBI. Patients re-
ceiving medical management with naltrexone, or with CBI, or with
both, showed improvement. No combination showed better efficacy
than medical management plus naltrexone or medical management
plus CBI. Medical management plus placebo had a greater effect, dur-
ing treatment, than CBI.
Normally, the liver metabolizes alcohol to acetaldehyde and then to
harmless acetic acid. The first of these two steps is the rate-limiting step,
the “bottleneck” of this two-step metabolic chain. Disulfiram (available
for 60 years) inhibits the liver enzyme of the second step, acetaldehyde
dehydrogenase. Because acetaldehyde (which is toxic) is metabolized
more slowly, blood levels increase (5–10 times higher) and the person
becomes ill, experiencing flushing, throbbing of the head and neck,
nausea, vomiting, and shortness of breath. However, with present-day
dosage (250 mg daily), people in good health are not in medical danger.
As one can predict from the physiological mechanism, the severity of
the alcohol–disulfiram reaction depends on the quantity of alcohol
ingested. For example, the reaction from the small amount of alcohol
that may be ingested from a vinegar salad dressing (or absorbed from
hair spray or deodorant) is not severe. Some persons with robust acetal-
dehyde dehydrogenase require an increased disulfiram dose.
In 1986, R.K. Fuller and colleagues, in a carefully done, controlled,
blinded study, concluded that disulfiram “may help reduce drinking
frequency after relapse” but “does not enhance counseling in aiding al-
coholic patients to sustain continuous abstinence or delay the resump-
tion of drinking” (Fuller et al. 1986, p. 1449).
Disulfiram can be useful for those patients who, although commit-
ted to not drinking, cannot trust themselves to abstain, especially early
in recovery. Because acetaldehyde dehydrogenase requires as long as
2 weeks to regenerate after cessation of disulfiram, taking disulfiram
once daily settles for that day the internal discussion, “Will I drink to-
day?” Disulfiram can be used episodically for protection during busi-
ness trips or vacation.
Case Example 11
Jerome is a distinguished professional, respected by his colleagues,
elected by them to chair an organization representing their interests.
However, at a lecture, he appeared intoxicated on stage. He could go for
weeks without drinking, but if he had a break in his schedule on a nice
day, he might step outside for a bit, and his favorite bar was just down
the block, across the street. Seldom could he pass it without going in.
Disulfiram gave him long-term sobriety.
Interestingly, disulfiram may be useful in treating cocaine addiction

through its inhibition of dopamine-β-hydroxylase (DBH), increasing
brain dopamine levels and thus producing an unpleasant sense of hy-
perstimulaton and discomfort in cocaine users. It is particularly effec-
tive for patients who are not also abusing alcohol (Carroll et al. 2004).
Current studies are looking at the effectiveness of disulfiram for treat-
ing cocaine-dependent individuals with different DBH genetic variants.
Topiramate seems to reduce appetitive behavior generally, includ-
ing overeating; it is one of the few psychotropic medications that tends
to reduce, rather than increase, weight (McElroy et al. 2003), alcohol
drinking and craving (Johnson et al. 2003), and cocaine use (Kampman
et al. 2004). Unfortunately, it often produces fatigue and cognitive dull-
ing (Salinsky et al. 2005).
Leaving aside the very important maintenance treatment of opioid
dependence (i.e., methadone, buprenorphine), our key addiction treat-
ments focus on psychologically influencing the emotions and thoughts
of our patients. We are a social species. Meeting with other people who
also are recovering from addiction, people who have had similar expe-
riences, people who come to know us and are not angry at us (as fami-
lies often are at the addicted person), people who seem to be decent
people other than the aberration of their alcohol- and drug-related be-
haviors, people who are succeeding at recovery—all this powerfully en-
courages and motivates. Learning that addiction is a disease, with pow-
erful genetic and environmental antecedents, not simply a hopeless
moral failing, can lift demoralizing guilt and lend credibility to strate-
gies for managing the disease, as we do with other chronic diseases. Pa-
tients learn these strategies from other recovering people, from counse-
lors, and from therapists. Although it is not the focus of this chapter,
treating comorbid psychiatric illness, and influencing destructive life
circumstances, can be vitally important.
Project MATCH disappointed many high hopes by not confirming
10 hypothesized treatment-effective “matches” between treatment type
and type of patient with alcoholism. The one “match” was between pa-
tients with low psychiatric severity and 12-Step facilitation therapy.
NIAAA Director Enoch Gordis remarked that these findings “challenge
the notion that patient–treatment matching is necessary in alcoholism
treatment” (National Institute on Alcohol Abuse and Alcoholism 1996).
Although it is hard to study 12-Step programs in a rigorous scientific
manner, clinicians who treat addiction have long respected 12-Step re-
covery. Dr. George E. Vaillant made a major scientific contribution, pub-
lishing in 1983 (with a revised edition in 1995) The Natural History of Al-
coholism, a prospective study of more than 700 individuals followed for
more than 40 years. Vaillant wrote that “multiple studies that collec-
tively involved a thousand or more individuals, suggest that good clin-
ical outcomes are significantly correlated with frequency of attendance
at Alcoholics Anonymous (AA) meetings, with having a sponsor, with
engaging in Twelve-Step work and with chairing meetings” (Vaillant
2005, p. 433, citing Emrick 1993). Dr. Vaillant (2005) observes that
Project MATCH found, during its first year, that AA alone was as ef-
fective as the two most effective professional alternatives: cognitive-
behavioral and motivational enhancement therapies.
Key Points
■ Alcohol and other drugs of addiction, interacting with personality
and circumstances, often increase the likelihood of violence.
■ Alcohol is the most widely used of addictive chemicals, and it is the
chemical most likely to induce aggression; therefore it has the
largest role in violence.
■ Patients often do not volunteer information about their drinking
or drug use. Every patient should be asked at least the single
screening question: How often in the past year have you had five
or more drinks in a day? (four or more drinks for a woman). Even a
single such day is reason for further evaluation.
■ Addictions treatment is effective. It should be tailored to each
patient, using supportive therapies (12-Step meetings, individual
therapy), counseling and cognitive-behavioral therapies, and
pharmacotherapy.
References
Abbey A, Zawacki T, Buck PO, et al: Alcohol and sexual assault. Alcohol Res
Health 25:43–51, 2001
Alcohol and Public Health Research Unit: Alcohol and violence: What’s the con-
nection? n.d. Available at: http://www.aphru.ac.nz/hot/violence.htm. Ac-
cessed March 6, 2007
Anton RF, O’Malley SS, Ciraulo DA, et al: Combined pharmacotherapies and
behavioral interventions for alcohol dependence. JAMA 295:2003–2017,
2006
Bahlmann M, Preuss UW, Soyka M: Chronological relationship between antiso-
cial personality disorder and alcohol dependence. Eur Addict Res 8:195–
200, 2002
Beck JC: Delusions, substance abuse, and serious violence. J Am Acad Psychia-
try Law 32:169–172, 2004
Benson BL, Rasmussen DW, Zimmerman PR: The Impact of Alcohol Control
Policies on the Incidence of Violent Crime. Final report submitted to Na-
tional Institute of Justice, Grant No. 99-IJ-CX-0041. 2001. Available from
NCJRS (NCJ 191199). Available at http://www.ncjrs.gov/pdffiles1/nij/
grants/191199.pdf. Accessed July 1, 2007
Bradshaw D, Spencer C: The role of alcohol in elder abuse cases, in Elder Abuse
Work: Best Practice in Britain and Canada. Edited by Pritchard J. London,
Jessica Kingsley, 1999, pp 332–353
Caetano R, Cunradi CB, Clark CL, et al: Intimate partner violence and drinking
patterns among white, black, and Hispanic couples in the U.S. J Subst
Abuse 11:123–138, 2000
Carroll KM, Fenton L, Ball S, et al: Efficacy of disulfiram and cognitive behavior
therapy in cocaine-dependent outpatients. Arch Gen Psychiatry 61:264–
272, 2004
Cloninger CR, Bohman M, Sigvardsson S: Inheritance of alcohol abuse: cross-
fostering analysis of adopted men. Arch Gen Psychiatry 38:861–868,1981
Czoty PW, Morgan D, Shannon EE, et al: Characterization of dopamine D1 and
D2 receptor function in socially housed cynomolgus monkeys self-admin-
istering cocaine. Psychopharmacology (Berl) 174:381–388, 2004
Dackis C: Drug Des Discov 2:79–86, 2005
Dackis C, O'Brien C: Neurobiology of addiction: treatment and public policy
ramifications. Nat Neurosci 8:1431–1436, 2005
Dawson DA, Grant BF, Li TK: Quantifying the risks associated with exceeding
recommended drinking limits. Alcohol Clin Exp Res 29:902–908, 2005
Department of Social Development: Mothers and fathers of the nation: the for-
gotten people: the Ministerial report on abuse, neglect and ill-treatment of
older persons. South Africa, Department of Social Development, 2001.
Available at www.info.gov.za/documents/subjectdocs/subject/so-
cial.htm. Accessed March 10, 2007
Drapkin ML, McCrady BS, Swingle JM, et al: Exploring bi-directional couple vi-
olence in a clinical sample of female alcoholics. J Stud Alcohol 66:213–219,
2005
Emrick CD, Tonigan JS, Little L: Alcoholics Anonymous: What is currently
known? In Research in Alcoholics Anonymous: Opportunities and Alter-
natives. Edited by McCready VX, Miller WR. Piscataway, NJ, Rutgers Cen-
ter for Alcohol Studies, 1993, pp 41–76
Fals-Stewart W: The occurrence of partner physical aggression on days of alco-
hol consumption: a longitudinal diary study. J Consult Clin Psychol 71:45–
52, 2003
Finger S, Stein DF: Brain Damage and Recovery. New York, Academic Press,
1982
Friedman RA: Violence and mental illness: how strong is the link? N Engl J Med
355:2064–2066, 2006
Fuller RK, Branchey L, Brightwell DR, et al: Disulfiram treatment of alcoholism:
a Veterans Administration cooperative study. JAMA 256:1449–1455, 1986
Gelles RJ: Family violence. Annu Rev Sociol 11:347–367, 1985
Gianoulakis C, Krishnan B, Thavundayil J: Enhanced sensitivity of pituitary
beta-endorphin to ethanol in subjects at high risk of alcoholism. Arch Gen
Psychiatry 53:250–257, 1996 [Erratum published in 53:555, 1996]
Goldstein PJ: The drugs/violence nexus: a tripartite conceptual framework.
J Drug Issues 15:493–506, 1985
Greenberg JR, McKibben M, Raymond JA: Dependent adult children and elder
abuse. J Elder Abuse Negl 2:73–86, 1990
Grusser SM, Wrase J, Klein S, et al: Cue-induced activation of the striatum and
medial prefrontal cortex is associated with subsequent relapse in abstinent
alcoholics. Psychopharmacology (Berl) 175:296–302, 2004
Hayward L, Zubrick SR, Silburn S: Blood alcohol levels in suicide cases. J Epi-
demiol Community Health 46:256–260, 1992
HBO: Addiction (television series). Home Box Office, 2007. Materials and DVD
available at http://www.hbo.com/addiction. Accessed April 22, 2007
Hunt GP, Laidler KJ: Alcohol and violence in the lives of gang members. Alco-
hol Res Health 25:66–71, 2001
Johnson BA, Ait-Daoud N, Bowden CL, et al: Oral topiramate for treatment of
alcohol dependence: a randomized controlled trial. Lancet 17:1677–1685,
2003
Johnson BA, Roache JD, Javors MA, et al: Ondansetron for reduction of drinking
among biologically predisposed alcoholic patients. JAMA 284:963–971,
2000
Kampman KM, Pettinati H, Lynch KG, et al: A pilot trial of topiramate for the
treatment of cocaine dependence. Drug Alcohol Depend 75:233–240, 2004
Koss MP: Hidden rape: sexual aggression and victimization in a national sam-
ple of students in higher education, in Rape and Sexual Assault, Vol 2. Ed-
ited by Burgess AW. New York, Garland, 1988, pp 3–25
Krug EG, Dahlberg LL, Mercy JA, et al (eds): World Report on Violence and
Health. Geneva, World Heath Organization, 2002
Krystal JH, Cramer JA, Krol WF, et al: Naltrexone in the treatment of alcohol de-
pendence. N Engl J Med 345:1734–1739, 2001
Langevin R, Ben-Aron M, Wortzman G, et al: Brain damage, diagnosis, and sub-
stance abuse among violent offenders. Behav Sci Law 5:77–94, 1987
Lindqvist P: Criminal homicide in northern Sweden 1970–1981: alcohol intoxi-
cation, alcohol abuse and mental disease. Int J Law Psychiatry 8:19–37, 1986
Lipsey MW, Wilson DB, Cohen MA, et al: Is there a causal relationship between
alcohol use and violence? A synthesis of evidence. In Recent Developments
in Alcoholism, Vol 13: Alcoholism and Violence. Edited by Galanter M.
New York, Plenum, 1997
Macdonald S, Wells S, Giesbrecht N, et al: Demographic and substance use fac-
tors related to violent and accidental injuries: results from an emergency
room study. Drug Alcohol Depend 55:53–61, 1999
Marzuk PM, Tardiff K, Leon AC, et al: Fatal injuries after cocaine use as a lead-
ing cause of death among young adults in New York City. N Engl J Med
332:1753–1757, 1995
McElroy SL, Arnold LM, Shapira NA, et al: Topiramate in the treatment of binge
eating disorder associated with obesity: a randomized, placebo-controlled
trial. Am J Psychiatry 160:255–261, 2003
McGlynn EA, Asch SM, Adams J, et al: The quality of health care delivered to
adults in the United States. N Engl J Med 348:2635–2645, 2003
Merrill J, Milker G, Owens J, et al: Alcohol and attempted suicide. Addiction
87:83–89, 1992
Miczek KA: The Pharmacology of Aggression, in Handbook of Psychopharma-
cology, Vol 19. Edited by Iversen LL, Iversen SD, Snyder SH. New York,
Plenum, 1987, p 257
Murdoch D, Pihl RO, Ross D: Alcohol and crimes of violence: present issues. Int
J Addict 25:1065–1081, 1990
National Institute on Alcohol Abuse and Alcoholism [NIAAA]: NIAAA reports

Project MATCH main findings. Dec. 17, 1996. Available at http://www
.niaaa.nih.gov/NewsEvents/NewsReleases/match.htm. Accessed March
10, 2007
National Institute on Alcohol Abuse and Alcoholism. Unpublished data from
the 2001–2002 National Epidemiologic Survey on Alcohol and Related
Conditions. 2004. Psychiatr Serv 58:424, 2007
National Institute on Alcohol Abuse and Alcoholism: Helping Patients Who
Drink Too Much: A Clinician’s Guide, Updated 2005 Edition. Available at
http://www.niaaa.nih.gov/guide
National Institute of Mental Health: Suicide Prevention. Available at http://
www.nimh.nih.gov/health/topics/suicide-prevention/index.shtml. Ac-
O’Malley SS, Jaffe AJ, Change G, et al: Naltrexone and coping skills therapy for
alcohol dependence: a controlled study. Arch Gen Psychiatry 49:881–887,
1992
Pescosolido BA, Fettes DL, Martin JK, et al: Perceived dangerousness of chil-
dren with mental health problems and support for coerced treatment. Psy-
chiatr Serv 58:619–625, 2007
Pihl RO, Lemarquand D: Serotonin and aggression in the alcohol-aggression re-
lationship. Alcohol Alcohol 33:55–65, 1998
Presley CA, Leichliter JS, Meilman PW: Alcohol and Drugs on American Col-
lege Campuses: A Report to College Presidents: Third in a Series, 1995,
1996, 1997. Carbondale, Core Institute, Southern Illinois University, 1998
Ramisetty-Mikler S, Caetano R: Alcohol use and intimate partner violence as
predictors of separation among U.S. couples: a longitudinal model. J Stud
Alcohol 66:205–212, 2005
Rammes G, Mahal B, Putzke J, et al: The anti-craving compound acamprosate
acts as a weak NMDA-receptor antagonist, but modulates NMDA-receptor
subunit expression similar to memantine and MK-801. Neuropharmacol-
ogy 40:749–776, 2001
Ray LA, Hutchison KE: Effects of naltrexone on alcohol sensitivityand genetic
moderators of medication response. Arch Gen Psychiatry 64:1069–1077,
2007
Regier DA, Farmer ME, Rae DS, et al: Comorbidity of mental disorders with al-
cohol and other drug abuse. JAMA 264:2511–2518, 1990
Roth JA: Psychoactive Substances and Violence. Washington, DC, National In-
stitute of Justice, 1994
Saitz R: Clinical practice: unhealthy alcohol use. N Engl J Med 352:596–607, 2005
Salinsky MC, Storzbach D, Spencer DC, et al: Effects of topiramate and gaba-
pentin on cognitive abilities in healthy volunteers. Neurology 64:792–798,
2005
Spunt BJ, Goldstein PJ, Bellucci PA, et al: Race/ethnicity and gender differences
in the drugs-violence relationship. J Psychoactive Drugs 22:293–303, 1990
Steadman HJ, Mulvey EP, Monahan J, et al: Violence by people discharged from
Swanson JW: Mental disorder, substance abuse, and community violence: an

epidemiological approach, in Violence and Mental Disorder: Develop-
ments in Risk Assessment. Edited by Monaghan J, Steadman HF. Chicago,
IL, University of Chicago Press, 1994, pp 101–136
Tinklenberg JR, Roth WT, Kopell BS, et al: Cannabis and alcohol effects on as-
saultiveness in adolescent delinquents. Ann N Y Acad Sci 282:85–94, 1976
U.S. Department of Justice: Family violence statistics: including statistics on
strangers and acquaintances. Washington, DC, U.S. Department of Justice,
Office of Justice Programs, Bureau of Justice Statistics, 2001. Available at
http://www.ojp.usdoj.gov/bjs/pub/pdf/fvs02.pdf. Accessed October 6,
2005
Vaillant GE: The Natural History of Alcoholism. Cambridge, MA, Harvard Uni-
versity Press, 1983
Vaillant GE: The Natural History of Alcoholism Revisited. Cambridge, MA,
Harvard University Press, 1995
Vaillant GE: Alcoholics Anonymous: cult or cure? Aust N Z J Psychiatry 39:431–
436, 2005
Virkkunen M, Linnoila M: Serotonin in early onset, male alcoholics with violent
behavior. Ann Med 22:327–331, 1990
Volavka J: Neurobiology of Violence, 2nd Edition. Washington, DC, American
Psychiatric Publishing, 2002
Volavka J, Dornbush R, Feldstein S, et al: Marijuana, EEG and behavior. Ann NY
Acad Sci 191:206–215, 1971
Volavka J, Levine R, Feldstein S, et al: Short-term effects of heroin in man. Arch
Gen Psychiatry 30:677–681, 1974
Volpicelli JR, Alterman AI, Hayashida M, et al: Naltrexone in the treatment of
alcohol dependence. Arch Gen Psychiatry 49:876–880, 1992
Widom CS, Hiller-Sturmhofel S: Alcohol abuse as a risk factor for and conse-
quence of child abuse. Alcohol Res Health 25:52–57, 2001
Witkiewitz K, Marlatt GA: Therapist’s Guide to Evidence-Based Relapse Pre-
vention. Boston, MA, Elsevier Academic, 2007, pp 50–54
World Health Organization: Facts on child maltreatment and alcohol. n.d.
Available at http://www.who.int/violence_injury_prevention/violence/
world_report/factsheets/ft_child.pdf. Accessed March 2007
Zukin SR, Sloboda Z, Javitt DC: Phencyclidine (PCP), in Substance Abuse: A
Comprehensive Textbook, 4th Edition. Edited by Lowinson JH, Ruiz P,
Millman RB, et al. Philadelphia, PA, Lippincott Williams & Wilkins, 2005,
pp 324–335
C H A P T E R 9
P ersonality disorders are enduring, inflexible, and pervasive patterns

of inner experience and external behavior that are maladaptive and det-
rimental to one’s overall level of functioning and are in contrast with
cultural expectations (American Psychiatric Association 2000). Al-
though personality disorders are associated with a guarded or poor
prognosis for change, lumping them into a single, global construct is
clinically inappropriate. It is more accurate, and more productive, to
view the personality disorders as heterogeneous, having symptoms
and behaviors best viewed on a continuum within each diagnosis.
There is a great deal of individual variation in the severity and perva-
siveness of symptoms. Many individuals with personality disorders
live relatively normal lives. Those whose disorders are more severe,
those with comorbid mental, physical, or substance abuse disorders,
and those whose lives are interrupted by periods of substantial internal
or external stress tend to display more severe symptoms and lower lev-
els of functioning.
The association, when there is any, between mental illness and vio-
lence has long been a subject of clinical and experimental debate. We do
not review in detail the data and arguments on either side of that issue.
We examine the assessment and management of violence potential in
people with personality disorders whose condition or behavior creates
concern. This chapter focuses on understanding and managing some
aspects of violence risk. It does not address “prediction” or “cure.”
161
Greater awareness of psychological and environmental characteristics

associated with violence potential—marked or subtle—in some people
with personality disorders should help clinicians and researchers iden-
tify patients or evaluees who are more likely than others to engage in
violent behavior. We hope that the present chapter will be helpful to
mental health professionals involved in forensic work and the criminal
justice system and to those who work with personality-disordered pa-
tients in inpatient, outpatient, and correctional settings.
Mental Disorders and Violence

Until a few years ago, various authors suggested that people with men-
tal disorders are no more likely to engage in violent behaviors than
those in the general population (e.g., Monahan 1981). Rabkin (1979),
however, reviewing studies published between 1922 and 1978, found
with great consistency that patients released from public psychiatric fa-
cilities had higher arrest and conviction rates for violent crimes than did
the general population. She concluded that “mental patients are more
likely to be arrested for assaultive and sometimes lethal behavior than
are other people” (p. 24). Swanson and colleagues’ (1990) epidemiolog-
ical study found that individuals diagnosed with a major mental disor-
der showed significantly higher rates of violent behavior than individ-
uals with no apparent psychiatric disorder.
Other studies have also supported the notion that, overall and very
generically, people with some type of mental disorder are more prone
to violent behavior than those in the general population (Krakowski et
al. 1986; Mulvey 1994; Nestor 2002; Swanson et al. 1990, 1996; Tardiff et
al. 1997). Monahan (1992) himself, in an apparent reversal of some of his
previous beliefs, wrote , “I now believe that there may be a relationship
between mental disorder and violent behavior, one that cannot be
fobbed off as chance or explained away by other factors that may cause
them both. The relationship, if it exists, is probably not large, but may
be important for both legal theory and…social policy” (p. 511).
Perhaps the most comprehensive search for associations between
psychiatric illness and violent behavior came from the MacArthur
Foundation study of mental disorders and violence (Monahan et al.
2001), a systematic, prospective, multisite examination of more than
1,100 male and female psychiatric patients. The research team used
carefully controlled diagnoses and multiple methods of data collection
to assess more than 130 potential risk factors for violence in subjects fol-
lowed for 20 weeks after discharge from psychiatric facilities. The forms
of violence studied included battery resulting in bodily harm, sexual
Personality Disorders ❘ 163
assault, assaultive acts involving weapons, and threats. It came as

a surprise to some that after 1 year, “major mental disorder” (such
as schizophrenia, major depressive disorder, or history of manic epi-
sodes)—in the absence of a substance abuse diagnosis—was associated with
a lower rate of violence (17.9%) than that found in socially and demogra-
phically matched control subjects. Dually diagnosed groups with
“major mental disorder/substance abuse” and “other mental disorder/
substance abuse” (including personality disorders) had higher overall
rates of violence (31.1% and 43.0%, respectively) than those with a sole
diagnosis of personality disorder.
Personality Disorders and Violence

The simplistic acceptance of a broad, general association between large
groups of patients and the many forms of violence must be parsed into
specific diagnoses, kinds and levels of symptoms, kinds and levels of
violence, and context. The MacArthur Foundation data does this to a
laudable extent, but there is a dearth of methodologically sound studies
that focus on the relationship between personality disorders per se and
violent behavior (Coid 2002; Otto 2000). Otto (2000) commented that the
lack of research examining possible relationships between personality
disorders and violence risk may, in part, reflect “(1) limitations of the
psychiatric nomenclature of personality disorders generally and (2) that
assessment of the psychopathy construct (via the PCL-R) is better re-
fined than assessment of any other personality disorder or personality
style” (p. 1248).
A review of those studies that do attempt to explore potential rela-
tionships between violent behavior and personality disorders suggests
a lack of conformity in operational definitions of personality disorder.
Some studies include subjects described as having various personality
“traits” or merely having endorsed a number of diagnostic criteria. Oth-
ers incorporate factor or cluster analysis of psychological test data (e.g.,
defining a personality disorder based on elevated scale scores from the
Millon Clinical Multiaxial Inventory, Personality Assessment Inventory,
or Minnesota Multiphasic Personality Inventory). Still others refer to
personality types or clusters that do not clearly correlate with any DSM-
IV-TR (American Psychiatric Association 2000) or ICD-10 diagnosis.
Another factor that can significantly affect research findings, and
should call into question the extent to which some findings can be gen-
eralized, is the point in time at which the diagnosis was given. Kra-
kowski et al. (1986) pointed out that the “fact that a patient is violent
may influence the type of diagnosis which he or she receives. This is
particularly true of personality disorder, where the violence itself may

serve as a basis for making the diagnosis” (p. 132).
A closer look at study designs suggests that readers must distin-
guish between violent behavior (based on self-report, criminal convic-
tion, and/or review of collateral data) and hostile/aggressive behavior.
The latter may or may not be the same as violent behavior but is easier
to examine in experimental settings.
The MacArthur Foundation project highlights the importance of
multiple measures of data collection in violence research. For example,
Steadman et al. (1998) observed 1-year violence rates of 4.5% when ex-
amining agency records of discharged mental patients, but violence
rates for the same subjects were six times greater (27.5%) when three
separate sources of information were used (agency records, interviews
with patients, and interviews with a collateral source familiar with the
subject’s behavior in the community).
Other validity and reliability concerns within the violence and risk
assessment literature include single-site studies, relatively small sam-
ple sizes, limited or nonrepresentative subject populations (including
various “extreme” populations), reliance on forensic and psychiatric in-
patient populations, and failure to adequately consider common con-
founding variables such as comorbid psychiatric disorders, substance
abuse, and prior history of violence. All of these methodological issues
limit the extent to which the results of any study can be viewed with
confidence, compared with other research, or accurately generalized to
community populations and society.
Lawmakers increasingly appear to be using skewed concepts of di-
agnosis to support legislation for the management of people whom
they consider a threat to society. Sex offenses, in particular, have engen-
dered special, often draconian, commitment statutes and procedures in
many states. Most of those procedures, different from ordinary civil
commitment of the mentally ill, rest not on a “psychiatric” disorder but
on a nonclinical concept of “behavioral abnormality” that often resem-
bles (and sometimes uses) “personality disorder” as evidence of predis-
position to sexually violent behavior (Kansas v. Crane 2002; Kansas v.
Hendricks 1997; Leong and Silva 2001).
Variations in Diagnosis-Related Behavior

The presence of a personality disorder diagnosis does not imply that ev-
eryone with that diagnosis has the same risk of violence. Although
broad similarities can be used for diagnosis and categorization, people
with personality disorders are heterogeneous in the variety, consis-
tency, and intensity of their symptoms. Costello (1996) noted that the
“use in research of such complex polythetic categories of personality
disorder made up of heterogeneous sets of experiences and behaviors
makes it very difficult to interpret research findings” (p. 1). In addition,
despite the definition and conceptualization of personality disorders as
an enduring pattern of inner experience and behavior that is inflexible,
pervasive, stable, and of long duration, some have argued against a
blanket position that all personality disorders are always stable (Coid
2003; Rogers and Shuman 2005).
Understanding the heterogeneous nature of violence is imperative
for clinicians and researchers attempting to develop effective manage-
ment and treatment approaches and for readers who review their find-
ings. Both ordinary experience and clinical and research findings lead
one to view violence itself as a multifaceted construct varying widely in
quality, severity, purpose, duration, and frequency. Even very specific
forms of violence can have a variety of causes (see Widiger and Trull
1994, p. 212). Swanson et al. (2006) commented on that heterogeneity in
their recent study of violence rates among schizophrenics, noting that
“[v]iolent behavior occurs within a social-ecological system, involving
a whole person with a particular life history and a state of health or dis-
ease, interacting with a particular social surround” (pp. 490–491). Each
individual’s risk level varies with the degree to which his or her person-
ality characteristics and dimensions may or may not be exposed to var-
ious environmental variables (Nestor 2002). In general, what accounts
for violent behavior in one individual with a given personality disorder
does not necessarily portend violent behavior in others with the same
diagnosis.
Legal Responsibility and Violence Associated With

Personality Disorder
The diagnoses and situations discussed in this chapter are not generally
associated with legal exoneration due to mental illness (e.g., being “not
guilty by reason of insanity”), but symptoms can mitigate criminal cul-
pability to some extent. Conditions that affect one’s intent are relevant
to legal matters. Perpetrators almost always bear at least some, usually
all, responsibility for their behavior, provided they are capable of un-
derstanding the harmfulness or illegality of the behavior and that other
means of dealing with the situation are more appropriate (such as call-
ing police to handle real or perceived threats, dealing nonviolently with
spouses suspected of having affairs, or choosing nonviolent, non-
destructive ways to experience personal stimulation or excitement).
Treatment and Management

The treatments discussed in this section are, in most cases, necessarily
generic. Psychotherapeutic repair of personality deficit is exceedingly
difficult and rarely available to patients even when they are willing to
tolerate the emotional, temporal, and financial costs involved. Highly
specialized, resource-intensive treatments for some characterological
disorders (such as severe borderline traits) may be effective for those
who qualify clinically and have access to innovative programs, but such
opportunities are a rare exception rather than the rule.
Equally relevant to this chapter, there is no psychiatric treatment or
other reasonable clinical answer to violence that is a discrete symptom
of a personality disorder (unlike some other forms of violence asso-
ciated with, for example, certain paraphilias, severe depression, the
mood instability of bipolar disorder, frank psychosis, or ictal or other
neurological “dyscontrol”). Treatment of problems such as anxiety and
depression, unstable mood or affect, psychotic thinking, and inade-
quate social or relationship skills may decrease the risk of violence as it
improves patients’ ability to cope with internal and external threat
without decompensating.
Setting aside direct treatment for the moment, much of our (mental
health professionals’) usefulness lies in assisting potential victims, fam-
ilies of real and potential perpetrators and victims, and others who have
frequent contact with personality-disordered people (including law en-
forcement personnel and corrections staffs). Helping members of these
groups to better understand potential perpetrators, to modify (or avoid)
contact with them, and sometimes to contain and prevent them from
harming others is an important part of decreasing the damage done by
violence and, in some cases, preventing violence in future generations.
Three Unavoidable Treatment Issues

• Personality disorders are rarely ego dystonic. Most people with person-
ality disorders or aberrant character traits do not seek psychiatric
help. Those who do are often trying to alleviate symptoms but are
unable or unwilling to address characterological issues. They may
comply with treatment aimed at anxiety or depression but usually
shun serious psychotherapeutic approaches to their maladaptive be-
haviors and adaptations. Many aspects of personality disorders are
simply not amenable to commonly available treatment.
• Most patients and situations of violence that come to professional attention in-
volve coexisting disorders and conditions. Treating or managing comorbid
or coexisting conditions (many often external to the patient) may alle-

viate some potential for violence. The presence of a personality disor-
der usually makes it more difficult to treat accompanying conditions.
• Violence and risk of violence, with or without a personality disorder, is often
associated with intoxication. Management of simple intoxication and
control of damaging behavior associated with it are generally out-
side the province of the mental health professions. Treatment of sub-
stance abuse disorders may reduce violence risk, but personality
disorders usually worsen the prognosis. We generally agree with the
common principle of first getting the patient clean and sober, then
reassessing for personality disorders and other psychiatric condi-
tions. Removing the substance issue often clarifies, and sometimes
erases, evidence of other mental disorders.
These bullet points suggest that nonpsychiatric, non–mental health

approaches are often more important than mental health professionals in
the prevention and management of violence by people with personality
disorders and the protection of potential victims from those people.
The discussions in the next section are predicated on personality
characteristics, generally without regard to other mental disorders or
intoxication. Our examples apply to many situations in the “real world”
in which violence related to personality disorders may not come to the
attention of mental health professionals or be very amenable to our in-
tervention.
Kinds of Violence Associated With

We have chosen not to parse violent behaviors by specific personality
disorder. The common thread for our purpose is the violence, not the di-
agnosis. Understanding similarities among kinds of violence is more
useful, and better related to common presentations, than separating be-
havioral and psychological issues by diagnosis (although consideration
of DSM-IV-TR “clusters” is sometimes helpful). The categories we have
developed and present here are experience-based and practical. They
are not mutually exclusive, nor are they intended to create an assess-
ment “decision tree.” All examples are taken from actual cases.
Purposeful, Instrumental Violence

Some personality disorders predispose one to violence for obvious per-
sonal gain. Disorders that decrease or eliminate a sense of empathy or
otherwise diminish the potential perpetrator’s thoughtful consider-

ation of other people increase this risk. Antisocial, narcissistic, and bor-
derline disorders are common examples. Such violence is targeted rather
than random. It includes acts in which violence is a means to a con-
scious, gainful end (such as a robbery or preventing apprehension after
a crime) or designed to manipulate or mislead another into some
wanted behavior (such as manipulative behavior by persons with anti-
social, narcissistic, or borderline personality). Violence for revenge and
violence for hire should be considered here, provided there is a charac-
terological deficit in the perpetrator that allows it to take place.
Case Example 1
A man who met diagnostic criteria for antisocial personality ap-
proached an elderly man in an isolated area of a park and demanded
money. When the robbery victim resisted giving up his valuables, the
perpetrator hit him repeatedly in order to force him to comply. After the
victim had been subdued, the perpetrator took his wallet and ran away.
Case Example 2
An injured worker suing his employer for millions of dollars lost his
lawsuit because of a somewhat technical judicial decision. His attorney
noted that the worker was very upset over the loss, and he referred the
client to a psychologist. During a brief course of treatment, the therapist
uncovered long-standing signs of paranoia but no frankly delusional
material. Believing the sessions to be completely confidential, the pa-
tient/plaintiff eventually admitted that since losing the lawsuit, he had
rehearsed sabotaging the defense attorney’s car, had actually entered
the lawyer’s property and examined the engine and brake lines, and
had a plan to murder the judge.
Case Example 3
A patient with borderline personality was distraught about the possibil-
ity of losing custody of her children during a divorce. She told her psy-
chiatrist that her estranged husband had beaten her in front of the
children and that she was afraid he would harm them as well. She of-
fered bruises on her neck and arms as evidence. The psychiatrist helped
her to contact police and obtain a restraining order prohibiting her hus-
band from visiting her or the children.
As the police and the state child protective agency investigated the
matter, attention began to shift to the patient herself. It eventually be-
came clear that she inflicted her own bruises, then invited her husband
to her house, started an argument in front of the children, and began hit-
ting him. When he did not hit back and began to leave, she dramatically
fell screaming to the floor and loudly proclaimed “Daddy hit me!
Daddy’s hurting me! Run, or he’ll hurt you, too!”
Risk Assessment
An experienced evaluator of antisocial, paranoid, or borderline persons
should recognize the presence of some risk, but the level of risk and the
probability of violence may be difficult to ascertain, even in patients
with substantial aberrant personality traits. Those with histories of vio-
lent behavior, paranoid or mildly psychotic thinking in a context of pos-
sible gain, and/or marked lack of empathy should raise additional
concern.
Treatment/Management of the Violent Behavior

Physical prevention of violence, such as by incarceration, removal of
potential victims, or alleviation of risk-laden situations, may be the
most practical approach, particularly in the short term. Treatment of
symptoms such as psychosis or intolerable anxiety usually decreases
risk but should not be considered a lasting solution to personality-
based risk. Specialized psychotherapy and psychosocial efforts to ad-
dress the personality deficit itself, such as helping a characterologically
borderline or paranoid patient develop internal alternatives to acting
out, are theoretically logical but require expertise, resources, and time
that are rarely available. Even in the best cases of psychiatric and psy-
chotherapeutic care, external or pharmacological management, if indi-
cated, should be considered while waiting for insight and change (if
insight and change are to be forthcoming at all).
Purposeful, Noninstrumental Violence

Noninstrumental violence may be purposeful, but the injury to others
is outwardly unnecessary; it is violence for the sake of excitement or
stimulation, as contrasted with that aimed toward tangible gain. It may
add parenthetically to the pleasure of a stimulating or antisocial activ-
ity, but actually injuring others is not integral to the activity’s purpose.
Bystanders may refer to the violent part of the overall behavior as
“senseless” or “random,” but it has an emotional purpose, such as stim-
ulation. The targets may be random, but the behavior that places others
in danger is intentional.
This concept should not be confused with violence in which the dan-
ger to others is unanticipated or not intentional, such as that incidental
to impulsively overreacting to an affective state associated with threat-
ened emotional survival (e.g., intolerable anxiety, stifling entrapment,
acute abandonment, or marked humiliation; see “Nontargeted, Impul-
sive Violence Incidental to Emotional Escape” later in this chapter).
Case Example 4
A man who met diagnostic criteria for antisocial personality but not for
any Axis I disorder broke into a home and stole several items while the
occupants were asleep. He then set fire to the house in order to hide his
crime, deflect blame from himself, and destroy evidence that might
have incriminated him. The occupants were awakened by a smoke
alarm and escaped, but they could easily have been injured or killed.
After his arrest several days later, the robber described setting the
fire as necessary to avoid being caught. He denied wanting to harm the
occupants, describing the fire as simply a means of avoiding arrest. It
was “nothing personal,” just “something I had to do whether they [the
occupants] were there or not.” Warning or awakening them so that they
could escape had not crossed his mind.
Case Example 5
A young man with a long history of relatively minor antisocial acts en-
gaged in a drag race on a city street. As he neared the end of the race, he
realized he was about to run a red light. Nevertheless, he continued to
accelerate, ran the red light, and struck another car. The driver of the
other car was killed. When testifying about his reckless behavior, he de-
scribed it entirely in terms of his taking a thrilling chance with his own
life, seeming oblivious to any responsibility for others’ safety. He under-
stood the chance of an accident, and the chance that someone might be
hurt or killed, but he described the danger of racing purely in terms of
a focus on himself, saying “I can live with those odds.”
Risk Assessment
This form of violence is overrepresented in those with substantial anti-
social and asocial character traits. In addition to lack of empathy or rec-
ognition of other’s needs and feelings (a common thread in many kinds
of violence), the risk of purposeful but noninstrumental violence may
be heralded by a potential perpetrator’s wish for pleasure or need for
stimulation that overshadows his judgment, impulse control, and ap-
preciation of future consequences.
Treatment/Management
When symptoms are recognizable and treatable, the general treatment
principles described in the last section apply (e.g., treatment for sub-
stance abuse or mood instability).
Purposeful, Targeted, Defensive Violence

Purposeful, targeted, defensive violence is generally a maladaptive at-
tempt to stop some intolerable affect, often associated with humiliation
or abandonment. The violent reaction to such a condition, which threat-

ens the integrity of the person’s ego, may be rapid (see also the sections
“Targeted, Impulsive Violence” and “Nontargeted, Impulsive Violence
Incidental to Emotional Escape” in this chapter) or it may be carefully
planned. The target may seem illogical to an observer (e.g., it may be re-
lated to paranoid ideation or some other idiosyncratic source). The level
of violence is often baffling until one realizes its internal meaning. Ex-
amples include the sometimes extreme behavior of paranoid stalkers,
who may create near-delusional scenarios of competition or abandon-
ment, and paranoid “defenders,” who believe they must defend them-
selves from imagined or exaggerated slights or threats. Dependent,
avoidant, and schizoid traits occasionally increase risk. When such
thinking becomes more than mildly delusional or other aspects of the
person’s function are significantly compromised, an Axis I disorder
should be considered.
Case Example 6
Frequent arguments between a middle-aged man and his wife, often in-
volving intoxication with alcohol, routinely led to his threatening or as-
saulting her, her threatening or briefly leaving him, and then his
successfully begging her to stay. Eventually, the wife resolved to ignore
his entreaties and promises and filed for divorce. He did not believe she
would go through with the divorce, but when he came home one night,
she had locked him out. He stayed with a friend for a few days, calling
her often and thinking she would change her mind.
After several days, he was served with the divorce papers. He drove
to her place of work and once again pleaded with her to reconsider. She
refused, adding (in front of her coworkers) that he had never been much
of a husband and had never satisfied her sexually. He returned to his car,
took a shotgun out of the trunk, went back into the building, killed her,
and then waited for police to arrive.
Risk Assessment
Characterological paranoia is among the most dangerous personality
traits. It is associated with both domestic and general violence. Many
people with paranoid personality routinely imagine and rehearse (men-
tally or literally) violent “solutions” to paranoia-created scenarios.
Truly delusional persons with Axis I disorders are much more likely to
be seen by a mental health professional than are those with paranoid
personality alone. Passive, dependent, or avoidant people do not antic-
ipate violence but may become dangerous when trapped or restrained
and unable to escape emotional pressure. However, they usually can
adjust their environments to decrease their anxiety (and concomitantly
lower their risk of violence). Threatened breach of narcissistic character

defenses carries risk as well.
Characterologically paranoid people, who go through life with an over-
determined expectation of trouble, are very difficult to manage (and
sometimes to recognize) unless or until their public behavior raises con-
cerns. Persons whose personality traits lead them to avoid confronta-
tion and anxiety can usually be relied on to avoid triggering situations
if they have the choice to do so. Once such a person is in a setting that over-
whelms even resilient characterological defenses, such as a perceived
inescapable threat or restraint, the best management approaches in-
volve quickly defusing the situation, isolating or containing the poten-
tially violent person, or removing potential victims.
Targeted, Impulsive Violence

Targeted, impulsive violence involves striking out, without planning,
at a perceived or psychological threat that others would not consider
to warrant the same quality or quantity of violence. The victim is spe-
cifically targeted, often in a desperate effort to eliminate (literally or
symbolically) the source of an acute psychic threat. Examples of such
violence, which erupts in order to escape an intractable situation by
eliminating the source, include enraged reactions to acutely perceived
humiliation or abandonment. Although severe examples are not com-
mon, people who are characterologically paranoid, narcissistic, or ex-
quisitely sensitive to loss (as found in borderline personality) are
predisposed, to a greater or lesser extent, to such actions given a trigger-
ing setting or environment. Dependent, obsessive-compulsive, and
avoidant persons are at less risk but may decompensate into violent be-
havior under remarkable circumstances (see Coid 2002 and some other
studies of prison populations).
Note that we are not referring here to violence whose victims are
incidental to uncontrolled rage or escape behavior by, for example, “be-
ing in the wrong place at the wrong time” (see “Nontargeted, Impulsive
Violence Incidental to Emotional Escape” later in the chapter).
Case Example 7
Dr. X, an otherwise competent abdominal surgeon, was known for both
his skill and his irritable, narcissistic manner. He led a regimented life,
with little warmth for family or friends, the barest superficial acknowl-
edgment of the roles of others in his cases and other achievements, and
no tolerance for criticism. The latest of many operating room incidents

involved his berating a nurse when she pointed out unacceptable ooz-
ing from the omentum as he began to close a laparotomy. Nevertheless,
he stopped and dealt with the bleeding before proceeding with the clos-
ing. Another nurse commented under her breath, “Saved by a nurse.”
The surgeon finished the closing, then calmly asked, “What did you
say?”
The nurse who had made the comment said something like “I didn’t
mean anything disrespectful. I wanted to compliment ‘J’ [the nurse who
noted the bleeding] for making a good catch. She probably saved the pa-
tient from reopening.”
The surgeon replied hotly, “J works for me. She did her job. Every
damned one of you works for me. I’ll let you know when you make a
good catch and I’ll be damned if I’ll tolerate anybody in this hospital
criticizing my surgery until you’ve been through medical school and
residency yourself.”
J came to her colleague’s defense. “No problem, Dr. X. We’ll just get
the patient out of here and awake.”
Dr. X then raised his voice and continued to rant in spite of verbal
efforts to calm him. Finally, one of the nurses, concerned about the situ-
ation and the patient’s safety, announced that she was calling for the
chief nurse of the surgery suite. Dr. X responded by tossing a tray of
bloody sponges in the nurse’s direction and storming out of the operat-
ing room.
Dr. X was disciplined by the medical staff. He protested their verdict
and retained a lawyer to sue for libel and expunge his record. The law-
suit was later dismissed. The medical staff matter was eventually
reported to the state medical licensing board, which added its own
censure.
Case Example 8
Mr. S was known as a nice, quiet fellow, the adult son of a very aggres-
sive, poorly liked father whose bullying controlled most people close to
him. Mr. S behaved in an almost opposite manner, passive and appear-
ing dependent on his father for income and a place to live. In over a de-
cade of adulthood, he had traveled and interacted socially with others
but had never held a meaningful job for more than a few weeks, had
never married, and had never lived away from his father. Privately,
Mr. S dreaded the thought of being like his father, who had abused him
during childhood and as an adult.
One night while both were intoxicated, his father began once again
to bully and humiliate Mr. S. At some point, the combination of physical
and emotional humiliation reached an intolerable level; Mr. S grabbed
his father’s arms to restrain them. The father laughed derisively, break-
ing the son’s hold, slapped him repeatedly in the face, and called him
“my little bitch.” Mr. S exploded, pummeled his father to the ground,
and finally shot him in the chest with a shotgun kept nearby in case of
intruders.
When his father lay obviously dead, Mr. S wrapped him in a bed-
sheet and bound the body with duct tape, then retreated to his bedroom,
locked the door, and went to bed. The next day, he called an attorney
and gave himself up to police. Asked later about what he did to the
body, Mr. S replied that although he knew his father was dead, he could
not feel truly secure until he had wrapped and bound him and locked
the bedroom door.
Risk Assessment
Many violent acts of this type occur when an external event threatens
poorly defended fears of inadequacy or abandonment. Some people
with severely dependent, paranoid, narcissistic, schizotypal, or obses-
sive-compulsive traits—characteristics that decrease one’s ability to
marshal and rely upon more efficient internal defenses when trapped in
emotionally intractable situations from which one cannot escape—can
be very dangerous. Such conditions, particularly inability to escape an
intolerable and anxiety-producing situation, increase the likelihood of a
violent reaction designed to stop the pain and escape the threat. When
conditions are extremely stressful, even schizoid and avoidant persons
may revert to primitive, violent actions to defend their egos. Intoxica-
tion is a substantial risk factor, as are some kinds of emotional attacks
and idiosyncratic emotional triggers (e.g., repeated, inescapable de-
meaning or “in your face” challenges during arguments with a spouse
or competitor).
“Treatment” of the characterologic vulnerability, when motivation and
resources are present, is described elsewhere. See also the general de-
fusing, separation, and containment principles described above, as well
as treatments for coexisting substance abuse, mood instability, and
other noncharacterologic factors mentioned earlier in this chapter.
Nontargeted, Impulsive Violence Incidental

to Emotional Escape
Impulsive violence incidental to emotional escape is generally nontar-
geted, although the person who triggers the intractable emotional state
may bear the brunt of the violence if he or she is in the path of egress.
The purpose of the behavior is rapid escape from a situation that has
created an acute, intolerable internal situation for which the personal-
ity-disordered person has inadequate emotional defenses and behav-
ioral alternatives. It is different from the type just discussed (targeted,
impulsive) in that the anxious or humiliated person does not seek to

mitigate or destroy the source of the pain, only to escape from it.
Case Example 9
Ms. T was a 43-year-old woman with borderline personality disorder
and very primitive attachment needs. She and her 24-year-old daughter
had an extraordinarily hostile-dependent relationship that was often
characterized by rather obvious manipulations designed to keep the
daughter physically and emotionally bound to the mother. The daugh-
ter had tried to move away on several occasions, but each time she had
changed her plans to meet her mother’s needs and continued to live on
her mother’s property. At one point, the daughter approached her
mother once again—by telephone, to avoid a personal confrontation—
to tell her she was moving in with a boyfriend who lived some distance
away. She called from her place of work.
The telephone conversation soon deteriorated into a volatile event.
Ms. T alternated among superficially rational “suggestions” that the
daughter reconsider and have her boyfriend move into the daughter’s
trailer on the mother’s property, pleas that the daughter consider the
mother’s health conditions (which were not particularly serious), sar-
castic comments that the boyfriend would probably leave her, and,
eventually, angry threats to rent the daughter’s house to someone else
so that she could never “come home.”
The daughter would not budge. She repeatedly told her mother that
she was indeed going to move away and parried each of the manipula-
tive comments and threats with sarcasm and threats of her own (such as
“You’ll never see your grandchildren” and “You’ve been sick for years;
let me know when it gets really serious”). The daughter finally hung up
in the middle of her mother’s tirade.
Ms. T got into her car to drive to the daughter’s workplace, shaking
with anger and anxiety. On the way, she drove very recklessly, failed to
yield at an intersection, and hit another car, injuring several people.
Case Example 10
A woman with severe borderline and paranoid traits was being told that
she had lost custody of her children. A social worker and a trainee were
trying to treat her as gently as possible while making it clear that she would
only be allowed to see her children, who had been removed from the
home, in a supervised setting. The woman listened for a moment, then be-
gan screaming that none of the things they were saying about her was true,
that she was a good mother, and that she refused to listen to their lies.
The social work trainee raised her voice and somewhat assertively
tried to confront the woman, recounting her past abusive acts in order
to make her understand why her parental rights were being terminated.
The woman only became more agitated, screamed louder, and bolted
from the room, pushing the senior social worker away from the door
and into an aquarium, which fell and broke, cutting her arm and neck.
Risk Assessment
This level of fragility and potential for decompensation is not typical of
most people with personality disorders and may suggest an Axis I dis-
order. Those prone to such reactions have marginal egos that are inade-
quately protected by sometimes superficially resilient, but inwardly
brittle, defenses. Their personalities may have substantial, poorly inte-
grated borderline, schizotypal, dependent, obsessive-compulsive, and/
or avoidant features. They often seem outwardly stable but have inner
worlds kept artificially free of mental controversy that might threaten
their emotional lives. They may show stilted or even ritualistic behav-
iors in order to control the impact of the external environment on those
inner lives, or they may simply choose isolation and other defenses as
means of avoiding stressors.
Careful examination of such persons’ lives may reveal reaction for-
mation, an extraordinary need to defend desperately against discovering
in oneself some frighteningly destructive core emotion or self-reviled de-
pendency. For some, that veneer can become dangerously weak under
stressful (often idiosyncratic) circumstances such as intoxication, loss, or
inescapable humiliation.
In case example 10 above, the trainee’s confrontational manner in-
creased the pressure on the woman’s already assaulted, fragile ego. A
less confrontive approach would likely have prevented the accidental
injury to her colleague.
Random But Purposeful Violence

People who perform acts of random but purposeful violence derive
pleasurable stimulation from violence itself, often to instill a feeling of
power. It is neither a means to some profitable end nor merely an ad-
junct to some other exciting activity (omitting primarily sexual sadism,
which we view as an Axis I disorder even though its roots are often
characterologic). A particular, repetitive style of violence, such as snip-
ing with a rifle or setting others on fire, is common, but careful review
usually reveals other violent or sadistic behavior.
The victim may be stalked or the situation carefully planned in order
to set up the violent act (and often to plan escape); however, the victim
usually has no direct relationship to the perpetrator, nor is the particu-
lar victim associated with revenge or personal gain. He or she is a target
of convenience.
Randomness of victim choice does not imply random, impulsive, or

uncontrolled action. The violence is not a result of neurological dyscontrol,
a psychogenic impulse control or explosive disorder, or a thought disorder
(better discussed as Axis I or Axis III conditions); rather, it is self-absorbed,
antisocial, and uncaring, without empathy or sympathy. A wish to exert or
establish power over others, and over the passive portion of one’s own
psyche, is commonly an important component of the violent purpose.
Case Example 11
Two men decided to play a deadly “urban war game” in which they as-
sumed the roles of assassins. They outfitted a small van in such a way
that one of them could drive to an “assassination” location and park the
truck while the other sat in the back with a high-powered rifle. The rifle
was equipped with a telescopic sight so that shots could be taken from
some distance. The driver would spot a faraway victim, chosen at ran-
dom according to opportunity, and give the shooter a signal. The
shooter would then open a side window, fire, and quickly close the win-
dow, after which the van would drive away.
The pair were caught after killing several people. Upon evaluation,
neither met criteria for any significant psychiatric diagnosis except per-
sonality disorder with antisocial and (in one) paranoid traits.
Risk Assessment
As in the case of several other conditions described in this chapter, a his-
tory of this kind of violence, in reality or in substantial fantasy (e.g.,
with “rehearsing” behavior), increases risk of future violence. Those
with disorders whose hallmark is a lack of empathy, responsibility,
and/or impulse control, such as antisocial, narcissistic, or paranoid per-
sonality, are of most concern.
Management of persons in this group is generally practical and societal
(e.g., judicial, correctional) rather than psychiatric. For those rare cases
in which a treatable disorder such as mood instability or substance
abuse is relevant, see the treatment comments found earlier in this
chapter and elsewhere in the broader treatment literature.
Violence Related to Perceived/Feared Loss

or Abandonment
Violence in response to perceived abandonment is a special case of tar-
geted, usually purposeful, and instrumental violence that may be either
impulsive or calculated.
Case Example 12
A man and a woman had a 4-month dating relationship that fluctuated
between superficial intimacy and loud arguments. The man, who
treated the relationship primarily as one of sexual convenience, grew
tired of the woman’s volatile emotions and demands for proof of his
love. The woman clung to the hope that the relationship would lead to
a fairy tale marriage. Finally, he stopped calling her and began dating
someone else.
The woman was hurt and angry at being abandoned and at losing
what she had viewed as a lasting future of love and security. Over time,
the loss became less and less tolerable. Her anger grew. She saw no life
for herself without (her fantasy of) the ex-boyfriend and convinced her-
self that their relationship would have a chance if his new girlfriend
were gone.
One night, when driving by her ex-boyfriend’s house, she noticed
the girlfriend’s car in the driveway. She stopped and smashed the wind-
shield with a hammer, then drove away. The ex-boyfriend suspected she
had broken the windshield and called to confront her. She denied it but
took the call as an opportunity to rekindle the relationship and as evi-
dence that she was still in his thoughts. He told her the police had been
called and hung up. He continued to date the new girlfriend.
Two weeks later, the woman saw the ex-boyfriend and his new girl-
friend in a bar. When they noticed her, they got up to leave. She yelled
at them to stop and began to berate the girlfriend. The couple again
started to leave, and the woman, who was somewhat intoxicated, at-
tacked the girlfriend. She was restrained and later arrested.
Psychiatric evaluation of the woman revealed these facts as well as
a history of significant abuse by a stepfather, very unstable adult rela-
tionships, and episodes of depression and self-injury associated with
relatively minor losses. No frankly delusional material was evident.
Risk Assessment
People with flagrant manifestations of borderline coping should be
viewed with concern. Those with paranoid personality disorder are rel-
atively common perpetrators as well. Severely dependent character
traits in the absence of borderline features should raise consideration
for violence in some settings and contexts, albeit to a lesser extent. Chil-
dren with markedly borderline or paranoid parents are especially at
risk of either direct abuse or exposure to violent moods and unstable
parenting.
Treatment of ancillary symptoms of anxiety, mood instability, and occa-
sional psychotic thinking is relatively straightforward and often suc-
cessful with cooperative patients. Psychotherapy should be designed to
teach and reinforce realistic assessment of perceived loss and abandon-

ment, appropriate forms of soothing and coping in lieu of destructive
acting out, and use of supportive external resources (including the ther-
apist) rather than reliance on destructive coping mechanisms. Treating
professionals who suspect substantial risk of violence should monitor
their patients frequently for impending (including symbolic) loss and
abandonment, coping difficulties, and deteriorating mood stability.
Violence Related to Chronic Paranoia

or Related Misconception
Although we are not addressing chronically delusional or otherwise
psychotic states, paranoid and severely narcissistic character features
are often associated with episodic violence and enduring levels of ten-
sion or threat to others. Some stalkers are paranoid, acting out of a sense
of fear or defense against threat rather than erotomania or other signs
of an Axis I delusional disorder. Narcissistic individuals may errone-
ously view others as attempting to undermine their positions (and,
more accurately, their highly defensive sense of competence) and react
with irritability or outbursts when assailed by reality. Schizotypal per-
sons, usually well defended with self-absorbed isolation when in stable
settings, nevertheless often misperceive the nature and purpose of
those around them.
Case Example 13
Mr. H was a Vietnam veteran with a stable but childless marriage to a
Vietnamese woman whom he had brought to the United States. He was
generally domineering, expecting her to be a submissive wife. Once in
the United States, she pursued an education, getting a graduate degree
and becoming a college teacher, while he remained relatively unedu-
cated and generally unsuccessful in his small business. At home, his
wife tried to tolerate his dominating style, dislike for socializing, and
noticeable paranoid traits.
One of Mr. H’s few hobbies was working with his several dogs, pit
bulls who required a large pen and considerable care. He was quite gen-
tle with them, and although he was reclusive and suspicious and had a
history of severe childhood beatings from his father, there was no evi-
dence that he had ever physically abused his wife.
As his business failed, Mr. H spent more and more time working on
an elaborate backyard structure, the walls of which were created from
thick metal plates salvaged from a construction site. He described the
structure as a shelter for his dogs. Police would later describe it as a
“bunker,” but there was no other evidence that he was preparing for
some fantasized attack, and he had no known association with antigov-
ernment or “survivalist” groups.
Mr. H legally owned several weapons, some of which he kept at his

place of business (which was in an unsavory part of town and vulnera-
ble to robbery). A year before the events that brought him to treatment,
he was caught driving with a loaded handgun without a concealed
carry permit. The weapon was confiscated and never returned, despite
his frequent requests. He told several people that the confiscation was
illegal, and although he had excellent relations with the local police
through his business, the confiscation remained a sticking point in his
interactions with them.
Mr. H’s wife finally left him, ostensibly because of the way he
treated her and the widening divergence of their interests. He was very
upset about her leaving and became noticeably depressed. A family
physician prescribed an antidepressant, which he took only sporadi-
cally.
Late one night soon thereafter, he was stopped for speeding. The
officer noticed a shotgun and a handgun on the passenger seat of the
vehicle. He drew his service weapon and retreated to the patrol car.
Reports differ at this point, but it appears from patrol car video that
soon after he called for a second officer, the first officer fired at the truck,
starting an intensive exchange of gunfire. When the second officer ar-
rived, more shots were fired.
No officer was ever wounded, in spite of the police cars’ being hit
dozens of times by Mr. H’s shots. Mr. H was wounded in several places.
When he was finally extracted from his truck, several other weapons
and a drum of gasoline were found behind the seat.
Mr. H was arrested for attempted murder of a police officer. His de-
fense was that he was trying to commit “suicide by cop” and had been
on his way to a police station for that purpose. He had planned, he said,
to fire the weapons into the air, then die at the hands of the police whom
he respected so much, one of whom would become a “hero” for ending
the incident. Criminologic and psychiatric evaluations suggested that
the “suicide-by-cop” plan might have been real, but the overall impres-
sion was one of paranoid personality with schizotypal features and
episodes of depression. It appeared likely that the violent behavior was
related more to his paranoia and misperception, triggered by the imme-
diate situation, than to depressive suicidal intent. The first officer’s
behavior was probably part of the final triggering event, although what
might have happened had Mr. H. not been stopped for speeding is
unclear.
Risk Assessment
Those with paranoid personality, especially, warrant concern and mon-
itoring, particularly when there is a history of violence or threat. Chil-
dren in the family are at risk of both direct abuse and exposure to
violent moods and cold or unstable parenting. Severe narcissistic and
schizotypal traits suggest increased risk as well.
There is no specific treatment for patients with histories similar to that
of Mr. H. The pain of depression may cause them to seek treatment, but
compliance is a difficult issue, and antidepressant approaches do not
address the paranoid or other personality traits (although they may al-
leviate paranoia that stems from a mood disorder). Antidepressant
treatment may also contribute to a switch to manic or hypomanic be-
havior, which, in the context of paranoia or other aberrant personality
traits, can be quite dangerous. Adequate monitoring is important. Al-
though one cannot be certain, alleviation of Mr. H’s depression might
have changed the outcome.
Conclusion
With a few exceptions (e.g., antisocial personality), the potential rela-
tionships between personality disorders and violent behavior are
poorly studied. We encourage clinicians to view personality disorder as
a heterogeneous construct. That heterogeneity raises diagnostic and
methodological concerns about the reliability and generalizability of
much of the available research. Individual consideration, including un-
derstanding setting and context, is vital to improving risk assessment.
It follows that recognizing increased risk can allow clinicians to match
the problems and needs of individuals to available treatment when ap-
propriate and, perhaps more often, to recommend practical manage-
ment approaches.
Mental health professionals working with personality-disordered
individuals should review critically the better studies of violence risk,
particularly those pertaining to personality disorders, then extend that
focus beyond diagnosis. The case examples in this chapter highlight
characterologic vulnerability but also feature environmental factors re-
lated to increased risk. Clinicians who understand the importance of
setting and context, and their relationship to the internal vulnerability
associated with different personality traits, will have an easier time rec-
ognizing and assessing risk than many of their colleagues and will be in
a better position to help manage that risk.
Key Points
■ Although personality disorders are associated with a guarded or
poor prognosis for change, they should not be lumped into a
single, global construct. They are heterogeneous, with a variety
of symptoms and behaviors within each diagnosis.
■ Personality disorders are rarely ego dystonic. Most people with
personality disorders or aberrant character traits who seek profes-
sional help have coexisting disorders or conditions.
■ This chapter focuses on understanding and managing some
aspects of violence risk. It does not address “prediction” or “cure.”
■ The presence of a personality disorder diagnosis does not imply
that everyone with that diagnosis has the same risk of violence.
■ The diagnoses and situations discussed in this chapter do not gen-
erally imply lack of responsibility for one's actions. They are rarely
associated with exoneration from blame for criminal acts.
■ Violence and risk of violence, with or without a personality disor-
der, is often associated with intoxication of some kind.
■ The authors present a new, eight-category typology of character-
ologic violence, not intended to be mutually exclusive, whose
common thread is violence, not diagnosis per se.
1. Purposeful, instrumental violence
2. Purposeful, noninstrumental violence
3. Purposeful, targeted, defensive violence
4. Targeted, impulsive violence
5. Nontargeted, impulsive violence incidental to emotional
escape
6. Random but purposeful violence
7. Violence related to perceived/feared loss or abandonment
8. Violence related to chronic paranoia or consequent
misconception
References
Coid JW: Personality disorders in prisoners and their motivation for dangerous
and disruptive behavior. Crim Behav Ment Health 12:209–226, 2002
Coid JW: Epidemiology, public health, and the problems of personality disor-
der. Br J Psychiatry 182(suppl):s3–s10, 2003
Costello CG: The advantages of focusing on the personality characteristics of
the personality disordered, in Personality Characteristics of the Personality
Disordered. Edited by Costello CG. New York, Wiley, 1996, pp 1–23
Kansas v. Crane 534 U.S. 407 (2002)
Kansas v. Hendricks 521 U.S. 346 (1997)
Krakowski M, Volavka J, Brizer D: Psychopathology and violence: a review of
literature. Compr Psychiatry 27:131–148, 1986
Leong GB, Silva JA: Sexually violent predator, II. J Am Acad Psychiatry Law
29:340–343, 2001
Monahan J: Predicting Violent Behavior: An Assessment of Clinical Techniques.
Beverly Hills, CA, Sage Press, 1981
Monahan J: Mental disorder and violent behavior: perceptions and evidence.
Am Psychol 47:511–521, 1992
sity Press, 2001
Mulvey EP: Assessing the evidence of a link between mental illness and vio-
lence. Hosp Community Psychiatry 45:663–668, 1994
Nestor PG: Mental disorders and violence: personality dimensions and clinical
features. Am J Psychiatry 159:1973–1978, 2002
Otto RK: Assessing and managing violence risk in outpatient settings. J Clin
Psychol 56:1239–1262, 2000
Rabkin JG: Criminal behavior of discharged mental patients: a critical appraisal
of the research. Psychol Bull 86:1–27, 1979
Rogers R, Shuman DW: Fundamentals of Forensic Practice: Mental Health and
Criminal Law. New York, Springer, 2005
Steadman HJ, Mulvey EP, Monahan J, et al: Violence by people discharged from
Swanson JW, Holzer CE III, Ganju VK, et al: Violence and psychiatric disorder
in the community: evidence from the Epidemiologic Catchment Area sur-
Swanson JW, Borum R, Swartz M, et al: Psychotic symptoms and disorders and
the risk of violent behavior in the community. Crim Behav Ment Health
6:317–338, 1996
Swanson JW, Swartz MS, Van Dorn RA: A national study of violent behavior in
persons with schizophrenia. Arch Gen Psychiatry 63:490–499, 2006
Tardiff K, Marzuk PM, Leon AC, et al: Violence by patients admitted to a private
psychiatric hospital. Am J Psychiatry 154:88–93, 1997
Widiger TA, Trull TJ: Personality disorders and violence, in Violence and Men-
tal Disorder: Developments in Risk Assessment (The John D. and Cathe-
rine T. MacArthur Foundation Series on Mental Health and Development).
Edited by Monahan J, Steadman HJ. Chicago, IL, University of Chicago
Press, 1994, pp 203–226
C H A P T E R 1 0
Neurological and Medical

Disorders
Violent behavior has long been associated with focal brain lesions as
well as with diffuse damage to the central nervous system (CNS). Any
condition producing psychosis or mania may have aggression as a con-
comitant symptom. Medical conditions that do not directly affect the
brain can also be a cause of aggressive behavior due to diffuse effects on
CNS function. Irritability and aggression are a major source of morbid-
ity for many neurological and medical patients and increase the burden
on their families and other caregivers. In this chapter, we discuss com-
mon neurological and medical etiologies of aggressive behavior (see
Table 10–1). Identification and treatment of the underlying cause is of-
ten the first step in treatment of violence related to a neurological or
medical disorder.
Dementia
Dementia, a progressive decline in function across multiple cognitive
domains, is common, affecting 5%–8% of those over age 65 and nearly
50% of those over 85 (Small et al. 1997). Behavioral disturbances are
common in people with many types of dementia. Rabins et al.’s (1982)
survey of family and primary caregivers found that the most serious
problem reported was aggressive behavior. Families are able to tolerate
185
TABLE 10–1. Characteristics of aggression associated with

neurological or medical disease
Triggered by seemingly inconsequential or previously benign stimuli
Usually not planned
Does not help to fulfill long-term goals or gains
Sudden onset, abrupt offset
Brief outbursts contrasted by long periods of relative calm
After outbursts, patients are upset, concerned, and embarrassed, and may
express regret
their relative being forgetful; it is more difficult to manage sudden an-

ger. Aggressive behavior is one of the main factors leading to placement
of a demented person in a nursing home (O’Donnell et al. 1992). Hamel
(1990) studied predictors of aggression and the reaction of caregivers to
aggression in 213 demented outpatients. Aggression was reported in
57.2% of patients; verbal aggression was the most common form, occur-
ring in 51% of cases. Physical aggression, including threatening ges-
tures, was reported in 34% of patients, and sexual aggression, such as in-
appropriate hugging and kissing, in only 7%. Aggression most often
occurred in a situation in which a patient was instructed to do some-
thing. A history of aggressive personality traits predicted aggression in
patients with dementia. This finding is in agreement with earlier studies
by Ryden (1988) and by Morrant and Ablog (1983). A troubled premor-
bid relationship between patient and caregiver also predicted aggres-
sive behavior. Aggression was endorsed by caregivers as influencing
whether they would decide to institutionalize patients, a finding also re-
ported by other groups (Balestreri et al. 2000). Patients with dementia
who develop psychosis and subsequent agitation early in the course of
the illness should be evaluated for dementia with Lewy bodies, a more
rapidly progressive and less common condition than Alzheimer’s dis-
ease. This is especially true in cases in which parkinsonism or extreme
sensitivity to neuroleptics is present.
In a study of agitation and cognitive impairment in nursing home
residents, Cohen-Mansfield et al. (1990) studied 408 residents of a large
suburban nursing home. Level of impairment in performing activities
of daily living (ADLs) and in cognition were associated with increased
aggression, with high levels of impairment correlating with problem-
atic behavior. Patel and Hope (1992) found that in 90 inpatients on an
extended-stay psychogeriatric unit, levels of aggressive behavior were
quite high. Most of the aggressive behavior occurred consistently in re-
lation to morning assistance with ADLs.
Neurological and Medical Disorders ❘ 187
Several predictors of aggressive and violent behavior have been

found specifically in Alzheimer’s disease, the most prevalent form of
dementia in the elderly. Devanand et al. (1992) looked at specific behav-
ioral disturbances in 106 outpatients with probable Alzheimer’s dis-
ease. Aggression and agitation were associated with greater functional
impairment but did not correlate with severity of cognitive decline.
Deutsch et al. (1991) found that psychotic symptoms were a predictor of
aggressive behavior in Alzheimer’s patients. The prevalence of delu-
sions and misidentifications was significantly higher in physically ag-
gressive patients.
Imaging studies have revealed correlations among structural and
functional deficits and aggression. Burns et al. (1990) studied 178 pa-
tients with Alzheimer’s disease. Wandering and aggression together
were significantly associated with cognitive impairment. Computed to-
mography scans of subjects’ brains were also examined, and a positive
correlation was found between temporal lobe atrophy and aggression.
A positron emission tomography (PET) study by Sultzer et al. (1995) of
21 Alzheimer ’s patients found that aggression was correlated with
frontal and temporal hypometabolism. Agitation and disinhibition
were significantly correlated with global hypometabolism.
Traumatic Brain Injury

Traumatic brain injury (TBI) accounts for substantial morbidity and
mortality, especially among those in their mid-teens to mid-twenties.
Behavioral symptoms, including aggression, are common both acutely
following the injury and as a long-term consequence (Silver et al. 2005).
In the acute phase after brain injury, patients often experience a period
of agitation and confusion lasting days to months, which is probably
best considered a delirium (Sandel and Mysiw 1996). Bipolar disorder
secondary to TBI can lead to impulsive aggression. Agitation usually
appears within the first 2 weeks of hospitalization and resolves within
2 weeks. Restlessness may appear after 2 months and may persist for
4–6 weeks (Brooke et al. 1992). Subsequently, patients may develop low
frustration tolerance and explosive behavior that can be set off by min-
imal provocation or occur without warning. These episodes range in
severity from irritability to outbursts that result in damage to property
or assaults on others. In a study of 89 patients assessed during the first
6 months after TBI, Tateno et al. (2003) found aggressive behavior in
33.7% of TBI patients compared with 11.5% of patients with multiple
trauma but no TBI. Aggressive behavior was significantly associated with
the presence of major depression, frontal lobe lesions, poor premorbid
social functioning, and a history of alcohol and substance abuse. In

severe cases, affected individuals cannot remain in the community or
with their families and often are referred to long-term psychiatric
or neurobehavioral facilities.
In the early recovery period, 35%–96% of patients are reported to
have agitated behavior (Levin and Grossman 1978; Rao et al. 1985). Af-
ter the acute recovery phase, irritability or bad temper is common. In a
survey of individuals with TBI who were in skilled nursing facilities,
Wolf et al. (1996) found that agitation was present in 45% of 140 pa-
tients. There has been only one prospective study of the occurrence of
agitation and restlessness that has been monitored by an objective rat-
ing instrument, the Overt Aggression Scale (Brooke et al. 1992). The au-
thors of this study (Brooke et al. 1992) found that out of 100 patients
with severe TBI (Glasgow Coma Scale score less than 8, more than
1 hour of coma, and greater than 1 week of hospitalization), only 11 pa-
tients exhibited agitated behavior. Only 3 patients manifested these be-
haviors for more than 1 week. However, 35 patients were observed to
be restless but not agitated.
Studies of mild TBI have evaluated patients for much briefer peri-
ods of time; 1-year estimates from these studies range from 5% to 70%.
Carlsson et al. (1987) examined the relationship between the number of
TBIs associated with loss of consciousness (LOC) and various symp-
toms and demonstrated that irritability increases with subsequent inju-
ries. Of those men who did not have head injuries with LOC, 21% re-
ported irritability; 31% of men with one injury with LOC and 33% of
men with two or more injuries with LOC admitted to this symptom.
An evidence-based review of the TBI literature suggested that TBI
patients with frontal lobe injury, a prior history of substance abuse, or
impulsive aggression may be at higher risk of developing post-TBI ag-
gression (Kim et al. 2007). These authors noted that, based on studies to
date, it is not possible to define whether there is a relationship between
cognitive function and development of aggression after TBI; this may
be due to lack of sensitivity to executive dysfunction in neuropsycho-
logical testing. They also concluded there was not enough evidence to
define the relationship between socioeconomic status and severity of
injury to development of post-TBI aggression.
Stroke
Cerebrovascular accidents are an extremely common condition. Despite
some decline in cases of stroke due to treatment of hypertension, stroke
continues to be a major source of morbidity and mortality, especially in
developed countries. Numerous neuropsychiatric symptoms have been

described after stroke, including violent behavior. Poststroke mania, a
rare behavioral change after stroke, may be associated with right hemi-
sphere lesions; left hemisphere damage is more likely to produce depres-
sion (Robinson et al. 1988). In particular, damage to right hemisphere re-
gions with limbic connections may make poststroke mania a more likely
phenomenon. Psychosis also occurs with a low prevalence and may be
associated with right-sided frontoparietal damage. Patients who have
seizures after a cerebrovascular accident may also be at higher risk of
developing psychosis. Subcortical atrophy may also play a role in devel-
opment of psychosis after stroke (Starkstein et al. 1992). Catastrophic re-
actions are another poststroke behavioral change associated with ag-
gression. The term was coined by Goldstein in 1939 to describe the
“inability of the organism to cope when faced with physical or cognitive
deficits.” Patients with this rare condition seem to lose control of their
behavior in a dramatic fashion after stroke. They often have a significant
history of prior psychiatric illness and are depressed at the time the be-
havior occurs. Contrary to earlier work, catastrophic reaction is not nec-
essarily more common in patients with aphasia (Starkstein et al. 1993).
Congenital Brain Disorders and

Developmental Disorders
As the life expectancy of persons with mental handicaps increases, man-
agement of behavioral problems in this group becomes a pressing issue.
Several studies have shown that people with intellectual deficiencies
who engage in aggression toward others or self-injurious behavior are
more likely to require intensive supervision and management (Hill and
Bruininks 1984). Aside from the severity of intellectual impairment, ag-
gression is the most important reason why patients are institutionalized.
Inpatients
Ghaziuddin and Ghaziuddin (1992) studied violent behavior in intel-
lectually impaired persons during 1 year at a 100-bed unit at a univer-
sity hospital. Of the 106 patients admitted to the unit during the study,
35% were involved in 145 violent incidents. Twelve patients who had
associated psychiatric diagnoses were responsible for 86% of the inci-
dents. Sigafoos et al. (1994) studied a population of 2,412 people with
intellectual disability in Queensland, Australia. Of the sample, 48%
were severely to profoundly mentally retarded, 24% were moderately
impaired, and 9.6% were mildly impaired. Most individuals (59%) were
male, and 16% of this population lived in institutions. Of the group liv-
ing in institutions, 35% exhibited aggressive behavior. For those living
in group homes, 17% were aggressive. Persons who were identified as
aggressive had more profound levels of retardation and lower verbal
abilities. Much of the aggressive behavior in this study was directed to-
ward other patients or at other patients and staff.
Outpatients
Variable rates of aggression are reported among mentally handicapped
people in outpatient settings. Bouras and Drummond (1992) conducted
a study in southeast London, England, of 318 people (190 men, 128
women) with intellectual deficiencies who live in a community setting
and were referred to the psychiatry department of a mental handicap
service. Most of the patients in the study (54.4%) had a mild mental
handicap, 28.6% had moderate mental handicap, and 17% had severe
mental handicap. Almost one-third of the patients demonstrated ag-
gression toward others, and 13% engaged in self-injurious behavior.
Those people with severe intellectual impairment were more severely
and frequently aggressive.
Davidson et al. (1994) studied 199 individuals who were referred to
an outpatient crisis intervention program during a 2.5-year period and
had an IQ below 70 and concomitant adaptive behavior deficits. All peo-
ple in the study had at least one severe behavior disorder, and all had be-
havioral problems that were severe enough to threaten their ability to
stay in an outpatient community setting. Half lived with family mem-
bers, 22% in community residences, and 9% in intermediate community-
based facilities for intellectually disabled persons. The remainder of
those studied were either in family care or living independently or semi-
independently. Intake evaluations, historical data from agency records,
and medical records were reviewed as a source of data on aggressive be-
havior. At the time of the study, 131 individuals were classified as ag-
gressive. This study found that aggressive and nonaggressive patients
had similar neurological histories and medical status. CNS disorders, in-
cluding seizures, were seen with a similar prevalence in both groups.
The study concluded that current aggressive behavior was best pre-
dicted by past aggressive behavior when found in males with lower cog-
nitive functioning who might have been previously institutionalized.
Epilepsy
Epilepsy has long been felt to be a cause of, or at least a contributor to,
acts of violence. However, the literature is far from clear in establishing
a strong link between epilepsy and aggression in the vast majority of

patients. Many cases of aggressive behavior during or after a seizure are
due to the patient’s confusion or transient psychosis. In analyzing the
occurrence of aggressive behavior in individuals with epilepsy, it is im-
portant to note when the behavior occurs. Aggressive acts can be ictal
(during a seizure), postictal (immediately after a seizure), or interictal (in
the period between seizures).
Ictal Aggression
Ictal aggression occurs most often when persons attempting to assist
the patient during a seizure restrain the patient and the patient resists.
Treiman (1986) gives several examples of “resistive violence” in re-
sponse to restraint at the end of a generalized tonic-clonic seizure. Re-
sistive violence has also been observed in animal seizure models. Ictal
aggression is rarely directed and does not show elements of planning or
premeditation.
Postictal Aggression
Postictal aggression involves violent acts that occur when a patient is
still confused following a seizure. Usually, postictal aggression is seen
after a general tonic-clonic seizure. It can occur less commonly after a
complex partial seizure. Attempts at restraint are the most common
cause of aggression during this time. Postictal psychosis is another
likely cause of much postictal aggression, especially if the patient expe-
riences frightening hallucinations or feels paranoid during that time
(Devinsky and Bear 1984).
Interictal Aggression
Aggressive behavior between seizures is more controversial than ictal
or postictal aggression because there is no direct relationship between
the aggression and the seizure event. Most epileptologists agree that the
majority of patients with epilepsy are psychologically normal between
seizures. It is still unclear whether a small subset of persons with epi-
lepsy behave differently between seizures as a result of brain alterations
caused by the ictal events. Devinsky et al. (1994) studied 61 adult pa-
tients with epilepsy (46 patients with temporal lobe epilepsy, 15 with
absence epilepsy) and compared this group with 17 neurologically
normal control subjects. This study found no pattern of aggressive or
hostile behavior among persons with epilepsy. However, the author
did find increased suspiciousness in patients with left temporal lobe
epilepsy and increased assaultive behavior in persons with bilateral

temporal lobe epilepsy. There was a lack of difference between absence
epilepsy and normal groups. These data are in agreement with earlier
studies that did not demonstrate an increase in aggressive behavior
among persons with epilepsy (Hermann et al. 1984). Mendez et al.
(1993) examined 44 patients with epilepsy who were referred for psy-
chiatric evaluation because of violence. They concluded that interictal
violence was associated more with underlying psychopathology, such
as schizophrenia, or with mental retardation, rather than seizure activ-
ity. Stevens and Hermann (1981) critically examined the scientific lit-
erature on the association between temporal lobe epilepsy and violent
behavior, concluding that damage or dysfunction in the limbic area of
the brain was the significant factor in predisposition toward violence.
Central Nervous System Infections

Encephalitis
The influence of viral and other forms of encephalitis on behavior first
came to notice during the pandemic of encephalitis lethargica during
World War I. Also known as von Economo’s encephalitis, the illness
was noted to produce a plethora of psychiatric symptoms, including be-
havioral changes such as aggression, in previously normal persons.
Other forms of encephalitis, most notably herpes encephalitis, are now
also known to result in aggressive behavior.
Encephalitis Lethargica
Encephalitis lethargica was first described in detail by Constantine von
Economo in 1917 (von Economo 1937). Besides producing physical
symptoms of an acute CNS infection, encephalitis lethargica could, at
times, progress to coma or death. Survivors were sometimes afflicted
with parkinsonism or bizarre behavioral disturbances. The agent that
causes encephalitis lethargica has not yet been isolated. Some sur-
vivors, mostly adolescents, experienced pseudopsychopathic states.
Sporadic cases of encephalitis lethargica are still seen infrequently
throughout the world.
Herpes Simplex Encephalitis

The herpes simplex virus (HSV-1) can produce a severe form of enceph-
alitis. It is probably the most common cause of nonepidemic encephal-
itis in temperate zones (Ho and Harter 1982). Mortality rates are as high
as 70%. However, with new antiviral treatments, many patients are

living longer. These survivors often have severe neurological sequelae,
including profound behavioral disturbances. For unclear reasons, the
herpes virus tends to produce focal temporal lobe destruction. This can
result in a Klüver-Bucy syndrome in which patients are hyperoral and
hypersexual and may have an abnormal desire to explore objects
(hypermetamorphosis) (Friedman and Allen 1969).
Klüver-Bucy syndrome generally produces passive behavior; how-
ever, some patients with temporal lobe damage from the encephalitis
are aggressive. The literature on aggressive behavior after herpes
encephalitis consists only of case reports. Greer et al. (1989) described a
14-year-old patient with bilateral damage to the temporal lobes (right
worse than left on computed tomography scan) due to herpes simplex
viral encephalitis. Along with severe intellectual deficits, the patient
also had severe, uncontrollable motor activity, including aggressive and
self-injurious behavior. He eventually required placement in a residen-
tial facility due to his violent behavior. Greenwood et al. (1983) de-
scribed four patients with herpes simplex encephalitis who exhibited
aggressive behavior. The patients all exhibited some bizarre eating and
chewing behaviors, mostly related to nonfood items such as bedding or
feces. None were hypersexual or sexually inappropriate. All patients
looked emotionless and had few facial expressions, but with question-
ing, three of the four patients flew into unpredictable rages. Greenwood
noted that the patient who was the most unpredictably and violently
aggressive was also the one with the least memory loss from the en-
cephalitis. When in control, he could sit and play simple board games
with staff. Yet when he became angered, he would throw food and fe-
ces, shout at staff, and swear. Greenwood suggested that because ag-
gression does not occur in monkeys with bilateral temporal lobectomy,
which produces classic Klüver-Bucy syndrome, the aggression in select
patients with herpes simplex encephalitis may be due to partial in-
volvement of limbic areas, whereas patients with a classical Klüver-
Bucy syndrome have complete destruction of both temporal lobes.
Other Forms of Encephalitis

Because herpes simplex encephalitis is the only encephalitis known to
localize to a particular brain area, it is the only encephalitis in which
neuropsychiatric symptoms are relatively predictable. Other forms of
encephalitis produce more diffuse CNS damage. Subacute encephalitis
of various other etiologies causes numerous behavioral syndromes in
patients, including marked aggression, sometimes due to psychosis.
These include autoimmune deficiency encephalitis (Beresford et al.

1986; Nurnberg et al. 1984; Snider et al. 1983) and limbic or “paraneo-
plastic” encephalitis due to remote effects of malignancy (Khan et al.
1994; Newman et al. 1990).
Central Nervous System Tumors

Prior to the widespread availability of neuroimaging, substantial be-
havioral changes, including aggression, were not infrequently seen
with CNS malignancy. With the advent of relatively affordable detailed
imaging studies in most developed countries, early diagnosis and treat-
ment are fortunately now the rule. Available data are thus from autopsy
studies done early in the last century. When behavioral changes occur
due to malignancies, several factors influence what type of behavioral
symptoms occur. These include interconnections of structures involved
in the pathology, the patient’s premorbid level of function, rapidity of
tumor growth (which may cause increased intracranial pressure due to
rapid expansion) and whether the malignancy produces a single lesion
or multiple sites of involvement. Rapidly growing malignancies and
those with multiple foci are most likely to cause acute behavioral
changes, including psychosis and concomitant aggression (Lishman
1987). Association between tumor histological type and behavioral
symptoms has not been shown (Frazier 1935; Keschner et al. 1936). Lo-
cation of the tumor was found in older, autopsy-based literature to have
little correlation with presence or type of psychiatric symptoms ob-
served in patients, due to the factors noted earlier (Keschner et al. 1936).
Frontal lobe tumors can produce irritable, labile behavior (McAllister
and Price 1987) and psychosis (Strauss and Keschner 1935). Temporal
lobe tumors may produce psychotic symptoms, although the literature
is conflicting as to whether these symptoms are particularly common in
patients with temporal lobe pathology (Davison and Bagley 1969; Mul-
der and Daly 1952). Parietal and occipital malignancies are relatively
less likely to cause psychosis and agitation. Malignancies affecting the
diencephalons (thalamus, hypothalamus, and structures surrounding
the third ventricle) typically affect the limbic system due to its close
proximity and may cause agitation.
Movement Disorders
Movement disorders may result in behavioral changes. Increased irrita-
bility and angry outbursts are reported in many movement disorders.
Psychotic symptoms, which exacerbate underlying aggression, are also
seen. Patients with essential tremor, dystonia, and hereditary ataxias

are generally not violent, although exceptions occur.
Parkinson’s Disease
Although aggression may not be a common manifestation of Parkin-
son’s disease, per se, it may develop as a result of treatment of motor
symptoms with dopaminergic medications, especially if psychosis is a
side effect. Patients who develop impulse control disorders—“hedon-
istic dopaminergic dysregulations” that are at times related to use of
dopaminergic medications to treat motor symptoms—may be agitated
or impulsive.
Wilson’s Disease
Wilson’s disease, or hepatolenticular degeneration, is an autosomal re-
cessive disorder involving dysregulation of copper metabolism by the
liver. Neurological, renal, and hepatic abnormalities are the usual find-
ings in the disease. In a study of 42 patients with Wilson’s disease, Akil
et al. (1991) noted that 24 of the patients had psychiatric symptoms as
the presenting complaint. Personality changes, including aggression
and irritability, were the most common presenting psychiatric com-
plaint. Dening and Berrios (1989) assessed multiple neuropsychiatric
symptoms in 195 patients with Wilson’s disease. Aggression was defi-
nitely present in 17 patients and assessed as possibly present in 11 indi-
viduals.
Huntington’s Disease
Huntington’s disease is an autosomal dominant movement disorder. It
typically features choreiform movements and/or psychiatric symp-
toms. Burns et al. (1990) assessed 26 patients with Huntington’s disease
and found that 59% of the patients scored significantly on an aggression
scale. Aggression and irritability were not correlated with apathy or
with each other. Marder et al. (2000) found in a large study of patients
at various stages of the disease that aggression was reported by more
than half of patients or caregivers.
Gilles de la Tourette Syndrome

Numerous authors have cited behavioral problems as part of the clini-
cal picture in some cases of Tourette’s syndrome. Obsessive-compulsive
disorder and attention-deficit/hyperactivity disorder are the most
commonly described psychiatric symptoms in Tourette’s syndrome,

but aggressive outbursts have also been described. Robertson et al.
(1988) studied the correlation between motor symptoms and behavioral
disorders in 90 patients with Tourette’s syndrome. They found that 28
patients had been physically aggressive toward people (most typically
family members), animals, or objects. Aggressive behavior was sig-
nificantly associated with symptoms of an urge to touch everything in
the immediate surroundings and with copropraxia. There was no sig-
nificant association between aggression and age at onset, personal or
family history of psychiatric illness, electroencephalographic or neuro-
logical abnormalities, medication, distribution of tics, hyperactivity, or
difficulty in concentration or attention as a child.
White Matter Disorders

White matter, which makes up slightly less than half the volume of the
adult human brain, is critical for normal communication between neu-
rons. It is therefore not surprising that pathology of the white matter
tracts can lead to aggression. Prediction of whether a particular individ-
ual with white matter disease will develop violent behavior is problem-
atic, because most disorders of white matter lead to wide-ranging ef-
fects on the brain. Multiple sclerosis, vascular disorders (Binswanger’s,
cerebrovascular accidents), metabolic conditions (cobalamin deficiency,
hypoxia), infections (AIDS),TBI, and neoplasms can all result in white
matter damage.
Multiple sclerosis, a demyelinating condition, is the most common
adult disorder of white matter. Patients with multiple sclerosis have in-
creased rates of bipolar disorder compared with the general population.
Mania is, of course, associated with increased risk of impulsive and ag-
gressive behavior. Temporal lobe demyelination may carry a particular
risk of mania in multiple sclerosis, consistent with data discussed ear-
lier regarding temporal lobe pathology and violence (Filley 1996).
Medical Disorders
Numerous medical conditions may result in diffuse brain dysfunction
and subsequent aggression. Several authors have discussed the fre-
quency of “minimal brain dysfunction” or poorly characterized “neu-
rological soft signs” in aggressive individuals (e.g., Elliot 1992; Monroe
1978). The most commonly seen disorders are described here; it is well
worth pursuing a complete medical workup in any patient who devel-
ops aggression suddenly or without a prior history of violent behavior.
Delirium
Delirium, or acute confusional state, is a transient global disorder of cog-
nition; it is a syndrome, not a disease, with multiple causes (see Table
10–2). The condition is a medical emergency associated with increased
morbidity and mortality. Decreased attention span and a waxing and
waning type of confusion are important features (American Psychiatric
Association 2000).
Agitated or violent behavior has been reported in delirium due to
many causes, including postoperative confusion (Lepouse et al. 2006).
Several conditions associated with delirium are discussed in the follow-
ing sections.
Toxins
Toxin exposure can produce various neurobehavioral changes, the most
common of which are sedation and memory deficits. Toxins that are as-
sociated with aggression include alkyltin, arsenic, lead (in adults), man-
ganese, and mercury. Solvents generally cause lethargy and confusion,
although toluene exposure may produce excitation and disinhibition.
Gas exposure, such as carbon monoxide, causes lethargy and impaired
cognition and may cause delirium. Nitrous oxide use is associated with
delusions and agitation (Bleecker 1994; Bolla and Roca 1994).
Medication Side Effects

Drug effects and side effects can cause disinhibition or irritability lead-
ing to aggression. By far the most common drug associated with aggres-
sion is alcohol, during both intoxication and withdrawal. Stimulating
TABLE 10–2. Common causes of delirium

Hypoxia
Metabolic disruption
Hypoglycemia
Hyperthermia
Alcohol or sedative withdrawal
Localized or systemic infections
Structural brain lesions
Postoperative confusion
Medications (may be at therapeutic doses in elderly patients), especially agents
with anticholinergic effects
drugs such as cocaine and amphetamines, as well as stimulating antide-

pressants, may produce agitation. Antipsychotic medications may in-
crease agitation through anticholinergic side effects. Agitation and irri-
tability usually accompany severe akathisia. Many other drugs may
produce confusional states, especially anticholinergic medications that
can cause delirium. Other drugs that may produce aggressive behavior
include steroids, which can cause psychosis (prednisone, cortisone, and
the anabolic steroids); quinolone antibiotics; analgesics (opiates and
other narcotics); and anxiolytics (barbiturates and benzodiazepines).
The latter two groups of agents may cause aggression due to disinhibi-
tion or medication withdrawal effects.
Rheumatic Diseases
Systemic lupus erythematosus is the autoimmune disorder most closely
associated with neuropsychiatric symptoms, including aggression. Be-
havioral effects of lupus can be due to either direct CNS involvement or
effects on other organ systems, such as uremia due to renal impairment,
leading to confusion and delirium. Other rheumatological disorders,
such as the vasculitides, Sjögren’s syndrome, and sarcoidosis, can all
produce dementia and psychosis, leading to aggressive behavior
(Ovsiew and Utset 1997). As noted earlier, steroids, which are com-
monly used to treat rheumatological diseases, may cause psychosis and
agitation.
Sleep Disorders
There have been a few reports of aggressive behavior during parasom-
nias, including violent attacks by patients with rapid eye movement be-
havior disorder (Mahowald et al. 2007). Recognition of this disorder is
particularly important due to its association with several other condi-
tions, including Parkinson’s disease, Lewy body dementia, and multi-
system atrophy. Sleep deprivation can, of course, worsen irritability in
many conditions, leading to exacerbation of underlying behavioral
problems.
Hypoglycemia
A series of studies conducted by Virkkunen et al. (2007) in Finland
has examined biological correlates of aggression in a group of violent
prisoners. One consistent finding has been that this group is prone to
hypoglycemia and that they have in increase in irritability during these
episodes.
Treatment
Assessment and Quantification of Aggressive Episodes
Before therapeutic intervention to treat violent behavior is initiated, cli-
nicians should document the baseline frequency and severity of the oc-
currences. It is essential to establish a treatment plan that uses objective
documentation of aggressive episodes to monitor the efficacy of inter-
ventions for both acute and chronic aggression. The Overt Aggression
Scale is an instrument of proven reliability and validity that can be used
easily and effectively to rate aggressive behavior in patients with a wide
range of medical or neurological disorders (Silver and Yudofsky 1991;
Yudofsky et al. 1986). The scale comprises items that assess verbal ag-
gression, physical aggression against objects, physical aggression
against self, and physical aggression against others. Behavior can be
monitored by staff or by family members utilizing this instrument.
Pharmacotherapy
Although no drug is approved by the U.S. Food and Drug Administra-
tion specifically for the management of acute or chronic aggression,
medications are widely used, and often misused, for this purpose. The
use of pharmacological interventions can be considered in two catego-
ries: 1) use of the sedating effects of medications, as required in acute
situations, so that the patient does not harm him- or herself or others,
and 2) use of nonsedating antiaggressive medications to treat for
chronic aggression when necessary. Some patients may not respond to
just one medication but may require combination treatment. There are
few double-blind, placebo-controlled trials conducted in this area to
guide the use of medication to treat aggressive behavior (Neurobehav-
ioral Guidelines Working Group et al. 2006). We suggest using the
guidelines published by the Expert Consensus Panel for Agitation in
Dementia (Alexopolous et al. 1998) as a framework for the assessment
and management of agitation and aggression in medical and neurolog-
ical illness.
Acute Aggression and Agitation

In the treatment of agitation and for treating acute episodes of aggres-
sive behavior, medications that are sedating, such as antipsychotic
drugs or benzodiazepines, may be indicated. However, because these
drugs are not specific in their ability to inhibit aggressive behaviors,
there may be detrimental effects on arousal and cognition. In addition,
due to the potential for interference with respiration and thermoregula-

tion, these drugs should be administered only under medical supervi-
sion. Therefore, the use of sedation-producing medications must be
time limited to avoid the emergence of seriously disabling side effects
ranging from oversedation to extrapyramidal side effects.
Chronic Aggression
If a patient continues to exhibit periods of agitation or aggression be-
yond several weeks, the use of specific antiaggressive medications
should be initiated to prevent future episodes. The choice of medication
may be guided by the underlying hypothesized mechanism of action
(i.e., effects on serotonin system, adrenergic system, kindling) or in con-
sideration of the predominant clinical features. Since no medication has
been approved for the treatment of aggression, the clinician must use
medications that have been approved for other uses (i.e., for seizure dis-
orders, depression, anxiety, mood stabilization, hypertension).
Table 10–3 summarizes our recommendations for the utilization of
various classes of drugs in the treatment of aggressive disorders. In
treating aggression, the clinician, when possible, should diagnose and
treat underlying disorders and use antiaggressive agents specific for
those disorders. When there is a partial response after a therapeutic trial
with a specific medication, adjunctive treatment with a medication with
a different mechanism of action should be instituted. For example, a pa-
tient with a partial response to β-blockers may show further improve-
ment with the addition of an anticonvulsant or a serotonergic antide-
pressant. Side effects may limit dosing, as in any patient population;
patients with disease affecting the brain are often more sensitive to
medication side effects. Among the more important side effects, akathi-
sia may occur, with concomitant restlessness and irritability, in patients
who are being treated with neuroleptics for suppression of chorea or for
psychiatric symptoms. This can potentially worsen aggression if not
recognized.
Clinicians should be aware of recent U.S. Food and Drug Adminis-
tration warnings that the use of atypical antipsychotics was associated
with an increased risk of death in a review of data from 5,106 elderly de-
mented patients in randomized, controlled clinical trials (see Kuehn
2005 for an excellent commentary on the “black box” warnings). A mor-
tality rate of 4.5% was seen in those elders receiving atypical agents
compared with a rate of 2.6% in those who were given placebo. Deaths
were predominantly due to cardiovascular and infectious illnesses. A
“black box” warning has been added to labeling of all atypical neuro-
leptics. As discussed earlier, many of the patients who are most in need
of treatment for aggression are elderly and may have memory loss; they
may thus be at higher risk of mortality associated with atypical neuro-
leptic use. Conversely, agitation and violence are associated with signif-
icant risk of increased morbidity and mortality. Clinicians should weigh
carefully the small possibility of increased mortality associated with
atypical antipsychotic use versus the many complications inherent in
leaving aggressive symptoms untreated in these patients.
Behavioral Treatment
It is clear that aggression can be caused and influenced by a combination
of environmental and biological factors. Because of the unpredictable
nature of aggression in neurological and medical disease, caregivers and
staff in institutional settings may overreact to aggression when it occurs.
Behavioral treatments have been shown to be highly effective in treating
patients with organic aggression and may be useful when combined
with pharmacotherapy. Behavioral strategies—including a token econ-
omy, aggression replacement strategies, and decelerative techniques—
may reduce aggression in the inpatient setting and can be combined ef-
fectively with pharmacological treatment.
Because irritability is often directed toward individuals known to the
patients, education of caregivers in how to identify and avoid situations
that trigger irritability and how to minimize its effects if it does occur is
crucial. Behavioral interventions may prove helpful in prevention of ag-
gression by removing precipitating factors. This includes adherence to a
schedule to avoid surprising the patient and provoking an outburst
(Moskowitz and Marder 2001). Caregivers should be advised to stop an
activity, such as assistance with ADLs, if aggressive behavior begins to
escalate. They should also be counseled not to argue with patients if the
behavior begins to escalate. If threats of physical aggression occur, they
should quickly remove themselves and other family members from the
area where the patient is and contact emergency medical services for as-
sistance. Any ammunition and weapons should be removed from the
home. The patient should be prevented from accessing alcohol and illicit
drugs because use of these substances, even in small quantities, may
contribute greatly to disinhibition in patients with neurological and
medical illness. If there is a history of severe aggression against persons
or property, law enforcement agencies may need to intervene to prevent
injury to the patient or caregivers. Evaluation by a medical professional
to rule out medical illness, delirium, medication toxicity, or physical dis-
comfort should be conducted, especially in patients who have not been
202
TABLE 10–3. Psychopharmacological treatment of chronic aggression
Examples of agents
❘
Class of agent used (dosing range) Potential side effects Overall comments
Selective serotonin Escitalopram (10–20 mg) Initially may cause Preferable to start with an SSRI, then add a neuroleptic if
reuptake inhibitors anxiety or agitation needed. However, in cases of acute or extreme
(SSRIs) aggression, a neuroleptic should be started first. SSRIs
Textbook of Violence Assessment and Management

may cause or exacerbate apathy in some patients.
Higher doses and longer treatment time are needed if
anxiety is part of the cause for aggression.
Sertraline (100–200 mg) May be more activating
than other SSRIs
Atypical neuroleptics Olanzapine (2.5–20 mg) Weight gain, sedation, Standard neuroleptics will provide the most rapid
possible EPS sedation but are more likely to cause EPS and sedation.
Recent FDA black box warnings were issued for use of
atypical neuroleptics in elderly demented patients.
Quetiapine Sedation, possible EPS
(100–400 mg, may go
higher in select cases)
Standard (typical) Haloperidol EPS, sedation Best for acute agitation or delirium; start with small
neuroleptics (0.5–10 mg) doses. Often used IM in acute situations in conjunction
with a benzodiazepine.
Neurological and Medical Disorders
TABLE 10–3. Psychopharmacological treatment of chronic aggression (continued)
Examples of agents
Class of agent used (dosing range) Potential side effects Overall comments
β-Blockers Propranolol May lower heart rate; Higher dosages may be needed in select cases. Latency
(200–600 mg) contraindicated in of 4–6 weeks for onset of treatment.
bronchospasm, diabetes,
thyroid disease, and
heart failure
Anticonvulsants/ Valproate May cause sedation; need May be particularly useful in traumatic brain injury and
Mood stabilizers (500–1,000 mg) to monitor liver function other conditions associated with increased rate of
tests. Blood levels can be seizures. Abrupt discontinuation of these agents may
monitored. precipitate seizures in individuals who do not have a
prior history of epilepsy.
Carbamazepine Possible bone marrow
(200–400 mg) suppression
Benzodiazepines Lorazepam (0.5–3 mg) Sedation, disinhibition Withdrawal from benzodiazepines may precipitate
further aggression. Clonazepam is longer acting,
causing fewer withdrawal symptoms.
Clonazepam (0.5–3 mg) Sedation, disinhibition
Note. EPS =extrapyramidal symptoms; FDA =U.S. Food and Drug Administration; IM=intramuscular.
❘
203
violent previously or those with impaired communication abilities. Un-

derlying psychiatric illness should also be considered as a mediating
factor, including depression, anxiety, or psychosis. A review of behav-
ioral interventions can be found in Moskowitz and Marder (2001) .
Discussion
Aggressive behavior in the presence of medical illness is common and
can be highly disabling. Neuroanatomical, neurochemical, and neuro-
physiological factors may have an etiological or mediating role in the
production of violence. The vignettes that follow illustrate some com-
mon clinical features of violence in the medical setting.
Case Example 1: Traumatic Brain Injury

Mr. T is a 25-year-old man who sustained a severe TBI in a motor vehicle
accident. Damage to his orbitofrontal cortex was visualized on magnetic
resonance imaging. When he returned home after a 1-month treatment
in a rehabilitation facility, his family noticed a significant change in his
temper. Whereas he previously was patient, he now had a “short fuse”
and would go into a verbal rage with minimal frustration. Thankfully,
he would only occasionally slam a door or throw an object down in an-
ger. The family had been walking around “on eggshells” because they
did not know what would provoke the next episode.
The family was educated on the common occurrence of aggression
after TBI and told that this is a sequelae of TBI, as is memory problems.
They were told to start keeping a diary of the episodes, so that we would
know how many occurred during the week. Mr. T was started on pro-
pranolol 60 mg/day, with monitoring of his pulse and blood pressure.
When the dosage was increased to 240 mg/day, he became much calmer
and had a longer “fuse.” He described his aggression as like being on a
beach and being surprised when a large wave hits you from behind;
with the medication, he could see the “wave” approaching and deal
with those feelings.
Case Example 2: Huntington’s Disease

Mr. H is a 45-year-old married man whose father died of Huntington’s
disease in his mid-60s after having symptoms of the disease for more
than 15 years. Mr. H, who works as a taxicab driver, has been suspended
from driving his cab due to several episodes of severe aggression. In one
episode he took a baseball bat, which he says he always keeps in his cab
for self-defense, and broke the windows on another car after he thought
the driver cut him off in traffic. In another episode, he threw a cup of hot
coffee at a customer who failed to give him a tip. He began to drink on
the job and made inappropriate sexual comments to female passengers.
On interview, he admitted he would previously never have reacted in
this way to fairly minor provocations and expressed regret that he had
behaved in such a manner. Mr. H has never been evaluated for Hunting-
ton’s symptoms clinically and has no wish to know if he has the gene.
This case illustrates the change in behavior of someone who may be

developing a neurological disorder. Behavioral changes can be a pre-
senting symptom of Huntington’s disease, before onset of motor symp-
toms. As is typical in aggression in the setting of neurological disease,
the patient responds out of proportion to the incident triggering the be-
havior and is puzzled as to why he reacted in such an extreme manner.
Key Points
■ Explosive and violent behavior has long been associated with both
focal brain lesions and diffuse damage to the CNS.
■ Irritability and/or aggression is a major source of morbidity for in-
dividuals with neurological or medical disease and a source of ad-
ditional stress to their caregivers and families.
■ Presence of aggression is often a primary factor when the decision
is made to place patients in an institution rather than provide care
in a home setting.
■ Low frustration tolerance and explosive behavior can be set off by
minimal provocation or occur without warning. It is essential that
all clinicians be aware of aggression and its assessment and treat-
ment in order to provide effective care to patients with this condi-
tion.
■ After appropriate evaluation and assessment of possible etiolo-
gies, treatment begins with the documentation of the aggressive
episodes. Psychopharmacological strategies may be divided into
those intended to treat acute aggression and those intended to
prevent episodes in the patient with chronic aggression.
■ The treatment of acute aggression involves the judicious use of se-
dation; the treatment of chronic aggression is guided by underly-
ing diagnoses and symptoms.
■ Behavioral strategies, including caregiver reassurance and educa-
tion, remain an important component in the comprehensive
treatment of aggression.
References
Akil M, Schwartz JA, Dutchak D, et al: The psychiatric presentations of Wilson's
disease. J Neuropsychiatry Clin Neurosci 3:377–382, 1991
Alexopoulos GS, Silver JM, Kahn DA, et al: Treatment of agitation in older per-
sons with dementia. The Expert Consensus Panel for Agitation in Demen-
tia. Postgrad Med (spec no):1–88, 1998
Balestreri L, Grossberg A, Grossberg GT: Behavioral and psychological symp-
toms of dementia as a risk factor for nursing home placement. Int Psycho-
geriatr 12(suppl):59–62, 2000
Beresford TP, Blow FC, Hall RC: AIDS encephalitis mimicking alcohol dementia
and depression. Biol Psychiatry 21:394–397, 1986
Bleecker M: Clinical presentation of selected neurotoxic compounds, in Occu-
pational Neurology and Clinical Neurotoxicology. Edited by Bleecker ML,
Hansen J. Philadelphia, PA, Williams & Wilkins, 1994, pp 207–233
Bolla KI, Roca R: Neuropsychiatric sequelae of occupational exposure to neuro-
toxins, in Occupational Neurology and Clinical Neurotoxicology. Edited
by Bleecker ML, Hansen JA. Philadelphia, PA, Williams & Wilkins, 1994, pp
133–159
Bouras N, Drummond C: Behaviour and psychiatric disorders of people with
mental handicaps living in the community. J Intellect Disabil Res 36 (pt
4):349–357, 1992
Brooke MM, Questad KA, Patterson DR, et al: Agitation and restlessness after
closed head injury: a prospective study of 100 consecutive admissions.
Arch Phys Med Rehabil 73:320–323, 1992
Burns A, Jacoby R, Levy R: Psychiatric phenomena in Alzheimer's disease, IV:
disorders of behaviour. Br J Psychiatry157:86–94, 1990
Carlsson GS, Svardsudd K, Welin L: Long-term effects of head injuries sus-
tained during life in three male populations. J Neurosurg 67:197–205, 1987
Cohen-Mansfield J, Marx MS, Rosenthal AS: Dementia and agitation in nursing
home residents: how are they related? Psychol Aging 5:3–8, 1990
Davidson PW, Cain NN, Sloane-Reeves JE, et al: Characteristics of community-
based individuals with mental retardation and aggressive behavioral dis-
orders. Am J Ment Retard 98:704–116, 1994
Davison K, Bagley CR: Schizophrenia-like psychoses associated with organic
disorders of the central nervous system: a review of the literature, in Cur-
rent Problems in Neuropsychiatry: Schizophrenia, Epilepsy and the Tem-
poral Lobe (Br J Psychiatry, special publication no 4). Edited by Harrington
RN. London, Headley Brothers, 1969, pp 126–130
Dening TR, Berrios GE: Wilson's disease: psychiatric symptoms in 195 cases.
Deutsch LH, Bylsma FW, Rovner BW, et al: Psychosis and physical aggression
in probable Alzheimer's disease. Am J Psychiatry 148:1159–1163, 1991
Devanand DP, Brockington CD, Moody BJ, et al: Behavioral syndromes in
Alzheimer's disease. Int Psychogeriatr 4 (suppl 2):161–184, 1992
Devinsky O, Bear D: Varieties of aggressive behavior in temporal lobe epilepsy.

Am J Psychiatry 141:651–656, 1984
Devinsky O, Ronsaville D, Cox C, et al: Interictal aggression in epilepsy: the
Buss-Durkee Hostility Inventory. Epilepsia 35:585–590, 1994
Elliot FA: Violence: the neurologic contribution. An overview. Arch Neurol
49:595–603, 1992
Filley CM: Neurobehavioral aspects of cerebral white matter disorders, in Neu-
ropsychiatry, Edited by Fogel BS, Schiffer RB, Rao SM. Baltimore, MD, Wil-
liams & Wilkins, 1996, pp 913–934
Frazier CH: Tumor involving the frontal lobe alone: a symptomatic survey of
105 verified cases. Arch Neurol Psychiatry 35:525–571, 1935
Friedman HM, Allen N: Chronic effects of complete limbic lobe destruction in
man. Neurology 19:679–690, 1969
Ghaziuddin M, Ghaziuddin N: Violence against staff by mentally retarded in-
patients. Hosp Community Psychiatry 43:503–504, 1992
Greenwood R, Bhalla A, Gordon A, et al: Behaviour disturbances during recov-
ery from herpes simplex encephalitis. J Neurol Neurosurg Psychiatry
46:809–817, 1983
Greer MK, Lyons-Crews M, Mauldin LB, et al: A case study of the cognitive and
behavioral deficits of temporal lobe damage in herpes simplex encephali-
tis. J Autism Dev Disord 19:317–326, 1989
Hamel M, Gold DP, Andres D, et al: Predictors and consequences of aggressive
behavior by community-based dementia patients. Gerontologist 30:206–
211, 1990
Hermann BP, Whitman S, Gordon AC: Psychopathology in epilepsy: how great
is the risk? Biol Psychiatry 19:213–236, 1984
Ho SU, Harter DH: Herpes simplex virus encephalitis. JAMA 247:337, 1982
Hill BK, Bruininks RH: Maladaptive behavior of mentally retarded individuals
in residential facilities. Am J Ment Defic 88:380–387, 1984
Keschner M, Bender MB, Strauss I: Mental symptoms in cases of tumor of the
temporal lobe. Arch Neurol Psychiatry 35:572–596, 1936
Khan N, Wieser HG: Limbic encephalitis: a case report. Epilepsy Res 17:175–
181, 1994
Kim E, Lauterbach E, Reeve A, et al: Neuropsychiatric complications of trau-
matic brain injury: a critical review of the literature (A report by the ANPA
Committee on Research). J Neuropsychiatry Clin Neurosci 19:106–127, 2007
Kuehn BM: FDA warns antipsychotic drugs may be risky for elderly. JAMA
293:2462, 2005
Lepouse C, Lautner CA, Liu L, et al: Emergence delirium in adults in the post-
anaesthesia care unit. Br J Anaesth 96:747–753, 2006
Levin HS, Grossman RG: Behavioral sequelae of closed head injury: a quantita-
tive study. Arch Neurol 35:720–727, 1978
Lishman WA: Organic Psychiatry: The Psychological Consequences of Cerebral
Disorder. New York, Oxford University Press, 1987
Mahowald MW, Schenck CH, Bornemann MA: Pathophysiologic mechanisms
in REM sleep behavior disorder. Curr Neurol Neurosci Rep 7:167–172, 2007
Marder K, Zhao H, Myers RH, et al: Rate of functional decline in Huntington's
disease. Huntington Study Group. Neurology 54:452–458, 2000 [Erratum
in: Neurology 54:1712, 2000]
McAllister TW, Price TR: Aspects of the behavior of psychiatric inpatients with
frontal lobe damage: some implications for diagnosis and treatment.
Compr Psychiatry 28:14–21, 1987
Mendez MF, Doss RC, Taylor JL: Interictal violence in epilepsy: relationship to
behavior and seizure variables. J Nerv Ment Dis 181:566–569, 1993
Monroe RR: Brain Dysfunction in Aggressive Criminals. Lexington, MA, Lex-
ington Books, 1978
Morrant JC, Ablog JR: The angry elderly patient: an approach to understanding.
Postgrad Med 74:93–95, 98–100, 102, 1983
Moskowitz CB, Marder K: Palliative care for people with late-stage Hunting-
ton’s disease. Neurol Clin 19:849–865, 2001
Mulder DW, Daly D: Psychiatric symptoms associated with lesions of the tem-
poral lobe. JAMA 150:173–176, 1952
Neurobehavioral Guidelines Working Group, Warden DL, Gordon B, et al:
Guidelines for the pharmacologic treatment of neurobehavioral sequelae of
traumatic brain injury. J Neurotrauma 23:1468–1501, 2006
Newman NJ, Bell IR, McKee AC: Paraneoplastic limbic encephalitis: neuropsy-
chiatric presentation. Biol Psychiatry 27:529–542, 1990
Nurnberg HG, Prudic J, Fiori M, et al: Psychopathology complicating acquired
immune deficiency syndrome (AIDS). Am J Psychiatry 141:95–96, 1984
O'Donnell BF, Drachman DA, Barnes HJ, et al: Incontinence and troublesome
behaviors predict institutionalization in dementia. J Geriatr Psychiatry
Neurol 5:45–52, 1992
Ovsiew F, Utset T: Neuropsychiatric aspects of rheumatic diseases, in The
American Psychiatric Press Textbook of Neuropsychiatry and Clinical
Neurosciences, 4th Edition. Edited by Yudofsky SC, Hales RE. Washington,
DC, American Psychiatric Press, 1997, pp 813–850
Patel V, Hope RA: Aggressive behaviour in elderly psychiatric inpatients. Acta
Psychiatr Scand 85:131–135, 1992
Rabins PV, Mace NL, Lucas MJ: The impact of dementia on the family. JAMA
248:333–335, 1982
Rao N, Jellinek HM, Woolston DC: Agitation in closed head injury: haloperidol
effects on rehabilitation outcome. Arch Phys Med Rehabil 66:30–34, 1985
Robertson MM, Trimble MR, Lees AJ: The psychopathology of the Gilles de la
Tourette syndrome: a phenomenological analysis. Br J Psychiatry 152:383–
390, 1988
Robinson RG, Boston JD, Starkstein SE, et al: Comparison of mania with depres-
sion following traumatic brain injury: causal factors. Am J Psychiatry
145:172–178, 1988
Ryden MB: Aggressive behavior in persons with dementia who live in the com-
munity. Alzheimer Dis Assoc Disord 2:342–355, 1988
Sandel ME, Mysiw WJ: The agitated brain injured patient, part 1: definitions,
differential diagnosis, and assessment. Arch Phys Med Rehabil 77:617–623,
1996
Sigafoos J, Elkins J, Kerr M, et al: A survey of aggressive behaviour among a
population of persons with intellectual disability in Queensland. J Intellect
Disabil Res 38 (pt 4):369–381, 1994
Silver JM, Yudofsky SC: The Overt Aggression Scale: overview and guiding
principles. J Neuropsychiatry Clin Neurosci 3:22–29, 1991
Silver JM, Yudofsky SC, Anderson KA: Aggressive disorders, in The American
Psychiatric Publishing Textbook of Traumatic Brain Injury. Edited by Silver
JM, McAllister TW, Yudofsky SC. Washington, DC, American Psychiatric
Publishing, 2005, pp 259–278
Small GW, Rabins PV, Barry PP, et al: Diagnosis and treatment of Alzheimer’s
disease and related disorders: consensus statement of the American Asso-
ciation of Geriatric Psychiatry, the Alzheimer’s Association and the Amer-
ican Geriatrics Society. JAMA 278:1363–1371, 1997
Snider WD, Simpson DM, Nielsen S, et al: Neurological complications of ac-
quired immune deficiency syndrome: analysis of 50 patients. Ann Neurol
14:403–418, 1983
Starkstein SE, Fedoroff JP, Price TR, et al: Anosognosia in patients with cere-
brovascular lesions: a study of causative factors. Stroke 23:1446–1453, 1992
Starkstein SE, Fedoroff JP, Price TR, et al: Catastrophic reaction after cerebrovas-
cular lesions: frequency, correlates, and validation of a scale. J Neuropsy-
chiatry Clin Neurosci 5:189–194, 1993
Stevens JR, Hermann BP: Temporal lobe epilepsy, psychopathology, and vio-
lence: the state of the evidence. Neurology 31:1127–1132, 1981
Strauss I, Keschner M: Mental symptoms in cases of tumor of the frontal lobe.
Arch Neurol Psychiatry 33:986–1005, 1935
Sultzer DL, Mahler ME, Mandelkern MA, et al: The relationship between psy-
chiatric symptoms and regional cortical metabolism in Alzheimer's dis-
ease. J Neuropsychiatry Clin Neurosci 7:476–484, 1995
Tateno A, Jorge RE, Robinson RG: Clinical correlates of aggressive behavior af-
ter traumatic brain injury. J Neuropsychiatry Clin Neurosci 15:155–160,
2003
Treiman DM: Epilepsy and violence: medical and legal issues. Epilepsia 27
(suppl 2):S77–S104, 1986
Virkkunen M, Rissanen A, Naukkarinen H, et al: Energy substrate metabolism
among habitually violent alcoholic offenders having antisocial personality
disorder. Psychiatry Res 150:287–295, 2007
von Economo C: Encephalitis Lethargica. New York, Oxford University Press,
1931
Wolf AP, Gleckman AD, Cifu DX, et al: The prevalence of agitation and brain in-
jury in skilled nursing facilities: a survey. Brain Inj 10:241–245, 1996
Yudofsky S, Silver JM, Jackson W, et al: The Overt Aggression Scale for the ob-
jective rating of verbal and physical aggression. Am J Psychiatry 143:35–39,
1986
C H A P T E R 1 1
Impulsivity and Aggression

Aggression is a perplexing phenomenon. It is influenced by many fac-

tors (e.g., culture, environment, biology, psychology, neurochemistry)
that shape the manner in which it is expressed and perceived. Aggres-
sion can have adaptive properties such as self-preservation, protection
of one’s young, or defense of territory. It is thought to help adolescents
develop autonomy, an independent identity, and mastery over their en-
vironment (Rome and Itskowitz 1990). In animal models, aggressive be-
havior has even been demonstrated to serve as a stress management
strategy (Williams and Eichelman 1971). Alternatively, aggression can
produce destructive behaviors that are directed against others through
physical violence or verbal attacks or toward oneself, leading to self-
injurious behaviors (SIBs) or suicidality.
Almost every one of us has engaged in an impulsive or aggressive
act at some point in our lives, whether it was saying an unkind word we
later regretted or making a rash purchase on an extravagant item. Felt-
hous and Barratt (2003) rhetorically posed the question, “Are we not all,
mentally disordered or not, capable of acts that are both impulsive and
aggressive?” followed by the decisive response “Of course!” (p. 133).
They went on to note that the distinguishing feature between patholog-
ical and non-pathological behavior is one of severity. Stone (1995) de-
veloped a four-zone continuum of aggressive behavior that illustrates
this very concept. In Stone’s model, severity ranges from culturally
211
sanctioned outbursts of anger and mild aggressive behaviors such as

those exhibited on the playing field or at a political rally to severe im-
pulsivity and aggression, which may lead to the commission of felonies.
Theoretical Models of Aggression

Nearly 40 years ago, Moyer (1968) proposed an early and influential
classification of seven categories of aggression: 1) fear-induced (aggres-
sion associated with fleeing or attacking a perceived threat); 2) maternal
(an attack as a means to protect one’s young); 3) inter-male (an attack by
a male toward another male in the immediate environment as a way to
establish dominance or status); 4) irritable (an attack directed toward
some source of frustration such as a threat, intimidation, or environ-
mental condition); 5) sex-related (sexual arousal is frequently associ-
ated with increased levels of hostility); 6) predatory (an aggressive act
aimed at taking down a prey); and 7) territorial (an attack on an in-
truder who enters into an area claimed by the attacker).
Contemporary definitions refer to aggression as “behavior directed
by one individual against another individual (or object or self) with the
aim of causing harm” (Bond 1992, p. 1). Other researchers have in-
cluded the notion of intent. For example, Anderson and Bushman
(2002) described aggression as behavior that is intended to cause imme-
diate harm to another individual when the intended victim desires to
avoid harm. Bjorkvist and Niemela’s (1992) definition of aggression not
only includes intent to hurt someone or damage something but also
adds affective arousal that has the potential to lead to an overtly aggres-
sive act and displays of intimidation.
In the process of developing the Aggression Questionnaire, Buss
and Perry (1992) conducted a series of factor analyses resulting in a
model with four separate but related categories of aggression. The first
category, Overt Physical Aggression, involves a physical attack on an-
other person that results in harm. In Overt Verbal Aggression, an indi-
vidual uses words to harm another. The third category, Anger, is the
emotional element of aggression, described as the “physiological
arousal and preparation for aggression” (p. 457). The fourth category,
Hostility, is “feelings of ill will and injustice” (p. 457).
The current literature consistently views aggressive behavior as a
dichotomous construct. Although the terminology varies greatly, con-
ceptually the two broad classifications result in premeditated aggression
(also referred to as predatory, instrumental, callous-unemotional, or
proactive) and impulsive aggression (often called affective, reactive, emo-
tional, hostile, or expressive) (Cornell et al. 1996; Stanford et al. 2003a).
Impulsivity and Aggression ❘ 213
Premeditated aggression is typically purposeful and aimed at obtain-

ing an object such as a reward or advantage for the aggressor (Hartup
1974). These types of behaviors tend to be carried out in a methodical
and deliberate fashion, with the aggressor demonstrating limited phys-
iological arousal (Stanford et al. 2003b). Meloy (2000) found that
perpetrators often possess a heightened sense of awareness that permits
them to effectively stalk their victim, gathering necessary information
in preparation for the aggressive act. Studies of incarcerated popula-
tions have also demonstrated that premeditated-predatory aggressors
are more psychopathic, as measured by the Psychopathy Checklist–
Revised (PCL-R), than those classified as impulsive-affective aggressors
(Cornell et al. 1996; Woodworth and Porter 2002).
The second prong of the aggression dichotomy is impulsivity, which
typically is a response to a perceived provocation with immediate and
destructive violence. Individuals who display impulsive-affective ag-
gressive behaviors are commonly labeled “unpredictable” and “short
fused.” Impulsive aggressive behaviors can be carried out involuntarily,
in a burst of rage, with no weighing of potential outcomes. Research has
found that individuals with impulsivity disorders tend to have lower
verbal scores and executive cognitive functioning impairments based
on neuropsychological testing (Villemarette-Pittman et al. 2003). It is
hypothesized that the limited cognitive resources of impulsive aggres-
sors allow them to become easily overwhelmed by competing stimuli,
which leads to feelings of frustration and helplessness. With limited per-
ceived options, the impulsive aggressor frequently acts before thinking
about the impact and likely consequences of the behavior (Meloy 2000).
The Etiology of Aggression and Impulsivity

To better understand aggressive and impulsive behaviors, it is helpful
to examine the various potential sources and pathways that can contrib-
ute to them. These behaviors have multiple causes (e.g., genetics, social
learning, environment, mental illness, substance abuse), each with a
substantial and valuable body of research. This chapter focuses on pre-
natal development, early childhood trauma, traumatic brain injury, and
neurochemistry as contributors to aggression and impulsivity.
Prenatal Development
Prenatal risk factors such as maternal use of alcohol, tobacco, or cocaine
and pregnancy/birth complications have been linked to developmental
delays and behavioral problems in children and to antisocial behavior
and violent offending in adults (Raine 2002). Exposure to these risk fac-
tors may directly or indirectly affect the structure and function of the
developing fetal brain, leading to long-term damage to central nervous
system neurotransmitter pathways (Ernst et al. 2001). For example, chil-
dren of mothers who smoked were found to be twice as likely to have a
criminal record by age 22 (Räsänen et al. 1999). In addition, 6- or 7-year-
old girls who had been exposed to cocaine in utero were significantly
more likely to score in the abnormal range on the Aggression subscale
of the Child Behavioral Checklist than control subjects (Sood et al.
2005). However, these risk factors are rarely found independent of other
psychosocial complications such as poverty, poor parenting skills, and
limited access to medical care and educational opportunities.
Childhood Trauma
Childhood trauma has been associated with impulsive and aggressive
behaviors, including self-destruction and suicidal behavior in later
years (Briere and Runtz 1990). Brodsky et al. (2001) examined 136 adults
with major depression and found that participants with a history of
physical or sexual abuse were more likely to have made a suicide at-
tempt and had higher levels of impulsivity, aggression, and comorbid
borderline personality disorder than participants with no abuse history.
The researchers assert that childhood trauma may constitute an envi-
ronmental risk factor that leads to the development of both suicidality
and impulsivity. Similarly, Roy (2005) studied 268 abstinent drug-
dependent patients and found a significant positive correlation between
impulsivity and risk-taking scores on the Barratt Impulsivity Scale and
scores of abuse and neglect on the Childhood Trauma Questionnaire.
Traumatic Brain Injury and Brain Dysfunction

Aggression and impulsivity have been associated with dysfunction in
various regions of the brain, most notably the temporal and frontal
lobes (Liu and Wuerker 2004). These regions of the brain regulate exec-
utive functions, and damage to these regions has been shown to lead to
intermittent emotional dyscontrol, an increase in impulsivity, a reduc-
tion in self-regulation, and the diminished capacity to consider the out-
comes of behaviors (Golden et al. 1996). Violent offenders have been
found to have poor functioning in the frontal and temporal regions of
the brain as evidenced by neuropsychological tests (Raine 2002), exces-
sive slow-wave electroencephalographic activity (Stoff et al. 1997), and
a reduced glucose metabolism in the prefrontal brain region as shown
in brain imaging studies (Raine et al. 1997). Grafman et al. (1996) stud-
ied Vietnam veterans who had penetrating head injuries and found
they had higher verbal aggression scores than control subjects and
patients with lesions in other areas of the brain. A study evaluating
89 patients with traumatic brain injury found that those who scored
high on the Overt Aggression Scale had a preinjury history of mood dis-
order, alcohol and drug abuse, and aggressive behaviors (Tateno et al.
2003). These findings indicate that postinjury behavior may be reflec-
tive of preexisting impulsive and aggressive tendencies.
Neurochemistry
Neurotransmitters are chemicals that send information between neurons
in the brain and help to regulate mood, thinking, and behaviors (Ber-
man and Coccaro 1998). Among the many known neurotransmitters,
the most studied in relation to aggression and impulsivity are seroto-
nin, norepinephrine, dopamine, and γ-aminobutyric acid (GABA). The
majority of the studies suggest that GABAergic and serotonergic sys-
tems inhibit predatory aggression, and the noradrenergic and dopa-
minergic systems stimulate affective aggression (Eichelman 1988). Low
levels of serotonin have been associated with increased rates of aggres-
sion, impulsivity, depression, and suicidality. Coccaro (1996) found
lower serotonin levels in suicide victims, particularly those who used a
violent suicide method, when compared with accident victims. In addi-
tion, Coccaro and Kavoussi (1997) examined 40 patients with person-
ality disorders and a history of impulsive aggression and found that
fluoxetine, a selective serotonin reuptake inhibitor, reduced scores of
aggression and irritability on the Overt Aggression Scale.
The noradrenergic system affects attention to stimuli, arousal levels,
and responses to stressors (Berridge and Waterhouse 2003) and is one
of the fastest-responding neurochemical systems (Haller et al. 1998).
Norepinephrine is involved in the fight-or-flight response and has been
linked to aggressive behavior (Haden and Scarpa 2007). In a study mea-
suring aggressive behavior and norepinephrine levels, Gerra et al.
(1997) experimentally induced aggression using a free-operant proce-
dure in 15 males with “low normal” and 15 males with “high normal”
basal aggressivity (based on scores on the Buss-Durkee Hostility Inven-
tory and other measures). They found no differences between the
groups in base rate plasma norepinephrine levels. However, during the
task, norepinephrine levels were significantly higher in the high group
than the low group. These findings suggest that high-aggressive indi-
viduals respond more intensely, and their norepinephrine increases
to higher levels when they are presented with frustrating situations,

than is the case in low-aggressive individuals.
The remainder of this chapter focuses on formal disorders (i.e., in-
termittent explosive disorder [IED], pyromania, intellectual disabilities,
and autism) and distinct symptoms (i.e., impulsive suicide and SIB)
within which aggression or impulsivity are core determinants.
Discussion
Intermittent Explosive Disorder
Case Example 1
J.W. is a 30-year-old man with a long history of fights and assaultive be-
havior. He was referred by a judge for anger management classes several
years ago. J.W. was walking down the street when he inadvertently
bumped into a stranger. The stranger said, “Why don’t you watch where
you’re going!” Infuriated, J.W. turned on the stranger and started pum-
meling him into unconsciousness. Minutes later, when onlookers pulled
him off of his victim, J.W. was upset and remorseful over his behavior.
IED is categorized in DSM-IV-TR (American Psychiatric Association

2000) as an impulse-control disorder not elsewhere classified and is the
only diagnosis with recurring acts of aggression as the primary symp-
tom. The inclusion criteria for IED consist of 1) distinct episodes of seri-
ous assault against others or destruction of property, 2) behavior that is
grossly out of proportion to any precipitating provocation or psychoso-
cial stressor, and 3) explosive episodes that are not better accounted for
by another mental disorder, substance use, or a medical condition. As
in most of the impulse-control disorders, the individual feels a sense of
tension or affective arousal before committing the explosive behavior,
may experience pleasure or gratification during the act, and may feel re-
lief or regret after the act.
IED is rare in terms of prevalence, and research has revealed similar
rates of lifetime incidence. For example, Coccaro et al. (2004) evaluated
253 participants for the Baltimore Epidemiologic Catchment Area Fol-
low-Up study and found lifetime rates of 4.0%. In a survey of 9,282 U.S.
adults, Kessler et al. (2006) found slightly higher rates ranging from
5.4% to 7.3%. IED behaviors typically become apparent in childhood,
often in the form of temper tantrums. Explosive outbursts tend to peak
during the teen years and to decline after age 30, with only about 7% of
new cases occurring after this age (McElroy et al. 1998). Research has
found incidents of IED to occur earlier for men (13 years of age) than for
women (19 years) (Coccaro et al. 2005).
In a study of 27 individuals who met the criteria for IED, partici-

pants described their aggressive impulses as “a need to attack,” “an
adrenaline rush,” “a need to defend oneself,” and “an urge to kill”
(McElroy et al. 1998). The aggressive episodes were associated with
physical or autonomic symptoms such as heart palpitations, chest tight-
ness, head pressure, a loss of awareness, and affective symptoms such
as irritability, euphoria, and racing thoughts. The outbursts were often
in response to an external stressor (typically a minor disagreement with
someone), but many reported that the aggressive episodes were spon-
taneous. The aggressive episodes occurred approximately nine times
per month, and although the duration of a specific episode was rela-
tively brief (22 minutes), the outcomes had devastating repercussions
resulting in destruction of property, serious assault on another person,
assault with a weapon, attempted homicide, and homicide. Not sur-
prisingly, individuals with IED have difficulties in maintaining employ-
ment, financial stability, and meaningful relationships.
The usefulness of classifying IED as a separate diagnosis has come
under criticism because aggressive impulses occur in a wide range of
psychiatric and medical disorders (Coccaro 2003). McElroy et al. (1998)
found a high comorbidity rate among IED patients and mood disorder,
anxiety disorders, and other impulse-control disorders. Coccaro et al.
(2005) found a substantial amount of lifetime comorbidity among IED
patients with mood disorders, anxiety disorders, and alcohol/drug dis-
orders. Nearly a quarter of the patients in a study conducted by Lejoy-
eux et al. (1999) who met the criteria for alcohol dependence also met
the criteria for IED.
Some researchers have raised concerns about the value of the criteria
for IED, noting several limitations and ambiguities. For example, Coc-
caro (2003) pointed out that DSM-IV-TR does not set parameters for the
frequency of the aggressive acts, the time span between episodes, or the
severity of the outbursts. In addition, it is difficult to determine whether
an aggressive outburst is more likely to be caused by another personal-
ity disorder, such as antisocial personality disorder. The current defini-
tion may be underestimating the number of individuals with IED by ex-
cluding individuals with frequent but less severe aggressive actions.
Pyromania
Case Example 2
R.J. is 23 years old and has had a fascination with fires since early child-
hood. In the past he has set many small, contained fires and enjoyed
watching the resultant blazes. Tonight, he is sitting in his room fondling
a book of matches. At his ease, he happily remembers the tension and

excitement he felt when he set fire to an abandoned garage the previous
night.
Pyromania is designated as an impulse-control disorder not else-

where classified in DSM-IV-TR, along with IED, kleptomania, patholog-
ical gambling, trichotillomania, and impulse-control disorder not other-
wise specified. The diagnosis itself has an unstable history in DSM: it
was included in DSM-I (American Psychiatric Association 1952) as an
obsessive-compulsive reaction; omitted from DSM-II (American Psy-
chiatric Association 1968); and reinstated in DSM-III (American Psychi-
atric Association 1980) as a distinct disorder of impulse control. In
DSM-IV-TR it is defined as repeated, deliberate, and purposeful fire set-
ting and is associated with tension before the act; fascination with fire;
and gratification when setting, witnessing, or putting out fires. The fire
setting is not committed for monetary gain, revenge, as an expression
of sociopolitical ideology, to conceal a criminal act, to express anger or
vengeance, to improve one’s living circumstances, in response to a de-
lusion or hallucination, or as a result of impaired judgment. Finally, the
impulse to set fires cannot be better accounted for by another diagnosis.
True pyromania is rare. Rasanen et al. (1995) studied arson defen-
dants in Finland from 1975 to 1993 and found only 4% of their sample
to have pyromania. Ritchie and Huff (1999) examined the mental health
records and/or prison files of 283 arsonists, and pyromania was diag-
nosed in only three cases (1.3%). A slightly higher rate was identified in
a study conducted by Repo et al. (1997) between 1978 and 1991 in which
14.2% of 304 male Finnish arsonists were diagnosed with pyromania.
The low rates of pyromania found in these fairly recent studies calls into
question the substantial number of arsonists who were diagnosed with
pyromania (39%) in Lewis and Yarnell’s (1951) classic work of nearly
1,500 pathological fire setters. The later results almost certainly reflect
the changes in diagnostic criteria, which have become more structured
and narrowly defined in the intervening decades.
Several risk factors associated with pyromania are consistently
found in the literature. Barker (1994) found men to be much more likely
than women to have a fascination with fire, and Kafry (1980) found
boys to be more interested than girls in fire setting. Large percentages
of pathological fire setters are unemployed and live alone (Ritchie and
Huff 1999). Lejoyeux et al. (2006) described people with pyromania as
individuals with a keen interest in fires who like watching fires and set-
ting off false fire alarms. Their fascination with fires often leads them to
seek employment as firefighters. In a study by Lindberg et al. (2005) of
90 arson recidivists, three were diagnosed with pyromania. All three of

these arsonists worked as volunteer firefighters.
There is a consistent link reported between fire setting and mental
illness. Ritchie and Huff (1999) found that nearly all (90%) of the sub-
jects in their study had a history of mental health issues; 36% of these
individuals also had a diagnosis of either schizophrenia or bipolar dis-
order. Two-thirds of the sample (64%) were abusing alcohol or drugs at
the time of the fire setting, and the fire-setting act of half of the sample
was judged “very impulsive” by the researchers. In a 5-year study con-
ducted by Leong and Silva (1999) of court-ordered outpatient forensic
psychiatric evaluation of individuals charged with arson, nearly half
(43.8%) were diagnosed as psychotic, 15.6% as mentally retarded, and
15.6% with alcohol abuse.
Aggressive Behavior in Individuals With

Intellectual Disabilities
Case Example 3
W.P. is a 35-year-old woman with moderate intellectual disability deriv-
ing from fetal alcohol syndrome. She was recently placed in a group
home after the death of her parents several months ago. She has never
needed psychiatric care in the past. She does well at her job placement
until it is time to leave. At that point, on a fairly consistent basis over the
past few weeks, this normally pleasant woman hits anyone who at-
tempts to persuade her to board her minivan for the ride back to her
group home.
In contrast to the aforementioned disorders, the following discus-

sion of impulsivity and aggression occurs in the context of other disor-
ders where sudden, unpredictable, and violent behavior may occur.
Intellectual disability has been defined as “significantly subaverage in-
tellectual functioning resulting in or associated with a concurrent im-
pairment in adaptive behaviour” (Strongman 1985, p. 202). Holland et
al. (2002) defined it as significantly impaired intellectual ability and
significantly impaired social functioning, with these conditions
present from childhood. DSM-IV-TR uses the term mental retardation
and has three inclusion criteria: a score of 70 or below on an individu-
ally administered IQ test, deficits in two adaptive functioning areas
(i.e., communication, self-care, home living, social/interpersonal
skills, use of community resources, self-direction, functional academic
skills, work, leisure, health, and safety), and the onset of the impair-
ment occurring before the age of 18 years.
Aggressive behavior can have a severe negative impact on individ-

uals with intellectual disabilities and their caregivers. Aggression often
becomes a barrier to less-restrictive residential options, educational op-
portunities, competitive employment, and general social acceptability
(Bruininks et al. 1994). In addition, aggressive behavior in individuals
with intellectual disability is also associated with greater service costs,
higher staff turnover rates (Sigafoos et al. 1994), more frequent referrals
to mental health professionals (Maguire and Piersel 1992), increased
risk for victimization (Rusch et al. 1986), and criminal activity (Crocker
and Hodgins 1997). Aggressive behavior may contribute to these indi-
viduals being admitted to institutions and being prescribed antipsy-
chotic and behavior control medication (Aman et al. 1987).
The concept of aggressive behavior in individuals with intellectual
disability is very broad, and a consistent definition is lacking. McClin-
tock et al. (2003) conducted a meta-analysis of research on aggressive
behavior in these individuals, reviewing 86 articles from 1968 to 1997.
The researchers found a wide range of terms used to describe aggres-
sive behavior, including physical aggression, threatening others, SIB,
destruction of property, and hitting. Deb et al. (2001) considered aggres-
sive behavior to encompass “aggression, destructiveness, self-injurious
behavior, temper tantrum, over-activity, screaming/shouting, scatter-
ing objects around, wandering, night-time disturbance, objectionable
personal habits, antisocial behavior, sexual delinquency, and attention-
seeking behaviors” (p. 507). A possible explanation for the variation in
describing aggressive behavior is that the descriptions are often based
on the perception of caregivers who must manage or endure the behav-
iors and may be influenced by their coping ability. In addition, the la-
beling of aggressive behavior may be dependent on the environment. In
other words, some behavior may be tolerated in an institutional setting
but not in a family setting.
Prevalence rates of aggressive behavior for individuals with intel-
lectual disabilities vary from 2% to 60% based on a variety of factors,
such as level of behavioral severity, age, gender, and type of residential
environment (Davidson et al. 1994). Males with intellectual disabilities
tend to have higher rates of aggressive behavior than females (Harris
1993), and aggressive behavior tends to peak around adolescence
(Davidson et al. 1994). Acts of aggression tend to increase with the se-
verity of disability (Davidson et al. 1994); however, higher-functioning
individuals tend to act aggressively toward others and lower-function-
ing individuals tend to engage in SIB (Emerson et al. 1997). Acts of ag-
gression were higher in institutional settings (38%) than in community
settings (11%) according to a survey of service providers (Harris 1993).
Notably, deinstitutionalization has not been found to reduce an individ-

ual’s aggressive behavior (Larson and Lakin 1989).
Individuals with an intellectual disability often have skill deficits in
a variety of areas (e.g., attention span, impulse control, memory, neuro-
logical functioning, communication skills, and social skills), which may
increase the probability of aggression (Allen 2000). For example, in a
study measuring the ability of aggressive and nonaggressive individu-
als with intellectual disabilities to label facial expressions, aggressive in-
dividuals were more likely than their nonaggressive peers to mislabel
“angry” and “sad” facial expressions, and to label “anger” when they
were unsure (Walz and Benson 1996).
Aggressive Behavior in Individuals With Autism
Case Example 4
M.R. is 28 years old and has been in psychiatric care throughout his life.
He is currently on a trial of a new anticonvulsant. He sits quietly rocking
back and forth, chewing on his already bleeding left wrist. When Jim,
his caregiver, attempts to intervene, this startles him. M.R. then furi-
ously swings his arms, hitting Jim repeatedly.
The term autism, derived from the Greek word for “self,” was coined
by Leo Kanner (1943), a child psychiatrist, to describe the extreme
aloneness he viewed as the central trait of the disorder. Autism spec-
trum disorder (ASD), as it is now called, is currently recognized as a
neurodevelopmental disorder (Aicardi 1998). According to DSM-IV-
TR, to be diagnosed with autistic disorder an individual must exhibit 6
or more of the 12 identified behaviors, with at least two from the social
interaction domain and one each from the communication, repetitive,
and stereotyped patterns domains. In addition, delays in social interac-
tion, language, or symbolic or imaginative play must be evident before
the age of 3 years.
Prevalence rates of ASD range from 0.7 to 72.6 per 10,000, depend-
ing on the diagnostic criteria used in the studies (Williams et al. 2006).
Fombonne (1999) reviewed 23 studies on autism and found that preva-
lence rates significantly increased with publication year, indicating im-
proved diagnostic criteria and methods along with greater availability
of services. For example, Croen and Grether (2003) found that 75% of in-
dividuals with autism had some level of intellectual disability and pro-
posed that the increase in the prevalence of autism may be attributable
to the reclassification of some individuals’ diagnoses from intellectual
disability to autism. Fombonne (2003) also found a higher rate of boys
than girls being affected; intellectual disabilities in about two-thirds of

the sample; and a relatively high rate of epilepsy.
Aggression is a common behavioral characteristic of autism and
may include impulsivity, aggression toward others, SIB, destruction of
property, disruption to the environment, stereotypy, and other socially
unacceptable behaviors (McDougle et al. 2003). Although the inappro-
priate or aggressive behavior may be objectionable, the intention of the
behavior is not necessarily malicious (Dewey 1991). The impairments in
brain functioning and neurological activity commonly found in autism
may reduce one’s capacity for social interaction, verbal and nonverbal
communication, and the ability to alter behavior and emotional states
in response to another’s action or perceived feelings. For example, Wil-
liams et al. (2005) compared the memories of non–intellectually dis-
abled adults with autism and a control group. The participants with
autism did not demonstrate any deficits in word pairs, stores, or verbal
working memory. However, on tests measuring immediate and de-
layed recall of faces and family scenes, there was significant impair-
ment. The research suggests that a lack of social connectiveness and
empathy may predispose individuals with autism toward acts of ag-
gression (Rogers et al. 2006).
Impulsive Suicide
Case Example 5
After 65 years of life, K.T. has coped with several serious illnesses and
repeated bouts of depression. He has been treated by a psychiatrist for
the past few years, with only modest symptomatic improvement. The
past 3 months have been filled with unremitting depression. K.T. feels
hopeless and has intermittently considered ending his life. The three
drinks he just had seem to make the decision easier. He hits the car’s ac-
celerator and aims head on for the gap in the guard rail, with the river
100 feet below.
Suicide ranks among the top 10 causes of death for individuals in all
age groups in several Western countries (World Health Organization
2006). Suicide attempts have been defined as “potentially self-injurious
behavior with a nonfatal outcome, for which there is evidence (either
implicit or explicit) that the person intended at some (nonzero) level to
kill himself/herself” (O’Carroll et al. 1996, p. 247). Terms such as para-
suicide, deliberate self-harm, and suicidal gestures are considered SIBs that
may have the appearance of a suicide attempt but may not have the as-
sociated intention of ending one’s life. Using data from the National Co-
morbidity Survey of 5,877 respondents, Nock and Kessler (2006) found
4.6% of the sample had made a suicide attempt, 2.7% reported doing so
with the intent to die, and 1.9% committed the act as a way to commu-
nicate distress to others with no intent to die.
To date, a prior suicide attempt is among the best predictors of even-
tual death by suicide (Goldstein et al. 1991). In a 5-year follow-up of
1,573 suicide attempters, Nordstrom et al. (1995) found the risk of recur-
rent suicidal behavior to be 11% for attempted suicide and 6% for ulti-
mately completed suicide. Rates were highest among young men.
Johnsson-Fridell et al. (1996) reported a 13% suicide completion rate
among inpatients within 5 years of attempted suicide. In a study exam-
ining lifetime history of suicide attempts and methods of 1,397 suicides
in Finland, Isometsä and Lonnqvist (1998) found that 56% of fatal sui-
cides occurred on the first attempt (62% of males and 38% of females),
and the risk of suicide completion was highest during the first year after
a suicide attempt. With such high fatality figures, particularly for men,
using previous suicide attempt as a predictor of suicide completion has
limited preventive value.
Risk factors associated with suicidal behavior include male gender,
fewer years of education, being young, and residence in the southern or
western regions of the United States (Nock and Kessler 2006). Based on
psychological autopsies, a 6-month prevalence rate of an Axis I diagno-
sis has been found in 88% of suicide completers (Lesage et al. 1994). Spe-
cifically, depressive (major depressive episode and mania), impulsive
(drug abuse and dependence), and aggressive (conduct disorder and
antisocial personality disorder) behaviors and psychiatric comorbidity
increase the risk of suicide attempts (Nock and Kessler 2006), with ma-
jor depression being the most common psychiatric disorder associated
with suicide and attempted suicide (Henriksson et al. 1993). Childhood
trauma has been associated with self-destruction and suicidal behavior
in later years (Briere and Runtz 1990) and contributes to a younger age
of onset of suicidal behaviors, often beginning in childhood or adoles-
cence (Brodsky et al. 2001).
Impulsivity has been conceptualized as action without planning or
reflection; it differs from premeditated behavior by having a short re-
sponse time, lack of reflection, and a dissociation between action and
consequence (Barratt et al. 1999). Impulsivity, along with other disin-
hibiting moderators such as substance use or significant current dis-
tress, is strongly associated with self-destructive behaviors, including
suicidal behavior (Dumais et al. 2005). Impulsive suicidal behavior has
been defined as a suicide attempt with less than 5 minutes of premedi-
tation (Simon et al. 2001). Using this definition of impulsive suicidality,
prevalence rates range from 24% for nearly lethal suicide attempts by
individuals 13–34 years old (Simon et al. 2001) to 40% for hospital pa-
tients treated for self-injury (Williams et al. 1980).
Individuals who engage in impulsive suicidal behavior tend to use
more violent methods, such as firearms, hanging, cutting, and jumping
(Simon et al. 2001), than their nonimpulsive counterparts. At the same
time, impulsive suicide attempters have lower expectations of dying
from their actions (Swann et al. 2005). Despite the impulsive attempters’
lower expectations of dying than nonimpulsive attempters, the destruc-
tive outcomes are comparable in terms of severity of injuries, reversibil-
ity of condition, and admission into intensive care (Simon et al. 2001).
The incongruous thought process involved in these attempts is consis-
tent with a defining construct of impulsivity: the disconnect between
action and intention (Swann et al. 2005).
Impulsivity does not appear to increase the risk of suicide inde-
pendently. Simon et al. (2001) suggested that suicidal behavior may be
associated with the inability to control aggression-related impulsive
behavior rather than with impulsivity in general. For example, the re-
searchers examined indicators of impulsive behavior such as prior ar-
rests, quitting a job without a source of income, having multiple sex
partners, and alcohol use within 3 hours of the suicide attempt and
found no relationship to impulsive suicidal behavior. However, being
in a physical fight in the past year was associated with impulsive sui-
cide attempts. Zouk et al. (2006) examined the psychiatric records of
164 suicide cases using the Barratt Impulsivity Scale. Individuals who
scored 75 or higher (which was the 70th percentile for the group) were
labeled impulsive and scored significantly higher on the Buss-Durkee
Hostility Inventory than their nonimpulsive peers (defined by a score
equal to or below the 30th percentile on the Barratt scale), suggesting
that aggression is a serious risk factor for impulsive suicidal behavior.
Dumais et al. (2005) found impulsive and aggressive behaviors to be as-
sociated with suicidality in 104 males diagnosed with major depression.
However, they asserted that the relation of aggressive and impulsive
behaviors and suicide may be better explained by Cluster B personality
disorder and alcohol/drug abuse.
Self-Injurious Behaviors
Case Example 6
B.D. has been in psychiatric care for a decade. Her problems have in-
cluded an eating disorder, emotional instability, intense unstable rela-
tionships, and SIBs. Finding a space on her inner thigh not already
scarred, the 25-year-old cuts herself. As she watches the blood flow, the
intense roiling emotions she had felt moments before give way to a
sense of relaxation and peace.
SIB involves deliberate and often repetitive harm to one’s own body
without suicidal intent (Favazza 1998). A typical pattern for SIB begins
with an overwhelming psychological distress such as anger, anxiety,
tension, fear, or a sense of loss. An individual often responds to the
overwhelming emotion by isolating and dissociating. In carrying out
SIB, there is an absence of suicidal intent and often a lack of pain. The
precipitating tension is relieved by the SIB, and individuals report feel-
ing a sense of calm, often followed by disgust and/or guilt (Suyemoto
1998).
SIBs can be very diverse in terms of specific behaviors, severity, and
frequency. Simeon and Favazza (2001) proposed four classifications of
SIBs—major, stereotypic, compulsive, and impulsive—as a way to help
understand and treat the disorder. Major SIBs tend to be severe, poten-
tially lethal, and irreversible, such as castration, eye enucleation, and
amputation of extremities. This category of SIB is relatively rare and is
associated with schizophrenia, intoxication, neurological conditions,
bipolar disorder, and severe personality disorders. The impetus for ma-
jor SIB is often associated with sin, religious delusions, sexual tempta-
tion, punishment, and salvation (DeMuth et al. 1983). Stereotypic SIBs
tend to be repetitive and lack symbolism or affect. The behaviors can be
occasional or chronic, such as head banging, eyeball pressing, and fin-
ger biting (Favazza and Simeon 1995). These types of SIBs are common
in individuals with mental retardation (Griffin et al. 1986), autism
(Christie et al. 1982), and Tourette’s syndrome (Robertson et al. 1989).
Compulsive SIBs are ritualistic and repetitive behaviors such as trichotil-
lomania, nail biting, skin picking, and skin scratching (Simeon 2006).
Individuals with this type of behavior often report that the behaviors
occur unintentionally. The behaviors are typically associated with
mounting anxiety followed by relief. Impulsive SIBs include skin cut-
ting, skin burning, poisoning, and self-hitting. These behaviors tend to
provide short-term relief from unbearable psychological states (Simeon
2006).
The prevalence of any SIB in the general public has been estimated
to be 4% (Briere and Gil 1998). However, prevalence rates vary greatly
based on selected populations. For example, in a study of 15- and 16-
year-old students in England, researchers (Hawton et al. 2002) found
that 6.9% of their sample had engaged in at least one act of deliberate
self-harm in the previous year. The primary method of harm was cut-
ting (two-thirds) and poisoning (less than one-third). Multiple acts of
SIB were reported by about half of those acknowledging SIB. Matsu-

moto et al. (2005) studied 201 adolescents in a juvenile detention center
in Japan and found 16.4% had cut their wrists or forearms at least once,
and 28.4% had burned themselves at least once. This was found to be
significantly higher than the incidence among Japanese university stu-
dents, where the overall rate was 3.3% (males 3.1% and females 3.5%;
Yamaguchi et al. 2004). Prevalence rates for U.S. college students have
ranged from 12% (Favazza et al. 1989) to 17% (Whitlock et al. 2006). In
a study of male prisoners, Shea (1993) found prevalence rates of 6.5%–
25%. SIB in adult psychiatric populations can range from 4% (Darche
1990) to 20% (Langbehn and Pfohl 1993). In adolescent inpatients the
prevalence rate can range from 40% (Darche 1990) to 61% (DiClemente
et al. 1991). SIB may occur in up to 60% of individuals with Tourette’s
syndrome (Eisenhauer and Woody 1987). The differences in prevalence
rates may be attributed to the various definitions of SIB, different study
populations, and the reporting mechanisms for SIB.
SIB can serve multiple functions simultaneously. Paris (2005) identi-
fied five psychological functions of SIB: 1) relief from negative mood
states; 2) distraction, encouraging the individual to refocus attention
from psychological pain to physical pain; 3) communication of distress,
as the behaviors come to the attention of significant others or therapists;
4) expression of emotions such as guilt or anger; and 5) dissociation
from the current state while engaging in SIB.
SIB has been associated with psychological disorders such as bor-
derline personality disorder (Paris 2005), antisocial behavior (Suyemoto
1998), and eating disorders (Paul et al. 2002). Other indicators of SIB in-
clude depression, anxiety, impulsivity, and low self-esteem (Herpertz et
al. 1997). In a study of adolescent students in England, researchers
(Hawton et al. 2002) found that SIB was more common in females than
in males, and the presence of SIB increased with greater consumption
of cigarettes, alcohol, and drugs. SIB has been associated with child-
hood adversities such as physical abuse, sexual abuse, and parental ne-
glect. It is also related to environmental factors such as being bullied,
having a family member who had attempted suicide, and knowing a
peer who had engaged in SIB. Individuals who engage in SIB have also
been found to have impulsive behavioral traits (Simeon et al. 1992). For
example, individuals who carried out SIB had fewer future-oriented
problem-solving abilities and were more likely to be involved with
other impulsive behaviors such as suicide attempts, substance abuse,
bingeing, and promiscuity (Herpertz et al. 1997).
Key Points
■ Impulsivity and aggression may be useful or destructive behav-
iors, depending on the context. They are characteristic of a range
of disorders and of symptoms of mental illness that may result in
significant functional impairment, morbidity, and mortality.
■ Intermittent explosive disorder (IED), an impulse-control disorder,
is the only DSM-IV-TR diagnosis with recurrent aggressive acts as
the primary symptom. People with IED describe “a need to attack,”
“an adrenaline rush,” “a need to defend oneself,” and “an urge to
kill.” Episodes may be infrequent and brief but can have devastat-
ing results such as destruction of property, serious assault, or even
homicide. People with IED have difficulties maintaining employ-
ment, financial stability, and meaningful relationships.
■ Pyromania is the repeated failure to resist the impulse to set
motiveless fires. Risk factors associated with pyromania are male
gender, unemployment, living alone, and a keen interest in fires.
■ For people with intellectual disabilities, aggressive behavior can
have a severe negative impact on residential, education, and
employment opportunities. Acts of aggression tend to increase
with severity of disability, but higher-functioning individuals tend
to act aggressively toward others and lower-functioning individu-
als to engage in self-injurious behavior (SIB). Skill deficits in atten-
tion span, impulse control, memory, neurological functioning,
communication skills, and social skills may be present and may
increase the probability of aggression.
■ In individuals with autism, aggression is common and may include
impulsivity, aggression toward others, SIB, destruction of property,
disruptiveness, stereotypy, and other socially unacceptable behav-
iors, not necessarily with malicious intent. Impairments com-
monly found in autism may reduce the capacity for social interac-
tion, communication, and ability to alter behavior and emotional
states in response to another’s action or perceived feelings.
■ Suicide is a major public health concern and is a leading cause of
death in several Western countries. A prior suicide attempt is one
of the best predictors for eventual death by suicide; however, the
preventive value of this is limited because more than half of fatal
suicides occur on the first attempt. Those who engage in impul-
sive suicide (with less than 5 minutes of premeditation) use more
violent methods, have lower expectations of dying from their
actions, and have comparable destructive outcomes to their non-
impulsive counterparts.
■ Self-injurious behavior is deliberate, repetitive harm (e.g., cutting,

burning, poisoning) to one’s own body without suicidal intent. A
typical pattern for SIB is an initial overwhelming distress with sub-
sequent emotional dissociation; a lack of pain while carrying out
the SIB; and relief of tension, possibly followed by a sense of guilt.
SIB can serve psychological functions, such as relief from negative
mood states, distraction, communication of distress, expression of
emotions, and dissociation from one’s current psychological state.
References
Aicardi J: Diseases of the Nervous System in Children, 2nd Edition. Cambridge,
England, Cambridge University Press, 1998
Allen D: Recent research on physical aggression in persons with intellectual dis-
ability: an overview. J Intellect Dev Disabil 25:41–57, 2000
Aman MG, Richmond G, Stewart AW, et al: The Aberrant Behavior Checklist:
factor structure and the effect of subject variables in American and New
Zealand facilities. Am J Ment Defic 91:570–578, 1987
American Psychiatric Association: Diagnostic and Statistical Manual: Mental
Disorders. Washington, DC, American Psychiatric Association, 1952
Disorders, 2nd Edition. Washington, DC, American Psychiatric Associa-
tion, 1968
tion, 1980
Anderson CA, Bushman BJ: Human aggression. Annu Rev Psychol 53:27–51,
2002
Barker AF: Arson: A Review of the Psychiatric Literature. Oxford, UK, Oxford
University Press, 1994
Barratt ES, Stanford MS, Dowdy L, et al: Impulsive and premeditated aggres-
sion: a factor analysis of self-reported acts. Psychiatry Res 86:163–173, 1999
Berman ME, Coccaro EF: Neurobiologic correlates of violence: relevance to
criminal responsibility. Behav Sci Law 16:303–318, 1998
Berridge CW, Waterhouse BD: The locus coeruleus-noradrenergic system: mod-
ulation of behavioral state and state-dependent cognitive processes. Brain
Res Rev 42:33–84, 2003
Bjorkvist K, Niemela P: Of Mice and Women: Aspects of Female Aggression.
New York, Academic Press, 1992
Bond AJ: Pharmacological manipulation of aggressiveness and impulsiveness
in healthy volunteers. Prog Neuropsychopharmacol Biol Psychiatry 16:1–
7, 1992
Briere J, Gil E: Self-mutilation in clinical and general population samples: prev-
alence, correlates, and functions. Am J Orthopsychiatry 68:609–620, 1998
Briere J, Runtz M: Differential adult symptomatology associated with three

types of child abuse histories. Child Abuse Negl 14:357–364, 1990
Brodsky BS, Oquendo M, Ellis SP, et al: The relationship of childhood abuse to
impulsivity and suicide behavior in adults with major depression. Am J
Psychiatry 158:1871–1877, 2001
Bruininks RH, Olson KM, Larson SA, et al: Challenging behaviors among per-
sons with mental retardation in residential settings, in Destructive Behav-
ior in Developmental Disabilities: Diagnosis and Treatment. Edited by
Thompson TE, Gray DBE. Thousand Oaks, CA, Sage, 1994, pp 24–48
Buss A, Perry M: The Aggression Questionnaire. J Pers Soc Psychol 63:452–459,
1992
Christie R, Bay C, Kaufman IA, et al: Lesch-Nyan disease: clinical experience
with nineteen patients. Dev Med Child Neurol 24:293–306, 1982
Coccaro EF: Neurotransmitter correlates of impulsive aggression in humans.
Ann NY Acad Sci 794:82–89, 1996
Coccaro EF: Intermittent explosive disorder, in Aggression: Psychiatric Assess-
ment and Treatment. Edited by Coccaro EF. New York, Marcel Dekker,
2003, pp 149–199
Coccaro EF, Kavoussi RJ: Fluoxetine and impulsive aggressive behavior in per-
sonality disorder subjects. Arch Gen Psychiatry 54:1081–1088, 1997
Coccaro EF, Schmidt CA, Samuels JF, et al: Lifetime and 1-month prevalence
rates of intermittent explosive disorder in a community sample. J Clin Psy-
chiatry 65:820–824, 2004
Coccaro EF, Posternak MA, Zimmerman M: Prevalence and features of intermittent
explosive disorder in a clinical setting. J Clin Psychiatry 66:1221–1227, 2005
Cornell DG, Warren J, Hawk G, et al: Psychopathy in instrumental and reactive
violent offenders. J Consult Clin Psychol 64:783–790, 1996
Crocker AG, Hodgins S: The criminality of noninstitutionalized mentally re-
tarded persons: evidence from a birth cohort followed to age 30. Crim Jus-
tice Behav 24:432–454, 1997
Croen LA, Grether JK: Response: a response to Blaxill, Baskin, and Spitzer on
Croen et al. (2002). The changing prevalence of autism in California. J Au-
tism Dev Disord 33:227–229, 2003
Darche MA: Psychological factors differentiating self-mutilating and non-self-
mutilating adolescent inpatient females. Psychiatr Hosp 21:31–35, 1990
Davidson PW, Cain NN, Sloane-Reeves JE, et al: Characteristics of community-
based individuals with intellectual disability and aggressive behavioral
disorders. Am J Ment Retard 98:704–716, 1994
Deb S, Thomas M, Bright C: Mental disorder in adults with intellectual disabil-
ity, 2: the rate of behaviour disorders among a community-based popula-
tion aged between 16 and 64 years. J Intellect Disabil Res 45:506–514, 2001
DeMuth GW, Strain J, Lombardo-Maher A: Self-amputation and restitution.
Gen Hosp Psychiatry 5:25–30, 1983
Dewey M: Living with Asperger’s syndrome, in Autism and Asperger Syn-
drome. Edited by Frith U. Cambridge, UK, Cambridge University Press,
1991, pp 184–206
DiClemente RJ, Ponton LE, Hartley D: Prevalence and correlates of cutting be-
havior: risk for HIV transmission. J Am Acad Child Adolesc Psychiatry
30:735–739, 1991
Dumais A, Lesage AD, Alda M, et al: Risk factors for suicide completion in ma-
jor depression: a case-control study of impulsive and aggressive behaviors
in men. Am J Psychiatry 162:2116–2124, 2005
Eichelman B: Toward a rational pharmacotherapy for aggressive and violent be-
havior. Hosp Community Psychiatry 39:31–39, 1988
Eisenhauer GL, Woody RC: Self-mutilation and Tourette’s disorder. J Child
Neurol 2:265–267, 1987
Emerson E, Alborz A, Reeves D, et al: The HARC Challenging Behaviour
Project, Report 2: The Prevalence of Challenging Behaviour. Manchester,
England, Hester Adrian Research Centre, University of Manchester, 1997
Ernst M, Moolchan ET, Robinson ML: Behavioral and neural consequences of
prenatal exposure to nicotine. J Am Acad Child Adolesc Psychiatry 40:630–
641, 2001
Favazza AR: The coming of age of self-mutilation. J Nerv Ment Dis 186:259–268,
1998
Favazza A, Simeon D: Self-mutilation, in Impulsivity and Aggression. Edited
by Hollander E, Simeon D. New York, Wiley, 1995, pp 185–200
Favazza AR, DeRosear L, Conterio K: Self-mutilation and eating disorders. Sui-
cide Life Threat Behav 19:352–361, 1989
Felthous AR, Barratt ES: Impulsive aggression, in Aggression: Psychiatric As-
sessment and Treatment. Edited by Coccaro EF. New York, Marcel Dekker,
2003, pp 123–148
Fombonne E: The epidemiology of autism: a review. Psychol Med 29:769–786,
1999
Fombonne E: The prevalence of autism. JAMA 289:87–89, 2003
Gerra G, Zaimovic A, Avanzini P, et al: Neurotransmitter-neuroendocrine re-
sponses to experimentally induced aggression in humans: influence of per-
sonality variable. Psychiatry Res 66:33–43, 1997
Golden CJ, Jackson ML, Peterson-Rohne A, et al: Neuropsychological correlates
of violence and aggression: a review of the clinical literature. Aggression
and Violent Behavior 1:3–25, 1996
Goldstein RB, Black DW, Nasrallah A, et al: The prediction of suicide: sensitiv-
ity, specificity, and predictive value of a multivariate model applied to sui-
cide among 1906 patients with affective disorders. Arch Gen Psychiatry
48:418–422, 1991
Grafman J, Schwab K, Warden D, et al: Frontal lobe injuries, violence, and ag-
gression: a report of the Vietnam Head Injury Study. Neurology 46:1231–
1238, 1996
Griffin JC, Williams DE, Stark MT, et al: Self-injurious behavior: a state-wide
prevalence survey of the extent and circumstances. Appl Res Ment Retard
7:105–116, 1986
Haden SC, Scarpa A: The noradrenergic system and its involvement in aggres-
sive behaviors. Aggression and Violent Behavior 12:1–15, 2007
Haller J, Makara GB, Kruk MR: Catecholaminergic involvement in the control
of aggression: hormones, the peripheral sympathetic and central noradren-
ergic systems. Neurosci Biobehav Rev 22:85–97, 1998
Harris P: The nature and extent of aggressive behaviour amongst people with
learning difficulties (mental handicap) in a single health district. J Intellect
Disabil Res 37 (pt 3):221–242, 1993
Hartup WW: Aggression in childhood: developmental perspectives. Am Psy-

chol 29:336–341, 1974
Hawton K, Rodham K, Evans E, et al: Deliberate self harm in adolescents: self
report survey in schools in England. BMJ 325:1207–1211, 2002
Henriksson MM, Aro HM, Marttunen MJ, et al: Mental disorders and comor-
bidity in suicide. Am J Psychiatry 150:935–940, 1993
Herpertz S, Sass H, Favazza A: Impulsivity in self-mutilation behavior: psycho-
metric and biological findings. J Psychiatry Res 31:451–465, 1997
Holland T, Clare IC, Mukhopadhyay T: Prevalence of criminal offending by
men and women with intellectual disability and the characteristics of of-
fenders: implications for research and service development. J Intellect Dis-
abil Res 46 (suppl 1):6–20, 2002
Isometsä ET, Lonnqvist JK: Suicide attempts preceding completed suicide. Br J
Psychiatry 173:531–535, 1998
Johnsson Fridell E, Ojehagen A, Träskman-Bendz L: A 5-year follow-up study
of suicide attempts. Acta Psychiatr Scand 93:151–157, 1996
Kafry D: Playing with matches: children and fires, in Fires and Human Behav-
ior. Edited by Canter D. Chichester, UK, Wiley, 1980, pp 47–62
Kanner L: Autistic disturbances of affective contact. Nerv Child 2:217–250, 1943
Kessler RC, Coccaro EF, Fava M, et al: The prevalence and correlates of DSM-IV
intermittent explosive disorder in the National Comorbidity Survey repli-
cation. Arch Gen Psychiatry 63:669–678, 2006
Langbehn DR, Pfohl B: Clinical correlates of self-mutilation among psychiatric
inpatients. Ann Clin Psychiatry 5:45–51, 1993
Larson SA, Lakin KC: Deinstitutionalization of persons with mental retarda-
tion: behavioral outcomes. J Assoc Pers Sev Handicaps 14:324–332, 1989
Lejoyeux M, Feuché N, Loi S, et al: Study of impulse-control disorders among
alcohol-dependent patients. J Clin Psychiatry 60:302–305, 1999
Lejoyeux M, McLoughlin M, Ades J: Pyromania, in Clinical Manual of Impulse-
Control Disorders. Edited by Hollander E, Stein DJ. Washington, DC,
American Psychiatric Publishing, 2006, pp 229–250
Leong GB, Silva JA: Revisiting arson from an outpatient forensic perspective.
J Forensic Sci 44:558–563, 1999
Lesage AD, Boyer R, Grunberg F, et al: Suicide and mental disorders: a case-
control study of young men. Am J Psychiatry 151:1063–1068, 1994
Lewis ND, Yarnell H: Pathological firesetting (pyromania). Nervous and Men-
tal Disease Monographs no 82, 1951
Lindberg N, Holi MM, Tani P, et al: Looking for pyromania: characteristics of a
consecutive sample of Finnish male criminals with histories of recidivist
fire-setting between 1973 and 1993. BMC Psychiatry 5:47, 2005
Liu J, Wuerker A: Biosocial bases of aggressive and violent behavior: implica-
tions for nursing studies. Int J Nurs Stud 42:229–241, 2004
Maguire KB, Piersel WC: Specialized treatment for behavior problems of insti-
tutionalized persons with mental retardation. Ment Retard 30:227–232,
1992
Matsumoto T, Yamaguchi A, Chiba Y, et al: Self-burning versus self-cutting: pat-
terns and implications of self-mutilation. A preliminary study of differ-
ences between self-cutting and self-burning in a Japanese juvenile
detention center. Psychiatry Clin Neurosci 59:62–69, 2005
McClintock K, Hall S, Oliver C: Risk markers associated with challenging be-

haviours in people with intellectual disabilities: a meta-analytic study. J In-
tellect Disabil Res 47:405–416, 2003
McDougle CJ, Stigler KA, Posey DJ: Treatment of aggression in children and ad-
olescents with autism and conduct disorder. J Clin Psychiatry 64(suppl):16–
25, 2003
McElroy SL, Soutullo CA, Beckman DA, et al: DSM-IV intermittent explosive
disorder: a report of 27 cases. J Clin Psychiatry 59:203–212, 1998
Meloy JR: Violence Risk and Threat Assessment: A Practical Guide for Mental
Health and Criminal Justice Professionals. San Diego, CA, Specialized
Training Services, 2000
Moyer KE: Kinds of aggression and their physiological basis. Commun Behav
Biol 2:65–87, 1968
Nock MK, Kessler RC: Prevalence of and risk factors for suicide attempts versus
suicide gestures: analysis of the National Comorbidity Survey. J Abnorm
Psychol 115:616–623, 2006
Nordström P, Samuelsson M, Asberg M: Survival analysis of suicide risk after
attempted suicide. Acta Psychiatr Scand 91:336–340, 1995
O’Carroll PW, Berman AL, Maris RW, et al: Beyond the Tower of Babel: a no-
menclature for suicidology. Suicide Life Threat Behav 26:237–252, 1996
Paris J: Understanding self-mutilation in borderline personality disorder. Harv
Rev Psychiatry 13:179–185, 2005
Paul T, Schroeter K, Dahme B, et al: Self-injurious behavior in women with eat-
ing disorders. Am J Psychiatry 159:408–411, 2002
Raine A: Annotation: the role of prefrontal deficits, low autonomic arousal, and
early health factors in the development of antisocial and aggressive behav-
ior in children. J Child Psychol Psychiatry 43:417–434, 2002
Raine A, Buchsbaum M, LaCasse L: Brain abnormalities in murderers indicated
by positron emission tomography. Biol Psychiatry 42:495–508, 1997
Räsänen P, Hakko H, Väisänen E: Arson trend increasing: a real challenge to
psychiatry. J Forensic Sci 40:976–979, 1995
Räsänen P, Hakko H, Isohanni M, et al: Maternal smoking during pregnancy
and risk of criminal behavior among adult male offspring in the Northern
Finland 1996 Birth Cohort. Am J Psychiatry 156:857–862, 1999
Repo E, Virkkunen M, Rawlings R, et al: Criminal and psychiatric histories of
Finnish arsonists. Acta Psychiatr Scand 95:318–323, 1997
Ritchie EC, Huff TG: Psychiatric aspects of arsonists. J Forensic Sci 44:733–740,
1999
Robertson MM, Trimble MR, Lees AJ: Self-injurious behavior and the Gilles de
la Tourette syndrome: a clinical study and review of the literature. Psychol
Med 19:611–625, 1989
Rogers J, Viding E, Blair RJ, et al: Autism spectrum disorder and psychopathy:
shared cognitive underpinnings or double hit? Psychol Med 36:1789–1798,
2006
Rome S, Itskowitz R: The relationship between locus of control and type of ag-
gression in middle-class and culturally deprived children. Pers Individ Dif
11:327–333, 1990
Roy A: Childhood trauma and impulsivity: possible relevance to suicidal be-
havior. Arch Suicide Res 9:147–151, 2005
Rusch RG, Hall JC, Griffin HC: Abuse-provoking characteristics of institution-

alized mentally retarded individuals. Am J Ment Defic 90:618–624, 1986
Shea S: Personality characteristics of self-mutilating male prisoners. J Clin Psy-
chol 49:576–585, 1993
Sigafoos J, Elkins J, Kerr M, et al: A survey of aggressive behaviour among a
population of persons with intellectual disability in Queensland. J Intellect
Disabil Res 38:369–381, 1994
Simeon D: Self-injurious behavior, in Clinical Manual of Impulse-Control Dis-
orders. Edited by Hollander E, Stein D. Washington, DC, American Psychi-
atric Publishing, 2006, pp 63–86
Simeon D, Favazza AR: Self-injurious behaviors: phenomenology and assess-
ment, in Self-Injurious Behaviors: Assessment and Treatment. Edited by
Simeon D, Hollander E. Washington, DC, American Psychiatric Publish-
ing, 2001, pp 1–28
Simeon D, Stanley B, Frances A, et al: Self-mutilation in personality disorders:
psychological and biological correlates. Am J Psychiatry 149:221–226, 1992
Simon OR, Swann AC, Powell KE, et al: Characteristics of impulsive suicide at-
tempts and attempters. Suicide Life Threat Behav 32 (1, suppl):49–59, 2001
Sood BG, Nordstrom Bailey B, Covington C, et al: Gender and alcohol moderate
caregiver reported child behavior after prenatal cocaine. Neurotoxicol Ter-
atol 27:191–201, 2005
Stanford MS, Houston RJ, Mathias CW, et al: Characterizing aggressive behav-
ior. Assessment 10:183–190, 2003a
Stanford MS, Houston RJ, Villemarette-Pittman NR, et al: Premeditated aggres-
sion: clinical assessment and cognitive psychophysiology. Pers Individ Dif
34:773–781, 2003b
Stoff D, Breiling J, Maser J: Handbook of Antisocial Behavior. New York, Wiley,
1997
Stone MH: Psychotherapy in patients with impulsive aggression, in Impulsivity
and Aggression. Edited by Hollander E, Stein D. New York, Wiley, 1995, pp
313–331
Strongman KT: Emotion in mentally retarded people. Aust NZ J Dev Disabil
10:201–213, 1985
Suyemoto KL: The functions of self-mutilation. Clin Psychol Rev 18:531–554,
1998
Swann AC, Dougherty DM, Pazzaglia PJ, et al: Increased impulsivity associated
with severity of suicide attempt history in patients with bipolar disorder.
Am J Psychiatry 162:1680–1687, 2005
Tateno A, Jorge RE, Robinson RG: Clinical correlates of aggressive behavior af-
ter traumatic brain injury. J Neuropsychiatry Clin Neurosci 15:155–160,
2003
Villemarette-Pittman NR, Stanford MS, Greve KW: Language and executive
function in self-reported impulsive aggression. Pers Individ Dif 34:1533–
1544, 2003
Walz NC, Benson BA: Labeling and discrimination of facial expressions by ag-
gressive and nonaggressive men with mental retardation. Am J Ment Re-
tard 101:282–291, 1996
Whitlock J, Eckenrode J, Silverman D: self-injurious behaviors in a college pop-
ulation. Pediatrics 117:1939–1948, 2006
Williams CL, Davidson JA, Montgomery I: Impulsive suicidal behavior. J Clin

Psychol 35:90–94, 1980
Williams DL, Goldstein G, Minshew NJ: Impaired memory for faces and social
scenes in autism: clinical implications of memory dysfunction. Arch Clin
Neuropsychol 20:1–15, 2005
Williams JG, Higgins JP, Brayne CE: Systematic review of prevalence studies of
autism spectrum disorders. Arch Dis Child 91:8–15, 2006
Williams RB, Eichelman B: Social setting: influence on the physiological re-
sponse to electric shock in the rat. Science 174:613–614, 1971
Woodworth M, Porter S: In cold blood: characteristics of criminal homicides as
a function of psychopathy. J Abnorm Psychol 111:436–445, 2002
World Health Organization: Suicide prevention, 2006. Available at http://www
.who.int/mental_health/prevention/suicide/suicideprevent/en/index
.html. Accessed February 5, 2007
Yamaguchi A, Matsumoto T, Odawara T, et al: Prevalence of self-mutilation in
Japanese university students: a study using self-reporting questionnaire.
Seishin Igaku 46:473–479, 2004
Zouk H, Tousignant M, Seguin M, et al: Characterization of impulsivity in sui-
cide completers: clinical, behavioral and psychosocial dimensions. J Affect
Disord 92:195–204, 2006
P A R T I I I
Treatment Settings
C H A P T E R 1 2
Outpatient Settings
For the vast majority of psychiatric patients potential violence is not an

issue, but when it is, the clinician is obliged to deal with it. This chapter
focuses on voluntary outpatient treatment, with brief comments, when
relevant, on involuntary or otherwise coerced outpatient treatment.
Violence in this chapter is defined according to the MacArthur group
criteria—purposeful physical assault against a person. Serious violence
refers to assault with a weapon, assault that causes significant injury, or
sexual assault (Monahan and Steadman 1994).
Potential violence is a topic that concerns every clinician, and it
frightens some clinicians more than others. In discussions of this issue,
people often look for clear answers to the question of whether there is a
risk of violence, and if there is a risk, whether there is a clear basis for
deciding what to do about it. Many clinicians would like to believe that
there are clear rules for making decisions or that “This risk assessment
tool will provide me with the guidance I need to proceed.” In this vein,
a clinician once said, “In our clinic it is policy that whenever a patient
makes a threat, we warn the threatened person.” The wish for certainty
is understandable, and probably harmless, but the rigid warning policy
of that clinic is pernicious. It carries the risk of serious harm to patients,
and responsible clinicians and agencies should avoid policies like these.
General Principles
There is no substitute for clinical judgment in assessing the risk
of violence and in making the many decisions involved in treatment.
237
Actuarial methods are helpful as background, but only as background.

For example, it helps to know that young people are more often violent
than older people, that men are more often violent than women, that
people brought up in violent circumstances are more likely to be violent
than those brought up in safety, and so on. However, knowing all the
actuarial data will not in the end serve to reduce the necessity for mak-
ing a clinical assessment of risk. The person in the clinical encounter is
unique; the circumstances of his or her life are unique and may be
changing, rapidly or slowly, in ways that affect risk. It is the clinician’s
responsibility, difficult as this may be, to learn about this person and the
circumstances of his or her life, and on this basis to make the best esti-
mate of risk. This is true in all clinical settings.
This chapter does not review the literature on risk assessment in-
struments or recommend one particular risk assessment method. In-
stead, it presents a conceptual framework that should help guide the
clinician’s assessment.
Whether the clinician is concerned about violence or not, two useful
screening questions are “Are you angry at anyone?” and “Have you
hurt anyone?” If the answer is “No” to both questions, the risk of vio-
lence is almost always low—unless the interview or other data suggest
contrary facts. This is an example of a general principle of clinical as-
sessment: to find out the answer to a question about a patient, ask the
patient the question. If concerned about potential violence, ask “Are
you thinking about hurting someone?” This does not imply that the cli-
nician should always believe the answer—here, as elsewhere, clinical
judgment is required. Yet this approach yields quite a lot of useful infor-
mation, and it can be far more efficient than sniffing around the prob-
lem by asking a number of supposedly related questions.
A corollary of this principle is that the more the clinician knows
about the person, the more confidence the clinician can have in his or
her assessment of the answers to these questions. An interview after
1 year of weekly psychotherapy is a different situation from an initial
interview in an emergency service.
If the interview develops evidence that the person is having violent
fantasies, or is thinking about hurting someone, it is useful to ask some-
thing like, “Do you think this is something you might actually do, or are
you just thinking about it?” (This question is useful in assessing risk of
suicide or other self-harmful behavior when a patient has expressed
suicidal ideation.) Patients are often able to answer this question in
ways that provide useful guidance to the clinician in assessing risk. Sec-
ond, assess the affect associated with the answer. Is the patient fright-
ened that he might actually hurt his separated wife? Or excited about
Outpatient Settings ❘ 239
possibly getting revenge on her and her new boyfriend? Or is this a fan-
tasy with little or no affective charge of any kind? It is essential to attend
to the affect associated with the answers to these questions and not only
to the content of the answer.
Conceptual Framework and Case Example

Kurt Lewin, often recognized as the father of social psychology, first
wrote the following simple equation, quoted in Hall and Gardner (1978):
B=f(P,E)
Expanded, it means that behavior (B) is a function of the person (P) and
the environment (E). Too often, clinical training focuses on the individual
person, P, ignoring or giving too little attention to the facts of the per-
son’s situation or environment, E. The description of a patient as a “vi-
olent person” is common but not terribly helpful. It is true that some
people are more likely to be violent than others, but the likelihood of vi-
olence depends to a substantial extent on the person’s environment or
situation, and the clinician should not neglect gathering the relevant
facts about the person’s situation. This conceptual framework is useful
not only for assessment, but also for treatment, as in this case example.
Case Example
Mr. A, a 37-year-old single white male, was first diagnosed with para-
noid schizophrenia at age 19. He receives disability payments and lives
with his mother, whom he believes is poisoning his food. He gives as his
reason that his food tastes funny. He adds that when he complains, his
mother tells him there is nothing wrong with the food—there must be
something wrong with him.
He tells his clinician that he is frightened and that he has begun to
think about killing his mother. He denies that he has any concrete idea
of how to kill her or any plan for what he might do, but he acknowledges
that this thought does not cause him any anxiety. He denies being angry
at anyone else, and he has no known history of violence. He denies any
substance abuse. He says that he is taking his medication and attending
outpatient treatment and a drop-in center regularly. His health is good.
His past history shows that he was a quiet child with few friends.
His chart contains no mention of fighting, truancy, or oppositional be-
havior. He had one brief romantic relationship at age 19, and he has a
long history of unemployment—he last worked 15 years ago, part time,
bagging groceries. In the past he has taken his medication inconsis-
tently, and when he has stopped his medicine he has also stopped com-
ing to outpatient appointments.
What is to be done for this patient? As a first step in assessing risk, the
clinician might review one of the published methods of risk assessment.
The HCR-20, a 20-item instrument, is a good choice (Webster 1997). It
serves as a reminder-checklist for things we need to know in weighing
risk. The authors of the HCR-20 suggest that each item be rated absent
(0), possibly present (1), or definitely present (2). The authors wrote, “Put
simply, it is reasonable for assessors to conclude that the more factors
present in a given case, the higher the risk for violence” (Webster 1997,
p. 22). But, they added, “Even here, though, assessors must be cautious”
(p. 22), implying that they recognized that a simple quantitative ap-
proach is not an adequate basis on which to assess risk. Published stud-
ies show correlations of about 0.30 between scores on the HCR-20 and fu-
ture violence (Webster 1997). This means that the HCR-20 predicts less
than 10% of the variance in future violence. It is useful for helping us
gather the data we need, but it is not a substitute for clinical judgment.
There are three scales in the HCR-20: 10 historical items (H), 5 clini-
cal items (C), and 5 risk items (R). Rating the H items for the patient in
the case example, we have the following:
• Major mental illness: Yes

• Early maladjustment: Possible
• Employment problems: Possible
• Five negatives: A history of violence, substance abuse, intimate rela-
tionship instability (this patient’s history is considered to show pau-
city rather than instability), psychopathy, and personality disorder
• Two not applicable: Age at first violence and failed supervision (“su-
pervision” refers to probation or other involuntary supervision)
Rating the clinical items for this patient, we have
• Three positives: Lack of insight, negative attitude (toward mother),

active symptoms of mental illness
• Two negatives: No evidence for impulsivity, and patient is respon-
sive to treatment currently. However, if past history of inconsistent
medication compliance is factored in, this might better be scored
“possible.”1
1
This illustrates that so-called objective rating scales often require clinical judgments in
making the ratings. However, since the inconsistent response is rated “possible” as a
risk-management factor as well, rating it as possible here may be double counting. Thus
“negative” is more accurate.
Rating the five risk-management items for Mr. A, we find that he has
• No positive indicators
• One negative: No feasible plan
• One possible: Variable past compliance with treatment
• Three items we have not evaluated: Exposure to destabilizers, lack of
personal support, and current stress
This exercise reveals that evaluation to this point has focused almost
exclusively on P (person) items in attempting to assess B—that is, future
violent behavior—and has almost totally ignored E (environment). In
the HCR-20, three of the five risk-management items refer to environ-
mental variables: exposure to destabilizers, lack of personal support,
and stress. At this point, these have not been adequately assessed.
This analysis points in the appropriate direction. The clinician can
either re-interview Mr. A or, if she will agree, interview his mother. Note
here I did not say to interview the mother if Mr. A agrees. Interview the
mother regardless of whether he agrees. Why? Because, as a general
rule, safety issues override confidentiality issues. When the clinician is
seriously concerned about the potential for harm to either the patient or
anyone else, he or she should attempt to gather whatever data is rele-
vant to the assessment of risk. Here, the mother is likely to provide use-
ful information, and the concern is about a potential killing, so the cli-
nician must interview her.
Too often, when a senior clinician consults about a patient’s poten-
tial for suicide or for violence toward others and asks what the family
says, the answer is, “Oh, we haven’t talked with the family. The patient
wouldn’t give us permission.” This occurs most often on inpatient units.
This is wrong. Safety is more basic than confidentiality. This does not
mean that the clinician has carte blanche to violate confidentiality. It does
mean that when the clinician has done a thorough assessment with the
patient and safety remains an important issue, the clinician must look
further to whatever other sources of information may be accessible.
To add one clinical note to this aside, the clinician should almost al-
ways, except in the most unusual circumstances (e.g., a delirious pa-
tient), explain to the patient what he or she is going to do and why.
There are data showing that therapy is disrupted not when the clinician
breaches confidentiality to seek additional sources of information but
rather when the clinician fails to inform the patient of what he or she is
doing (Beck 1981). When the clinician goes behind the patient’s back to
breach confidentiality, the patient feels betrayed. The result is that the
patient distrusts the clinician, will end the contact if possible, and is
likely to avoid the mental health system in the future (Beck 1981; Tara-
soff v. Regents of the University of California 1976).
Returning to Mr. A, and recalling the missing information, here are
two alternative scenarios.
In the first scenario, Mr. A and his mother live alone with no immediate
family nearby. Mr. A is not sure whether his mother is angry with him
or not. When the clinician interviews the mother, she says that she is
quite annoyed with her son—she cannot understand why he is com-
plaining about her cooking, and she is quite unsympathetic. She says,
“With all I do for him, he is really ungrateful. I tell him if he doesn’t like
the food here he should ‘go to another hotel.’ ” Mr. A is worried about
his situation at home. He wonders if his mother means he should leave.
Asked about outside social supports, Mr. A says his only regular contact
is the drop-in center and that he has recently had an argument with one
of the other patients and has not been back for several weeks.
In terms of the missing data, each of the three risk-management vari-

ables converts to positive. Mother’s suggestion that he can “go to another
hotel” destabilizes the living situation because the patient is actively
worried about it. We have no evidence for other social supports, and his
argument at the drop-in center is evidence of additional current stress.
In the alternate scenario, the clinician learns Mr. A’s divorced sister lives
downstairs with her 10-year-old son and Mr. A has a good relationship
with them. He often eats there, especially if there has been an argument
at home. Mother (or sister) says that Mr. A has been talking for years
about his fear of being poisoned but that his sister is able to reassure
him, and the family takes all this in stride as part of Mr. A’s illness. Mr.
A has no close friends, but he does have coffee almost every day at the
local Dunkin’ Donuts, and he is doing fine in his drop-in center.
The assessment of risk in these two scenarios is quite different. In the

first scenario, the patient’s mother is a destabilizing influence with her
angry implied threat that he should leave. There are no other balancing
family members who can be a source of support, and he has an unstable
social situation at the drop-in center that is an additional source of
stress. In this case, hospitalization seems clearly to be indicated. In the
second scenario, however, a number of positive features are present: a
relationship with other family members that provides both support and
respite; a family that is apparently able to deal with this member with-
out becoming alarmed; and a stable social situation at the drop-in cen-
ter. In this case some clinicians might still be inclined to hospitalize, but
others might conclude that there is no risk of imminent harm and that
a longer-term plan might be contemplated while the patient continued
in the community. Here the issue of prior knowledge is important. If the

assessing clinician has been seeing this person for medication manage-
ment for several years and knows the family, outpatient management is
more likely. If the mother has brought the patient to an emergency de-
partment where he is unknown, and the outpatient psychiatrist is un-
available, hospitalization is more likely.
It may be necessary to hospitalize this man acutely and perhaps ad-
just his medication. Medication may reduce the strength of the delusion
or it may not. He may take his medication as prescribed, or he may not.
For the long term, the critical intervention that will reduce risk, under
the first, more dangerous set of facts, is to separate Mr. A and his
mother. If the patient is agreeable to moving to a group home or other
setting, and if this can be accomplished, the risk of violence can be re-
duced to tolerable levels. This man has never threatened anyone except
his mother as far as we know, and he has one very specific paranoid de-
lusion—thus, removing him from the source of perceived threat may be
adequate to control risk.
This example illustrates a fundamental point. If B=f(P,E), and the B
(behavior) in question is serious violence, we can reduce the risk of vi-
olence by changing either P (the individual) or E (the environment).
Changing P may involve depot medication or new psychosocial treat-
ment such as day treatment, a group, or a rehabilitation program.
Changing E may involve changing the patient’s physical environment
or interpersonal situation or, as in Mr. A’s example, changing both.
Assessing Risk in Patients With a Past History

of Violence
When called on to assess potential future violence in the case of a pa-
tient who has been violent, a simple but often helpful rule is, “The best
predictor of what will happen in the future is what happened in the
past, unless something is different.” This rule is not absolute, but it focuses
attention where it should be placed—on what has happened and what
has or has not changed. If someone has been violent in this or a similar
environment, then the immediate question is whether the person or the
situation has changed sufficiently that the risk is acceptably low. The
critical question to ask is, “What has changed?”
The situation often arises in outpatient practice that the clinician is
asked to accept an about-to-be discharged inpatient into some kind of
outpatient follow-up treatment. If the patient had been hospitalized in
part because of violent behavior, the clinician should be satisfied, before
accepting the patient, that the risk of violence has been reduced to an
acceptable level. At the present time, when managed care dictates brief
hospitalizations, very little if anything is likely to have changed be-
tween pre-admission and discharge except that a week has passed since
the patient was removed from his or her environment to the hospital.
Now the hospital is proposing to discharge the patient to outpatient
care. If the outpatient clinician is part of an organizational network car-
ing for such patients, she or he may have something to say about
whether enough has changed that this patient is safe for outpatient
treatment. The value of B=f(P,E) in this situation is that it helps the cli-
nician ask a broad range of questions and consider a range of poten-
tially useful interventions or changes that may reduce the risk.2
Assessing Imminence of Risk

Assume that the clinician has assessed risk and decided that the patient in
this situation is at risk for violence. In this context, assessing imminence of
risk is critical. If risk is imminent, the clinician must make a definitive in-
tervention now. If risk is not imminent, there is the luxury of time in which
to try to engage the patient in treatment or to otherwise intervene.
As with everything else, assessment of imminence requires assess-
ment of both P and E. For example, a patient who insists that he plans
to blow up the post office but who appears to have no access to, or ex-
perience with, explosives and no knowledge of how to build them is
probably not at imminent risk. The clinician may decide to hospitalize
this patient, but unless there is reason to believe the patient may be con-
templating some other imminent violent act, the choice is not forced.
When the clinician is convinced that there is a risk of imminent
violence, and the patient is mentally ill, then action is necessary, and
usually the best choice is hospitalization. If the patient agrees to be
hospitalized, the decision is easy. If the patient refuses voluntary
hospitalization and the clinician judges that failure to hospitalize would
create a likelihood of serious harm by reason of mental illness, then the
clinician has the legal power to hospitalize the patient involuntarily.
2 Although the focus of this analysis is on danger to others, it is equally useful in assess-
ing risk of self-harmful behavior. For example, a man who worked in the same office
with his girlfriend became suicidally depressed after she dumped him. This patient
improved significantly after he changed his job situation. Not seeing the ex-girlfriend
every day gave the patient and the clinician time to provide treatment addressing the
person’s vulnerability to loss and related depression. Changing E-related risk reduced
the risk of self-harmful B and gave the psychiatrist the freedom to work on enduring P
variables with her patient.
This implies the responsibility to use one’s best judgment in deciding

how to use that power.
Involuntary hospitalization involves not only treatment but also a
deprivation of liberty. This deprivation is a very serious matter. In the
United States, imminent risk of serious harm by reason of mental illness is
the only condition under which a person may be deprived of liberty with-
out having been charged with a crime. Clinicians are likely to focus on the
good they can do by providing treatment, whether voluntary or not; they
are much less likely to weigh the seriousness of depriving the patient of
his or her liberty. Good clinical practice is to hospitalize involuntarily only
if the risk of serious harm is imminent. Yet a question arises for which
scales or algorithms provide no guidance: how to define imminent risk?
According to the Oxford English Dictionary (1991), imminent means
“impending, threateningly, overhanging (almost always of evil in-
tent)…close at hand…coming on shortly.” A useful operational defini-
tion of imminent is “within 24 hours.” In other words, if in best clinical
judgment this patient is likely to act violently within the next 24 hours,
then there is a basis to hospitalize involuntarily. If the estimated time
frame is longer or more vague—there is a risk of violence, but not to-
morrow, maybe the day after tomorrow or next week—then it is wrong
to hospitalize involuntarily because there are treatment choices that do
not involve depriving this person of his or her liberty.
When violence is a concern but judged not to be imminent, then it is
possible, for example, to start the patient on medicine and have him or
her come back tomorrow or start day treatment. Where this gets tricky is
when the patient refuses to come back tomorrow. Then, if the clinician is
worried about potential violence the day after tomorrow but thinks the pa-
tient will not come back to be reevaluated tomorrow, this gets into a gray
area in which the clinician will not be able to evaluate imminence going
forward. If violence is judged not imminent today, but the clinician’s best
estimate is that she or he will not be able to reevaluate imminence tomor-
row, then involuntary hospitalization today may be called for.
Preventive Action to Reduce Risk

When the clinician knows that the patient has threatened someone else,
and the clinical assessment is that these threats are not just talk or fan-
tasy but behavior for which the patient is at risk, then the clinician has
a responsibility to do whatever is possible to prevent the future vio-
lence. The clinician’s first responsibility is to make sure the patient un-
derstands the limits on confidentiality—that the clinician is obligated to
do whatever is judged to be necessary to prevent threatened violence.
Preventive action can include breaching the patient’s confidentiality to

warn the victim or the authorities, and the patient needs to understand
this as a condition of his or her further discussions with the clinician.
This does not mean that breach of confidentiality is always called for;
often it is not, but the patient needs to know the ground rules.
If the clinician judges the threatened violence not to be imminent,
there are several options. As always, the basic principle is to treat the
threatened violence primarily as a clinical problem, and only second-
arily as a legal issue. Just as the clinician would not stand by when judg-
ing a patient to be a risk to self, so, equally, the clinician must try to pre-
vent violence when a patient is judged to be a risk to others. The first effort
should be to involve the patient in a discussion of these impulses and of
possible ways to avoid violence. It is critical to assess the emotional
charge associated with the thoughts, beliefs, fantasies, or threatened be-
havior that has raised concern about future violence. If risk assessment
were only a matter of asking, “Are you thinking of hurting anyone?” then
interviewers with minimal training could be hired to screen for this.
When patients perceive that the clinician is interested in their welfare
and not just in protecting the potential victim, it is often possible to use
the alliance to reduce risk. If a man is threatening his significant other, he
may agree that the clinician should let the threatened partner know how
angry he is and what he is thinking about doing. If the potential victim
and the patient will agree, perhaps both can meet with the clinician. In
the context of threatened violence, this sounds like an extreme interven-
tion. In more usual circumstances, it is called “couples therapy.”
This brief discussion illustrates that there are no bright-line rules for
deciding what to do when potential violence is an issue. Potentially vi-
olent patients present difficult clinical problems. In the end, it is the cli-
nician’s best judgment based on careful clinical assessment that the pro-
fession must rely on. It is tempting for clinicians, when in doubt about
imminence, to hospitalize the patient, even when they are in serious
doubt, because this is the “conservative” thing to do. It is a safer choice
for the clinician in regard to the threat of being sued. However, the cost
of this conservative approach is the deprivation of another’s liberty. If
successful, this chapter will help clinicians assess these risks more com-
fortably so that they are less often in the kind of doubt that leads to un-
necessary hospitalization.
Legal Aspects of Potential Violence

Since the Tarasoff decision, clinicians have been concerned about the
risks of being sued for future harm done by a patient who is currently
in treatment. There is an extensive literature on this topic that is not re-

viewed here, but see, for example, Herbert (2002) and Walcott et al.
(2001). Here, it is enough to comment that when the clinician focuses on
the clinical issues and does a careful risk assessment, the likelihood of
being found negligent for a patient’s future violent acts is very small.
There is a standard for risk assessment. There is no standard for risk
prediction. Therefore, if the clinician has done a careful risk assessment
along the lines suggested in this chapter, then there is no basis to find
the clinician negligent. If violence occurs, it is an unfortunate outcome
for which the clinician cannot be held legally responsible.
Currently, there are laws on the books in 29 states that sharply limit
the basis on which a clinician can be sued for harm done by a patient.
These laws are all broadly similar. They state that the clinician’s duty to
protect is limited to cases in which the patient has made a threat to the
therapist (or, in California, to a credible third party who informs the
therapist) or in which there is other reason, based on the patient’s his-
tory of serious violence, to believe the patient constitutes a threat. The
clinician then discharges the duty either by hospitalizing the patient or
warning the victim or appropriate authorities. An equally important
provision is that if the clinician’s communication to third parties has
been made in good faith, then the clinician is immune from suit for
breach of confidentiality. In the past 10 years, almost no clinician has
been found liable for a breach of the duty to protect.
This is not to say that there are no legal risks. In America, anyone can
sue anyone else for almost anything, and there are underemployed law-
yers who will take bad cases. Being a defendant is always a difficult ex-
perience, and there is no guarantee that good clinical work will protect
you from all eventualities. Good clinical work will, however, keep you
safe from all reasonable risks associated with the assessment of poten-
tially violent patients. It is true that that there are risks, however small,
associated with potential violence that are not present in other clinical
work. These come with the territory, and a clinician who is not at least
tolerably comfortable with them should think about working in situa-
tions where patients who present such risks are rare.
Threats to Patients
The focus of this chapter is on assessing risk of possible violence that the
patient may commit in the future, and how to assess that. Much of the
relevant data is elicited from the patient in the clinical interview. This
section addresses risk assessment when the patient is worried about
being a victim of violence, for example if she is being stalked. For an
excellent discussion of threats as risk factors for violence related to

stalking, see McEwan et al. (2007) and Mullen et al. (2006).
In a recent lecture, Paul Mullen, a distinguished Australian professor
of forensic psychiatry, commented on a relevant finding. His group asked
persons who had been threatened with violence whether the threats had
seriously frightened them. Only 3% of respondents said that they had
been seriously frightened. Clinicians should take this seriously in assess-
ing risk. It is true that some people have a much lower threshold for being
frightened than others; but, just as assessing the affect of the potentially
violent person is important, this result points to the importance of assess-
ing the affect of the threatened victim. One should try to make allowances
for potential victims who appear to be somewhat dramatic or hysterical,
but in general it is prudent to take more seriously a threat that the poten-
tial victim takes more seriously. As with assessing patients as possible ac-
tors, so in assessing patients as possible victims it is critical to assess the
emotional charge associated with the thoughts, beliefs, fantasies, or
threatened behavior that has raised the patient’s concern.
Risk of Violence to the Clinician

Very occasionally the therapist may be concerned about being assaulted
by the patient. In that context, physical safety for the therapist is an es-
sential consideration. Location and time of day are the critical issues to
consider related to the therapist’s potential isolation.
Outpatient locations vary widely in the extent to which they provide
physical safety. At one potentially dangerous extreme, the therapist in
individual practice may have a private office in an isolated building—
for example, a home office or an office in a small building with few or
absent tenants. Not much better, from a safety standpoint, are single of-
fices in large buildings where there are no organizational connections
between the many offices and their occupants. Safer are therapist offices
within an institutional structure—outpatient clinics or offices physi-
cally related to inpatient services.
An office is only as safe as the nearest available help. Any office is
unsafe at times of day when no one else is around. Similarly, any situa-
tion is unsafe if the clinician fails to use good sense. Not to belabor the
obvious, but where the clinician and the patient sit can be important. A
young resident once was working in an emergency department inter-
viewing a rather agitated patient. The resident was sitting on a low stool
in front of the patient, who was in a chair. It suddenly occurred to this
resident that the patient could easily kick him in the face. At that mo-
ment he moved to a more prudently placed chair.
With a paranoid patient the question of who sits nearer the door
may be important. If the clinician is not worried about violence toward
him- or herself, it may be best to let the patient have a clear path to the
door, so that the patient can leave without going through the clinician if
he or she becomes frightened. On the other hand, if the clinician is con-
cerned about violence, there may be greater risk to placing the patient
between the door and the clinician.
When in doubt about your own physical safety, ask the patient. That
is, raise a concern about the patient’s current volatility and your own
safety in the patient’s presence and try to negotiate an agreement on
how to further conduct the interview. This could mean leaving the door
open. In extreme cases it could mean arranging to meet in an emergency
department or other hospital setting. The point is that the clinician
should never meet with a patient in a situation in which the clinician
feels unsafe.
This last point leads to another critical assessment—namely, the cli-
nician’s assessment of his or her own comfort in dealing with potential
violence. Clinicians vary widely in how tolerant they are of dealing
with patients with histories of serious violence and in how comfortable
they are working in settings such as emergency services or prisons in
which potentially violent patients are a significant part of the caseload.
It is important to know one’s own level of comfort or discomfort with
the issue of violence and to try to arrange a professional life in which
the expected level of assessment of potential violence is within a zone
of comfort.
Again, no physical location is safe if no one else is present. It is es-
sential when seeing patients identified as potentially violent that other
people are within earshot and that they are aware of the issue. If the cli-
nician is worried, he or she can arrange to keep the office door partly
open if this is the best that can be done in a particular setting. Clinics can
install “panic buttons” that the clinician can use to alert staff to difficul-
ties. A clinician once interviewed a patient in a prison setting in which
the guards placed a small tower-shaped buzzer on the desk. “If this gets
knocked down,” they said, “40 people will come running in.” In emer-
gency services, weapons checks or metal detectors may be appropriate.
The recent tragic death of a senior psychiatrist—killed by a patient
whom he saw in a physically isolated office on a Sunday morning—
illustrates the potential dangers of seeing patients in isolation. It is not
good for patients to be violent, and it is certainly not good for therapists
to be victims. Commitment to patients is good, but professional services
should be provided only in a context that keeps the clinician safe.
Axis I Diagnosis and Violence

Clinicians need to know the evidence relating violence to particular
mental disorders as part of their own knowledge base when conducting
risk assessments. However, they must also be able to discuss this issue
knowledgably with members of the public. Here it is important to know
the evidence, but it is also important to know what the public believes.
Several studies have documented that the general public believes men-
tal disorder and violence are associated (Wahl 2003). This belief is sup-
ported by portrayals of mentally ill people on television (Diefenbach
2007) and also by the media attention to random killings perpetrated
by mentally ill persons. Our understanding of the relationship between
mental disorders and violence has evolved significantly over the past
15 years, and for that reason it is summarized here.
Swanson et al.’s 1990 paper providing self-report data on violence in
relation to mental disorder marks the beginning of the current under-
standing of the relationship between violence and mental disorder. Rely-
ing on interview data in a community sample of more than 10,000 peo-
ple, they found that 2% of respondents without evidence of mental
disorder reported having been violent in the past year. In comparison, vi-
olence was reported by 10%–12% of people with schizophrenia or bipo-
lar disorder; 25% of people who abused alcohol; and 33% of people with
substance abuse. The data also showed a steady increase in self-reported
violence for people with multiple diagnoses (Swanson et al. 1990).
That paper has been followed by epidemiological studies on four
continents (Arseneault et al. 2000; Brennan et al. 2000; Corrigan and
Watson 2005; Stueve and Link 1998; Wessely et al. 1994) and by case reg-
ister (Wessely et al. 1994) and other studies of clinical samples, almost
all of which have produced consistent findings: people with schizophre-
nia are more likely to be violent or seriously violent than comparison
samples. Comorbidity of substance abuse with schizophrenia increases
these risks. Similar risks of violence apply for persons with bipolar dis-
order (Buchanan et al. 1993; Commander et al. 2005).
More recent work suggests that much of the risk of violence associ-
ated with schizophrenia is actually related to the presence of comorbid
antisocial personality or conduct disorder, both of which are more com-
mon in association with schizophrenia than in the general population
(Bland et al. 1987; Swanson et al. 1990). For example, Mueser et al.
(2006) studied patients with schizophrenia or bipolar disorder and co-
morbid substance abuse. Of those who also met criteria for comorbid
antisocial personality disorder, more than two-thirds had a criminal his-
tory of one or more violent offenses. Similar but less dramatic results
held for patients who had a comorbid conduct disorder.
Hodgins et al. (1999) also reported that antisocial personality was as-
sociated with violence among patients with schizophrenia, but in contrast
to Mueser et al. (2006), Hodgins’s group did not find that antisocial per-
sonality was associated with violence among patients with bipolar disor-
der. For the bipolar patients, substance abuse was associated with vio-
lence, and compliance with treatment decreased the risk of violence.
Recent data establish quite clearly that the violence associated with
schizophrenia is strongly related to the presence of comorbid conduct
disorder or antisocial personality disorder (Hodgins et al. 2005). In a
sample of 248 men with schizophrenia or schizoaffective disorder, ap-
proximately 20% had comorbid conduct disorder, and almost all men
with conduct disorder also met criteria for antisocial personality disor-
der. “In childhood and adolescence, conduct disorder was associated
with poor academic performance, physical abuse, substance misuse, in-
stitutionalization, and being raised in a family characterized by crimi-
nality and substance misuse” (Hodgins et al. 2005, p. 323). Conduct dis-
order was associated with an increased risk of violent crime such that
for each additional symptom of conduct disorder, the risk of violent
crime increased by 1.2. This was true after controlling for diagnoses of
drug or alcohol misuse. Notably, there was no association between vio-
lence and positive or negative symptom level in this sample, and cur-
rent substance abuse was equally common in conduct disorder patients
and others. The clinical implication of these findings is clear: the clini-
cian should routinely assess for a history of symptoms of conduct dis-
order or antisocial personality.
A body of work exists showing that delusions of threat (paranoia) or
control-override (delusions of thought insertion, or that one’s behavior
is controlled by outside forces) are associated with violence (Hodgins et
al. 2003; Link et al. 1992, 1998). These symptoms are associated with vi-
olence after controlling for gender, age, ethnicity, antisocial personality
disorder, and years of education. This association is another example of
the general principle that violence is associated with anger. It seems
likely that most people would be angry if they thought someone was
trying to harm them or was inserting thoughts into their mind.
Inquiring about the emotional charge associated with delusions is
critical (Buchanan et al. 1993). Buchanan et al. (1993) studied delusional
patients with the aim of better understanding why patients acted on de-
lusions. They found that patients were more likely to act on delusions
when feeling sad, anxious, or frightened as a result of the delusion. The
authors noted that Bleuler, in 1924, had already found affect to be an im-
portant determinant of delusionally driven action.
In a multisite/multinational study of 128 men with schizophrenia
discharged from the hospital and followed up for 2 years, Hodgins et al.
(2003, 2005) found that aggressive behavior was associated with severe
positive symptoms and in particular with threat or control-override
symptoms. This relationship held true when psychopathy and sub-
stance abuse were held constant. Depot medication did not reduce the
risk of violence for these patients. Again, the clinical implication is
clear: inquire about threat or control-override symptoms with care in
patients who show evidence of a psychotic disorder.
Fava and Rosenbaum (1999; Fava et al. 1993) studied anger attacks
as a symptom of depressive disorders. More than one-third of depressed
outpatients reported anger attacks, and 30% of those said that they
threw things or destroyed property. Personality disorders were more
common in patients with anger attacks than in those without. A small
study of women with psychosis in the community (Dean et al. 2006)
found more Cluster B personality disorders (impulsive, dissocial, histri-
onic, and borderline) in violent compared with nonviolent patients.
These data show that clinicians should not imagine that patients are too
depressed to be violent. They reinforce also the importance of assessing
personality characteristics as risk factors for violence.
What to conclude from all this? First, that the risk of violence asso-
ciated with schizophrenia in the absence of antisocial personality or
conduct disorder or substance abuse is probably not much higher than
the risk for persons without mental disorder—an informed guess is that
the risk is perhaps doubled. Risk for future violence is greater when, in
addition to schizophrenia, there is evidence for antisocial personality
disorder, conduct disorder, substance abuse, symptoms of threat or con-
trol override, past history of victimization, or exposure to violence in
the current environment. These are variables that are critical to evaluate
when assessing any patient for risk of future violence.
Personality Traits and Risk of Violence

Some personality traits may increase the risk of violence, regardless of
whether the person meets criteria for a personality disorder. Narcissism
and impulsivity are both of concern, because narcissistic people can feel
entitled to do whatever they want in order to get their way, and impul-
sive people may not have the usual controls on violent behavior when
they are angry. Conversely, traits that militate against violence include
capacity for empathy, and strongly held religious beliefs.
Gender and Risk of Violence

In the community at large, men are more violent than women, but this
relationship is significantly attenuated in patient samples. Women pa-
tients are more likely to be violent than women in general. This is clini-
cally important. The clinician cannot assume that just because a patient
is a woman she is unlikely to have been violent. Many people come to
evaluation or treatment in part because they have been violent. Thus it
is important for the clinician not to assume that risk assessment has to
do mostly with men. For example, Stueve and Link (1998) examined the
relationship between gender and violence in an Israeli sample of 2,700
persons with mental disorder. In the sample as a whole, men were four
times as likely as women to report fighting in the past 5 years (10% vs.
2.5%), but for people who had recently been psychiatric patients, men
were only twice as likely as women to report violence.
Environmental Variables
Swanson et al. (2002) reported on the relationship between environ-
mental variables and violence in a study of 802 inpatients and outpa-
tients with psychotic or major mood disorders. They found that 1-year
rates of serious violence were related to three variables: 1) substance
abuse, 2) past history of victimization, and 3) exposure to violence in the
current environment. When all three were present, 30% of patients re-
ported serious violence; when none were present, almost no patients re-
ported serious violence. When one of these factors was present, 2%
reported violence, and when two were present, 8%–10%.
Treatment Variables
As a general principle, there is no specific treatment for aggression. One
exception to this may be the treatment of impulsive aggression. Barratt
et al. (1997) showed that in a prison sample, phenytoin decreased im-
pulsive aggression but not planned aggression. Medications that reduce
affect dysregulation may reduce the likelihood of aggression. Examples
of these medications include oxcarbazepine, gabapentin, lamotrigine,
and other medications initially developed to treat seizure disorders.
Although there is no treatment for violence per se, there is treatment
for mental disorders, and to the extent that aggression is associated
with active symptoms of mental disorder, successful treatment of the
mental disorder may reduce the risk of violence. Clozapine has been
shown to reduce the risk of violent behavior among inpatients, when
compared with olanzapine and haloperidol (Krakowski et al. 2006).

However, in an earlier study of atypical versus typical antipsychotic
medications, Swanson et al. (2004) found that clozapine, olanzapine,
and risperidone all reduced the rate of violent behavior in 229 outpa-
tients with schizophrenia spectrum disorders, when compared with
treatment with typical antipsychotics. However, the study was not suf-
ficiently powered to determine whether one atypical agent was more ef-
fective than another.
Violence in the community among psychotic patients was strongly re-
lated to perceived need for treatment in a study of 1,011 outpatients with
major mental disorder in treatment in public sites in five states (Elbogen
et al. 2006). Patients were asked if they needed treatment, whether treat-
ment helped, and whether they had been compliant. Of those who an-
swered yes to all three, only 8% reported having behaved violently; of
those who answered no to all three, 40% reported violence. Interestingly,
7.5% of men and 5% of women reported serious violence, but more
women (17%) than men (11%) reported non-serious violence.
In a sample of recidivist inpatients, those randomly assigned to
involuntary outpatient treatment who received at least 6 months of
involuntary treatment were less likely to be violent in the year after dis-
charge than patients in treatment as usual (Swanson et al. 2000). Vio-
lence was much less likely for patients who were compliant with
regular outpatient treatment and who did not abuse substances.
In the study of anger attacks in depressed outpatients, these attacks
diminished or disappeared in more than half of patients treated with se-
lective serotonin reuptake inhibitors (Fava and Rosenbaum 1999; Fava
et al. 1993).
Key Points
■ Risk assessment is first and foremost a process of clinical assess-
ment. Applying an algorithm or using a rating scale may be a use-
ful part of the process, but in the end it is the clinician’s assess-
ment of the unique facts that characterize this person in this
situation that will determine the recommended course of action.
■ The clinician should always, always, always attend to his or her
own safety. Never work in isolation with patients about whom
there are any safety concerns.
■ Assessing the person is important, but it is not the whole story—
behavior (B) is a function of the person (P) and the environment (E).
■ A past history of violence is a strong predictor of future violence,

especially if there has been no change in either P or E since the
violent B.
■ Current mental status—anger, impulsiveness, delusions held with
strong affect—is a more important determinant of near-term
behavior than enduring characteristics of the person such as age,
gender, or diagnosis.
■ Antisocial personality disorder and conduct disorder, whether
alone or as comorbid conditions, are associated with violence.
■ Treat violence as a clinical issue primarily. Clinical assessment, not
legal concerns, should drive the decisional process.
References
Arseneault L, Moffitt TE, Caspi A, et al: Mental disorders and violence in a total
birth cohort: results from the Dunedin Study. Arch Gen Psychiatry 57:979–
986, 2000
Barratt ES, Stanford MS, Felthous AR, et al: The effects of phenytoin on impul-
sive and premeditated aggression: a controlled study. J Clin Psychophar-
macol 17:341–349, 1997
Beck JC: When the patient threatens violence: an empirical study of clinical
practice after Tarasoff. Bull Am Acad Psychiatry Law 10:199–201, 1981
Bland RC, Newman SC, Orn H: Schizophrenia: lifetime co-morbidity in a com-
munity sample. Acta Psychiatr Scand 75:383–391, 1987
Brennan PA, Mednick SA, Hodgins S: Major mental disorders and criminal vi-
olence in a Danish birth cohort. Arch Gen Psychiatry 57:494–500, 2000
Buchanan A, Reed A, Wessely S, et al: Acting on delusions, II: the phenomeno-
logical correlates of acting on delusions. Br J Psychiatry 16:377–381, 1993
Commander M, Sashidharan S, Rana T, et al: North Birmingham assertive out-
reach evaluation: patient characteristics and clinical outcomes. Soc Psychi-
atry Psychiatr Epidemiol 40:988–993, 2005
Corrigan PW, Watson AC: Findings from the National Comorbidity Survey on
the frequency of violent behavior in individuals with psychiatric disorders.
Psychiatry Res 136:153–162, 2005
Dean K, Walsh E, Moran P, et al: Violence in women with psychosis in the com-
munity: prospective study. Br J Psychiatry 188:264–270, 2006
Diefenbach DL: The portrayal of mental illness on prime time television. J Com-
munity Psychol 25:289–302, 2007
Elbogen EB, Van Dorn RA, Swanson JW, et al: Treatment engagement and vio-
lence risk in mental disorders. Br J Psychiatry 189:354–360, 2006
Fava M, Rosenbaum JF: Anger attacks in patients with depression. J Clin Psy-
chiatry 60(suppl)15:21–24, 1999
Fava M, Rosenbaum JF, Pava JA, et al: Anger attacks in unipolar depression,
part 1: clinical correlates and response to fluoxetine treatment. Am J Psy-
chiatry 150:1158–1163, 1993
Hall G, Gardner L: Theories of Personality. New York, John Wiley, 1978
Herbert PB: The duty to warn: a reconsideration and critique. J Am Acad Psy-
chiatry Law 30:417–424, 2002
Hodgins S, Hiscoke UL, Freese R: The antecedents of aggressive behavior
among men with schizophrenia: a prospective investigation of patients in
community treatment. Behav Sci Law 21:523–546, 2003
Hodgins S, Lapalme M, Toupin J: Criminal activities and substance use of pa-
tients with major affective disorders and schizophrenia: a 2-year follow-up.
J Affect Disord 55:187–202, 1999
Hodgins S, Tiihonen J, Ross D: The consequences of conduct disorder for males
who develop schizophrenia: associations with criminality, aggressive be-
havior, substance use, and psychiatric services. Schizophr Res 78:323–335,
2005
Krakowski MI, Czobor P, Citrome L, et al: Atypical antipsychotic agents in the
Link BG, Andrews H, Cullen FT: The violent and illegal behavior of mental pa-
tients reconsidered. Am Sociol Rev 57:275–292, 1992
Link BG, Stueve A, Phelan J: Psychotic symptoms and violent behaviors: prob-
ing the components of “threat/control-override” symptoms. Soc Psychia-
try Psychiatr Epidemiol 33(suppl):S55–S60, 1998
McEwan T, Mullen PE, Purcell R: Identifying risk factors in stalking: a review of
current research. Int J Law Psychiatry 30:1–9, 2007
Monahan J, Steadman HJ: Violence and Mental Disorder: Developments in Risk
Assessment. Chicago, IL, University of Chicago, 1994
Mueser KT, Crocker AG, Frisman LB, et al: Conduct disorder and antisocial per-
sonality disorder in persons with severe psychiatric and substance abuse
disorders. Schizophr Bull 12:626–636, 2006
Mullen PE, Mackenzie R, Ogloff JR, et al: Assessing and managing the risks in
the stalking situation. J Am Acad Psychiatry Law 34:439–450, 2006
Oxford English Dictionary. Oxford, UK, Oxford University Press, 1991
Stueve A, Link BG: Gender differences in the relationship between mental ill-
ness and violence: evidence from a community-based epidemiological
study in Israel. Soc Psychiatry Psychiatr Epidemiol 33(suppl):S61–S67,
1998
Swanson JW, Swartz MS, Borum R, et al: Involuntary out-patient commitment
and reduction of violent behaviour in persons with severe mental illness.
Br J Psychiatry 176:324–331, 2000
Swanson JW, Swartz MS, Essock SM, et al: The social-environmental context of
violent behavior in persons treated for severe mental illness. Am J Public
Health 92:1523–1531, 2002
Swanson JW, Swartz MS, Elbogen EB: Effectiveness of atypical antipsychotic
medications in reducing violent behavior among persons with schizophre-
nia in community-based treatment. Schizophr Bull 30:3–20, 2004
Tarasoff v. Regents of the University of California. 17 Cal 3d. 425, 551 P 2d. 334 (1976)
Wahl OF: News media portrayal of mental illness. Am Behav Sci 46:1594–1600,
2003
Walcott DM, Cerundolo P, Beck JC: Current analysis of the Tarasoff duty: an evo-
lution towards the limitation of the duty to protect. Behav Sci Law 19:325–
343, 2001
Webster CD: HCR-20: Assessing Risk for Violence. Vancouver, BC, Canada, Si-
mon Fraser University, 1997
Wessely SC, Castle D, Douglas AJ, et al: The criminal careers of incident cases
of schizophrenia. Psychol Med 24:483–502, 1994
C H A P T E R 1 3
Inpatient Settings
Inpatient violence is widely recognized as a serious problem in need of

evidence-based solutions. Among mental health professionals working
in outpatient settings, psychiatrists are the most likely to be assaulted.
In inpatient settings, however, the vast majority of assaults target the
nursing staff working most closely with psychiatric patients (Tardiff
1995). Public psychiatric nursing is a hazardous occupation; each year,
one in four nurses suffers a disabling injury from a patient assault (Love
and Hunter 1996). Assaulted staff experience emotional distress, as ev-
idenced by high rates of substance misuse, anxiety disorders, poor mo-
rale, and job burnout (Quintal 2002). Aggressive psychiatric inpatients
are adversely affected as well. Seclusion and restraints are frequently
used to manage violent behavior, even though their use poses physical
risks to patients and can be psychologically damaging (Frueh et al.
2005; Kaltiala-Heino et al. 2003). Past research in a variety of inpatient
settings has consistently shown that a small percentage of patients are
responsible for the majority of assaults (Kraus and Sheitman 2004); this
subset of repetitively assaultive patients is 10 times more likely than
other patients to inflict serious injuries (Convit et al. 1990).
Characteristics of Assaultive Inpatients

Past behavior is the best predictor of future behavior. A history of as-
saultive behavior or violent crime is the most robust long-term predictor
259
of inpatient violence (Steinert 2002). Research has shown that men

are more violent than women in the community (Swanson et al. 1990);
however, this gender difference is not observed in a psychiatric inpa-
tient setting. Women are just as likely as men to assault, and their risk
of violence should not be discounted (Lam et al. 2000). Other diagnos-
tic, historical, developmental, and neurological factors associated with
inpatient violence are shown in Table 13–1.
Short-Term Risk Factors for Inpatient Assault

The characteristics of assaultive inpatients just described help the clini-
cian identify which inpatients are at highest risk for assault. Once such
patients are identified, it becomes critically important to be cognizant of
the symptoms and behaviors that indicate an inpatient is at increased
risk for assault so that preventative measures can be taken. Although
psychiatric clinicians have been criticized for the inability to predict
long-term violent recidivism, the ability of these clinicians to correctly
identify which inpatients will be aggressive in the short term (i.e.,
hours, days) is somewhat accurate (Nijman et al. 2002). Clinical, rather
than sociodemographic, risk factors have been shown to best predict
aggression in the short term (McNiel et al. 2003).
A number of studies have shown the following symptoms and be-
haviors to be short-term risk factors for inpatient aggression: 1) recent
physical violence and threats of violence (McNiel and Binder 1989; Mc-
Niel et al. 1988); 2) poor therapeutic alliance (Beauford et al. 1997), for
instance, failure to cooperate with an initial assessment (Swett and Mills
1997); 3) a hostile attitude and irritable mood (Linaker and Busch-
Iversen 1995; McNiel and Binder 1994); 4) psychomotor agitation (Lanza
et al. 1996; Whittington and Patterson 1996); and 5) attacks on objects or
property damage. In patients with a psychotic disorder, severe positive
symptoms and thought disorder are short-term violence risk factors
(Hoptman et al. 1999; Krakowski et al. 1999; Nolan et al. 2005).
Environmental Risk Factors

Certain situational or environmental factors have been associated with
inpatient assaults. Assaults are more likely to occur during the week,
when activity demands are higher (Smith et al. 2005). They occur more
frequently during times of transition and increased staff–patient inter-
action, such as meal times, changes of shift, and medication administra-
tion (Carmel and Hunter 1989). The early afternoon is a peak time for
assault (Manfredini et al. 2001). Assaults occur most often in crowded,
Inpatient Settings ❘ 261
TABLE 13–1. Long-term risk factors for inpatient aggression

Axis I diagnosis (Binder and McNiel 1988; Hoptman et al. 1999; Lehmann et
al. 1999; Miller et al. 1993; Tardiff and Sweillam 1982)
Schizophrenia
Mania
Substance misuse disorder
Dementia and other organic mental disorders
Axis II diagnosis (Hill et al. 1996; Miller et al. 1993; Soliman and Reza 2001;
Tardiff and Sweillam 1982)
Antisocial personality disorder and psychopathy
Borderline personality disorder
Mental retardation
Past suicide attempts (using violent methods) (Convit et al. 1988; Soliman and
Reza 2001)
Developmental factors (Convit et al. 1988; Hoptman et al. 1999)
Parental substance misuse or psychiatric illness
Childhood physical abuse
Placement into foster care
School truancy
Neurological abnormalities (Barratt et al. 1997; Krakowski and Czobor 1997;
Krakowski et al. 1989)
Abnormal P300 wave amplitude
Impairment in integrative sensory and motor functions (in those with
schizophrenia)
Frontal lobe deficits
high-traffic areas such as hallways, dayrooms, bedrooms, and in front

of the nursing station (Chou et al. 2002).
Staffing ratios and ward atmosphere are also linked to rates of as-
sault. There is an inverse relationship between the number of nursing
staff and patient assaults (Lanza et al. 1994). Different wards or units
have different rates of violence. Higher rates of violence have been ob-
served on units where staff functions are not clearly defined and the
schedule of activities is unpredictable; conversely, units with strong
psychiatric leadership, clearly defined roles, and a predictable, repeti-
tive routine have lower rates of violence (Katz and Kirkland 1990).
Motivations for Assault

Until recently, aggressive inpatients have been viewed as a homoge-
neous group. Recent research, however, indicates that it is important to
differentiate between acts of aggression and to categorize assaults

based on their various motivating factors. Two primary subtypes of ag-
gression have been identified in both animals and humans (Weinshen-
ker and Siegel 2002): 1) an uncontrolled outburst of aggressive behavior
driven by intense emotion (varyingly termed impulsive, affective, reac-
tive, or overt) or 2) a controlled, purposeful, and planned act of aggres-
sion (organized, predatory, premeditated, psychopathic, or covert).
Impulsive aggression refers to spontaneous, unplanned aggressive acts
that are externally provoked; feelings of remorse and confusion often
follow the violent act. Persons who are impulsively aggressive often are
described as “hot-blooded” or are said to have “anger control prob-
lems.” In contrast, premeditated aggressive acts are not usually consid-
ered to have a large emotional component but are more “cold-blooded”
in nature; the aggression is goal directed and requires a degree of fore-
thought or planning. This type of aggression has been linked with indi-
viduals described as more antisocial and/or psychopathic (Woodworth
and Porter 2002).
A study conducted in a New York psychiatric state hospital exam-
ined inpatient assaults to determine whether these two types of aggres-
sion were adequate in describing inpatient violence (Nolan et al. 2003).
Assaults occurring in a common area of the inpatient unit were video-
taped so that motivations and triggers for the aggression could be ex-
amined. After the assault, the assailant and victim were interviewed
separately in an attempt to identify the underlying reason for the ag-
gressive act. Three primary factors motivating assaults were identified:
1) disordered impulse control, 2) psychopathy, and 3) symptoms of psy-
chosis. Psychotic assaults were generally committed by an individual
acting under the influence of delusions, hallucinations, or disordered
thinking.
A recent study attempted to determine the types of aggression oc-
curring at a large psychiatric state hospital (Quanbeck et al. 2007).
Nearly a thousand assaults committed by a large sample of randomly
selected chronically aggressive inpatients were categorized as impul-
sive, organized, or psychotic in motivation. Impulsive assaults were the
most common (54%) and the most likely to target staff. Assaults on staff
were precipitated by the following interpersonal interactions: 1) at-
tempting to change a patient’s behavior, such as by enforcing unit rules,
and 2) refusing a patient’s request. Organized or planned assaults (29%)
and psychotic assaults (17%) were less common but were more likely to
target other patients in the facility. Acts of organized aggression were
frequently motivated by the patient’s desire to retaliate against or “get
even with” another patient or staff member. Psychotic aggression was
usually committed by patients acting under the paranoid ideation that

the victim intended to harm (e.g., by poisoning), was stealing from, or
was talking about/laughing at the assailant. This research suggests that
the affective/reactive and predatory/planned subtyping of aggressive
behavior is also valid when examining institutional violence. However,
for a more accurate characterization, a third type should be included:
psychotically motivated aggression. It is critical to characterize aggres-
sion exhibited on inpatient units so that appropriate and effective inter-
ventions can be developed. The different characteristics of the three
types of aggression are summarized in Table 13–2.
The following three case examples illustrate these various forms of
aggression. Each case is followed by a discussion of appropriate clinical
interventions.
Case Example 1
Ms. Smith is a 23-year-old woman with borderline personality disorder.
She was brought by police into an acute psychiatric hospital after cut-
ting her wrists in an apparent suicide attempt. Ms. Smith had become
distraught after discovering her boyfriend was involved with another
woman. On the unit, she exhibits agitated behavior. She is seen pacing
rapidly though the hallways of the unit, her emotions ranging from in-
tense anger to hysterical crying. She exclaims repeatedly, “How could
he do this to me?!” Suddenly, she approaches the nursing station and
demands to use the phone so she can call her boyfriend. The nurse on
duty sternly responds, “No! Calling him right now is not a good idea,
you just need to go to your room right now and calm down!” Ms. Smith,
not pleased with the nurse’s response, becomes more hostile and yells,
“Give me your phone! I need to call him right now! I’m gonna kill you,
bitch!” The nurse, attempting to control the situation, stands up and
barks out an order: “Go to your room immediately!” Ms. Smith then
jumps over the desk and begins to strike the nurse repeatedly in the
head and face.
This is an example of an impulsive assault, the most common type

of aggression in an inpatient setting and the most likely to target staff
(Quanbeck et al. 2007). This assault occurred immediately after the
nurse refused a patient request; the vast majority of assaults on staff are
preceded by an aversive interpersonal stimulus to the patient (Whit-
tington and Wykes 1996). Put simply, patients lash out in frustration af-
ter being told to do something they do not want to do or being told they
cannot have something they want. Because the assault is precipitated
by an interpersonal situation, the staff–patient interaction that preceded
the assault warrants close examination. Nurses with an authoritarian
attitude, a tendency to externalize blame, high levels of anxiety, less
264
TABLE 13–2. Characteristics of impulsive, organized, and psychotic assaults
Assault type
❘
Impulsive Organized Psychotic
Precipitant Spontaneous reaction to Opportunity arises to use Paranoia peaks in intensity and
interpersonal provocation violence to advantage assailant feels compelled to act
Behaviors preceding assault Psychomotor agitation, hostile Distinct lack of emotional Paranoid delusions accompanied

and irritable mood, verbal display; controlled; assault by anger and fear; command
abuse and threats occurs with little warning auditory hallucinations
Motivation for assault Threat reduction A desire to assert social A desire to protect oneself against
advantage or obtain items of a perceived persecutor
value
Insight regarding assaultive Remorse may follow act; Remorse lacking; little concern Insight is typically poor, but
behavior recognizes that control for victim; may deny or fluctuates with level of psychotic
was lost justify violence symptoms
experience, and less formal training are the most likely to be assaulted
(Flannery et al. 2001; Ray and Subich 1998). Furthermore, when asked
about the underlying reasons for staff assaults, patients cite as precipi-
tants poor communication with staff or the perception that staff are too
controlling (Duxbury 2002).
Research has shown that if psychiatric clinicians learn the clinical
skills needed to “de-escalate” emotion in patients who are agitated,
staff assaults can be reduced (Forster et al. 1999). A key component of
this technique is the ability to recognize early signs that indicate a pa-
tient is “escalating” in a process that may ultimately result in physical
violence (Maier 1996). The escalating process has been characterized as
follows: 1) tension in minor muscles; 2) verbal abuse, verbal threats, and
hostility; 3) tension in major muscles; 4) physical violence; and, finally,
5) relaxation and exhaustion. Experienced nurses viewed as experts in
de-escalation techniques cite the importance of intervening in the first
two stages of the process in order to avert an attack (Johnson and
Hauser 2001). Effective de-escalation techniques are summarized in
Table 13–3. Interventions clinicians can use to help gain control of a
patient’s dangerous behavior are outlined in Table 13–4. (For a more
detailed discussion of seclusion and restraints, see Chapter 17.) If de-
escalation fails and a staff assault occurs, it is important to engage the
clinicians involved in a debriefing session (Secker et al. 2004). The goals
of this debriefing session should be to 1) determine what triggered the
assault; 2) review the interventions taken and why they failed; and
3) examine what was learned and what can be done differently next
time. Finally, the physical and psychological impact on the involved
staff should be evaluated; a significant decline in rates of assault has
been noted after psychiatric facilities implement a program that pro-
vides psychological support for staff who have been assaulted (Flan-
nery et al. 1998).
Case Example 2
Mr. Jones is a 34-year-old man with schizophrenia, paranoid type. He
was transferred from an acute facility to a psychiatric state hospital for
long-term treatment. The first several months of his stay were unevent-
ful. Nursing staff then began to note deterioration in his clinical condi-
tion. Mr. Jones began to complain that other patients were saying nega-
tive things about him behind his back. He began to isolate himself in his
room throughout the day and was often observed talking to himself. He
accused nursing staff of putting poison in his medications and occasion-
ally refused to take his scheduled antipsychotic. While waiting in line
for medications a few weeks later, he became very angry and agitated
and exclaimed, “Stop giving out drugs and syringes to kids; it’s illegal!”
TABLE 13–3. De-escalation techniques to prevent an impulsive

assault: approaching an agitated patient
Notice early signs of agitation and read the situation.
Identify what is upsetting the patient.
Verbally connect with the patient using a calm voice.
Listen to the patient and attempt to understand and empathize with the
patient’s perspective.
“I notice you are upset; what is bothering you?”
“What might help you calm down?”
Agree with the patient if possible.
“Yes, your medication does cause some annoying side effects.”
Avoid taking an authoritarian stance with the patient; getting into an
argument will only fuel the escalating process.
Interventions designed to give the patient a sense of control can be helpful.
Show respect by asking permission to speak with the patient.
Divide energy by giving the patient choices:
“Would you like to move to a different area and talk?”
“Would you like to take some medication [as needed]?”
“Would you like to listen to some music?”
Increase the personal space between yourself and the patient; do not make
an escalating patient feel cornered.
Locate an escape route and summon help if necessary.
Accompany the patient to a calmer space and observe patient for signs of
relaxation.
He then threw his medications at nursing staff. A few days later, he fran-
tically dialed 911 and demanded an ambulance be sent to get him be-
cause his life was in jeopardy. Early the next day, he approached the unit
psychiatrist from behind and struck him in the head while yelling, “You
poured kerosene on me last night!”
This long-term psychiatric inpatient committed an assault while act-

ing under the influence of a paranoid delusion. Persons with schizo-
phrenia are the most likely to act on persecutory delusions (Swanson et
al. 2006). There are multiple other factors associated with violent behav-
ior and delusions. Inpatients are more likely to act on delusions if they
feel frightened, anxious, unhappy, or angry as a consequence of the de-
lusional belief (Appelbaum et al. 1999). A history of acting violently in
response to a delusion increases the future likelihood of acting on delu-
sions (Monahan et al. 2001). Inpatients who are both psychotic and vio-
lent have poorer insight into their psychotic symptoms than those who
TABLE 13–4. Measures of control in inpatient settings

Medications (oral or intramuscular)
One-to-one staff observation
Seclusion (open or closed)
Restraints
are not violent (Arango et al. 1999). Inpatients with treatment-resistant

schizophrenia exhibit a significant increase in the positive symptoms of
psychosis 3 days prior to the occurrence of an aggressive incident (No-
lan et al. 2005).
A patient experiencing an auditory hallucination that issues a com-
mand to harm others is at increased violence risk. Command hallucina-
tions double the risk a patient will be assaultive (McNiel et al. 2000). An
inpatient is more likely to act on command hallucinations if the voice is
familiar to the patient, if there is a delusion associated with the halluci-
nation, and if coping strategies to diminish the hallucination are not
successful (Cheung et al. 1997). Indicators that a patient with psychosis
is at increased risk for assault are summarized in Table 13–5.
Management strategies for preventing psychosis-related aggression
can include separating the inpatient from the perceived persecutor and
the use of cognitive therapies that may reduce the patient’s conviction
that the delusional belief is true (Turkington et al. 2006). The most effec-
tive approach to reducing violence in individuals with a psychotic dis-
order is pharmacological; clozapine is superior to other antipsychotic
agents in reducing aggressive behavior (Volavka et al. 2004).
TABLE 13–5. Psychotic violence: indicators of impending assault

Persecutory or paranoid delusions are closely linked to violence
Delusions causing fear, unhappiness, or anger are more likely
to be acted on
Past aggression based on delusions is predictive of future similar acts
An increase in positive symptoms has been observed in the days
before an assault
Command hallucinations increase the risk of violent behavior
Patients are more likely to behave aggressively if
—The voice is familiar
—A delusion is associated with the hallucination
—Coping strategies are not successful
Case Example 3
Mr. Green is a 28-year-old man who was admitted to a Veterans Affairs
psychiatric hospital after going to the emergency department and re-
porting that voices were telling him to kill himself and others. His urine
toxicology screening was positive for cocaine. His preliminary diagnosis
was psychotic disorder not otherwise specified. On the unit, Mr. Green
frequently came to the nursing station complaining of anxiety and re-
ceived several doses of lorazepam (Ativan) as needed. Several days
later, Mr. Jackson (another patient on the unit) approached the charge
nurse and reported that Mr. Green was threatening to harm patients un-
less they “cheek” their benzodiazepines and give them to him. The
nurse notified the unit psychiatrist, and Mr. Green’s order for lorazepam
was discontinued based on the suspicion that he was abusing it.
Later in the day, Mr. Green approached the nursing station com-
plaining of agitation and requested lorazepam. The nurse on duty in-
formed Mr. Green that his lorazepam had been discontinued by the psy-
chiatrist. When Mr. Green asked why it was stopped, the nurse replied,
“I don’t know, you’ll have to ask the doctor tomorrow during rounds.”
Visibly irritated, Mr. Green walked away and entered his room. He re-
turned to the nursing station 1 hour later and calmly told the nurse,
“You have a really pretty face, I’d hate to see it get all cut up, but you
never know what could happen around here. Now, I want you to work
on getting my Ativan reordered.” Later, on the night shift, nurses
rushed to Mr. Jackson’s room after hearing a lot of commotion and yell-
ing. They found Mr. Jackson (the patient who had reported Mr. Green’s
extortion scheme to nursing staff) with a bloodied and broken nose. Mr.
Jackson immediately exclaimed, “He hit me when I was sleeping!” and
pointed to Mr. Green standing in the hallway. Mr. Green smiled at the
nursing staff and said, “The voices made me do it.”
Predatory violence is dangerous because it usually occurs without

warning and is difficult to predict and prevent (Meloy 1987). Persons
who engage in organized aggression may have a limited capacity to em-
pathize with others. Thus, they are comfortable using violence as a tool
to gain control over others, assert dominance, and obtain desired goals.
In this case, the patient made a “cold threat” against the unit nurse
(Quanbeck et al. 2007). A cold threat is defined as a threat of future vio-
lence delivered in an unemotional, controlled manner and intended to
frighten another into doing what the threatener wants. This type of
threat is seen with increasing frequency in psychiatric facilities and can
distress and demoralize clinicians (Flannery et al. 1995). Cold threats
should be managed differently than the “hot threats” that foreshadow
an impulsive assault (Maier 1996). Because of the personal nature and
gravity of the threat, psychiatric clinicians have a tendency to hide the
threat from colleagues and may isolate themselves from peers. This re-
sponse, however, is exactly what the threatener desires, because it only
serves to increase the level of control the patient holds over the clini-
cian. To deal with these threats, psychiatric units can maintain a “threat
book” in which staff can document threats received that day. At the end
of the shift, staff meet as a group to assess the clinical meaning of the
threat in the context of the patient’s clinical condition. Two staff mem-
bers then later confront the patient who made the threat, attempt to get
him or her to accept responsibility for making the threat, and suggest
more effective ways of getting what the patient wants. Clinicians work-
ing in long-term and forensic settings must develop approaches to this
type of aggression because, unlike in short-term community settings,
immediate discharge is not an option.
A controversial approach to inpatient assaults is criminal prosecu-
tion (Appelbaum and Appelbaum 1991). Filing criminal charges against
violent inpatients is a recent phenomenon, with the first case report ap-
pearing in the literature in 1978 (Schwarz and Greenfield 1978). In the
past, prosecution has not been considered a viable option because of the
prevailing belief that hospitalized psychiatric patients are, by virtue of
their situation, not responsible for their actions (Norko et al. 1991). Fur-
thermore, taking an action that moves a patient out of a therapeutic mi-
lieu and into the punitive atmosphere of a jail or prison creates an ethi-
cal dilemma. Mental health professionals are expected to act in patients’
best interests and respect their autonomy, and physicians have a duty
to “do no harm” (Appelbaum and Appelbaum 1991).
Over recent years, however, the number of inpatients prosecuted for
assaults has increased on the basis of several countervailing ethical
viewpoints (Dinwiddie and Briska 2004):
• The Supreme Court has determined that psychiatric inpatients are

entitled to safe conditions, and statutes based on the Tarasoff case
obligate mental health professionals to protect endangered third
parties (Appelbaum and Appelbaum 1991). Organized or predatory
inpatients preferentially target other patients in the facility (Quan-
beck et al. 2007). An involuntarily confined patient who is being vic-
timized by another patient is unable to escape the situation. Thus,
removing the dangerous patient from the hospital is ethically justi-
fied because it creates a safer environment for all patients.
• Psychiatric institutions also have an interest in providing for the
safety of hospital personnel. Ignoring assaults can lead to poor staff
morale and performance, which interferes with the therapeutic aims
of the hospital (Coyne 2002).
• If patients who engage in acts of aggression that are planned and
designed to meet their own needs go unpunished, the antisocial
behavior may be positively reinforced (Dinwiddie and Briska 2004).

Many argue that a psychiatric patient, even though he or she is liv-
ing in a hospital environment, should be held responsible for his or
her conduct and live by societal rules. Criminal prosecution can set
firm limits on antisocial behaviors and result in positive change
(Coyne 2002; Miller and Maier 1987). Allowing such behavior to
continue unchecked may ultimately harm these patients when they
reenter society with the belief that acts of violence do not result in se-
rious consequences (Dinwiddie and Briska 2004).
Psychiatric facilities should develop clear, consistent guidelines for

prosecuting inpatients (Norko et al. 1991). Before a policy is imple-
mented, gaining the cooperation of local law enforcement is essential.
The district attorney may be reluctant to view inpatient assaults as a
criminal matter because the community is not endangered; hospital ad-
ministrators can meet with local law officials to educate them about the
issue and determine the type of information needed to pursue prosecu-
tion. A prosecution policy should give inpatients an opportunity to
change their behavior though treatment before resorting to the filing of
criminal charges. At admission, when patients are provided written ma-
terial explaining their rights and responsibilities, a statement should be
included that notifies the patient that respect for others and the law is
expected in the hospital and that a failure to conform could result in
criminal charges. Other elements of a model prosecution policy are
summarized in Table 13–6. Note that these policy recommendations are
intended for repetitively assaultive patients who have not yet disrupted
the milieu to a substantial degree. There may be instances when a single
act of violence is so egregious that prosecution is a viable option even
without a history of assaults, such as an assault causing severe physical
trauma or sexual assaults on patients and staff (Dinwiddie and Briska
2004).
TABLE 13–6. Key elements and clinical considerations in determining

whether prosecution is appropriate
After an inpatient assault, perform a thorough assessment of the assailant’s
state of mind and motivation for the violence.
Determine whether the assault was motivated by impulsivity,
psychopathy, or psychosis, and document your findings in detail
(see Table 13–2).
—Assaultive behavior stemming from a patient’s illness (severe
mood dysregulation or psychosis) may respond to clinical
intervention.
—Because violence motivated by a patient’s antisocial characteristics
is less likely to respond to treatment, prosecution (or the threat to
prosecute) may be the best strategy (Reid and Gacono 2000).
On the basis of the assessment, implement appropriate clinical
interventions in the patient’s individualized treatment plan.
If the interventions fail and the assaultive behavior continues, contact a
hospital administrator to determine whether further acts of violence
should be prosecuted.
—The administrator could have an independent clinician or forensic
specialist review the case.
—If a decision is made to pursue prosecution, the assaultive patient
should be placed on notice; the creation of “probationary status”
itself may be effective (Hoge and Gutheil 1987).
A decision to prosecute a particular assault should be based on careful
consideration of the following factors:
—Nature and severity of the violence
—Adequacy of prior treatment attempts and results
—Likelihood of response to further treatment
—Clinical impact of incarceration on the patient
—Chances of conviction
—Potential negative effects and risks to staff and other patients if
assailant remains in the hospital
Key Points
■ Inpatient aggression is an important problem because it has
damaging psychological and physical effects on both psychiatric
patients and staff.
■ A small minority of psychiatric inpatients are responsible for the
majority of inpatient assaults, including the most serious as-
saults; this subset of repetitively assaultive patients warrants
greater attention in the form of systematic study.
■ The most robust long-term risk factor for inpatient violence is a
history of inpatient assaults or violent crime.
■ Certain psychiatric symptoms and behaviors indicate an inpatient
is at increased risk for assault in the short term; recognizing these
clinical risk factors and making appropriate interventions can
help prevent aggression.
■ The clinical management of an aggressive inpatient should be
guided by the type of violence in which a patient engages. It is im-
portant to characterize the primary factor motivating aggressive
behavior.
■ Impulsively aggressive inpatients should be observed for signs of
escalation so that measures can be taken to de-escalate patients
early in the process. Because most impulsive assaults are precipi-
tated by an aversive interpersonal interaction, clinicians should
critically examine how they approach patients.
■ In inpatients with a history of assaults motivated by psychosis,
increased assault risk is indicated by paranoid or persecutory
delusions (with or without commanding auditory hallucinations)
and accompanying anger, fear, or sadness.
■ Among inpatients whose aggression is motivated by antisocial or
self-serving interests, it is important to confront the behavior and
to attempt to get the patient to assume responsibility for his or
her actions. Criminal prosecution may be ethically justified in
managing this type of patient.
■ When a clinician is assuming care of an inpatient with a history of
violence, time spent investigating past motivations for aggressive
behavior through record review and clinical interview can be valu-
able. Knowing the precipitants, situations, and symptoms that
have led to violence in the past can be useful in preventing future
assaults. After an inpatient assault, a debriefing session can yield
critical information that can be used to develop different ap-
proaches effective in preventing subsequent violence.
References
Appelbaum KL, Appelbaum PS: A model hospital policy on prosecuting pa-
tients for presumptively criminal acts. Hosp Community Psychiatry
42:1233–1237, 1991
Appelbaum PS, Robbins PC, Roth LH: Dimensional approach to delusions: com-
parison across types and diagnoses. Am J Psychiatry 156:1938–1943, 1999
Arango C, Calcedo Barba A, Gonzalez, et al: Violence in inpatients with schizo-
phrenia: a prospective study. Schizophr Bull 25:493–503, 1999
Barratt ES, Stanford MS, Felthous AR, et al: The effects of phenytoin on impul-
sive and premeditated aggression: a controlled study. J Clin Psychophar-
macol 17:341–349, 1997
Beauford JE, McNiel DE, Binder RL: Utility of the initial therapeutic alliance in
evaluating psychiatric patients’ risk of violence. Am J Psychiatry 154:1272–
1276, 1997
Binder RL, McNiel DE: Effects of diagnosis and context on dangerousness. Am
J Psychiatry 145:728–732, 1988
Carmel H, Hunter M: Staff injuries from inpatient violence. Hosp Community
Psychiatry 40:41–46, 1989
Cheung P, Schweitzer I, Crowley K, et al: Violence in schizophrenia: role of hal-
lucinations and delusions. Schizophr Res 26:181–190, 1997
Chou KR, Lu RB, Mao WC: Factors relevant to patient assaultive behavior and
assault in acute inpatient psychiatric units in Taiwan. Arch Psychiatr Nurs
16:187–195, 2002
Convit A, Jaeger J, Lin SP, et al: Predicting assaultiveness in psychiatric inpa-
tients: a pilot study. Hosp Community Psychiatry 39:429–434, 1988
Convit A, Isay D, Otis D, et al: Characteristics of repeatedly assaultive psychi-
atric inpatients. Hosp Community Psychiatry 41:1112–1115, 1990
Coyne A: Should patients who assault staff be prosecuted? J Psychiatr Ment
Health Nurs 9:139–145, 2002
Dinwiddie SH, Briska W: Prosecution of violent psychiatric inpatients: theoret-
ical and practical issues. Int J Law Psychiatry 27:17–29, 2004
Duxbury J: An evaluation of staff and patient views of and strategies employed
to manage inpatient aggression and violence on one mental health unit: a
pluralistic design. J Psychiatr Ment Health Nurs 9:325–337, 2002
Flannery RB Jr, Hanson MA, Penk W: Patients’ threats: expanded definition of
assault. Gen Hosp Psychiatry 17:451–453, 1995
Flannery RB Jr, Hanson MA, Penk WE, et al: Replicated declines in assault rates
after implementation of the Assaulted Staff Action Program. Psychiatr Serv
49:241–243, 1998
Flannery RB Jr, Stone P, Rego S, et al: Characteristics of staff victims of patient
assault: ten year analysis of the Assaulted Staff Action Program (ASAP).
Psychiatr Q 72:237–248, 2001
Forster PL, Cavness C, Phelps MA: Staff training decreases use of seclusion and
restraint in an acute psychiatric hospital. Arch Psychiatr Nurs 13:269–271,
1999
Frueh BC, Knapp RG, Cusack KJ, et al: Patients’ reports of traumatic or harmful
experiences within the psychiatric setting. Psychiatr Serv 56:1123–1133,
2005
Hill CD, Rogers R, Bickford ME: Predicting aggressive and socially disruptive
behavior in a maximum security forensic psychiatric hospital. J Forensic Sci
41:56–59, 1996
Hoge SK, Gutheil TG: The prosecution of psychiatric patients for assaults on
staff: a preliminary empirical study. Hosp Community Psychiatry 38:44–49,
1987
Hoptman MJ, Yates KF, Patalinjug MB, et al: Clinical prediction of assaultive be-
havior among male psychiatric patients at a maximum-security forensic fa-
cility. Psychiatr Serv 50:1461–1466, 1999
Johnson ME, Hauser PM: The practices of expert psychiatric nurses: accompa-
nying the patient to a calmer personal space. Issues Ment Health Nurs
22:651–668, 2001
Kaltiala-Heino R, Tuohimaki C, Korkeila J, et al: Reasons for using seclusion
and restraint in psychiatric inpatient care. Int J Law Psychiatry 26:139–149,
2003
Katz P, Kirkland FR: Violence and social structure on mental hospital wards.
Psychiatry 53:262–277, 1990
Krakowski M, Czobor P: Violence in psychiatric patients: the role of psychosis,
frontal lobe impairment, and ward turmoil. Compr Psychiatry 38:230–236,
1997
Krakowski M, Convit A, Jaeger J, et al: Neurological impairment in violent
schizophrenic inpatients. Am J Psychiatry 146:849–853, 1989
Krakowski M, Czobor P, Chou JC: Course of violence in patients with schizo-
phrenia: relationship to clinical symptoms. Schizophr Bull 25:505–517, 1999
Kraus JE, Sheitman BB: Characteristics of violent behavior in a large state psy-
chiatric hospital. Psychiatr Serv 55:183–185, 2004
Lam JN, McNiel DE, Binder RL: The relationship between patients’ gender and
violence leading to staff injuries. Psychiatr Serv 51:1167–1170, 2000
Lanza ML, Kayne HL, Hicks C, et al: Environmental characteristics related to
patient assault. Issues Ment Health Nurs 15:319–335, 1994
Lanza ML, Kayne HL, Pattison I, et al: The relationship of behavioral cues to as-
saultive behavior. Clin Nurs Res 5:6–27, 1996
Lehmann LS, McCormick RA, Kizer KW: A survey of assaultive behavior in
Veterans Health Administration facilities. Psychiatr Serv 50:384–389, 1999
Linaker OM, Busch-Iversen H: Predictors of imminent violence in psychiatric
inpatients. Acta Psychiatr Scand 92:250–254, 1995
Love CC, Hunter ME: Violence in public sector psychiatric hospitals: bench-
marking nursing staff injury rates. J Psychosoc Nurs Ment Health Serv
34:30–34, 1996
Maier GJ: Managing threatening behavior: the role of talk down and talk up.
J Psychosoc Nurs Ment Health Serv 34:25–30, 1996
Manfredini R, Vanni A, Peron L, et al: Day-night variation in aggressive behav-
ior among psychiatric inpatients. Chronobiol Int 18:503–511, 2001
McNiel DE, Binder RL: Relationship between preadmission threats and later vi-
olent behavior by acute psychiatric inpatients. Hosp Community Psychia-
try 40:605–608, 1989
McNiel DE, Binder RL: The relationship between acute psychiatric symptoms,
diagnosis, and short-term risk of violence. Hosp Community Psychiatry
45:133–137, 1994
McNiel DE, Binder RL, Greenfield TK: Predictors of violence in civilly commit-
ted acute psychiatric patients. Am J Psychiatry 145:965–970, 1988
McNiel DE, Eisner JP, Binder RL: The relationship between command halluci-
nations and violence. Psychiatr Serv 51:1288–1292, 2000
McNiel DE, Gregory AL, Lam JN, et al: Utility of decision support tools for as-
sessing acute risk of violence. J Consult Clin Psychol 71:945–953, 2003
Meloy JR: The prediction of violence in outpatient psychotherapy. Am J Psy-
chother 41:38–45, 1987
Miller RD, Maier GJ: Factors affecting the decision to prosecute mental patients
for criminal behavior. Hosp Community Psychiatry 38:50–55, 1987
Miller RJ, Zadolinnyj K, Hafner RJ: Profiles and predictors of assaultiveness for
different psychiatric ward populations. Am J Psychiatry 150:1368–1373, 1993
Monahan J, Steadman HJ, Silver E, et al: Violence and the clinician: assessing
and managing risk, in Rethinking Risk Assessment: The MacArthur Study
of Mental Disorder and Violence. New York, Oxford University Press, 2001,
pp 129–143
Nijman H, Merckelbach H, Evers C, et al: Prediction of aggression on a locked
psychiatric admissions ward. Acta Psychiatr Scand 105:390–395, 2002
Nolan KA, Czobor P, Roy BB, et al: Characteristics of assaultive behavior among
psychiatric inpatients. Psychiatr Serv 54:1012–1016, 2003
Nolan KA, Volavka J, Czobor P, et al: Aggression and psychopathology in treat-
ment-resistant inpatients with schizophrenia and schizoaffective disorder.
J Psychiatr Res 39:109–115, 2005
Norko MA, Zonana HV, Phillips RT: Prosecuting assaultive psychiatric inpa-
tients. Hosp Community Psychiatry 42:193–194, 1991
Quanbeck CD, McDermott BE, Lam J, et al: Categorization of aggressive acts
committed by chronically assaultive state hospital patients. Psychiatr Serv
58:521–528, 2007
Quintal SA: Violence against psychiatric nurses: an untreated epidemic? J Psy-
chosoc Nurs Ment Health Serv 40:46–53, 2002
Ray CL, Subich LM: Staff assaults and injuries in a psychiatric hospital as a func-
tion of three attitudinal variables. Issues Ment Health Nurs 19:277–289,
1998
Reid WH, Gacono C: Treatment of antisocial personality, psychopathy, and other
characterologic antisocial syndromes. Behav Sci Law 18:647–662, 2000
Schwarz CJ, Greenfield GP: Charging a patient with assault of a nurse on a psy-
chiatric unit. Can Psychiatr Assoc J 23:197–200, 1978
Secker J, Benson A, Balfe E, et al: Understanding the social context of violent
and aggressive incidents on an inpatient unit. J Psychiatr Ment Health
Nurs 11:172–178, 2004
Smith GM, Davis RH, Bixler EO, et al: Pennsylvania State Hospital system’s se-
clusion and restraint reduction program. Psychiatr Serv 56:1115–1122, 2005
Soliman AE, Reza H: Risk factors and correlates of violence among acutely ill
adult psychiatric inpatients. Psychiatr Serv 52:75–80, 2001
Steinert T: Prediction of inpatient violence. Acta Psychiatr Scand Suppl 412:133–
141, 2002
Swanson JW, Swartz MS, Van Dorn RA, et al: A national study of violent behav-
Swett C, Mills T: Use of the NOSIE to predict assaults among acute psychiatric
patients: Nurses’ Observational Scale for Inpatient Evaluation. Psychiatr
Serv 48:1177–1180, 1997
Tardiff K: The risk of being attacked by patients: who, how often, and where? in
Patient Violence and the Clinician. Edited by Eichelman BS, Hartwig AC.
Washington, DC, American Psychiatric Press, 1995, pp 13–32
Tardiff K, Sweillam A: Assaultive behavior among chronic inpatients. Am J Psy-
chiatry 139:212–215, 1982
Turkington D, Kingdon D, Weiden PJ: Cognitive behavior therapy for schizo-
phrenia. Am J Psychiatry 163:365–373, 2006
Volavka J, Czobor P, Nolan K, et al: Overt aggression and psychotic symptoms
in patients with schizophrenia treated with clozapine, olanzapine, risperi-
done, or haloperidol. J Clin Psychopharmacol 24:225–228, 2004
Weinshenker N, Siegel A: Bimodal classification of aggression: affective defense
and predatory attack. Aggression and Violent Behavior 7:237–250, 2002
Whittington R, Patterson P: Verbal and non-verbal behavior immediately prior
to aggression by mentally disordered people: enhancing the assessment of
risk. J Psychiatr Ment Health Nurs 3:47–54, 1996
Whittington R, Wykes T: Aversive stimulation by staff and violence by psychi-
atric patients. Br J Clin Psychol 35:11–20, 1996
Woodworth M, Porter S: In cold blood: characteristics of criminal homicides as
a function of psychopathy. J Abnorm Psychol 111:436–445, 2002
C H A P T E R 1 4
Emergency Services
Anzalee Khan, M.S.
P sychiatric emergency services (PESs) usually provide a systematic

care process in which patients who present psychiatric emergencies are
optimally evaluated and managed. A psychiatric emergency is defined as
a disturbance in thoughts, feelings, or behaviors for which immediate
assessment and treatment are necessary. An emergency can not only be
declared by the care delivery system but also by the patient, family,
community, or a friend who may present the patient as having an emer-
gency. Staff in the emergency setting will then assess and attend to the
patient and decide on a course of action to further ascertain the problem
and its origins. The systematic care process includes sequential steps
such as assessment, problem identification, treatment planning, inter-
ventions, ongoing monitoring, and discharge. However, resolution of
the emergency usually does not occur in the emergency setting itself;
most often it is post-emergency care that leads to resolution.
PESs usually consist of acute hospital-based psychiatric service de-
livery models that are available for mental health emergencies. They are
generally open 24 hours a day in the United States and in some form or
other in most developed countries. In the United States the 1963 Com-
munity Mental Health Act mandated emergency psychiatric care as one
of “five essential services” in all federally funded community mental
health service systems (Gerson and Bassuk 1980). Such a service was be-
lieved to be critical “to prevent unnecessary [re]hospitalizations that
277
might, in turn, foster chronicity and dependence on institutional care”

(Solomon and Gordon 1986–1987, pp. 119–120).
Bassuk and Gerson (1979) suggested that the PES’s role was to “rec-
oncile the complex needs of the local population with the traditional or-
ganizational structure of [local treatment options]” (p. 35), suggesting
that definitions of appropriate use of such services should include a
broad array of treatments. A competing and classically medical-surgical
viewpoint is that these costly services ought to be reserved for those
who legitimately cannot wait for psychiatric intake appointments. This
perspective would suggest that a narrow definition of “emergency”
conditions is more appropriate to justify help-seeking in the PES, per-
haps including only illness episodes “characterized by surprise, time
constraints, high stakes, and pressure for action” (Murdach 1987, pp.
268–269). That definition is supported, in part, by a 1988 review of psy-
chiatric decision making in the emergency department (Marson et al.
1988), which concluded that acuteness of symptoms and inherent dan-
gerousness, such as acute aggressive or agitated states, were the vari-
ables that most strongly predicted the decision to hospitalize.
The service delivery model for providing hospital-based PES ap-
pears to vary widely by site and state. Some hospital emergency depart-
ments in large urban areas have designated separate areas for handling
psychiatric patients; such areas often are linked to a 24- or 72-hour hold-
ing facility for patients requiring extended observation. As an example
of such expanded services, New York State has introduced the Compre-
hensive Psychiatric Emergency Program (CPEPs), which provides a
range of psychiatric emergency services including brief overnight ob-
servation stays. These programs coordinate the delivery of a full range
of psychiatric crisis and emergency care within a distinct geographic
area. Such programs are required to provide four components of ser-
vice: hospital-based crisis intervention, extended-observation beds,
mobile crisis outreach services, and crisis residences. These integrated
programs attempt to alleviate the overcrowding in emergency rooms,
provide alternatives to inpatient admissions, and maintain a commu-
nity-based focus. The objectives are to provide crisis intervention in the
community—consisting of timely triage, assessment, intervention, and
links to other community-based mental health services—and to control
inpatient admissions.
Smaller facilities utilize referral to an on-call mental health specialist
as needed but maintain the patient in the general emergency depart-
ment population. Regardless of the service delivery model used, re-
search indicates that hospital-based PES facilities receive a broad array
of service requests, many of which at times appear to be nonemergent
Emergency Services ❘ 279
(Kooiman et al. 1989; Oyewumi et al. 1992; Vaslamatzis et al. 1987;

Vigiser et al. 1984), such as requests for social services, requests for
treatment of substance abuse, or referral requests for psychological
treatments.
The following types of services are included in PES:
• Psychiatric treatment to stabilize and/or ameliorate acute symp-

toms of mental illness/emotional crisis
• Evaluation and referral for inpatient psychiatric hospitalization
• Medical screening and referral to acute medical services
• Continued observation and assessment in the Extended Observa-
tion Unit
• Transfer to other facilities for further assessment and/or care
• Referral to an outpatient facility and/or treatment program
• Referral to assistance in resolving a situational crisis
• Evaluations for patients of private practitioners, psychiatrists, social
workers, counselors (second opinion) in crisis situations. The pa-
tient is referred back to the primary therapist.
Case Example 1:
An Agitated Patient and the
Delivery of Emergency Psychiatric Care
Initial evaluation. A 24-year-old male is brought by the local police and
ambulance to an emergency psychiatric facility within a medical emer-
gency setting for bizarre and uncontrolled behavior at home. His
mother had contacted the police because she was afraid for her life and
that of her son. She reported that he had not slept for the past 3 nights,
was not eating, and was pacing the apartment “damning people to hell”
and stating “God is here.” He had begun to throw plates, glasses, and
furniture around the apartment. The patient also admitted to visual hal-
lucinations of seeing God in the shape of a white male. Both mother and
patient denied any drug use. The patient had no prior medical history,
was not on any medications, and had no allergies. He was initially eval-
uated by the emergency psychiatric nurse, who also interviewed the
mother. The nurse reported that the patient did not have any significant
past psychotic episodes or hospitalizations, and she referred him to be
seen by the medical doctor and the psychiatrist.
Medical evaluation. The patient’s physical examination results
were within normal limits, as was his laboratory workup.
Psychiatric evaluation. The patient was examined by the psychia-
trist and psychiatric resident. The psychiatric assessment resulted in the
diagnosis of acute schizophrenic episode. Following the diagnosis, the
patient was referred to a social worker and case manager on staff for fur-
ther evaluation of available support structures and past history, which
he refused, saying, “The people here are going to kill me.” One staff
member with whom he had bonded convinced the patient that no harm
would come to him. However, the patient refused to be further exam-
ined and instead paced up and down the hallway outside yelling, “God
is here, come and get me.” The emergency department psychiatrist rec-
ommended that the patient be admitted to the inpatient psychiatric fa-
cility for observation because he was thought to present a risk of harm
to himself or others. A short-acting intramuscular antipsychotic for his
agitated behavior was prescribed. The patient eventually agreed to take
the medication and to go to the inpatient psychiatric unit.
Psychiatric Emergency Services Delivery Models

Psychiatric Emergency Room
The psychiatric emergency room is a key component of a PES. Its pri-
mary function is the care of “true emergency” psychiatric patients who
are acutely distressed and disturbed, with rapidly changing mental sta-
tus situations. Brief intake evaluation, crisis intervention, and appropri-
ate subsequent referral are the primary tasks of this service.
The emergency room is usually housed in easily accessible, fairly
spacious quarters. An appropriately structured psychiatric emergency
room will usually provide a quiet environment and procedures to man-
age patients who may want to escape; who would require restraint or
seclusion because of dangerous, assaultive behavior; and who may
want to harm themselves (Allen 2002). A typical emergency room has
several different areas, each specialized for patients with particular se-
verities of psychiatric illness, as described in the following paragraphs.
In the triage area, patients receive a preliminary evaluation by psy-
chiatry-trained nurses and/or psychiatric social workers. After triage,
patients are usually taken directly to one of four functional treatment
areas based on the nature of the emergency: nonthreatening psychiatric
emergencies, life-threatening psychiatric emergencies, extremely agi-
tated behavior requiring physical restraint, or extremely agitated be-
havior requiring seclusion. Suicidal patients may bypass triage and are
seen directly by the emergency room psychiatrist.
The seclusion area is an important area of an emergency room. There
may be separate rooms for voluntary and unlocked seclusion, often
termed a “quiet room,” as well as locked seclusion rooms and available
restraint. The seclusion rooms should have no furniture or accessories
and have visual observational capacity for the staff.
The general medical area is for stable patients who still need to be fol-
lowed-up but may not pose a serious physical threat. This area usually
contains several interviewing rooms that offer privacy but are in close
proximity to ancillary staff as well as a physical examination room with
life safety equipment. The surrounding area is often very busy, filled
with patients with a wide range of psychiatric problems. Many will re-
quire further investigation and possible admission. Patients who are
not in need of immediate treatment are sent to a different area to await
disposition or discharge.
Generally, a PES within a hospital should be designed to increase
overall efficiency while providing a calming, open setting that mini-
mizes the crowded environment and resulting anxiety that often char-
acterize emergency visits. Emergency rooms should be focused on max-
imum patient privacy and optimum comfort because these features are
essential for an emergency psychiatric patient. This can be enhanced by
aesthetically pleasing design features throughout the emergency room,
such as an interior glass wall between waiting and patient registration
areas to provide a sense of security to the staff and a monitoring capa-
bility of the entire emergency area.
An efficient emergency room layout is important to the rapid ad-
ministration of services. The following design features should be avail-
able: 1) separate entrances for life-threatening emergencies, such as
threats of suicide and self-harm; 2) entry/registration/waiting/triage
functions located in close proximity to one another to decrease distance;
and 3) entrances to patient interview rooms that allow maximum access
by psychiatric and medical staff for patients to be taken directly to the
treatment areas. It is very important to note that all doors should have
locks but that doors are usually kept open to avoid creating a sense of
crowding for paranoid patients. All sharp objects should be out of the
reach of patients and their families. Articles of furniture should be in-
stalled so as to prevent them from being used as a weapon. Additional
comfort is provided by appropriate reading materials, televisions, and
availability of telephones. Televisions and telephones should be se-
cured to their respective structures, and the television should be out of
the reach of patients.
There also needs to be an area where safety personnel, usually local
police officers, can wait while the paperwork for patients whom they
have brought to the emergency room is being processed. A cardinal rule
is that no loaded weapons can be brought in by law enforcement offi-
cers. Most emergency departments also have their own safety officer
supervising the waiting and examination areas.
Goals of the Psychiatric Emergency Room

The psychiatric emergency room acts as a central assessment and acute
treatment agency with the possibility of referrals to various mental,
medical, and social services. The emergency services branch out to the
inpatient admissions service, all outpatient clinics, day programs, and,

if possible, to case management systems, transitional housing, and sub-
stance abuse treatment programs. The mere presence of a receptive and
helpful PES is often a tremendous relief to these other services and their
patients. Thus the psychiatric emergency room is able to facilitate pa-
tients’ connecting or reconnecting with various services without the
complexity of an inpatient admission. However, in many circumstances
inpatient admission is still indicated. An admission to inpatient services
should not be construed as a failure of the emergency team.
Staff
Most large emergency rooms include a number of psychiatrists (M.D.s)
and at times psychiatric residents, registered nurses (R.N.s), psycholo-
gists (Ph.D.s), social workers (M.S.W.s), psychiatric technicians (e.g.,
hospital orderlies), clerical staff, and security officers. PES staff should
be trained in making thorough assessments of patients’ problems and
in identifying appropriate dispositions and referrals. To conduct these
assessments, Summers and Happell (2002) argued that the psychiatric
nurses may be a core service provider in the emergency room. Similarly,
Osborne (2003) and McDonough et al. (2004) reported that the use of a
mental health triage nurse reduced lengthy waiting times and crowd-
ing in waiting rooms. In addition, studies found that the use of psychi-
atric triage scales contributed to reduced wait times, more efficient and
effective treatment of mental health patients, and improved referral to
appropriate resources (Broadbent et al. 2004; Happell et al. 2002).
A key focus of PES staff is the safe management of the psychiatric
and behavioral emergencies. Staff should be trained in emergency pro-
cedures, including seclusion and physical restraint. Trained safety offic-
ers should also be present and should be under the supervision of the
psychiatrist and medical and nursing personnel.
The psychiatric emergency team is led by the psychiatrist. This lead-
ership position is built on the specific medical-psychiatric diagnostic
background and the psychopharmacological expertise of the psychia-
trist. However, many psychiatric emergency rooms, particularly those
that do not offer comprehensive services, cannot provide continuous
coverage by a psychiatrist (Allen 1999). Consequently, psychiatric nurses,
social workers, and at times trained psychologists will provide crisis in-
tervention and emergency psychotherapy, if necessary.
Most psychiatric emergency rooms also have a physician available to
evaluate psychiatric emergencies with a medical component. This is
particularly important in medical emergencies that may have been mis-
identified as a psychiatric emergency by the patient, friend, or family

member. A general medical evaluation should be assured for all regis-
tered patients. When specific medical conditions are identified, it is im-
portant to have rapid access to appropriate medical care. Thus it is ad-
vantageous to have the psychiatric emergency room contiguous to the
medical emergency department and to have policies in place concerning
the movement of patients and consultants between these two services.
Good communication between the medical staff and the psychiatric
staff is also important (McClelland 1983).
Flow of Assessments in the Emergency Room

The process for determining patient needs in psychiatric emergency
rooms optimally includes six steps (Coristine et al. 2007):
1. A person arriving in the PES undergoes triage for mental health

complaints by an emergency room nurse using standardized risk
assessment criteria (e.g., Mental Health Triage Scale or the National
Triage Scale [Dreyfus 1987]) or another standardized triage proce-
dure.
2. If the person threatens to hurt him- or herself or others, or if the per-
son is at risk for escape or for violence to others, he or she should be
seen immediately by the emergency room psychiatrist.
3. If none of the risks in item 2 is present, a triage assessment is con-
ducted, including recording a complete set of vital signs, history of
significant mental illness, medical history, recent history of alcohol
or substance abuse, disorientation, clouding of consciousness, ap-
pearance of intoxication with alcohol or drugs, malnourishment,
unkempt appearance, or any other concerns of a medical nature.
4. If the assessment confirms the presence of any of these signs, the
person is triaged to be medically assessed by the emergency depart-
ment physician. Once the person is medically cleared, the emer-
gency medical doctor will refer him or her for more intensive men-
tal health assessment by the psychiatrist in conjunction with the
PES team. This team then provides a full mental assessment and the
most appropriate disposition, such as discharge, referral to an out-
patient program, admission to an inpatient unit, or transfer to an-
other facility.
5. If all observations are negative, the patient is fast-tracked to the case
worker or social worker, who screens the patient about problems
with housing, finances, legal issues, or social supports. The case
worker or social worker conducts an assessment and reports to the
emergency room psychiatrist regarding referral to appropriate com-

munity resources.
6. Case worker or social worker referrals may be made to any of sev-
eral community mental health agencies that provide case manage-
ment, housing, social supports, and crisis management services.
Extended Services
Research on PES delivery systems has documented persistent growth in
demand, with concomitant increased pressure on psychiatric emer-
gency room operations and personnel. Factors contributing to this de-
mand are the shift to the community mental health service delivery
model and the reduction of available long-term inpatient beds in state
psychiatric facilities; insufficient community mental health supports;
and recidivism among people with persistent mental illness identified
as socially disadvantaged (Ellison and Blum 1986; Smart et al. 1999; So-
lomon and Davis 1985). The delineation of these factors provides the
arena in which to test new PES delivery strategies, such as the introduc-
tion of Assertive Community Treatment teams and Intensive Case Man-
agement models, liaison with community agencies (Sundheim and
Ryan 1999), and utilization of psychiatric nurses to create an integrated
care pathway (Wynaden et al. 2003).
Case Example 2:
Psychiatric Evaluation in an Emergency Setting
Initial symptoms. A day after his prison release, Mr. A became agitated
at home and began staring at others and not communicating. Upon reg-
istration at the PES, and after initial review by the nurse, he was seen by
the psychiatric resident. He reported that other people could read his
mind and broadcast his thoughts. He also reported that he heard voices
outside his window at home telling him that he was “not a good man”
and would “not succeed.” Mr. A also reported that he heard his neigh-
bors talking about him. He had threatened and confronted some indi-
viduals, which resulted in an altercation.
Past history. Mr. A had no history of psychiatric symptoms or legal
problems until his early 40s, when he was incarcerated for a nonviolent
offense for 6 months. During the incarceration, his mental status deteri-
orated. He began exhibiting depression, ideas of hopelessness, and
paranoid delusions. He was hospitalized in the prison infirmary and
treated with venlafaxine and haloperidol, with subsequent stabilization.
He was released after serving his full sentence.
Initial examination. Mr. A was noted to be staring blankly without
looking at the interviewer. His speech was monosyllabic. He showed
marked psychomotor retardation, and he made negativistic statements
about his life. He denied experiencing auditory hallucinations. He was

oriented to time, place, and person but had poor insight and judgment.
Treatment and course of illness. Mr. A was given an evaluation in-
cluding the Mini-Mental State Examination and the Structured Guide
for the Assessment of Violence. His cognitive status was reported as
“fair.” Mr. A was seen by the emergency psychiatrist, who conferred
with the nurse. A diagnosis of psychotic depression was made. Mr. A
was hospitalized in the 72-hour bed service program at the emergency
psychiatric facility and began antidepressant and antipsychotic phar-
macotherapy. Mr. A’s mental status improved progressively after the
first day, and he became cooperative and well groomed. His condition
began to deteriorate at day 3, when he again reported that he was “not
a good person” and that people were threatening him. He also reported
auditory hallucinations, primarily in the evening. Mr. A was admitted
to the inpatient unit. After 3 weeks of treatment, his condition stabilized
and he was discharged to a residential facility.
Extended Psychiatric Observation Services

Extended psychiatric observation services typically use a designated
short length of stay to stabilize and observe patients with unclear and
unstable psychiatric presentations. These units are most often inte-
grated in the PES, are small in size, and may offer a length of stay be-
tween 24 hours and 3 days. Several studies have suggested that most
patients admitted to these programs show improvement in the severity
of their psychiatric symptoms, are able to be discharged in the desig-
nated time frame, express high satisfaction with the program (Schnei-
der and Ross 1996), and have a low rehospitalization rate (Allen 2002;
Rhine and Mayerson 1971; Weisman et al. 1969). These observation
beds are often used as an adjunct to the initial psychiatric emergency
room evaluation; patients needing definitive psychiatric hospitaliza-
tion will be transferred to a separate inpatient facility for hospitalization
of several days or weeks.
The option of offering brief admission to short-stay beds within a
PES provides a model that meets a variety of patient and system needs.
Brief admission within a PES 1) allows emergency staff more time to de-
velop alternatives to hospitalization or gain diagnostic clarity; 2) en-
ables difficult patients to remain in the community by offering respite
for both the patient and community providers; 3) provides selected pa-
tients with a setting that does not gratify dependency needs in the same
manner as a hospital stay might; and 4) makes available a targeted treat-
ment modality for patients whose presenting symptomatology can
be ameliorated within a brief period of time (e.g., those who become
disorganized following an acute stressor). However, a key admission
criterion for such short-term PES units is that the patient be cooperative
and voluntary. Involuntary admission can only be done to an acute in-
patient admissions unit. The main emphasis of such units is on main-
taining patients’ functioning in and ties to the community. Herz et al.
(1979), in a series of papers, developed the idea that the results of brief
hospitalization can be comparable with standard hospitalization, with
the advantage of fostering less regression and better maintenance of
community survival skills.
Observations
Several different models exist for short-term observation assessment
and stabilization units, including the 23-hour observation bed, the crisis
stabilization unit, and the 72-hour observation bed. Such units are usu-
ally in close proximity to PESs.
23-Hour Observation Bed, Psychiatric

A 23-hour observation bed is a facility-based crisis stabilization unit
that provides a medically safe environment for a limited period of up to
23 hours for individuals experiencing a crisis or acute psychiatric emer-
gency condition. Individuals are monitored, assessed, and evaluated to
ensure appropriate care and disposition within the given time period.
The observation facility is located in the emergency department and is
configured to provide primary emergency care during periods of peak
demand. It provides rapid resolution of many crises (e.g., filtering of
substance use emergencies [Breslow et al. 1996]).
Crisis Stabilization Unit

The crisis stabilization unit is generally a small unit located adjacent to
the emergency department. The service provides extended 24-hour ob-
servation, treatment, and support up to a total of 72 hours for patients
seen in the emergency department. The purpose of a crisis stabilization
unit is to stabilize and redirect a client to the most appropriate and least
restrictive community setting available, consistent with the client’s
needs. Crisis stabilization units may screen, assess, and admit for stabi-
lization those persons who present themselves to the unit on a volun-
tary basis or who are brought to the unit. Patients are referred to the cri-
sis stabilization unit by a physician or psychiatrist at the emergency
department. A multidisciplinary treatment team including physicians,
registered nurses, licensed clinicians, and mental health technicians
provide the patient with the following:
• Crisis intervention/stabilization
• Psychiatric nursing assessment
• Physical assessments
• Medication/somatic services
• 24-hour observation
• Individual and group counseling
• Linkage and referrals to longer-term services
• Education for safe return to the community
A client’s discharge from the crisis stabilization unit is based on his

or her self-assessment and a clinically appropriate disposition reached
by the treatment team.
Case Example 3: Extended Observation

Initial assessment. Mr. B is a 38-year-old man with three prior psychi-
atric hospitalizations with the diagnosis of major depression, the last
having been 2 months prior to his admission to the extended observa-
tion unit. A conflict with his mother more than a year ago led to a sui-
cidal episode in which Mr. B took various prescription drugs and was
subsequently hospitalized. After discharge, and prior to the emergency
department presentation, Mr. B had frequent episodes of anxiety and
suicidal ideation during which he reported wanting to die. He would
buy “all the drugs he could find.” In addition to his psychiatric follow-
up, Mr. B was followed up by a social worker who immediately took the
patient to the psychiatric emergency room, where he was again evalu-
ated. His physical and neurological evaluation, magnetic resonance im-
aging scans, and electroencephalogram were all negative.
Psychiatric evaluation. The emergency room psychiatrist con-
cluded that Mr. B would be a danger to himself and decided to keep him
in the 72-hour extended observation unit. During his time in the unit,
Mr. B received antidepressant pharmacotherapy. He responded well to
both the antidepressant and psychotherapy. After discharge he was fol-
lowed up at the outpatient psychiatric clinic. Two months after dis-
charge Mr. B developed an acute paranoid episode with suicidal and ho-
micidal ideation during which he believed that his mother was coming
to find him and hurt him for not calling her while he was in the hospital.
Mr. B’s psychiatrist again recommended admission to the extended ob-
servation unit that had previously worked so well for him. Mr. B was
admitted to the unit, and antipsychotic medication was added to his
regimen. He was greatly relieved by the availability of this additional
support (psychotherapy, psychiatric follow-up, and observation) to
help him control himself. During his stay in the unit, his paranoid symp-
toms markedly improved and the suicidal and homicidal ideas re-
solved. Mr. B agreed to remain on his antipsychotic medication and was
discharged to return to his psychiatric follow-up.
72-Hour Observation Bed, Psychiatric

Only designated personnel in a psychiatric emergency setting can place
a person in a 72-hour mental health hold. Such personnel include police
officers, members of a “mobile crisis team,” or other mental health pro-
fessionals authorized by their county. One of three conditions must be
present for an individual to be placed on a 72-hour hold. The desig-
nated personnel must believe there is probable cause that, due to a psy-
chiatric disorder, the individual is 1) a danger to him- or herself; 2) a
danger to others; or 3) gravely disabled (unable to provide for his or her
basic personal needs for food, clothing, or shelter).
The person placed in a 72-hour hold must be advised of his or her
rights. Most facilities require an application stating the circumstances
under which the person’s condition was called to the attention of the
professional; what probable cause there is to believe the person is a dan-
ger to self or others or is gravely disabled (due to a mental disorder);
and the facts upon which this probable cause is based. Mere conclusions
without supporting facts are not sufficient. When a person is detained
for up to 72 hours, the hospital is required to do an evaluation, taking
into account the patient’s medical, psychological, educational, social, fi-
nancial, and legal situation. The hospital does not have to hold the pa-
tient for the full 72 hours if it is thought that the patient no longer re-
quires evaluation or treatment. By the end of the 72 hours, one of the
following must happen:
• The person is released;

• The person signs in as a voluntary patient to the hospital; or
• The person is put on a 14-day involuntary hold (a “certification for
intensive treatment”), the structure of which will depend on the lo-
cal mental hygiene state laws.
A mental health patient being held involuntarily must be informed

of his or her rights in a language or manner that he or she can under-
stand, in accordance with the local laws.
In addition, the patient has the right to be informed fully of the risks
and benefits of the proposed treatment and to give his or her informed
consent to the treatment. A patient has the right to refuse medication
unless there is an emergency condition or the patient is found to lack ca-
pacity to make an informed decision after a judicial hearing. If a patient
is found in a hearing to lack capacity to consent to medication, the judge
may then order medication over the patient’s objection (Oldham and
DeMasi 1995; Weisman et al. 1969). In recent years, in New York State,
13%–15% of all patients in an emergency psychiatric service have been

admitted to extended observation units (Allen 2002).
Psychiatric Emergency Services in the Community

Community services providers may offer short-term case management
services to psychiatric patients who have been seen in the psychiatric
emergency department. These services (usually funded by municipal,
county, private, or hospital-affiliated providers) include
• Availability of crisis residences and mobile response teams

• Outreach
• Help with basic needs (food, clothing, emergency housing, identifi-
cation)
• Assistance in connecting or reconnecting with healthcare and men-
tal healthcare providers and programs
• Counseling and support
Crisis Residence
A crisis residence offers a supervised residential setting for persons re-
quiring extended stabilization during a mental health crisis. The ex-
pected length of stay could be up to 21 days. Crisis beds are usually
linked to local psychiatric emergency rooms and acute inpatient pro-
grams. Follow-up care is provided after discharge by community re-
sources and supports. The major goal of crisis residences is to stabilize
the situation and return the patient to his or her home quickly, rather
than to provide long-term care. Emphasis is on maintaining the rela-
tionships the patient has in the community, with family, the referral
agency, and with those resources that have provided services previ-
ously. Services that can be provided in the community will not be du-
plicated in the residence. Each program provides a highly structured,
individually designed intervention for each resident in accordance with
the needs of patients.
Crisis Response Services

Persons with serious and long-term psychiatric illness may experience
recurrent crises even when comprehensive and continuous community
support services are available. As a result, the capacity to provide crisis
assistance is a critical aspect of a community support system. Crisis re-
sponse services have 24 hours a day, 7 days a week telephone services
to provide counseling and support to relieve a crisis situation. If a face-
to-face contact is indicated, these services refer a trained professional to

visit the patient/client in the community.
Crisis response services assist individuals to alleviate and resolve
emotional distress or situational disturbances that affect their ability to
cope. The goals of a crisis response service are to use the least intrusive,
most effective intervention to provide immediate support, information,
and referrals; to facilitate problem solving to assist in the alleviation of
a mental health crisis; and to develop an intervention plan with individ-
uals in crisis that meets their needs, mobilizes their strengths and re-
sources, and averts hospitalization and contact with police.
Crisis Respite
Crisis respite is the lowest level of treatment intensity in the crisis resi-
dence program. The crisis respite programs serve patients with housing
problems. Sledge et al. (1996) described an arrangement linking respite
with day hospitalization programs designed for the severely and per-
sistently mentally ill. The respite component provides housing for up to
four clients, using mental health workers and Master’s-level program
directors.
Mobile Crisis Intervention Team

A mobile crisis intervention team (MCIT) program partners a mental
health professional and a police officer who respond to 911 emergency
and police dispatch calls involving emotionally disturbed persons. The
goal of the MCIT is to enable individuals experiencing a psychiatric crisis
or distress to access a range of crisis intervention services in a timely and
effective manner in their own environment or the environment of their
choice. An additional goal is to provide a consistent integrated response
to a psychiatric crisis in the community, regardless of which service iden-
tifies the individual in crisis. MCIT also serves to improve overall capac-
ity of the community to address the concerns of individuals experiencing
a mental health crisis at their living site, through provision of support, in-
formation, and education to caregivers. MCIT usually coordinates re-
sponse to an emergency call with the PES to support and triage over the
phone and through mobile visits. The services offered include
• Assessment of the presenting crisis, current supports, and resources

• Supportive, collaborative planning for solution-focused options
• Referral to appropriate follow-up services
• Consultation/advocacy with existing supports and services
• Short-term crisis management as necessary
Although mobile crisis models overlap to some extent, they also

differ in terms of readiness, tactical training, equipment, and cross-train-
ing of police in psychiatric techniques and vice versa. Stroul (1993) found
that 80% of mobile crisis models were accessible on a 24-hour basis.
These services can cover wide areas and may be particularly useful in ru-
ral communities where mental health services are distant (Allen 2002).
Assessment Issues in the Emergency Setting

Safety Considerations
One of the main goals of PESs is safety; the evaluating clinician and pa-
tients must be safe. At a minimum, patients must have been searched
and disarmed before meeting with the evaluating clinician in the psychi-
atric emergency room. A clear route of rapid egress from the examina-
tion room must be ensured, and security personnel must be available,
ideally through a panic button or other communication means. Safety
considerations may require that the patient be in restraints or that a phys-
ical barrier be present between patient and clinician. In addition to the
patient’s history and presentation, the clinician’s own experience and
anxiety level ought to be among the determining factors in deciding the
extent of safety precautions in place during a particular evaluation.
The importance of a full assessment of the patient to ensure safety
cannot be overstated. In the case of a suicidal patient, a determination
of what will keep the patient safe must be made as soon as possible. In
the patient with acute psychosis, medical comorbidities and substance
abuse should be considered early in the differential diagnosis and
treated (Buckley 1994).
Seclusion and Restraints

Seclusion
Seclusion can be useful for agitated patients and external stimulation; it
also allows the patient a period of “time-out” to regain behavioral con-
trol. A seclusion room must be safe and free of objects that could be used
to injure self or others. Medical conditions that are unstable and require
close physical interactions or monitoring preclude the use of seclusion.
Only staff who have been trained in using seclusion techniques are au-
thorized to implement the procedure. At first the door can remain open,
but if the patient continues to be agitated, the door is locked for safety.
The patient must always be aware of the consequences of his or her
behavior and be given periodic opportunities to comply with defined
behavioral parameters in order to be released from seclusion. Medica-

tions can be offered to avoid further restrictive measures. Patients in
seclusion should be monitored by closed-circuit television if possible.
Seclusion is always time limited, depending on the local mental health
policies; usually the duration does not exceed 4 hours, after which the
patient has to be reevaluated by the physician and a new order has to be
written. For patients under age 18 this duration is reduced to 2 hours.
During seclusion or restraints, staff is required to conduct 15-minute
checks to assess vital signs, any signs of injury, and the patient’s psycho-
logical state and readiness to discontinue seclusion or restraints. Staff
must clearly document in the medical record the need for seclusion, in-
tervening steps, and medications given (Hill and Petit 2002).
Restraints
The implementation of restraints is a difficult procedure, but an impor-
tant option, generally reserved for those situations in which there is the
potential for imminent harm to patient or staff through patients’ behav-
ior and where other interventions of a lesser degree of intensity and re-
strictiveness have been unsuccessful. The specific definition by the Joint
Commission for the Accreditation of Healthcare Organizations (JCAHO;
2002, p. 123) for use of restraint is “a direct application of force without
permission to restrict freedom of movement.”
Once the decision to use restraint is made, the overriding principle
is that it be done swiftly and humanely and that the patient be reassured
that it is done in his or her best interest. As is the case for seclusion, only
staff members trained in applying restraints are allowed to use them.
The implementation of physical restraints is a dangerous procedure,
both for staff and patient. It should never be attempted unless there is
sufficient staff present to ensure that it can be done with a minimum of
struggle. Using at least five staff members is recommended, one for each
limb and an extra person as team leader. The presence of a critical group
of staff may also assist in calming the patient, thus aborting the need for
restraints. Once the decision is made to proceed, implementation must
be completed and negotiations temporarily suspended. The team
leader, just as in team resuscitation, oversees the staff and ensures safety.
It is usually best that the physician avoid physical participation in sub-
duing a combative patient because this may interfere with the therapeu-
tic relationship. The same time limitations are placed on restraints as
they are on seclusion. During restraints, staff is required to conduct 15-
minute checks to assess vital signs, any signs of injury, and the patient’s
psychological state and readiness to discontinue seclusion or restraints.
If possible, the patient or family should be provided with an ongoing ex-

planation of the reasons for the procedure and what to expect.
After a seclusion or restraint episode is resolved, a debriefing note
needs to be added to the medical record that documents precipitants to
the incident and alternative treatments, modifications of the treatment
plan, and the patient’s psychological and physical well-being after the
intervention. Debriefing is also important to ascertain any trauma to
staff as a result of the intervention and the offering of support to staff
after a serious violent incident (Lindenmayer et al. 2002).
JCAHO’s (2002, p. 123) recommendations describe procedures of se-
clusion and restraints as “aversive experience with potential for serious
physical and emotional consequences including death.” Organizations
are “required to continually explore ways to decrease and eliminate use
through training, leadership commitment and performance improve-
ment.” (See also Chapter 17, “Seclusion and Restraint.”)
Case Example 4:
Seclusion in Emergency Psychiatric Services
Initial assessment. A 31-year-old male university graduate was es-
corted by police officers involuntarily because of delusional thinking
and aggressive behavior during the previous 24 hours. The patient vio-
lated a restraining order from his ex-wife, whom he had harassed and
threatened to physically assault. He had bizarre and rigid thinking that
was sexually inappropriate; he was noted to walk 10–25 miles daily ex-
posing himself to others. The initial screening evaluation completed by
the emergency room triage nurse indicated that he was disheveled and
showed grandiose delusions.
Medical evaluation. The patient did not have any significant medi-
cal history, and he had not been in psychiatric treatment. The neurolog-
ical examination was normal. There was no history of alcohol or illicit
drug use. Blood and urine tests were normal.
Psychiatric assessment. Mental status examination showed the pa-
tient to be disheveled and to have marked flight of ideas and belliger-
ence. The most striking features were grandiose delusional thinking and
significant paranoia centering on the police in his community. His affect
was inappropriate during the interview; at times he would laugh when
asked a serious question and at times he would cry. He was fully ori-
ented but would not cooperate with formal mental status testing. The
initial diagnosis was acute mania with psychosis.
Patient behavior. Upon admission to the emergency room, the pa-
tient was extremely uncooperative, verbally threatening staff members
and other patients and making sexually explicit remarks to a female
nurse, with occasional verbal outbursts. During the time of his out-
bursts, less restrictive methods of modifying the patient’s behavior,
such as “talking him down,” had failed. The doctrine of “the least
restrictive method of restraint” was employed. The patient was pro-

vided with options for modifying his behavior. Four emergency room
nursing aides were in clear view of the patient but remained 10 feet dis-
tant. The emergency room psychiatrist spoke to the patient in a firm but
nonthreatening voice, stating that the continuation of the patient’s un-
controlled and disruptive behavior would not be allowed and that the
patient would be restrained by staff unless he cooperated with the med-
ical and psychiatric staff. He was told that he could choose whether
he wanted to be restrained or secluded as a result of his behavior. He
agreed to seclusion over physical restraints. He also received emergency
intramuscular haloperidol. He was calmed by the medication to some
extent, as well as by a brief seclusion episode of 15 minutes. He then al-
lowed further diagnostic evaluation.
Suicidal Risk
Suicide and suicide attempts are among the most serious outcomes of
psychiatric illness, and the most extreme intervention (e.g., involuntary
hospitalization) may have to be used if these events are at high risk. The
national rate of suicide has remained fairly consistent at 1.1%–1.4%.
Suicidal ideation and behavior are the most common presenting com-
plaints of patients seeking treatment at psychiatric emergency facilities,
and these patients are at a considerable risk of subsequent suicide
(Dhossche 2000). Substance use disorders have been consistently recog-
nized as chronic risk factors for suicide (Pages et al. 1997). Patients pre-
senting to the emergency room with complaints of suicidal ideation or
suicidal command hallucinations, or presenting after a suicide attempt,
have to be carefully evaluated. Both state-related risk factors, such as
ideas of hopelessness or worthlessness, and trait-related risk factors,
such as age or a previous serious suicide attempt, need to be fully as-
sessed. The intensity of current suicidal ideation should be explored,
and the presence of protective factors against suicidal acting out should
also be assessed. The result of this in-depth evaluation will dictate the
final treatment decision.
Case Example 5:
Use of Restraints for Suicidal Ideation
in Emergency Service Settings
Initial assessment. Mr. R is a 50-year-old man with a long history of re-
current depressions and multiple hospitalizations dating back to age 19.
The patient had been discharged from an inpatient psychiatric facility
3 months earlier and voluntarily presented himself to the hospital emer-
gency room indicating that his medication was not helping him and that
he was having frequent thoughts of wanting to hurt himself. The patient
had a history of three suicide attempts in the past: at age 29 he took an

overdose of hypnotics, at age 38 he attempted to overdose on a combi-
nation of medications, and he made a final attempt at age 49, when he
lacerated his arm with a pin. The patient was followed at the local out-
patient department by a psychiatrist and a social worker on a monthly
basis and was receiving antidepressant medication. However, he had
not visited the clinic since his hospital discharge.
Psychiatric evaluation. Mr. R was assessed by the psychiatrist and
diagnosed with acute depression with suicidal ideation. Mr. R indicated
that he was feeling tense and depressed and was considering hurting
himself by hitting his head on the bathroom wall. The patient was eval-
uated on the InterSept Scale for Suicidal Thinking (Lindenmayer et al.
2003) and showed a score of 18 (out of a total of 24). During this time the
patient was monitored one to one. Mr. R later also reported that he was
hearing command voices telling him to hurt himself. During the psychi-
atric evaluation, Mr. R began hitting his head on the desk and saying he
no longer wanted to live. He became extremely agitated. Hospital emer-
gency staff immediately tried to restrain him; however, he broke loose
and proceeded to hit his head against the wall. The staff utilized wrist-
to-belt restraints, and Mr. R was given intramuscular ziprasidone 20 mg
and lorazepam 4 mg for severe agitation secondary to his psychotic de-
pressive disorder. He was then maintained on one-to-one monitoring
for suicidal behavior and ideation. After approximately 24 hours he was
able to calm down. He was admitted to the inpatient unit and continued
on one-to-one observation.
Key Points
■ Psychiatric emergency services (PESs) comprise a large spectrum
of acute psychiatric service delivery systems that are available for
the assessment, acute stabilization, and initial treatment of
mental health emergencies.
■ PESs usually function on a 24-hour-a-day basis and provide
extremely important clinical services.
■ The psychiatric emergency room is traditionally the main venue
for the delivery of emergency services; however, PESs cover exten-
sive and comprehensive mental health delivery systems and do
not function in isolation.
■ Such delivery systems are critical to prevent unnecessary hospital-
izations that might, in turn, foster chronicity and dependence on
institutional care.
■ The crisis situation leading up to an emergency presentation by
a patient is rarely completely resolved after evaluation and treat-
ment in the PES.
■ Often, the services provided by PESs represent patients’ entry

portal or referral place for longer-term care in an appropriate
inpatient or outpatient setting where various psychiatric and
social supportive services can be delivered and contribute to
the resolution of the crisis.
References
Allen MH: Level 1 psychiatric emergency services: the tools of the crisis sector.
Psychiatr Clin North Am 22:713–734, 1999
Allen MH (ed): Emergency Psychiatry. Washington, DC, American Psychiatric
Publishing, 2002
Bassuk EL, Gerson S: Into the breach: emergency psychiatry in the general hos-
pital. Gen Hosp Psychiatry 1:31–45, 1979
Breslow RE, Klinger BI, Erickson BJ: Characteristics of managed care patients in
a psychiatric emergency service. Psychiatr Serv 47:1259–1261, 1996
Broadbent M, Jarman H, Berk M: Emergency department mental health triage
scales improve outcomes. J Eval Clin Pract 10:57–62, 2004
Buckley PF: Dualism in psychiatry. Lancet 343:1102, 1994
Coristine RW, Hartford K, Vingilis E, et al: Mental health triage in the ER: a qual-
itative study. J Eval Clin Pract 13:303–309, 2007
Dhossche DM: Suicidal behavior in psychiatric emergency room patients. South
Med J 93:310–314, 2000
Dreyfus JK: Nursing assessment of the ED patient with psychiatric symptoms:
a quick reference. J Emerg Nurs 13:278–282, 1987
Ellison J, Blum NR: Repeat visitors in the psychiatric emergency service: a crit-
ical review of the data. Hosp Community Psychiatry 37:37–41, 1986
Gerson S, Bassuk EL: Psychiatric emergencies: an overview. Am J Psychiatry
137:1–11, 1980
Happell B, Summers M, Pinikahana J: The triage of psychiatric patients in the
hospital emergency department: a comparison between emergency depart-
ment nurses and psychiatric nurse consultants. Accid Emerg Nurs 10:65–
71, 2002
Herz MI, Endicott J, Spitzer RL: Brief hospitalization: a two-year follow-up.
Hill SL, Petit J: The violent patient. Emerg Med Clin North Am 18:301–315, 2002
Joint Commission on the Accreditation of Healthcare Organizations: Hospital
Accreditation Standards. Oakbrook Terrace, IL, Joint Commission Re-
sources, 2002
Kooiman CG, Van de Wetering BJ, Van der Mast RC: Clinical and demographic
characteristics of emergency department patients in The Netherlands: a re-
view of the literature and a preliminary study. Am J Emerg Med 7:632–638,
1989
Lindenmayer JP, Crowner M, Cosgrove V: Emergency treatment of agitation
and aggression, in Emergency Psychiatry. Edited by Allen MH. Washing-
ton, DC, American Psychiatric Publishing, 2002, pp 115–149
Lindenmayer JP, Czobor P, Alphs L, et al: The InterSePT scale for suicidal think-
ing reliability and validity. Schizophr Res 63:161–170, 2003
Marson DC, McGovern MP, Pomp HC: Psychiatric decision making in the
emergency room: a research overview. Am J Psychiatry 145:918–925, 1988
McClelland PA: The emergency psychiatric system. Psychiatr Clin North Am
6:225–232, 1983
McDonough S, Wynaden D, Finn M, et al: Emergency department mental
health triage consultancy service: an evaluation of the first year of the ser-
vice. Accid Emerg Nurs 12:31–38, 2004
Murdach AD: Decision making in psychiatric emergencies. Health Soc Work
12:267–274, 1987
Oldham JM, DeMasi ME: An integrated approach to emergency psychiatric
care. New Dir Ment Health Serv 67:33–42, 1995
Osborne J: Mental health triage. Emerg Nurs 11:14, 2003
Oyewumi LK, Odejide O, Kazarian SS: Psychiatric emergency services in a Ca-
nadian city, I: prevalence and patterns of use. Can J Psychiatry 37:91–95,
1992
Pages KP, Russo JE, Roy-Byrne PP, et al: Determinants of suicidal ideation: the
role of substance use disorders. J Clin Psychiatry 58:510–517, 1997
Rhine MW, Mayerson P: Crisis hospitalization within a psychiatric emergency
service. Am J Psychiatry 127:1386–1391, 1971
Schneider SE, Ross IM: Ultra-short hospitalization for severely mentally ill pa-
tients. Psychiatr Serv 47:137–138, 1996
Sledge WH, Tebes J, Rakfeldt J, et al: Day hospital/crisis respite care versus in-
patient care, part I: clinical outcomes. Am J Psychiatry 153:1065–1073, 1996
Smart D, Pollard C, Walpole B: Mental health triage in emergency medicine.
Aust N Z J Psychiatry 33:57–69, 1999
Solomon P, Davis J: Meeting community service needs of discharged psychiatric
patients. Psychiatr Q 57:11–17, 1985
Solomon P, Gordon B: The psychiatric emergency room and follow-up services
in the community. Psychiatr Q 58:119–127, 1986–1987
Stroul BA: Psychiatric Crisis Response Systems: A Descriptive Study. Rockville,
MD, Center for Mental Health Services, National Institutes of Mental
Health, 1993
Summers M, Happell B: The quality of psychiatric services provided by an Aus-
tralian tertiary hospital emergency department: a client perspective. Accid
Emerg Nurs 10:205–213, 2002
Sundheim ST, Ryan RM: Amnestic syndrome presenting as malingering in a
man with developmental disability. Psychiatr Serv 50:966–968, 1999
Vaslamatzis G, Kontaxakis V, Markidis M, et al: Social and resource factors re-
lated to the utilization of emergency psychiatric services in the Athens area.
Acta Psychiatr Scand 75:95–98, 1987
Vigiser D, Apter A, Aviram U, et al: Overutilization of the general hospital
emergency room for psychiatric referrals in an Israeli hospital. Am J Public
Health 74:73–75, 1984
Weisman G, Feirstein A, Thomas C: Three-day hospitalization: a model for in-
tensive intervention. Arch Gen Psychiatry 21:620–629, 1969
Wynaden D, Chapman R, McGowan S, et al: Emergency department mental
health triage consultancy service: a qualitative evaluation. Accid Emerg
Nurs 11:158–165, 2003
P A R T I V
Treatment
and Management
C H A P T E R 1 5
Psychopharmacology and
Electroconvulsive Therapy
T he use of medications and other somatic treatments in the care of

patients who exhibit violent behavior is complex. Prior chapters have
outlined the different diagnostic entities that may be associated with vi-
olent behavior. In addition, comorbidity is common: patients with
schizophrenia or bipolar disorder can also have a substance use disorder
or a personality disorder. There may be an additional superimposed non-
psychiatric medical disorder that has been overlooked. Careful attention
to the differential diagnosis will lead to the formation of a list of the med-
ication approaches most likely to be effective in treating the underlying
core disorder—and also in ameliorating the violent behavior, if it is a con-
sequence of the disorder. If this approach fails, consideration should be
given to the possibility of a missed diagnosis or of having underesti-
mated the importance of a secondary diagnosis as a cause of the noxious
behavior. Problems can also arise when the indicated treatment for the
underlying disorder has adverse behavioral consequences. An example
of this would be akathisia secondary to the use of first-generation anti-
psychotics. Akathisia can be mistaken for anxiety and agitation, resulting
301
in an increase in the dosage of the antipsychotic, thus leading to more se-

vere akathisia, which can ultimately result in violent behavior.
In acute emergencies, where the goal is the rapid resolution of agitated
behavior, there are many different effective pharmacological interven-
tions. A careful differential diagnosis is still important—for example, a pa-
tient whose violent behavior is secondary to alcohol withdrawal would be
better served by a drug treatment that has cross-tolerance with alcohol.
This chapter discusses the options currently available for the short-term
management of agitation, including the newly available intramuscular
formulations of second-generation antipsychotics. Emphasis is placed on
these agents particularly because they are relatively new and have re-
ceived regulatory approval in many countries for the specific indication of
agitation associated with schizophrenia and/or bipolar mania and thus
are being marketed extensively by their manufacturers for this purpose.
Once the acute episode is safely managed, longer-term treatment is
necessary to decrease the likelihood of future episodes and to diminish
the intensity of future outbursts should they still occur. Here, attention
to the underlying disorder (or disorders) is crucial in achieving this
goal. Addressing all comorbid conditions and environmental stressors
is essential.
This chapter emphasizes data from randomized studies, preferably
double blind. Where such information is not available, references to
naturalistic studies or case series are judiciously made.
Case Example
John is a 40-year-old white male, diagnosed at different times as having
schizophrenia, bipolar disorder, or schizoaffective disorder. His first
psychiatric hospitalization was at age 18 when he was brought to the
hospital emergency department in an acute psychotic state with para-
noid delusions. He has had several known hospitalizations since then,
the most recent after he was arrested for assault. In the emergency de-
partment he was acutely agitated and required intramuscular medica-
tion to control his behavior. He was given an intramuscular injection of
haloperidol 5 mg combined with lorazepam 2 mg in the same syringe.
After 45 minutes he appeared calmer, but he complained of stiffness in
his neck and tongue and was drooling. He received diphenhydramine
50 mg intramuscularly for this dystonic reaction, followed by oral benz-
tropine 2 mg. Toxicology screen in the emergency department came
back positive for cocaine and cannabis.
John was admitted to the psychiatric inpatient unit and refused all
oral medications, saying he was “allergic to Haldol” and that “nothing
really works.” On the second day of hospitalization he asked to smoke
a cigarette, and when told the unit was now “smoke free,” he began
shouting at the staff, threatened to sue them, and made a fist. Because
Psychopharmacology and Electroconvulsive Therapy ❘ 303
he was not taking any oral medications, a decision was made to give
him a “stat” dose of ziprasidone 20 mg intramuscular. After the injec-
tion, John was substantially calmer. He was able to articulate that he felt
less anxious, and he acknowledged he did not feel any restlessness or
stiffness that he usually felt after receiving intramuscular haloperidol.
He agreed to continue taking the “new” medication.
Despite the initial success with ziprasidone, John continued to feel
paranoid and uncomfortable. He was unable to sleep. He was subse-
quently prescribed a number of different other antipsychotics, includ-
ing quetiapine and olanzapine. Although he was free of any extrapyra-
midal side effects, symptom relief was incomplete and he continued to
be intermittently agitated, often with little provocation. John was ulti-
mately placed on clozapine, which did decrease the frequency of his
outbursts. John’s aggression became exclusively verbal, never physical,
and he was more easily reassured. Adjunctive valproate and adjunctive
lamotrigine treatment were also attempted but did not make a substan-
tial difference in his impulsivity.
John’s diagnostic history is confusing: it is unclear whether he has

schizophrenia or bipolar disorder. The diagnostic uncertainty is accen-
tuated by his comorbid substance use and nonadherence to medication
treatments. John presents the clinician with two distinct problems: 1)
how to best manage his acute behavioral dyscontrol and 2) how to pre-
vent these behavioral problems from happening again. John is an expe-
rienced patient; he knows what has not worked in the past and is keenly
aware of the discomforts he has had with different medications. This
plays into his paranoid outlook on life and makes establishing a thera-
peutic alliance a significant challenge for the clinician treating him. For-
tunately for John, there are new pharmacological tools to use that were
not available when he first became ill. For acute emergencies there are
medications that are better tolerated than haloperidol, and for longer-
term use there are medications that have greater efficacy.
Medications for Psychiatric Emergencies

Psychiatric emergencies such as acute states of agitation and overt ag-
gressive behavior are commonly treated with the use of sedating
agents. These treatments have evolved over the years from the use of
agents such as sodium amytal to the administration of benzodiazepines
and antipsychotics (often simultaneously). Intramuscular administra-
tion of medications yields higher maximum plasma concentrations
than that achieved with oral formulations. Moreover, these maximum
concentrations are reached much more quickly with intramuscular for-
mulations. Table 15–1 outlines several intramuscular options for the
pharmacological treatment of acute agitation.
304
TABLE 15–1. Selected intramuscular medications for the treatment of acute agitation
Half-life in
❘
Agent Dose, mg nonelderly adults, h Comments
Lorazepam 0.5–2.0 10–20 The only benzodiazepine that is reliably absorbed intramuscularly.
Useful for symptoms of alcohol withdrawal.
Haloperidol 5–10 12–36 The most commonly used anti-agitation intramuscular antipsychotic,

but associated with substantial risk for acute dystonia, akathisia, and
tremor. Other first-generation antipsychotics are also available in
intramuscular preparations but are associated with hypotension
(chlorpromazine) and a decrease in seizure threshold (all).
Droperidol 2.5–5 2 Association with QT prolongation has led to its removal from the U.K.
market and a “black box” warning in U.S. labeling.
Ziprasidone 10–20* 2.2–3.4 Lower risk of extrapyramidal adverse events than first-generation
agents.
Olanzapine 10* 34–38 Lower risk of extrapyramidal adverse events than first-generation
agents.
Aripiprazole 9.75* 75 Lower risk of extrapyramidal adverse events than first-generation
agents.
*Recommended dose in U.S. product labeling
Head-to-head double-blind comparisons of these intramuscular

agents for the treatment of agitation are not generally available, with
the exception of recent registration studies (see Table 15–2) comparing
olanzapine or aripiprazole with either haloperidol (in patients with
schizophrenia) or lorazepam (in patients with bipolar mania). A quan-
titative review of these registration studies can be found elsewhere (Cit-
rome 2007). In terms of combinations, the extant controlled evidence is
one three-arm study that compared haloperidol intramuscular 5 mg,
lorazepam intramuscular 2 mg, or both haloperidol and lorazepam in
combination, in 98 psychotic and agitated patients (Battaglia et al.
1997).
Sodium amytal entails the risk of respiratory depression, and es-
sentially this agent has been supplanted by lorazepam. Although
lorazepam also can result in a decrease in respiratory drive, particularly
in patients with a history of lung disease or sleep apnea, it is otherwise
relatively well tolerated. Advantages include its relatively short half-
life, lack of active metabolites, and cross-tolerance to alcohol, making it
the dominant choice in patients whose agitation is secondary to acute
alcohol withdrawal. Disadvantages include the potential for behavioral
disinhibition that may paradoxically increase agitation (Dietch and Jen-
nings 1988). Another disadvantage to using intramuscular lorazepam is
its lack of substantial antipsychotic effect; thus lorazepam is inadequate
in treating any underlying core psychotic disorder. Long-term use of a
benzodiazepine will also result in physiological tolerance, leading to
potential rebound anxiety or agitation in between doses or when doses
are missed. Abrupt withdrawal of benzodiazepines is associated with a
risk for epileptic seizures.
Given the limitations of intramuscular lorazepam for patients with
schizophrenia or bipolar disorder, intramuscular antipsychotics may be
preferred for treating acute agitation. The combination of intramuscular
haloperidol and lorazepam is commonly used, with the rationale that
combining these agents will improve the sedative effect as well as de-
crease the likelihood of extrapyramidal adverse events such as acute
dystonia, akathisia, or tremor. Our patient in the above case example re-
ceived the combination of haloperidol and lorazepam and had marked
reduction in agitation but developed an acute dystonic reaction that re-
quired additional interventions, including an additional injection. This
complicated course in the emergency department set him up to be
overtly noncompliant with medications once admitted to the psychiat-
ric inpatient unit. There are now three second-generation antipsychot-
ics available in rapidly acting intramuscular formulations that can be
considered. All have a lower propensity for extrapyramidal adverse
306
TABLE 15–2. Registration studies for intramuscular formulations of second-generation antipsychotics
Results vs. placebo or
❘
Reference Agent and indication N Study arms (N) placebo equivalent
Lesem et al. 2001 Ziprasidone; 117 Ziprasidone 2 mg (54), 10 mg (63) 10 mg superior on Behavioral Activity
schizophrenia Rating Scale at 0–2 hours, but not by
Clinical Global Impression–Severity

Daniel et al. 2001 Ziprasidone; 79 Ziprasidone 2 mg (38), 20 mg (41) 20 mg superior on Behavioral Activity
schizophrenia Rating Scale at 0–4 hours and by Clinical
Global Impression–Severity at 4 hours
Breier et al. 2002 Olanzapine; 270 Olanzapine 2.5 mg (48), 5 mg (45), All doses of olanzapine superior to placebo
schizophrenia 7.5 mg (46), 10 mg (46); on PANSS-EC; effect larger and more
haloperidol 7.5 mg (40); consistent for 5, 7.5, and 10 mg
placebo (45)
Wright et al. 2001 Olanzapine; 311 Olanzapine 10 mg (131); Olanzapine superior to placebo on
schizophrenia haloperidol 7.5 mg (126); PANSS-EC
placebo (54)
Meehan et al. 2001 Olanzapine; bipolar 201 Olanzapine 10 mg (99); Olanzapine superior to placebo on
manic or mixed lorazepam 2 mg (51); PANSS-EC
placebo (51)
Meehan et al. 2002 Olanzapine; dementia* 272 Olanzapine 2.5 mg (71), 5 mg (66); Both olanzapine doses superior to placebo
lorazepam 1 mg (68); on the PANSS-EC
placebo (67)
Psychopharmacology and Electroconvulsive Therapy
TABLE 15–2. Registration studies for intramuscular formulations of second-generation antipsychotics (continued)
Results vs. placebo or
Reference Agent and indication N Study arms (N) placebo equivalent
Andrezina et al. 2006a Aripiprazole; 448 Aripiprazole 9.75 mg (175); Aripiprazole superior to placebo on
schizophrenia haloperidol 6.5 mg (185); PANSS-EC
placebo (88)
Tran-Johnson et al. Aripiprazole; 357 Aripiprazole 1 mg (57), 5.25 mg All but the 1-mg dose of aripiprazole
2007 schizophrenia (63), 9.75 mg (57), 15 mg (58); were superior to placebo on PANSS-EC
haloperidol 7.5 mg (60);
placebo (62)
Zimbroff et al. 2007 Aripiprazole; bipolar, 301 Aripiprazole 9.75 mg (78), 15 mg Both doses of aripiprazole superior to
manic or mixed (78); lorazepam 2 mg (70); placebo on PANSS-EC
placebo (75)
Note. PANSS-EC=Positive and Negative Syndrome Scale, Excited Component.
*Not FDA approved for this indication.
❘
307
effects compared with the older antipsychotics and are discussed in the
following paragraphs.
Ziprasidone
Ziprasidone mesylate was approved in 2002 by the U.S. Food and Drug
Administration (FDA) for the indication of acute agitation in patients
with schizophrenia, on the basis of two 1-day, double-blind trials
(Daniel et al. 2001; Lesem et al. 2001) of agitated hospitalized subjects
with a primary diagnosis of schizophrenia, schizoaffective disorder, bi-
polar disorder with psychotic features, delusional disorder, or psy-
chotic disorder not otherwise specified (DSM-IV-TR; American Psychi-
atric Association 2000). Approximately 80% of the subjects had
schizophrenia or schizoaffective disorder. Doses tested were ziprasi-
done 10 mg versus 2 mg (Lesem et al. 2001) and 20 mg versus 2 mg
(Daniel et al. 2001). There was no placebo arm, per se, nor were active
comparators such as haloperidol or lorazepam used. The 2-mg dose of
ziprasidone can be considered as a placebo-equivalent. The 20-mg dose
yields a higher percentage of responders and a greater degree of re-
sponse in terms of reduction of agitation than the 10-mg dose; however,
product labeling recommends the range of 10–20 mg per injection.
Safety concerns specific to intramuscular ziprasidone, as noted in
product labeling, include caution in patients with impaired renal func-
tion because the cyclodextrin excipient is cleared by renal filtration. Be-
cause of ziprasidone’s dose-related prolongation of the QT interval and
the known association of fatal arrhythmias with QT prolongation by
some other drugs, ziprasidone is contraindicated in patients with a
known history of QT prolongation (including congenital long QT syn-
drome), recent acute myocardial infarction, or uncompensated heart
failure. However, more than 5 years of clinical availability has not re-
sulted in evidence that ziprasidone by itself poses a substantial clinical
problem in this regard (Zimbroff et al. 2005). Comparative intramuscu-
lar antipsychotic data on QTc are available; the product information
(Pfizer 2005) includes details of a study evaluating the QTc-prolonging
effect of intramuscular ziprasidone, with intramuscular haloperidol as a
control, and reveals a mean increase in QTc from baseline for ziprasi-
done of 4.6 msec following the first injection and 12.8 msec following the
second injection, compared with 6.0 msec and 14.7 msec for haloperidol,
and with no patients having had a QTc interval exceeding 500 msec.
The patient in the case example did not have any history of cardiac
conduction problems. When the need developed for an intramuscular
injection of an anti-agitation medication, ziprasidone was selected over
the combination of haloperidol and lorazepam because of his past ad-

verse experience with an acute dystonic reaction. John did not develop
akathisia or any other extrapyramidal symptoms after the injection. Be-
cause of the improved immediate tolerability over haloperidol, he
agreed to continue an oral preparation of this “new” medication.
Olanzapine
Olanzapine was approved in 2004 by the FDA for the indication of agi-
tation associated with schizophrenia and bipolar I mania, on the basis
of three 1-day, placebo-controlled inpatient trials with active compara-
tors (Breier et al. 2002; Meehan et al. 2001; Wright et al. 2001). A fourth
pivotal trial was done in patients age 55 or older with agitation asso-
ciated with dementia, though regulatory approval was not pursued
(Meehan et al. 2002). Superior onset of efficacy for intramuscular olan-
zapine 10 mg was demonstrated compared with intramuscular halo-
peridol 7.5 mg in patients with schizophrenia (Wright et al. 2001) and
intramuscular lorazepam 2 mg in patients with bipolar mania (Meehan
et al. 2001). In the bipolar trial, olanzapine was superior to lorazepam at
all time points up to and including 2 hours postinjection. In the schizo-
phrenia trial examining olanzapine 10 mg (Wright et al. 2001), olanza-
pine was superior to haloperidol at 15, 30, and 45 minutes postinjection.
In the study comparing multiple fixed doses of intramuscular olanza-
pine with intramuscular haloperidol 7.5 mg (Breier et al. 2002), patients
treated with 5.0, 7.5, or 10.0 mg of olanzapine had greater mean im-
provement in agitation than those given placebo at all time points, but
the groups given 2.5 mg of olanzapine or haloperidol did not show
greater mean improvement compared with those given placebo until
60 minutes after the first injection. In the pivotal trials, no adverse event
was significantly more frequent for intramuscular olanzapine com-
pared with intramuscular haloperidol or intramuscular lorazepam. The
recommended dose in product labeling is 10 mg (with lower doses of
2.5–5.0 mg for vulnerable patients such as the elderly or medically in-
firm) (Eli Lilly 2006).
Safety concerns specific to intramuscular olanzapine, as noted in
product labeling, include hypotension, bradycardia with or without hy-
potension, tachycardia, and syncope as reported during the clinical tri-
als. As per the product label, patients should remain recumbent if
drowsy or dizzy after injection until examination has indicated that
they are not experiencing postural hypotension, bradycardia, and/or
hypoventilation. Simultaneous injection of olanzapine intramuscular
and parenteral benzodiazepines is not recommended. Data from the
first 21 months of post-marketing safety experience with olanzapine in-

tramuscular were presented in a poster (Sorsaburu et al. 2006) in which
29 fatalities were reported among an estimated worldwide patient ex-
posure of 539,000. The fatalities were complicated by multiple concom-
itant medications, including benzodiazepines or other antipsychotics,
and medically significant risk factors.
Aripiprazole
Aripiprazole intramuscular was approved by the FDA in late 2006 for
the indication of agitation associated with schizophrenia or bipolar ma-
nia, on the basis of three 1-day, placebo-controlled inpatient trials with
active comparators (Andrezina et al. 2006a; Tran-Johnson et al. 2007;
Zimbroff et al. 2007). The schizophrenia studies utilized haloperidol in-
tramuscular as an active comparator, with mixed results in terms of rel-
ative efficacy. In the study comparing intramuscular aripiprazole
9.75 mg with intramuscular haloperidol 6.5 mg (Andrezina et al. 2006a),
analysis according to the non-inferiority hypothesis indicated that ari-
piprazole was non-inferior to haloperidol. However, for the aripiprazole
group, decrease in agitation differed significantly from placebo at 1 hour
after the first injection, whereas a significant difference was achieved at
45 minutes in the haloperidol group. There was no significant difference
in the improvement in the agitation scores between the aripiprazole and
haloperidol groups at these time points, nor at 30 minutes or 2 hours;
however, the difference at 90 minutes was significant in favor of halo-
peridol (P = 0.022). Aripiprazole performed somewhat better in the
study that compared multiple fixed doses of intramuscular aripiprazole
with intramuscular haloperidol 7.5 mg (Tran-Johnson et al. 2007). In that
study, changes in agitation scores were statistically significant as early as
45 minutes for the aripiprazole 9.75-mg group, whereas a significant dif-
ference between haloperidol and placebo was first seen at 105 minutes.
In the study comparing aripiprazole versus lorazepam and placebo in
agitated patients with bipolar disorder, lorazepam evidenced superior-
ity over placebo as early as 45 minutes after injection and aripiprazole at
60 minutes (Zimbroff et al. 2007). In product labeling, the usual recom-
mended dose is 9.75 mg (Bristol-Myers Squibb 2006).
Safety concerns specific to intramuscular aripiprazole, as noted in
product labeling, include greater sedation and orthostatic hypotension
with the combination of lorazepam and aripiprazole as compared with
that observed with aripiprazole alone.
The pivotal registration trials of the intramuscular formulations of
the second-generation antipsychotics suffer from the limitation that the
subjects were generally not as severely ill as some patients commonly

seen in clinical practice. Moreover, patients with comorbid medical con-
ditions and prescribed multiple psychotropic medications are generally
excluded from registration trials. In addition, the studies did not enroll
children or adolescents. Thus, generalizability of these studies may be
limited. Some information is now available in terms of naturalistic stud-
ies for ziprasidone (Preval et al. 2005) and olanzapine (San et al. 2006)
that enrolled more severely agitated patients than the registration stud-
ies. In the ziprasidone mesylate naturalistic study, 119 patients who pre-
sented to a psychiatric emergency department received either intra-
muscular ziprasidone 20 mg (n = 110) or conventional intramuscular
antipsychotics (n = 9). Ziprasidone was effective in reducing agitation
among patients with and without alcohol and substance abuse. In the
olanzapine naturalistic study, 92 patients attending psychiatric emer-
gency settings were enrolled, all receiving intramuscular olanzapine
10 mg; however, patients with active drug and alcohol use were
screened out. Olanzapine was effective in reducing agitation, with 96%
receiving a single injection and 4% receiving two. In a retrospective
chart review of 100 hospitalized patients younger than 18 years of age
treated with intramuscular ziprasidone or intramuscular olanzapine,
both agents resulted in similar reductions of agitation or aggression
(Khan and Mican 2006).
For the second-generation antipsychotics that are available in intra-
muscular form, several studies have been published that describe the
transition from intramuscular to oral administration for ziprasidone
(Brook et al. 2000, 2005; Daniel et al. 2004), olanzapine (Wright et al.
2003), and aripiprazole (Andrezina et al. 2006b).
For patients whose level of agitation does not mandate the use of in-
tramuscular medication, oral administration can be considered first line.
For patients who are actively refusing oral medication, such as our case
patient, this is not an option. Controlled clinical trials have been reported
on the use of risperidone liquid (Currier et al. 2004) and olanzapine tab-
lets (Baker et al. 2003) for patients with agitation. The risperidone trial
enrolled patients with schizophrenia, schizoaffective disorder, mania
with psychotic features, acute paranoid reaction, or delusional disorder.
The olanzapine trial enrolled patients with schizophrenia, schizoaffec-
tive disorder, schizophreniform disorder, or bipolar I disorder, manic or
mixed episode (not necessarily with psychotic features). Regarding
olanzapine, dosages that exceeded the product label recommended
maximum of 20 mg/day were found to be useful in the short-term man-
agement of agitation (Baker et al. 2003).
Medications for Long-Term Treatment

Long-term treatment of violent behavior is geared toward the preven-
tion of future episodes of agitation and aggression. For this goal to be
attained, future episodes must be reduced in both frequency and inten-
sity. If the aggressive behavior is secondary to uncontrolled psychosis,
treating these symptoms with an antipsychotic will have the desired ef-
fect in reducing aggressive behavior. However, violent behavior is often
multifactorial in origin, with contributing factors such as the influence
of street drugs, an underlying problem with poor impulse control, and
environmental triggers. Another level of complexity exists when the vi-
olent behavior is instrumental—that is, premeditated—and a means for
the aggressor to obtain an advantage of some sort. John, the patient in
our case example, exhibited aggression related to both an uncontrolled
psychotic disorder refractory to first-generation antipsychotics and
poor impulse control that persisted even during times he was free of
hallucinations and delusions. These characteristics make treatment
planning difficult. In an effort to address the psychosis, antipsychotics
have been prescribed with varying success based on efficacy and toler-
ability issues. To address the impulsivity, and possibly the substance
abuse, John received a trial of an anticonvulsant, valproate, with little
success. He ultimately was placed on clozapine, a second-generation
antipsychotic commonly reserved for patients with treatment-refractory
schizophrenia but also approved for the indication of recurrent suicidal
behavior in patients with schizophrenia. The available controlled evi-
dence is reviewed for these long-term treatment options.
Antipsychotics
In the United States several second-generation antipsychotics are avail-
able: clozapine, risperidone, olanzapine, quetiapine, ziprasidone, ari-
piprazole, and paliperidone. All seven are approved by the FDA for the
treatment of schizophrenia, and all, except for clozapine and paliperi-
done, are also approved for the treatment of bipolar mania. Second-
generation antipsychotics have also been used off-label for a variety of
conditions (Tremeau and Citrome 2006). Of special interest is the possi-
bility that these agents have specific anti-hostility effects, with cloza-
pine having the strongest evidence supporting this.
Clozapine’s usefulness for patients with aggressive behavior was
initially suggested by case series and retrospective studies in which a
reduction in the number of violent incidents and/or a decrease in the
use of seclusion or restraint was observed among inpatients once they
began clozapine treatment (Citrome et al. 2004b). The reductions in hos-

tility (Volavka et al. 1993) and aggression (Buckley et al. 1995) after clo-
zapine treatment were selective in the sense that they were (statistically)
independent of the general antipsychotic effects of clozapine. This was
confirmed in two double-blind, randomized clinical trials. The first was
a 14-week study that compared the specific anti-hostility effects of clo-
zapine with those of olanzapine, risperidone, or haloperidol in 157 in-
patients with schizophrenia or schizoaffective disorder (Citrome et al.
2001) and found that clozapine had significantly greater anti-hostility
effect than haloperidol or risperidone. This effect on hostility was
specific: it was independent of antipsychotic effect on delusional think-
ing, formal thought disorder, or hallucinations, and independent of se-
dation. Further analyses of these data, including measures of overt ag-
gression (Volavka et al. 2004), showed that patients exhibiting overt
aggression had less overall improvement of psychopathology but that
antipsychotic efficacy of clozapine was greatest in aggressive patients,
whereas the opposite was true for risperidone and olanzapine. A key
finding was that a therapeutic dosage of clozapine was necessary to
achieve superior effects on the frequency and the severity of overt ag-
gression. Because it can take many days to titrate clozapine to a thera-
peutic dosage, it is important not to terminate a clozapine trial prema-
turely. However, this study enrolled patients who were not necessarily
aggressive, which limits its generalizability. A second study was under-
taken that enrolled 110 patients who had been physically aggressive
and subsequently randomly assigned to receive double-blind cloza-
pine, olanzapine, or haloperidol for up to 12 weeks (Krakowski et al.
2006). Patients assigned to clozapine had statistically significant lower
endpoint aggression scores than patients assigned to either olanzapine
or haloperidol. Patients in the olanzapine group had statistically signif-
icant lower endpoint aggression scores than patients in the haloperidol
group. However, no differences were seen among the three groups in
terms of reduction of psychopathology as measured by the total Posi-
tive and Negative Syndrome Scale (PANSS) score, suggesting that clo-
zapine’s advantage was related to a specific anti-aggressive effect.
No other double-blind, randomized clinical trials are available that
report on the efficacy of other second-generation antipsychotics in re-
ducing aggressive behavior among patients specifically selected be-
cause of such behavior. In the absence of these studies, post hoc analyses
have been done using data gathered during other studies for risperi-
done (Czobor et al. 1995), olanzapine (Kinon et al. 2001), quetiapine
(Arango and Bernardo 2005; Chengappa et al. 2003), ziprasidone (Cit-
rome et al. 2006), and aripiprazole (Volavka et al. 2005). Results varied
from superiority to haloperidol (for risperidone, olanzapine, quetia-

pine, and ziprasidone) to equivalency to haloperidol (for aripiprazole)
in terms of anti-hostility or anti-aggressive effect. Compared with halo-
peridol, the second-generation antipsychotics were associated with
fewer extrapyramidal effects and thus were considered more tolerable
and overall more effective. Methodologies varied, however, and spe-
cific anti-hostility or anti-aggressive effect was not always determined
(for olanzapine [Kinon et al. 2001]) or was inconsistently demonstrated
(for quetiapine [Arango and Bernardo 2005; Chengappa et al. 2003]).
Anticonvulsants
Mood stabilizers such as lithium and anticonvulsants are extensively
used, including off-label use among patients with a diagnosis of schizo-
phrenia (Citrome et al. 2002). There is an expectation that adjunctive
mood stabilizers can reduce aggressive and impulsive behavior (Cit-
rome 1995). There are expert consensus guidelines suggesting the use of
adjunctive mood stabilizers in those with schizophrenia with agitation,
excitement, aggression, or violence (McEvoy et al. 1999), but the sup-
porting evidence for this indication is based almost entirely on uncon-
trolled studies and case reports. The most commonly used mood stabi-
lizer is valproate (Citrome et al. 2000, 2002). Our case patient had a trial
of adjunctive valproate, but it did not have a substantial impact on his
psychopathology or degree of impulsivity. A review of the use of val-
proate in violence and aggressive behaviors in a variety of diagnoses
(Lindenmayer and Kotsaftis 2000) did reveal a 77.1% response rate (de-
fined by a 50% reduction in target behavior) based on 17 reports (164
patients, approximately one-half with dementia). Double-blind con-
trolled studies that tested and support this are few in number but do in-
clude a varied array of diagnoses, including borderline personality dis-
order (Hollander et al. 2001, 2005), Cluster B personality disorders as a
group (Hollander et al. 2003), and children and adolescents with explo-
sive temper and mood lability (Donovan et al. 2000).
Positive symptoms were reduced with adjunctive valproate in a 28-
day double-blind, randomized study with olanzapine and risperidone
among 249 patients with an acute episode of schizophrenia (Casey et al.
2003). A post hoc secondary analysis from this study found that combi-
nation therapy with divalproex had significantly greater anti-hostility
effect at 3 days and at 7 days than antipsychotic monotherapy (P< 0.05),
as measured by the PANSS hostility item (Citrome et al. 2004a). The ef-
fect on hostility was statistically independent of antipsychotic effect on
other PANSS items that reflect delusional thinking, a formal thought
disorder, or hallucinations. Adequate dosing may be important and

may explain why valproate 480 mg/day did not differentiate from pla-
cebo on measures of aggressive-behavior patients with dementia (Sival
et al. 2002). Other negative data come from an 84-day study that failed
to replicate the 28-day double-blind, randomized study with olanza-
pine and risperidone (Abbott Laboratories 2007) and from a random-
ized, open-label label study of risperidone with and without valproate
that did not demonstrate an advantage for combination therapy on psy-
chopathology rating scales or measures of aggression (Citrome et al.
2007).
The strategy of adding lamotrigine to antipsychotics was supported
by promising results from a double-blind trial of adjunctive lamotrigine
in patients with treatment-refractory schizophrenia unresponsive to
clozapine monotherapy (Tiihonen et al. 2003). Although specific effect
on hostility was not reported, improvement in positive symptoms was
seen. Subsequent studies have not been encouraging; the usefulness of
adjunctive lamotrigine in patients with schizophrenia was not sup-
ported by two large trials undertaken by its manufacturer (GlaxoSmith-
Kline 2005, 2006; Goff et al. 2007).
Evidence supporting the use of carbamazepine for persistent ag-
gressive behavior is limited (Volavka 2002), with the largest random-
ized clinical trial of carbamazepine failing to detect a significant im-
provement on the total Brief Psychiatric Rating Scale but showing
differences in suspiciousness, uncooperativeness, and excitement
(Okuma et al. 1989). Studies have also been done in nursing home pa-
tients with agitation and dementia (Tariot et al. 1998), with significant
short-term efficacy of carbamazepine for agitation and aggression and
with generally good safety and tolerability.
In the absence of mania, lithium may not be efficacious in reducing
aggressive behavior, as evidenced in a study in which lithium was
added to antipsychotics for the treatment of patients with resistant
schizophrenia who were classified as “dangerous, violent or criminal”
(Collins et al. 1991). However, there are case reports of lithium being
helpful in cases of akathisia among patients with schizophrenia (Shalev
et al. 1987). There are also case reports of patients with paranoid schizo-
phrenia with aggressive or disorderly behaviors who responded to the
addition of lithium to their antipsychotic treatment, deteriorated after
the lithium was discontinued, and subsequently improved when it was
reinstituted (Prakash 1985). In another population, lithium treatment
reduced the number of violent infractions in 66 nonpsychotic, impul-
sively aggressive prisoners in a double-blind, placebo-controlled study
(Sheard et al. 1976).
Other Medication Approaches

The use of β-adrenergic blockers has been reported in randomized clinical
trials in several different disease states to reduce violent behavior (Vol-
avka 2002). Although propranolol has been the agent most studied, others
such as nadolol may be simpler to titrate (starting at 40 mg at bedtime and
then up to 80 mg and 120 mg at bedtime over the span of several days, de-
pending on parameters such as heart rate and blood pressure). Other pos-
sible medication choices include serotonin-specific reuptake inhibitors, for
which one double-blind, randomized clinical trial in patients with schizo-
phrenia and aggressive behavior revealed an advantage with the adjunc-
tive use of citalopram (Vartiainen et al. 1995). This is consistent with an
open-label study of citalopram in patients with DSM-IV-TR Cluster B per-
sonality disorder or intermittent explosive disorder (Reist et al. 2003) and
a double-blind trial of fluoxetine in the treatment of impulsive aggressive
behavior in non–major depressed, non-bipolar or schizophrenic, person-
ality-disordered individuals (Coccaro and Kavoussi 1997).
The prolonged use of benzodiazepines for aggression and schizo-
phrenia is discouraged because of the problems with physiological tol-
erance and dependence. For example, missing scheduled doses of
lorazepam may result in withdrawal symptoms that can lead to agita-
tion or excitement as well as irritability and a greater risk for aggressive
behavior. Moreover, a controlled study of adjunctive clonazepam in pa-
tients with schizophrenia demonstrated no additional therapeutic ben-
efit, and several patients demonstrated violent behavior during the
course of clonazepam treatment (Karson et al. 1982).
The Role of Electroconvulsive Therapy

Adjunctive electroconvulsive therapy (ECT) may help individuals who
have inadequately responsive psychotic symptoms (Fink and Sackeim
1996), in particular, patients with persistent aggressive behavior. Con-
trolled studies have not been reported, but an open trial of ECT in com-
bination with risperidone in male patients with schizophrenia and ag-
gression resulted in a reduction in aggressive behavior for 9 of the 10
patients (Hirose et al. 2001).
Nonpsychotic Patients Who Exhibit

Violent Behavior
A nonpsychotic outpatient may present with episodic violent behavior.
A differential diagnosis and workup are required to rule out a somatic
cause of the aberrant behavior. Psychological testing may also be help-

ful in discerning the impact of a personality disorder. Intermittent
explosive disorder (IED) is also a possibility, and its lifetime prevalence
was noted to be 7.3% in a recent report of a nationally representative
sample in the United States (Kessler et al. 2006). Despite this high prev-
alence, there are no positive published double-blind, randomized clini-
cal trials of medication treatments for DSM-IV-TR–defined IED. Even
so, the use of the agents described earlier can be considered; for exam-
ple, the off-label use of clozapine has been described in nine adolescents
with IED (Kant et al. 2004). Another option is the use of serotonin-
specific reuptake inhibitors, for which case reports for IED can be found
(Feder 1999) as well as a double-blind clinical trial that enrolled patients
with a personality disorder and impulsive aggressive behavior (Coc-
caro and Kavoussi 1997).
Outpatients with personality disorders and impulsive aggressive
behavior may benefit from treatment with mood stabilizers such as val-
proate at mean modal dosages of about 1,500 mg/day (Hollander et al.
2003, 2005). Notably, the study that found improvement with valproate
versus placebo in Cluster B personality disorders did not find the same
advantage in the entire enrolled sample, which also included patients
with IED and posttraumatic stress disorder (Hollander et al. 2003).
Although long-acting benzodiazepines are sometimes used in an at-
tempt to manage aggressive behavior, there are no supporting con-
trolled clinical trials and many cautionary reports regarding behavioral
disinhibition.
Behavioral and psychotherapeutic interventions, including cogni-
tive-behavioral modification (Meichenbaum and Goodman 1971) and
dialectical behavior therapy (Linehan 1987), remain important compo-
nents in the treatment of aggressivity, particularly among nonpsychotic
individuals (Citrome et al. 2004b).
Key Points
■ Attempts to manage violent behavior by using medication geared
to the primary diagnosis may fail if the violent behavior is due to
an unidentified comorbid psychiatric or medical disorder.
■ Medication options for management of an acute episode of agita-
tion have expanded to include several different rapid-acting for-
mulations of second-generation antipsychotics. These agents are
less likely than first-generation antipsychotics to cause extrapyra-
midal side effects. Extrapyramidal symptoms especially relevant
in the emergency setting are akathisia (introduces an iatrogenic
cause for worsening of agitation) and acute dystonia (complicates
treatment course and impairs the therapeutic alliance).
■ Long-term treatment requires addressing all comorbidities. For
patients with schizophrenia, use of clozapine appears to be the
best option to decrease aggressivity. The evidence base for the
other second-generation antipsychotics is not as compelling as
that for clozapine. Adjunctive use of anticonvulsants, β-adrenergic
blockers, and serotonin-specific reuptake inhibitors can be consid-
ered, as can electroconvulsive therapy. The long-term use of ben-
zodiazepines is discouraged.
■ Psychotherapeutic approaches remain an important part of man-
aging patients with impulsive aggressive behavior, particularly for
outpatients with personality disorders, although the use of cer-
tain agents such as valproate and serotonin-specific reuptake
inhibitors shows promise in controlled clinical trials.
References
Abbott Laboratories: ABT-711 M02–547 clinical study report. Available at http://
www.clinicalstudyresults.org/documents/company-study_782_0.pdf. Ac-
Andrezina R, Josiassen RC, Marcus RN, et al: Intramuscular aripiprazole for the
treatment of acute agitation in patients with schizophrenia or schizoaffec-
tive disorder: a double-blind, placebo-controlled comparison with intra-
muscular haloperidol. Psychopharmacology (Berl) 188:281–292, 2006a
Andrezina R, Marcus RN, Oren DA, et al: Intramuscular aripiprazole or halo-
peridol and transition to oral therapy in patients with agitation associated
with schizophrenia: sub-analysis of a double-blind study. Curr Med Res
Opin 22:2209–2219, 2006b
Arango C, Bernardo M: The effect of quetiapine on aggression and hostility in

patients with schizophrenia. Hum Psychopharmacol 20:237–241, 2005
Baker RW, Kinon BJ, Maguire GA, et al: Effectiveness of rapid initial dose esca-
lation of up to forty milligrams per day of oral olanzapine in acute agita-
tion. J Clin Psychopharmacol 23:342–348, 2003
Battaglia J, Moss S, Rush J, et al: Haloperidol, lorazepam, or both for psychotic
agitation? A multicenter, prospective, double-blind, emergency depart-
ment study. Am J Emerg Med 15:335–340, 1997
Breier A, Meehan K, Birkett M, et al: A double-blind, placebo-controlled dose–
response comparison of intramuscular olanzapine and haloperidol in the
treatment of acute agitation in schizophrenia. Arch Gen Psychiatry 59:441–
448, 2002
Bristol-Myers Squibb: Abilify product information (revised October 2006).
Available at http://www.abilify.com. Accessed December 13, 2006
Brook S, Lucey JV, Gunn KP: Intramuscular ziprasidone compared with intra-
muscular haloperidol in the treatment of acute psychosis. J Clin Psychiatry
61:933–941, 2000
Brook S, Walden J, Benattia I, et al: Ziprasidone and haloperidol in the treatment
of acute exacerbation of schizophrenia and schizoaffective disorder: com-
parison of intramuscular and oral formulations in a 6-week, randomized,
blinded-assessment study. Psychopharmacology (Berl) 178:514–523, 2005
Buckley P, Bartell J, Donenwirth K, et al: Violence and schizophrenia: clozapine
as a specific antiaggressive agent. Bull Am Acad Psychiatry Law 23:607–
611, 1995
Casey DE, Daniel DG, Wassef AA, et al: Effect of divalproex combined with
olanzapine or risperidone in patients with an acute exacerbation of schizo-
phrenia. Neuropsychopharmacology 28:182–192, 2003
Chengappa KN, Goldstein JM, Greenwood M, et al: A post hoc analysis of the
impact on hostility and agitation of quetiapine and haloperidol among pa-
tients with schizophrenia. Clin Ther 25:530–541, 2003
Citrome L: Use of lithium, carbamazepine, and valproic acid in a state-operated
psychiatric hospital. J Pharm Technol 11:55–59, 1995
Citrome L: Comparison of intramuscular ziprasidone, olanzapine, or aripipra-
zole for agitation: a quantitative review of efficacy and safety. J Clin Psychi-
atry 68:1876–1885, 2007
Citrome L, Levine J, Allingham B: Changes in use of valproate and other mood
stabilizers for patients with schizophrenia from 1994 to 1998. Psychiatr
Serv 51:634–638, 2000
Citrome L, Volavka J, Czobor P, et al: Effects of clozapine, olanzapine, risperi-
done, and haloperidol on hostility among patients with schizophrenia.
Psychiatr Serv 52:1510–1514, 2001
Citrome L, Jaffe A, Levine J, et al: Use of mood stabilizers among patients with
schizophrenia, 1994–2001. Psychiatr Serv 53:1212, 2002
Citrome L, Casey DE, Daniel DG, et al: Effects of adjunctive valproate on hostil-
ity in patients with schizophrenia receiving olanzapine or risperidone: a
double-blind multi-center study. Psychiatr Serv 55:290–294, 2004a
Citrome L, Nolan KA, Volavka J: Science-based treatment of aggression and ag-
itation, in The Science, Treatment, and Prevention of Antisocial Behaviors,
Vol 2. Edited by Fishbein D. Kingston, NJ, Civic Research Institute, 2004b
Citrome L, Volavka J, Czobor P, et al: Efficacy of ziprasidone against hostility in

schizophrenia: post hoc analysis of randomized, open-label study data.
Citrome L, Shope CB, Nolan KA, et al: Risperidone alone versus risperidone
plus valproate in the treatment of patients with schizophrenia and hostility.
Int Clin Psychopharmacol 22:356–362, 2007
Coccaro EF, Kavoussi RJ: Fluoxetine and impulsive aggressive behavior in per-
sonality-disordered subjects. Arch Gen Psychiatry 54:1081–1088, 1997
Collins PJ, Larkin EP, Shubsachs AP: Lithium carbonate in chronic schizophre-
nia: a brief trial of lithium carbonate added to neuroleptics for treatment of
resistant schizophrenic patients. Acta Psychiatr Scand 84:150–154, 1991
Currier GW, Chou JCY, Feifel D, et al: Acute treatment of psychotic agitation: a
randomized comparison of oral treatment with risperidone and lorazepam
versus intramuscular treatment with haloperidol and lorazepam. J Clin
Psychiatry 65:386–394, 2004
Czobor P, Volavka J, Meibach RC: Effect of risperidone on hostility in schizo-
phrenia. J Clin Psychopharmacol 15:243–249, 1995
Daniel DG, Potkin SG, Reeves KR, et al: Intramuscular (IM) ziprasidone 20 mg
is effective in reducing acute agitation associated with psychosis: a double-
blind, randomized trial. Psychopharmacology (Berl) 155:128–134, 2001
Daniel DG, Zimbroff DL, Swift RH, et al: The tolerability of intramuscular
ziprasidone and haloperidol treatment and the transition to oral therapy.
Int Clin Psychopharmacol 19:9–15, 2004
Dietch JT, Jennings RK: Aggressive dyscontrol in patients treated with benzodi-
azepines. J Clin Psychiatry 49:184–188, 1988
Donovan SJ, Stewart JW, Nunes EV, et al: Divalproex treatment for youth with
explosive temper and mood lability: a double-blind, placebo-controlled
crossover design. Am J Psychiatry 157:818–820, 2000
Eli Lilly: Zyprexa product information (revised November 13, 2006). Available
at http://pi.lilly.com/us/zyprexa-pi.pdf. Accessed December 13, 2006
Feder R: Treatment of intermittent explosive disorder with sertraline in 2 pa-
tients. J Clin Psychiatry 60:195–196, 1999
Fink M, Sackeim HA: Convulsive therapy in schizophrenia. Schizophr Bull
22:27–39, 1996
GlaxoSmithKline: A multicenter, double-blind, placebo-controlled, randomized,
parallel group evaluation of the efficacy of a flexible dose of lamotrigine ver-
sus placebo as add-on therapy in schizophrenia. Available at http://
ctr.gsk.co.uk/Summary/lamotrigine/III_SCA30926.pdf. Accessed Novem-
ber 5, 2005
GlaxoSmithKline: A multicenter, randomized, double-blind, parallel group
study to evaluate the efficacy and safety of a flexible dose of lamotrigine
compared to placebo as an adjunctive therapy to an atypical antipsychotic
agent(s) in subjects with schizophrenia. Available at http://ctr.gsk.co.uk/
Summary/lamotrigine/III_SCA101464.pdf. Accessed May 27, 2006
Goff DC, Keefe R, Citrome L, et al: Lamotrigine as add-on therapy in schizo-
phrenia: results of 2 placebo-controlled trials. J Clin Psychopharmacol
27:582–589, 2007
Hirose S, Ashby CR, Mills MJ: Effectiveness of ECT combined with risperidone
against aggression in schizophrenia. J ECT 17:22–26, 2001
Hollander E, Allen A, Lopez RP, et al: A preliminary double-blind, placebo-

controlled trial of divalproex sodium in borderline personality disorder.
Hollander E, Tracy KA, Swann AC, et al: Divalproex in the treatment of impul-
sive aggression: efficacy in Cluster B personality disorders. Neuropsycho-
pharmacology 28:1186–1197, 2003
Hollander E, Swann AC, Coccaro EF, et al: Impact of trait impulsivity and state
aggression on divalproex versus placebo response in borderline personal-
ity disorder. Am J Psychiatry 162:621–624, 2005
Kant R, Chalanani R, Chengappa KN, et al: The off-label use of clozapine in ad-
olescents with bipolar disorder, intermittent explosive disorder, or post-
traumatic stress disorder. J Child Adolesc Psychopharmacol 14:57–63, 2004
Karson CN, Weinberger DR, Bigelow L, et al: Clonazepam treatment of chronic
schizophrenia: negative results in a double-blind, placebo-controlled trial.
Am J Psychiatry 139:1627–1628, 1982
Kessler RC, Coccaro EF, Fava M, et al: The prevalence and correlates of DSM-IV
intermittent explosive disorder in the National Comorbidity Survey Repli-
cation. Arch Gen Psychiatry 63:669–678, 2006
Khan SS, Mican LM: A naturalistic evaluation of intramuscular ziprasidone ver-
sus intramuscular olanzapine for the management of acute agitation and
aggression in children and adolescents. J Child Adolesc Psychopharmacol
16:671–677, 2006
Kinon BJ, Roychowdhury SM, Milton DR, et al: Effective resolution with olan-
zapine of acute presentation of behavioral agitation and positive psychotic
symptoms in schizophrenia. J Clin Psychiatry 62:17–21, 2001
Krakowski M, Czobor P, Citrome L, et al: Atypical antipsychotic agents in the
Lesem MD, Zajecka JM, Swift RH, et al: Intramuscular ziprasidone, 2 mg versus
10 mg, in the short-term management of agitated psychotic patients. J Clin
Psychiatry 62:12–18, 2001
Lindenmayer JP, Kotsaftis A: Use of sodium valproate in violent and aggressive
behaviors: a critical review. J Clin Psychiatry 61:123–128, 2000
Linehan MM: Dialectical behavior therapy for borderline personality disorder:
theory and method. Bull Menninger Clin 51:261–276, 1987
McEvoy JP, Scheifler PL, Frances A: The expert consensus guideline series,
treatment of schizophrenia 1999. J Clin Psychiatry 60:43, 1999
Meichenbaum DH, Goodman J: Training impulsive children to talk to them-
selves: a means of developing self-control. J Abnorm Psychol 77:115–126,
1971
Meehan K, Zhang F, David S, et al: A double-blind, randomized comparison of
the efficacy and safety of intramuscular injections of olanzapine, lorazepam,
or placebo in treating acutely agitated patients diagnosed with bipolar ma-
nia. J Clin Psychopharmacol 21:389–397, 2001
Meehan KM, Wang H, David SR, et al: Comparison of rapidly acting intramus-
cular olanzapine, lorazepam, and placebo: a double-blind, randomized
study in acutely agitated patients with dementia. Neuropsychopharmacol-
ogy 26:494–504, 2002
Okuma T, Yamashita I, Takahashi R, et al: A double-blind study of adjunctive

carbamazepine versus placebo on excited states of schizophrenic and
schizoaffective disorders. Acta Psychiatr Scand 80:250–259, 1989
Pfizer: Geodon product information (revised May 2005). Available at http://
www.pfizer.com/pfizer/download/uspi_geodon.pdf. Accessed Decem-
ber 13, 2006
Prakash R: Lithium-responsive schizophrenia: case reports. J Clin Psychiatry
46:141–142, 1985
Preval H, Klotz SG, Southard R, et al: Rapid-acting IM ziprasidone in a psychi-
atric emergency service: a naturalistic study. Gen Hosp Psychiatry 27:140–
144, 2005
Reist C, Nakamura K, Sagart E, et al: Impulsive aggressive behavior: open-label
treatment with citalopram. J Clin Psychiatry 64:81–85, 2003
San L, Arranz B, Querejeta I, et al: A naturalistic multicenter study of intramus-
cular olanzapine in the treatment of acutely agitated manic or schizo-
phrenic patients. Eur Psychiatry 21:539–543, 2006
Shalev A, Hermesh H, Munitz H: Severe akathisia causing neuroleptic failure:
an indication for lithium therapy in schizophrenia? Acta Psychiatr Scand
76:715–718, 1987
Sheard MH, Marini JL, Bridges CI, et al: The effect of lithium on impulsive ag-
gressive behavior in man. Am J Psychiatry 133:1409–1413, 1976
Sival RC, Haffmans PM, Jansen PA, et al: Sodium valproate in the treatment of
aggressive behavior in patients with dementia: a randomized placebo con-
trolled clinical trial. Int J Geriatr Psychiatry 17:579–585, 2002
Sorsaburu S, Hornbuckle K, Blake DS, et al: The first 21 months of safety expe-
rience with post-marketing use of olanzapine’s intramuscular formulation.
Poster NR432 presented at the 159th Annual Meeting of the American Psy-
chiatric Association, Toronto, ON, Canada, May 2006
Tariot PN, Erb R, Podgorski CA, et al: Efficacy and tolerability of carba-
mazepine for agitation and aggression in dementia. Am J Psychiatry
155:54–61, 1998
Tiihonen J, Hallikainen T, Ryynanen OP, et al: Lamotrigine in treatment-resis-
tant schizophrenia: a randomized placebo-controlled trial. Biol Psychiatry
54:1241–1248, 2003
Tran-Johnson TK, Sack DA, Marcus RN, et al: Efficacy and safety of intramus-
cular aripiprazole in patients with acute agitation: a randomized, double-
blind, placebo-controlled trial. J Clin Psychiatry 68:111–119, 2007
Tremeau F, Citrome L: Antipsychotics for patients without psychosis? What
clinical trials support. Current Psychiatry 5:33–44, 2006. Available at http://
www.currentpsychiatry.com/pdf/0512/0512CP_Article2.pdf. Accessed
March 19, 2007
Vartiainen H, Tiihonen J, Putkonen A, et al: Citalopram, a selective serotonin re-
uptake inhibitor, in the treatment of aggression in schizophrenia. Acta Psy-
chiatr Scand 91:348–351, 1995
Volavka J: Neurobiology of Violence, 2nd Edition. Washington, DC, American
Psychiatric Publishing, 2002
Volavka J, Zito JM, Vitrai J, et al: Clozapine effects on hostility and aggression in
schizophrenia. J Clin Psychopharmacol 13:287–289, 1993
Volavka J, Czobor P, Nolan KA, et al: Overt aggression and psychotic symptoms
in patients with schizophrenia treated with clozapine, olanzapine, risperi-
done, or haloperidol. J Clin Psychopharmacol 24:225–228, 2004
Volavka J, Czobor P, Citrome L, et al: Efficacy of aripiprazole against hostility in
schizophrenia and schizoaffective disorder: data from 5 double-blind stud-
ies. J Clin Psychiatry 66:1362–1366, 2005
Wright P, Birkett M, David SR, et al: Double-blind, placebo-controlled compar-
ison of intramuscular olanzapine and intramuscular haloperidol in the
treatment of acute agitation in schizophrenia. Am J Psychiatry 158:1149–
1151, 2001
Wright P, Meehan K, Birkett M, et al: A comparison of the efficacy and safety of
olanzapine versus haloperidol during transition from intramuscular to oral
therapy. Clin Ther 25:1420–1428, 2003
Zimbroff DL, Allen MH, Battaglia J, et al: Best clinical practice with ziprasidone
IM: update after 2 years of experience. CNS Spectr 10(suppl):1–15, 2005
Zimbroff DL, Marcus RN, Manos G, et al: Management of acute agitation in pa-
tients with bipolar disorder: efficacy and safety of intramuscular aripipra-
zole. J Clin Psychopharmacol 27:171–176, 2007
C H A P T E R 1 6
Psychotherapeutic
Interventions
John R. Lion, M.D.
This chapter describes general psychodynamic principles applicable

to the inpatient or outpatient therapy of patients who are physically vi-
olent. This quite heterogeneous group includes spousal abusers, those
who hurt animals, temper-prone individuals, persons convicted of
criminal acts of violence such as murder, aggressive paraphiliacs, and
arsonists. The common denominator among these patients is that they
all translate the affect of anger into dangerous behavior, be it assaultive
or destructive. Diagnostically, aggression can be seen within a wide va-
riety of mood, thought, and personality disorders (Stone 1995). What is
described here focuses more on patients who are characterologically
disordered than those who are psychotic and may, for example, have
command hallucinations urging them to hurt someone. Although psy-
chotic patients may well benefit from psychotherapy, the main effort
with such patients is pharmacological suppression of pathology rather
than insight-driven treatment. The following comments refer to tradi-
tional, individual therapy. For a description of group therapy for violent
patients—the mainstay of institutional care and offender-based clinics—
the reader is referred to descriptions of prison-based programs such as
those at Patuxent Institution in Jessup, Maryland (Coldren 2004), or
Herstedvester in Denmark (Sturup 1968).
325
Clinical Inexperience
A word should be said about aggression as a “stepchild” entity in clin-
ical psychiatry. Despite its commonplace occurrence in society and its
frequent occurrence in hospital inpatient settings, violence largely re-
mains a behavior of the criminal realm and a subject of interest to soci-
ologists or students of law enforcement. Indeed, few clinicians enter the
field of psychiatry anticipating that the patients they treat will be vio-
lent or dangerous; the discipline is, after all, a “talking” specialty with
verbal interventions. A resident’s early exposure to the emergency de-
partment may correct some of this distortion, but denial is still possible,
particularly because most aggressive patients are managed by nursing
staff and technicians. Restraint and seclusion, although medical inter-
ventions, are not routinely taught to physicians. Indeed, in residency
training programs little formal education is provided about the man-
agement of violent patients when compared with teachings about the
psychotherapy of anxiety or depression or the pharmacological treat-
ment of the schizophrenias (Dubin and Lion 1992). The elective of fo-
rensic psychiatry delves into the subject of violence, but few psychiatric
residencies have formal links with jails or prisons.
A therapist thus can easily emerge from training with limited
awareness of the world of antisocial conduct. Many psychiatrists have
never examined a rapist or a murderer, let alone treated one. This inex-
perience often leads the beginning therapist to avoid confrontation with
the patient’s aggressiveness. The therapist may not properly delve into
past behaviors or may avoid probing subjects or feelings that evoke an-
ger; a false delicacy may descend on the therapy. It is recommended
that clinicians who undertake work with violent patients peruse the
journals of the forensic sciences and criminology. Psychiatrists rarely
see these periodicals and should be acquainted with the vocabularies
and concepts of this entirely different clinical world.
The phenomenology of violence leads to further complexity. Unless
the patient is still in the throes of a manic illness or has otherwise dem-
onstrated frequent outbursts on a hospital unit, he or she generally is
not violent when seen by the clinician. For example, jails and prisons
house men and women who have been very violent. However, that vi-
olence is over; when seen by the clinician, they are typically nonviolent.
This is not the case with a depressed patient who enters treatment in the
midst of an observable melancholic state. Not viewing the pathological
or deviant behavior makes it difficult for a clinician to appreciate the
gravity of a patient’s case or his or her dangerousness. Because violent
patients easily disclaim their violent propensities even when confronted
Psychotherapeutic Interventions ❘ 327
with overwhelming data and documentation, therapists can find them-

selves colluding with the patient’s denial. This is particularly true with
paraphiliac males who minimize their aggressive predatoriness.
Case Example 1
A therapist was sent to a local prison to review the case of a child mo-
lester. The patient presented as a mild-mannered man who rejected any
idea that he was guilty of multiple assaults on children and who insisted
that he had been falsely incarcerated. The therapist easily believed him
until such time as he began to read a very thick chart filled with ac-
counts of violent paraphiliac behavior.
Spouse abusers also often insist that they are no longer violent and
that they love those whom they have hurt and would never harm them
again. One-time murderers may describe a rich circumstantiality that
absolves them of their actions. A therapist, faced with such disavowal
of violent tendencies, may dismiss the act as a “one-time” event and ex-
cuse future propensities. Thus it is vital to review the violent act. If a po-
lice report exists, an attempt should be made to procure it and read it. If
the patient committed murder, the autopsy report of the victim should
be studied. These efforts bring home the seriousness of the behavior.
Another common reaction on the part of unseasoned therapists is to
boast about the patient’s violent deeds, as if they were treating individ-
uals of distinction (e.g., “He’s the one who brutally murdered all those
children…”). Such a statement converts fear into awe, diminishes re-
pugnance, and rationalizes the clinician’s involvement in a case that
other colleagues might view as belonging in the sphere of antisocial be-
havior and more fittingly dealt with by a prison psychologist than by
someone in a suburban practice.
Fears and Liabilities

An issue that arises early in treatment is the problem of agency. Many
violent patients are court mandated to attend therapy. This immedi-
ately poses problems of agency for the patient, creating a negative view
of the therapist that is a powerful deterrent to trust and to the revelation
of intimate thoughts. This skew must be periodically acknowledged,
for only after some time has passed will the patient come to believe in
the therapist’s sincerity. In the meantime, the clinician has to deal with
his or her own anxieties about treating someone who can harm others.
This transcends the usual worries felt in the therapy of suicidal patients,
where it is more or less accepted that failure might occur. In effect, the
risk of death by suicide parallels the risk involved in treating any inher-
ently fatal disease. However, in the case of outwardly directed aggres-
sion, the patient can hurt or kill others; the violence thus moves beyond
the boundaries of therapy and can affect more or less innocent bystand-
ers. Liability, always on the mind of today’s clinician, has the potential
to escalate dramatically. The average psychotherapy patient talks about
becoming violent or dreams of violence but does not act on those urges.
The violent patient, in contrast, has “crossed the line” from contempla-
tion to response. Therapists must worry about things going very wrong.
They must concern themselves with the safety of potential victims, with
weaponry, with Tarasoff issues, with liability, and with the patient’s use
of alcohol or other disinhibitory or stimulant substances.
Vectors of Violence
A common misconception is that the violent patient is exempt from sui-
cide and vice versa. However, violent patients can hurt others and
themselves as well; they can commit both homicide and suicide. The
disorder of unregulated aggression, then, is a disease unto itself, and
there is some evidence that it reflects central nervous system serotonin
deficits (Markowitz and Coccaro 1995). It is the vector, then, that often
becomes a focus of treatment. The psychological burden of treating vi-
olent patients, whether they be suicidal or homicidal, is significant, and
the therapist must worry about patients’ inwardly and outwardly di-
rected aggression. No homicidal patient is ever immune from commit-
ting suicide, just as no suicidal patient with violent tendencies is ex-
empt from channeling anger outwardly. Vectors of aggression can shift
abruptly, depending on the availability of the victim or of a weapon
such as a handgun.
Case Example 2
A man who was being treated for an ear infection became delusional
and thought that his doctor was poisoning him. He decided to kill him-
self in front of the doctor in order to illustrate his plight. He brought a
loaded pistol to his appointment, took it out, and pointed it at his head,
but at the last moment he turned the muzzle toward the doctor and shot
and killed him.
In one sense, the development of depression is a goal of treating vi-

olent patients. Therapy aims at helping patients tolerate painful affects
associated with loss and with injuries to self-esteem. Improperly mod-
ulated, however, the despair may become overwhelming.
Case Example 3
A young adolescent girl stabbed and killed her abusive mother during an
argument. The court mandated inpatient psychiatric treatment, and the
girl remained on a university teaching unit for more than a year. Because
her case was an unusual one, she became the subject of intensive psycho-
therapy. Upon her release, she continued outpatient treatment but became
increasingly depressed and ultimately committed suicide by hanging.
Postpsychotic depression has been described as emerging in the

therapy of schizophrenia at a point when the patient comes to relin-
quish his or her psychosis and realizes how ill and dysfunctional he or
she has been (McGlashan and Carpenter 1976). Violent patients can go
through the same phase. Therapists should predict this to the patient
and should outline the goals of fostering introspection. Patients given
to behaviorally “acting out” may find the passivity of therapy alien and
difficult, because they may not understand how talking and feeling can
possibly help them in their lives.
Goals and Strategies of Therapy

The goals of psychotherapy of violent patients are relatively straight-
forward: they are to help patients understand the origins of their need to
destroy or hurt and wound and to appreciate their inner affective state be-
fore it erupts. This urge springs from severe injury in childhood, neglect to
the point of rage, a long-standing pathological detachment, or an aberra-
tion in the choice of object, as seen in paraphiliac disturbances. Some ori-
gins are easier to identify than others. Spouse abuse, for example, often re-
flects intolerable ambivalence toward a lover and can often be traced back
to earlier abuse or neglect by a parental figure. Aggressive pedophilia, on
the other hand, has as yet undetermined biological origins; although some
components such as love-seeking can be identified, a clear resolution of
dynamics usually remains elusive. However, the therapist can still help
the patient identify emotional trigger points and inner yearnings that fuel
a need to act. Treatment of fire setting, which often is not identified as vi-
olence, requires that patients understand when they are feeling angry or
empty. Temper proneness, as seen in intermittent explosive disorder, re-
quires identification of recurring precipitants that lead to the rage.
Case Example 4
An investment broker entered treatment after erupting in rage at a coun-
try club wedding and loudly calling his girlfriend a whore. In the past,
he had once begun to choke her. It became apparent that a recurring
stressor was abandonment; at the country club, she had left his side to
interact with other family members, and he recalled feeling isolated and
becoming furious.
The therapy of these conditions is a mixture of insight-oriented and

cognitive therapy. The task of eliciting affect can be a very pedestrian
one. It requires that the clinician continually probe what the patient is
feeling and has the quality of learning a new language of expression.
Affective recognition is the most crucial task of treatment. To the extent
that the therapist begins to probe deeper issues, exploring the various
degrees of sadism is a central effort as well. Patients who break glass,
draw the blood of others, or create pain all receive some satisfaction
from the intimidation they create. Beginning therapists can easily view
the violent acts as “mistakes” by the patient, rather than as the products
of an intrinsic derangement of empathy. That is, it may be that a patient
not only has the capacity to harm or destroy but also is gratified by the
act and will seek to do it again unless a repair takes effect. It is necessary
to explore with a patient all the various manifestations of sadism, even
to the tiniest detail. The results are often surprising.
Case Example 5
A schoolteacher stabbed her husband during an argument. He later
died, and she adamantly maintained that she had not intended to kill
him. Indeed, when the police arrived, she was administering cardiopul-
monary resuscitation to him. She presented as a demure woman and
was much prone to intellectualize. It was only after the therapist ex-
plored her history of violence for some 9 hours that she admitted,
sheepishly, that she disliked ants in her house and would pour flamma-
ble fluids on them and light them on fire. In time, she also related that
she had burned herself on the ankle with cigarettes, something she hid
with ankle bracelets.
Even the most hardened criminal may be ashamed of admitting to

behaviors that reflect sadism, brutality to children or animals, or sexual
excitement in connection with the infliction of pain. In the elicitation of
such pathology, the therapist must bridge a gap between expressing
revulsion and exhibiting detachment. Some dismay can be shown,
together with a therapeutic desire to help the patient overcome such
malignant behaviors. The origins of sadism must be explored if they are
ever to be changed; such exploration once again involves a review of
childhood events that unleashed the need to be cruel.
This issue of sadism raises a larger question of whether it is ever
unrealistic to treat someone whose crimes or behaviors are extremely
heinous. The therapist obviously must decide whether therapeutic inter-

vention is feasible, but this is not a simple task. As with gravely injured
patients in a wartime battle situation, some patients cannot be salvaged.
This is particularly the case with intractably aggressive persons who
lack a conscience or use projection and denial to the point of therapeutic
impenetrability. Sometimes, a trial of therapy is needed to confirm
whether or not the patient is introspective or responsive to insight. Re-
morse, guilt, sorrow, and the capacity for empathy may not emerge for
a long time.
Deranged Transferences
Assuming that the violent person becomes a willing patient and mean-
ingfully partakes in treatment, the transference must be carefully mon-
itored. Physical rage usually reflects the most desperate helplessness,
and it can easily be rekindled, particularly within the nurturing process
of psychotherapy. Patients with primitive character structures can be-
gin to relate to a therapist in the same way they responded to parents or
authority figures. Small prohibitions on the part of the therapist may
become magnified; for example, the clinician who will not renew a ben-
zodiazepine drug may be perceived as very cruel and withholding. Per-
haps the greatest danger stems from the vicissitudes of intimacy. As the
violent patient comes to feel closer to the therapist, an intolerable yearn-
ing may arise that is frustrated by the constraints of treatment. Loss of
the therapist may arise as a risk. The simple dependency of the patient
on the therapist can become overwhelming. These dynamics must be
regularly explored, as must the specific fantasies that the patient has
about the therapist.
Case Example 6
A social worker treated a woman with a borderline personality disorder
in a public clinic setting. The patient was demanding and often para-
noid but came regularly to sessions and developed a clear dependency
on the therapist. In time, this dependency changed into wishes to be
with the therapist outside of the hour. These urges were not fully ex-
plored. One evening, the patient appeared outside the therapist’s home
wielding a loaded handgun. She fired a shot that missed the therapist,
at which time the therapist tackled her and subdued her until the police
arrived.
Deranged transferences are frightening events, seemingly appear-

ing out of thin air; however, there are usually warning signs, such as a
patient’s view of the therapist as cold and uncaring, a perception that
mirrors childhood neglect or violence. This perception can lead to anger

and an emerging desire to harm the therapist. Conversely, the intimacy
of treatment may spawn a desire in the patient to be closer to the thera-
pist, manifested as driving by the therapist’s home or engaging in stalk-
ing behavior (Lion and Herschler 1998). Even a patient’s spending ex-
cess time in the therapist’s waiting room can signal some form of
attachment that warrants exploration and intervention. The issue here
is pathological attachment, but such an attachment is often a silent one.
The patient will not talk about it because he or she senses it is not proper
and fears alienating the clinician. As the patient’s needfulness spirals
out of control, the unattainability of the sought closeness liberates an-
ger. This is the point at which violence can erupt. The only way to guard
against the development of a deranged transference is to periodically ask
the patient about it; the question takes the form of asking the patient how
he or she is feeling about therapy in general and about the clinician in
particular.
Where therapy is conducted bears mention. Many therapists see pa-
tients in the evening or on weekends when office buildings are empty.
This may create too intimate a setting for seeing some patients, and the
clinician generally would be safer working in a more public space
where there is traffic and other signs of external controls. At the very
least, the first intake session should occur when other staff are nearby.
Very little attention is given to the safety of the settings in which the cli-
nician practices (Lion et al. 1996). Most clinics, for example, have no
buzzer alarm system in place, nor are staff trained in how to summon
help in an emergency. Interview rooms may be poorly configured, plac-
ing the clinician behind a desk against a wall rather than near the door
so that, in a crisis, he could exit. Weapons screening practices, although
more in keeping with emergency departments, still warrant consider-
ation when the population being treated is very ill or the clinician serves
an offender population.
Countertransference Issues
It can happen that a therapist becomes irrationally fearful of a violent
patient, as the following case illustrates.
Case Example 7
A psychiatrist sought consultation because he found himself frightened
of a violent patient without good cause. No threat had been made, nor
was the patient menacing or otherwise intimidating. As he discussed
the case and the consultant asked him about his own experiences with
violence, the clinician stated that he once had gone on vacation and
parked at an overlook atop a mountain. Another car drove up and a
man got out of the car, walked toward the psychiatrist, took out a re-
volver, and aimed it at him. He managed to push the man over the
ledge in self-defense and called for the police. The man was found and
returned to a prison from which he had escaped. The psychiatrist had
actually not remembered this upsetting event, and talking about it re-
solved his uneasiness with the patient who had obviously kindled the
recollection.
Such an event is rare in the lives of most clinicians, but less dramatic
instances of violence may be evoked during treatment, leading to skews
in the perception of the patient (Lion 1998). For certain therapists, sex-
ual deviance such as that shown by a child molester is so repulsive that
no meaningful treatment can be undertaken. Clinicians must make an
inner appraisal of their own experiences with aggression as they labor
with the patient before them.
Victims
Treating a violent patient without consideration of an existing victim is
an error. In spouse abusers, for example, the therapist must monitor
how the patient is behaving at home, and this can only be done by ask-
ing the wife about the patient. A useful strategy with identified victims
is to invite them into an evaluative session with the patient present. This
minimizes any breach of confidentiality while allowing the therapist to
observe how the two parties interact. It is sometimes the case that the
would-be victim’s behaviors are so provocative that treatment is indi-
cated. In cases where the patient has been violent toward children or an-
imals, a corroborating source of information should be sought, such as
a case worker or relative. Again, the therapy of the violent patient is
conducted with less privacy than that operative in more traditional in-
dividual treatment. In some sense, it is like the treatment of an actively
suicidal patient, in which the therapist comes to rely on the family for
feedback and participation.
Warning victims about incipient violence is now a standard of care
under most Tarasoff-based state statutes. Although imminence of a
threat prevails as the qualifier for warning, there should be some con-
cept of reasonableness in the mind of the treating clinician. Certainly, a
patient who wishes to harm a non–family member presents a dire situ-
ation in which warning may be indicated. More frequently, however,
the patient is already enmeshed in a troubled relationship with a clearly
identified other such as a lover or spouse. If a threat is made, the clini-
cian should seek, whenever possible, to invite the spouse or lover into
a treatment session. This is far superior to issuing a sudden formal
warning by phone or letter. Such a warning is a clearly alienating event
and leads to little or no possibility of therapeutic resolution. In general,
it makes little clinical sense to wait until a major interpersonal crisis oc-
curs before intervening; the therapist must engage in prophylactic work
and introduce the potential victim into the treatment setting early on.
Otherwise, that person is distanced from the treatment and cannot eas-
ily enter into it to inform the therapist about the patient’s behavior.
Controversy about warning has recently arisen in the case of college
students whose suicidal or homicidal thoughts come to the attention of
administrators. However, institutions can come to see their role as a pri-
vate one, and misguided notions of confidentiality can prevail, leading
the college to withhold notification of a patient’s violent urges at possi-
ble risk to the public or to the patient him- or herself. Very few patients
are “imminently” homicidal (or even suicidal) to the point of equipping
themselves with a loaded weapon. Rather, they talk first about emerg-
ing thoughts, much as any patient describes angry feelings about some-
one else. Unchallenged, the thoughts escalate. The point here is that the
clinician should adopt a lower threshold for “warning” a victim or no-
tifying family so that preventive action can be taken.
The Uses of Medication

The pharmacological treatment of violence is discussed in Chapter 15 of
this book. Some principles, however, bear mention. No specific antiag-
gressive drug exists. Rather, it may be helpful to use a drug that affects
a target symptom conducive to violence. If the patient, for example, be-
gins to ruminate about a perceived insult made by his spouse and be-
comes agitated in the process, he may benefit from a benzodiazepine
taken upon recognition of his escalating anxiety. This may short circuit
the buildup of anger. Anxiety, in fact, is a key factor in the production
of rage. Just as the condition of anxiety increases the risk of suicide in a
despondent patient, so panic states can fuel rage in someone so predis-
posed. In general, “fuse lengthening” is a goal of treatment, but it can
only be effective if the patient comes to identify the affect that swirls
within. Similar rationales support the use of anticonvulsant drugs that
attenuate “ictal” expressions of explosive violence such as those seen in
intermittent explosive disorder. Some violent patients appear to have
behavioral lability that mirrors mood changes; in such instances, an an-
tidepressant or mood stabilizer may be of benefit. In other instances, a
quenching agent can be useful, such as a short-acting benzodiazepine
that the patient keeps in his or her wallet and uses when the anger starts
building up. Perhaps the only exception to the nonspecificity of treat-
ment is the use of antiandrogens or hormonal agents to treat the height-
ened drive state seen among pedophiles or rapists.
Assessing the efficacy of a drug is often difficult. A patient may
claim that he or she is no longer violent when the spouse reports other-
wise, and a dosage adjustment may be necessary. This is another reason
why it is useful to periodically (once a month, for example) solicit infor-
mation from a victim. One matter to be cognizant of is a reduction in
alertness resulting from too high a dosage of medication. Because vio-
lent patients are often hypervigilant, they may be noncompliant with
any drug that makes them feel vulnerable to the world around them.
Small dosages, titrated upward, is the rule of thumb. Patients can be
told that the initial amount of medication given may be ineffective and
that the clinician will slowly increase it.
Supervision
Psychiatry tends to be practiced in isolation. With violent patients, it is
useful to talk things over with another clinician, provided that the latter
has some experience in the management of aggression.
Case Example 8
A resident treated a paranoid patient who became threatening to her,
threw a pillow at her during a session, and ultimately made a homicidal
threat. The resident had been reporting this to her supervisor, a psycho-
analyst. The latter was dismissive of the danger involved and suggested
that the resident was both unconsciously eliciting the threats and mag-
nifying the risk of them. The resident sought consultation with a foren-
sic psychiatrist, who became appropriately alarmed. He recommended
that the resident notify the police of the threat and helped the resident
hospitalize the woman and halt therapy with her.
Unstable Treatment Situations

There may come a time, as the previous example illustrates, when the
clinician should consider disengagement from the patient. This is obvi-
ously not a simple matter.
Case Example 9
A young man prone to temper smoked marijuana frequently. He lived
by himself but was pressured by his family to find a stable job and get
married. Arguments often erupted between the patient and his father,
an attorney. However, when the parents went on vacation, the patient

typically became anxious and used more drugs. During one such holi-
day, the patient complained bitterly to his therapist about the parents.
The therapist acknowledged his dependence and anger, at which point
the patient took out a large pocket knife, opened the blade, and held it
to his own abdomen. The therapist immediately stated that he was
frightened by the patient’s actions and asked that he put away the knife.
The patient complied, ran out of the office, went home, and began to de-
stroy his apartment. Police were called and the patient was hospitalized.
The therapist decided that the patient posed a danger and could no
longer be safely managed in a private outpatient setting but should be
treated in a more public setting. He arranged for the patient to be trans-
ferred to a hospital-based clinic.
It is clear that there may come a time when limits with a patient are
reached and it becomes prudent to relinquish treatment. Assuming that
the situation does not reflect a countertransference element as described
earlier, the clinician should shift care to another setting. This is not sim-
ple to accomplish, because a truthful revelation of what occurred will
deter many prospective therapists from taking on the case. It is useful
in these instances to hospitalize the patient and meet on several occa-
sions with the treatment team and the future therapist. It is important
to explain to the patient why transfer is taking place—that is, to admit
to the patient directly that he or she has become too frightening. Pa-
tients are often surprised to hear this, but the comment has clear thera-
peutic value, particularly in a critical situation when physical confron-
tation by the patient occurs. If a patient becomes threatening, not
revealing the impact of the threat can lead the patient to become more
menacing because he or she senses no response. This advice seems
counterintuitive to security and law enforcement personnel who would
never admit to others that they are fearful. However, in the clinical
realm, the message can halt dangerous behavior.
Key Points
■ Many therapists lack experience with aggressive patients. It is rec-
ommended that clinicians who undertake work with violent
patients peruse the journals of the forensic sciences and
criminology.
■ Violent behavior typically is not on display during therapy, and
this can make it difficult for a clinician to appreciate a patient’s
dangerousness. Thus it is important to review the patient’s past
violent acts.
■ Therapists treating violent patients must concern themselves

with the safety of potential victims, with weaponry, with Tarasoff
issues, with liability, and with the patient’s use of alcohol or other
drugs that may contribute to violence.
■ No homicidal patient is immune from committing suicide. Like-
wise, suicidal patients may turn aggression outward.
■ The goals of psychotherapy of violent patients are to help patients
understand the origins of their need to destroy or hurt and to
appreciate their inner affective state before it erupts.
■ To guard against the development of a deranged transference, the
therapist must periodically ask the patient how he or she is feeling
about therapy in general and about the clinician in particular.
■ The therapist must be aware of duty to warn potential victims of a
violent patient. Warning victims about incipient violence is now a
standard of care under most Tarasoff-based state statutes.
■ Supervision is useful to therapists who are treating violent
patients, and the possibility that the best course will be disen-
gaging from the patient should be recognized.
■ Psychotic patients may benefit from psychotherapy, but the main
effort with such patients is pharmacological suppression of
pathology.
References
Coldren JR: Patuxent Institutions: An American Experiment in Corrections.
New York, Peter Lang Publishing Group, 2004
Dubin WR, Lion JR (eds): Clinician Safety. Task Force Report No 33. Washing-
ton, DC, American Psychiatric Press, 1992
Gellerman DM, Suddath R: Violent fantasy, dangerousness, and the duty to
warn and protect. J Am Acad Psychiatry Law 33:484–495, 2007
Lion JR: Countertransference in the treatment of the antisocial patient, in Coun-
tertransference Issues in Psychiatric Treatment. Edited by Gabbard GO (Re-
view of Psychiatry Series, Vol 18; Oldham JM and Riba MB, series eds).
Lion JR, Herschler JA: The stalking of clinicians by their patients, in The Psy-
chology of Stalking: Clinical and Forensic Perspectives. Edited by Meloy
JR. San Diego, CA, Academic Press, 1998, pp 165–172
Lion JR, Dubin WR, Futrell DE (eds): Creating a Secure Workplace. Chicago, IL,
American Hospital Publishing, 1996
Markowitz PH, Coccaro PI: Biological studies of impulsivity, aggression, and
suicidal behavior, in Impulsivity and Aggression. Edited by Hollander E,
Stein DJ. Chichester, UK, Wiley, 1995, pp 71–90
McGlashan TH, Carpenter WT: Postpsychotic depression in schizophrenia.
Stone MH: Psychotherapy in patients with impulsive aggression, in Impulsivity

and Aggression. Edited by Hollander E, Stein DJ. Chichester, UK, Wiley,
1995, pp 313–332
Sturup GK: Treating the “Untreatable”: Chronic Criminals at Herstedvester.
Baltimore, MD, Johns Hopkins University Press, 1968
C H A P T E R 1 7
Seclusion and Restraint

John R. Lion, M.D.
Seclusion and restraint are “hands-on” techniques that can be used to

manage violent patients within psychiatric inpatient units and emer-
gency rooms. Being often the sole means of initially controlling very
combative and assaultive individuals before medications become effec-
tive, and even when used in conjunction with drugs, restraint and se-
clusion are destined to remain controversial. This is because the tech-
niques so dramatically infringe on the physical freedom of patients and
because they are historically associated with the inhumane practices of
older times. Seclusion and restraint have thus received intense scrutiny
by the lay public as well as by psychiatrists themselves. The first effort
by the psychiatric profession to address the utility of restraint and se-
clusion was undertaken by a task force of the American Psychiatric As-
sociation during the years 1981–1985. The findings of this group were
published in 1985 and still remain a standard of care (Tardiff 1984), al-
though a second task force met during the years 2003–2005 and issued
a revised report that has not yet been fully approved by the American
Psychiatric Association at the time of this writing (American Psychiatric
Association 2006). Portions of the report, however, will appear as a cor-
rectional mental health commentary in the Journal of the Academy of Psy-
chiatry and the Law (J.L. Metzner, personal communication, October 14,
2007). The most recent task force (herein called “the APA task force”)
was convened to address new directives issued by Medicaid and Medi-
care agencies and the Joint Commission for the Accreditation of Health-
339
care Organizations (JCAHO). These directives are commented upon in

the following discussion.
Studies of Seclusion and Restraint

The 1984 task force studied the extent of the use of restraint and seclu-
sion in various states and found great variability in terms of how struc-
tured and specific the guidelines were. Some states had very detailed
formal policies; others had virtually no guidelines. In some published
studies in which restraint and seclusion were used primarily to halt
physical violence toward other persons, rates of usage ranged from 2%
to 10% (Mattson and Sacks 1978; Soloff and Turner 1981; Tardiff 1981;
Wells 1972). In other studies done within acute psychiatric inpatient
units, there was a greater frequency of restraint and seclusion, ranging
from 18% to 37% (Binder 1979; Convertino et al. 1980; Oldham et al.
1983; Plutchik et al. 1978; Schwab and Lahmeyer 1979). The latter set-
tings were found to be more likely to use physical intervention for other
reasons in addition to halting physical violence, such as to manage agi-
tation and anger, poor impulse control, threats, or property damage.
In 1986, Way conducted a survey of all state hospitals in New York.
He found that 59% of such facilities used restraints and 46% used seclu-
sion. Most of the hospitals used only one or the other technique for the
control of disruptive behaviors. Emergency medication was used with
seclusion or restraint in 56% of the hospitals, but there was great vari-
ability in the use of such drugs. Restraint or seclusion was used more
frequently in response to physical attacks toward staff (30%) or other
patients (21%) than for threatening behavior (16%) or agitated behavior
(16%). Later, in 1992, Ray and Rappaport (1995) surveyed 125 state hos-
pitals and psychiatric units in general hospitals within New York. Some
33% of general hospitals and 14% of state hospitals used seclusion or re-
straints, but they used these for fewer than 1% of their patients. Swett
(1994) studied all admissions to the only state hospital in New Hamp-
shire during a 1-year period in 1991–1992. A total of 114 (31%) of admis-
sions had at least one episode of seclusion or restraint, usually because
the patients were harmful to themselves or others. Crenshaw and Fran-
cis (1995; Crenshaw et al. 1997) conducted a survey of state hospitals in
44 states and the District of Columbia in 1991 and repeated the survey
in 1994. They found that smaller hospitals had greater use of seclusion
and restraint than larger hospitals. These authors attributed their find-
ings to the fact that smaller hospitals treated more acutely ill patients for
shorter time periods, whereas larger hospitals had more chronic pa-
tients. Lavoie (1992) studied the use of seclusion and restraint in the
Seclusion and Restraint ❘ 341
emergency department of an urban university hospital in Kentucky and

found that 9% of patients were secluded, whereas 26% were restrained.
Studying the overall use of restraint and seclusion, Busch and Shore
(2000) reviewed the literature from 1994 through 1999 and found a vari-
ation in the frequency of restraint and seclusion that appeared indepen-
dent of the socioeconomic and clinical characteristics of patients. Staff
decision making was inconsistent, but other nonpharmacological inter-
ventions (such as behavior modification) lowered the use of seclusion
and restraint. Sailas and Fenton (2000), in reviewing the literature,
found no controlled studies of the use of seclusion or restraint as an in-
tervention in the treatment of psychiatric emergencies and no studies
regarding the effects of seclusion or restraint upon patients with schizo-
phrenia or other serious psychiatric disorders.
Other countries have looked at the use of restraint and seclusion.
Ahmed and Lepnurm (2001) studied the use of seclusion in a Canadian
forensic hospital and found that patients with acute psychosis repre-
sented 11.8% of seclusions, whereas those with agitation or disruptive
behavior represented 12.7% of episodes of seclusion. Suicide threats or
self-harm gestures were the reasons for secluding a patient in 27.4% of
episodes. Diagnostically, substance-related disorders accounted for
41% of seclusion episodes, whereas patients with schizophrenia ac-
counted for 28% of seclusion episodes. Kaltiala-Heino et al. (2003) stud-
ied the reasons for the use of seclusion and restraint in the psychiatric
inpatient units within Finland and found results similar to those of the
Canadian study by Ahmed and Lepnurm.
New Governmental Guidelines

In 1999, the Health Care Financing Authority, now known as the Cen-
ters for Medicare and Medicaid Services (CMS), defined rules for the
use of seclusion and restraint in facilities that participated in Medicare
and Medicaid (U.S. Department of Health and Human Services 1999,
2001). CMS maintains a narrow indication for restraint and seclusion,
stating that these interventions can be used only to manage severely ag-
gressive or destructive behavior that places the patient or others in im-
minent danger. CMS has published directives on who can order re-
straint and seclusion, who can review the process, and when such a
review should take place. However, JCAHO has also entered the field
and defined its own parameters (Joint Commission on Accreditation of
Healthcare Organizations 2002). Table 17–1, from the unpublished re-
port of the most recent American Psychiatric Association Task Force on
Restraint and Seclusion [2006], provides a synthesis of these parameters
TABLE 17–1. Time parameters for restraint or seclusion interventions

Length of time
or frequency
Time from initial order to face-to-face evaluation

by physician (preferable) or LIP 1 hour
Subsequent face-to-face evaluations by LIP Every 12 hours
Maximum length of time before new order is
required
Adult 4 hours
Age 9–18 years 2 hours
Child 1 hour
Maximum length of time before chief physician
must review 24 hours
Visual observations by trained staff Every 15 minutes;
continuous if in four-
point restraints or
restraint plus seclusion
Face-to-face evaluations by clinical staff Every 2 hours
Note. This table synthesizes JCAHO and CMS recommendations. These two agencies
have differing definitions of staff.
A licensed individual practitioner (LIP) is a clinician state licensed to write orders for
restraint and seclusion and trained in emergency care techniques.
A clinical staff member is a degreed nurse or nursing assistant.
A trained staff member has some degree of mental health training, such as a mental
health assistant.
Source. Adapted from American Psychiatric Association Task Force on Seclusion and
Restraint 2006.
and definitions. Unfortunately, the fact that these two institutions have
differing standards has made the subject of restraint and seclusion more
confusing. Added to this complexity is the fact that states vary in which
regulations they follow. Thus the clinician must know the written
guidelines applicable to the state in which he or she practices.
Indications for Seclusion and Restraint

Described here are the principles on the use of seclusion and restraint
as detailed in the revised APA task force report. The main changes
made since 1985 pertain to the matter of who authorizes the restraint
and seclusion, who reviews its use or continuation, and when such re-
views occur. Also addressed in the revised report is how long a patient
can be kept in restraints and seclusion (see Table 17–1). Although CMS
guidelines state that seclusion or restraint can be used only in emer-

gency situations, the recent task force expands on this, emphasizing the
maintenance of the milieu and specifically adding that seclusion and re-
straint can be used to prevent imminent harm to other persons as well
as the patient and to prevent serious disruption of the treatment envi-
ronment. For example, a patient may be a danger to him- or herself in
two ways: first, in terms of deliberate suicidal acts or self-mutilation, or
second, by a degree of excitement or behavioral dyscontrol that, if it
continues, will result in exhaustion or injury. The patient can be a dan-
ger to another by deliberately trying to assault that person or by unin-
tentional violence as a result of marked disorganization of behavior,
such as that seen in an agitated paranoid delusional state. Under certain
circumstances, seclusion of a patient may be indicated both for the pa-
tient’s benefit and that of the environment. The delicate balance of com-
peting interests between the patient, other patients, and the milieu is of-
ten difficult to achieve. Patients who are seriously disruptive to the
environment or who are seriously interfering with the rights of other
patients generally do so because of the underlying disease process. Cer-
tain events, such as uncontrollable screaming or abuse, public mastur-
bation, nude behavior, uncontrolled intrusiveness on others, or fecal
smearing may indeed constitute indications for seclusion or restraint.
Before using seclusion or restraint, the staff should have considered
or tried other means of control, particularly verbal and environmental
interventions. Appropriate use of antipsychotic, mood-stabilizing, or
anxiolytic drugs should be considered as well. With appropriate docu-
mentation, staff may rely on the patient’s known history of violent epi-
sodes and their known predecessors, such as escalating, excited motor
behavior, increase in muscle tone or generalized tension, pacing, or
loud or profane speech. The key to the medically appropriate use of re-
straint and seclusion is documentation and review. Why such interven-
tions were used should always be stated in the record, together with
what other interventions failed. Any institution using restraint and se-
clusion must keep records of usage and review those records on an an-
nual basis. Although there exists no “average” use of these modalities,
changes in usage can signify a changing population, staffing problems,
or a change in treatment philosophy.
Initiation of Seclusion and Restraint

Portions of the following sections are taken from the current Task Force
Report on Seclusion and Restraint (American Psychiatric Association
2006) submitted to the American Psychiatric Association.
Initiation of a restraint procedure or placement of a patient in seclu-

sion is usually an emergency procedure carried out by nursing and
other professional staff in accordance with established hospital policy
for seclusion and restraint. CMS guidelines now specify the need for a
“licensed independent practitioner” (LIP) to initiate and monitor re-
straint and seclusion. Such clinicians are described as being trained in
emergency care techniques and licensed by their state to write orders
for restraint and seclusion. JCAHO guidelines make no such comments
about manpower and qualifications. According to CMS, a patient
should be seen face-to-face by the LIP within 1 hour after initiation of
restraint and seclusion and at least every 4 hours thereafter. If a patient
is released from seclusion before the initial assessment, the LIP must
still render an evaluation within that first hour’s time. JCAHO is more
lenient in its requirements, stating that an initial face-to-face evaluation
must be done within 4 hours in the case of an adult and within specified
shorter periods for younger patients. The task force has adopted the
CMS 1-hour interval as a maximum permissible time period between
the initiation of restraint or seclusion and an in-person evaluation. In
the first edition of the APA task force report, the physician was identi-
fied as the person needed to make such an assessment. Neither CMS
nor JCAHO makes this requirement, and the task force acknowledges
the LIP as qualified to assess the patient. However, the task force con-
tinues to view restraint and seclusion as medical procedures and iden-
tifies the physician as the preferred person responsible for evaluating
the patient within the first hour as well as countersigning the initiation
and subsequent orders. Furthermore, the task force states that the phy-
sician should maintain a leadership role in formulating all policies and
practices of restraint and seclusion. If a clinician other than one on the
patient’s treatment team orders restraint or seclusion, the patient’s
treating doctor should be notified.
CMS and JCAHO requirements vary considerably. The initial order
for restraint or seclusion, according to CMS, cannot exceed a duration
of 4 hours for adults, 2 hours for adolescents between the ages of 9 and
18, and 1 hour for children under 9 years of age. After the first specified
time period , a new order for another term of restraint or seclusion is re-
quired. On the basis of clinical information that can be conveyed by
telephone, CMS allows additional restraint or seclusion orders without
face-to-face evaluations for up to 24 hours. JCAHO mandates a face-
to-face evaluation every 8 hours for patients 18 and older and every
4 hours for younger patients, but it does not comment on total time lim-
its. In its first edition, the APA task force described the initial order for
restraint or seclusion as valid for 12 hours and orders of longer than
72 hours as requiring higher administrative approval. The task force

adheres to its original recommendation of 12 hours for the period be-
tween initiation of restraint or seclusion and subsequent in-person eval-
uations after the first 1-hour check and now recommends 24 hours as
the time limit after which higher administrative approval is required by
a chief physician (see Table 17–1).
The task force recognizes that there may be individuals so disturbed
that more time in seclusion or restraint will be needed beyond 24 hours.
In such instances, the chief physician of the institution or his or her des-
ignee must review the treatment plan and concur with additional re-
straint or seclusion. Some patients will require more frequent face-to-
face visits than every 4 hours. Examples are patients with concurrent
medical problems, those with organic brain syndrome such as related
to drugs or alcohol, and situations in which hyperthermia may occur.
CMS has commented that in cases of unusual self-mutilation, such as
Lesch-Nyhan syndrome, a patient can be kept in long-term restraints as
part of medical management, not behavioral management. The task
force further recommends that orders be time and behavior specific,
with a stated goal (e.g., “Four-point restraints for 1 hour or until patient
is no longer agitated and combative”). Standing orders for restraint or
seclusion are not allowed. The clinician must document in the patient’s
record the failure of less restrictive alternatives and the justification for
continued seclusion or restraint. This decision takes into account the
mental and physical status of the patient, his or her degree of agitation,
and the adverse effects of seclusion (both physical and emotional). De-
briefing at the end of each restraint or seclusion episode is mandatory.
The patient should be asked about the experience and appropriate com-
ments documented.
Techniques for Seclusion and Restraint Maneuvers

The APA task force makes the following recommendations concerning
the maneuvers for restraint and seclusion, based on the realization that
uniform techniques and standards are lacking nationwide.
First, the technique of restraint practiced within a particular facility
should be rehearsed and approved by the hospital staff, including the
chief of service of the institution. If the particular technique and modal-
ity, such as four-point leather restraints, is viewed as normal practice,
that should be specifically noted in the policy manual of the hospital.
Details of the specific technique should be disseminated to all members
of the clinical staff as part of the service training. Written instructions,
photographs, and videotapes are desirable. A certification process
should be in place, with documentation that each clinical staff member

has been taught restraint and seclusion and been recertified on an an-
nual basis. Hands-on training is requisite. Identified instructors should
be designated within a hospital facility to teach these skills to both new
and existing clinical staff and as part of in-service training. Physicians
should also take part in training procedures. This matter, along with the
general subject of staff protection, is extensively covered in an Ameri-
can Psychiatric Association monograph on the subject of clinician safety
(Dubin and Lion 1993).
A variety of restraint devices are marketed, including Velcro and
leather limb restraints, body vests, and full body jackets (Lion and
Danto 1996). Presently, no data exist concerning the relative efficacy of
any devices. JCAHO regulations recommend the gathering of such data
in keeping with “evidence-based” medical practices, but the task force
realizes that such an effort is exceedingly difficult. The choice of which
apparatus to use is left up to the individual institution. Legal represen-
tatives for the institution should be consulted regarding the use of the
particular restraint methods and their acceptability within the prevail-
ing regulations and law of the hospital and state. JCAHO, but not CMS,
recommends that families be informed of the institution’s practices and
policies regarding restraint and seclusion. Both agencies concur, how-
ever, on the need for restraint and seclusion practices to be a part of the
patient’s treatment plan. The APA task force concurs with some of these
recommendations. For instance, family knowledge about an institu-
tion’s policy on restraint and seclusion appears reasonable. However,
on the matter of the treatment plan, it should be recognized that re-
straint and seclusion are emergency procedures. Thus, their employ-
ment cannot be anticipated in most treatment plans unless the patient’s
history of previous restraint needs is known to staff. On the other hand,
new patients should be informed about existing restraint and seclusion
techniques and policies on the unit to which they are assigned.
Specific Techniques of Seclusion and Restraint

Once the decision has been made to proceed with seclusion or restraint
of an agitated or disruptive patient, a seclusion or restraint “leader” is
chosen among available clinical staff. Sufficient personnel consists of at
least one person per limb. Staff should gather in a “show of force.”
Rather than appear “combat ready,” the supporting staff should convey
an air of confidence and calm, a measured control, reflecting a detached
and professional approach to a routine and familiar procedure. A seclu-
sion monitor is designated to clear the area of other patients and phys-
ical obstructions to entering the seclusion room. In addition, the moni-

tor stands clear of the physical action, noting any and all injuries or
difficulties with physical technique, thus allowing for an accurate cri-
tique of the seclusion procedure after the event.
Any confrontation with the patient begins with a clear statement of
purpose and rationale for the seclusion or restraint. The patient is given
few and clear behavioral options without undue verbal threat or prov-
ocation. For example, the patient is told that his or her behavior is out
of control and that a period of seclusion is required to assist the patient
to regain control. The patient is then asked to walk quietly to the seclu-
sion room accompanied by staff. Because the decision for seclusion has
already been made, negotiation or psychodynamic interpretation at this
juncture is superfluous and leads only to an escalation of disruptive be-
havior, potentially aggravating the violence of the event. At this point
the team should position itself around the patient in such a manner as
to allow rapid access to the patient’s extremities. At a predetermined
signal from the leader, physical force commences, with each staff mem-
ber seizing and controlling the movement and each limb restrained at
the joint by a member of the team. The patient’s head must be controlled
to prevent him or her from biting. With the patient completely re-
strained on the ground, additional staff may be called to secure the
limbs and to prepare to move the patient to the seclusion room or to ap-
ply mechanical restraints. In the most violent of cases, staff may need to
carry the patient into the seclusion room. This involves physically lift-
ing the patient in the recumbent position with arms pinned to sides,
legs held tightly at the knees, head controlled, and force applied uni-
formly to the back, hips, and legs.
If the patient is taken to seclusion, he or she should be positioned on
his or her back with the head toward the seclusion door and feet in the
opposite direction. An assessment should be made regarding transfer to
a hospital gown, and special attention should be paid to rings, belts,
shoelaces, and other potentially destructive objects. Medication may be
injected at this time while the patient is physically restrained. For the
most violent patients, the cross-arm-vise maneuver is again established,
allowing attendants to control the head and both arms in preparation
for leaving the seclusion room. The staff exit in a coordinated fashion,
one at a time, releasing legs first and arms last.
Mention should be made as to whether the patient is restrained face
up or face down. The face-down posture is safer because the patient is
less apt to bite or aspirate. However, the task force is aware of instances
in which patients have suffocated from being so restrained. This is a
particular risk in obese patients or in instances where there is a medical
illness such as a goiter or another medical condition that can obstruct

breathing. Such patients should be restrained face up. In any event, cau-
tion should be exercised in the placing of knees on any patient’s back so
as to avoid compromising breathing. The monitoring of breathing ade-
quacy is critical to any restraint process.
A debriefing follows each seclusion or restraint maneuver to review
the technique and progress of the event and allow an emotional release
of tension for the staff members. The restraint or seclusion should be
discussed openly among the patient population to allay or uncover
fears associated with the eruption of violence and staff use of force. The
patient should also be asked about the experience and whether it con-
tributed to or worsened his or her sense of control. Documentation of
the restraint episode is undertaken and written in the chart or on an in-
cident form.
Observation
During the period of time the patient is in restraints or seclusion, obser-
vations regarding behavior should be made every 15 minutes by appro-
priately trained nursing staff. For those patients in four-point restraints
or in lesser degrees of restraint in combination with seclusion, observa-
tions must be continuous. In the first edition of its report, the APA task
force considered 15-minute observations as satisfied by visual checks
and recommended in-person evaluation every 2 hours minimally.
JCAHO mandates 15-minute checks and allows seclusion monitoring to
be done by video camera, whereas CMS makes no comment on tech-
nique or frequency of observations, deferring to parameters spelled out
in hospital policy. The APA task force adheres to the 15-minute observa-
tion rule, although it recommends more frequent monitoring if clinically
indicated, for example, in instances of self-destructiveness that may in-
volve headbanging. The task force recognizes that continuous television
monitoring of patients in seclusion is common, and it approves such
methods of observation provided that appropriate use is made of the
monitor. The screen itself should be placed in an area of the nursing sta-
tion conducive to privacy. Visual observation checks (as opposed to in-
person assessments) should comment on the patient’s behavior while as-
certaining that the patient is not injuring him- or herself. Observations
should also determine that the patient is not at risk for physical exhaus-
tion or hyperpyrexia due to exertion while taking psychotropic drugs.
Patients should be seen in person every 2 hours—more often if clin-
ically indicated, such as if the patient is banging his or her head. If agi-
tated patients are to be approached in the seclusion room, the same
number of staff should enter the room as were required in the first in-
stance to subdue the patient (e.g., one for each extremity). Once a pa-
tient is quiet, direct observation with the seclusion room door open
should be made so that the state of the patient and a description of ver-
bal interchange can be documented on the patient’s chart. If the patient
is in restraints, the pulse, blood pressure, and range of motion of ex-
tremities should be assessed. Table 17–1 summarizes times regarding
orders and observations.
Care of the Patient in Seclusion or Restraints

Toileting of the patient should be allowed at least every 4 hours. The de-
sign of some seclusion room facilities is such that the patient may have
to physically exit to accomplish this, or the patient may have to be re-
moved from restraint devices. In situations in which this cannot be car-
ried out for reasons of danger, toileting can be done through use of the
urinal or bedpan. Privacy should be considered. Meals should be
brought to the patient at regular intervals when the remainder of the
ward is served. All articles should be blunt; plastic knives and forks can
be used as weapons. Mealtime can be dangerous for belligerent pa-
tients, who can use food as a weapon. In certain rare instances with se-
verely regressed patients, the food tray may be placed within the room
and the patient allowed access to it without staff persons being present.
However, the rationale for this solitary meal should be strictly docu-
mented in nursing notes. Whenever possible, feeding should be a time
of interaction between patient and staff.
The proper administration of fluids is vital for patients in restraint
or seclusion, particularly for those who perspire profusely and are
prone to dehydration. Documentation of fluid intake, although often
difficult with regressed patients, is still requisite.
Patients in restraint and seclusion may exhaust themselves from the
physical activity of pushing or pulling against restraint devices or walk-
ing or running around the seclusion room itself. Some severely re-
gressed patients may be menstruating or prone to fecal soiling. Al-
though not dangerous, such behaviors are often sufficiently repugnant
to others to cause avoidance. Hence, the patient is ignored and ap-
proached with trepidation. Negligence is thus a potential hazard in the
seclusion of such patients.
There is the possibility of a worsening of a psychosis due to de-
creased sensory stimulation inherent in seclusion room use. The patient
may become more delusional as a function of being isolated. It has been
suggested that the emotional impact of seclusion is severe and that
some debriefing is necessary following removal from seclusion to miti-

gate painful memories (Wadeson and Carpenter 1976).
Seclusion Room Design

A full awareness should exist regarding the hazards of the seclusion
room. Presently, no standard seclusion-room architecture exists. Theo-
retically, the seclusion room is an empty cubical with a high ceiling and
recessed lamp fixtures. All walls and ceilings should be made of mate-
rial that cannot be gouged out by a patient intent upon self-harm. For
example, plasterboard walls are not acceptable. Padded walls can be
used, provided that the integrity of the material used is high and the
surfaces are clean; however, there are insufficient data available to make
this a formal recommendation of the APA task force. Although fire laws
sometimes dictate otherwise, the door to the room should open out so
that a patient cannot barricade him- or herself inside. Protuberances
within the room, such as oxygen jets, are dangerous. Windows must be
constructed of safety Plexiglas or otherwise shielded from breakage.
The mattress should be the only furnishing in the room; a full bed, even
when bolted to the floor, poses a danger because the patient can jump
from it and injure him- or herself. The mattress should be constructed
of durable foam and not fiber or another substance that the patient
could conceivably use for hanging or self-suffocation. The mattress
should not be flammable. Patients should always be searched before be-
ing placed alone in seclusion. The issue here is that a violent patient
may become self-destructive when placed in isolation. Self-mutilative
acts can occur, headbanging can occur, or a patient can throw him- or
herself against a wall if unrestrained.
Removal From Seclusion and Restraint

Patients may be released from seclusion when the goals of the treatment
have been achieved—that is, when a patient’s behavior is under control
and no longer poses a threat to self or others or a further disruption to
the therapeutic milieu. The relative ability of a patient to control his or
her behavior is observable many times during the course of seclusion.
At each exit from and reentry into the seclusion room for the purpose of
feeding, bathing, or examining the patient, responsiveness to verbal di-
rection can be judged. The patient can be asked about his or her control
of feelings. Cooperation with physical examinations are also important.
All these parameters form a data base for deciding to wean the patient
from restraints and seclusion. However, removal from restraint and se-
clusion does not have to be abrupt. Indeed, graduated steps toward

freedom are often safer. Patients can also have their restraints partially
removed first and then be observed for a period of time, or the door of
the seclusion room can be opened. Reactions to these events then form
the basis for further release procedures.
Emergency Medication
Confusing definitions and recommendations regarding medication
have been put forth by CMS and JCAHO. CMS allows medications to
be used “as part of an approved treatment plan for the patient’s diagno-
sis.” If, however, a medication is not so used, the intervention becomes
a “chemical restraint” and is thus viewed the same as a physical re-
straint and subject to CMS regulations. JCAHO considers the use of
medication to restrict a patient’s freedom of movement to be improper.
The APA task force believes that psychotropic medication may or may
not restrict movement but that it can be a powerful aid to patients who
are struggling to control behavior. The task force does not endorse the
term chemical restraint because it is both a misnomer and pejorative. The
use of medication obviously depends on the nature of the patient’s con-
dition, the degree of agitation, and the qualitative nature of the aggres-
siveness. If the patient is flagrantly psychotic or in an extremely agitated
manic state, medication may be indicated. Medication, if rationally
used, may shorten the length of stay in seclusion by helping the patient
to gain mastery over aggressive urges. One hazard, however, of medi-
cating assaultive patients is that the patient may be rendered so lethar-
gic that he or she becomes disorganized and combative as a function of
organic impairment. Thus clinicians need to navigate between under-
and over-medication and document specific target symptoms that re-
spond to psychotropic agents.
Some patients in restraint and seclusion can be offered medication
orally, a tactic more conducive to dignity than other routes. On the other
hand, it is well known that there are some patients so flagrantly ill that
a parenteral injection is needed. Parenteral medication is rarely curative
of an underlying psychosis but is used basically to induce symptomatic
improvement. Further and more vigorous treatment must ensue before
the core symptoms such as delusions and hallucinations abate.
“As-needed” dosing of drugs should be avoided. If drugs are used
during the restraint and seclusion process, the goals of administration
should be spelled out in specific orders, for example, “Haldol 5 mg po
q4h until belligerence and aggressiveness abates.” Clinicians who are
involved in the care of violent patients should be familiar with a variety
of parenteral drugs, including benzodiazepines, for use in the manage-

ment of aggression (Tardiff 1996).
Unique Restraints
It is possible to use various restraint devices in a creative fashion that
allows the patient to mingle with others on a ward or within the room.
The use of garments that restrain extremities or that bind older patients
to a wheelchair may allow the individual to participate in group meet-
ings and receive milieu enrichment that would not occur in a seclusion
room. The use of PADS (Protective Aggression Devices) allows a belt-
like device to be applied to the waist and wrists or to the ankles but in
a manner that permits some range of motion, including ambulation
(Van Rybroek et al. 1987). An advantage of these devices is that they en-
able the patient to take part in ward activities while still restricting po-
tentially dangerous arm or leg movements. There is flexibility in the use
of PADS, so that the nondominant hand can even be released from the
belt line as a function of improved behavior.
Contraindications to Seclusion and Restraint

Restraint and seclusion may be contraindicated on the basis of the pa-
tient’s clinical condition. For example, unstable medical status resulting
from infection, cardiac illness, disorders of thermoregulation, or meta-
bolic illness may make restraint a preferable intervention to the isola-
tion of seclusion. In some conditions of delirium or dementia, the pa-
tient’s vulnerability to sensory deprivation as a pathogenic force may
lead to worsening of the total clinical state. Patients prone to serious
and uncontrollable self-abuse and self-mutilations are also at risk in se-
clusion. With physical restraints, circulatory obstruction is always a
hazard, but this can be minimized by temporarily releasing one of the
four-point restraints every 15 minutes. When a patient is lying on his or
her back while restrained, aspiration is always a risk.
Seclusion of a patient as a purely punitive response is contraindi-
cated, nor should a patient ever be secluded for the convenience of the
staff or because of staffing difficulties or shortages.
Danger and Injury

The implementation of seclusion and restraint procedures places staff
and patients at high risk for injury. One-half of all assaults on staff occur
during the process of secluding or restraining disruptive patients or in

the initiation of seclusion (Mattson and Sacks 1978; Tardiff 1981). Well-
rehearsed restraint and seclusion techniques are the best safeguards
against this. Patient injuries can occur from improperly worn restraint
devices that can potentially restrict breathing. Thus, high chest vests
should be avoided. Even when placed in restraints that restrict arm
movement, some very thin patients can wriggle their hands and arms
with sufficient mobility to engage in self-mutilation such as scratching
of their eyes.
Forensic Aspects of Seclusion and Restraint

Seclusion and restraint are high-profile techniques, and both the public
and some clinicians have high hopes of abolishing these hands-on treat-
ments in the future. Indeed, some psychiatric facilities pride themselves
on using virtually no restraint and seclusion. Although in these in-
stances diligent use of behavior modification plans may be effective, a
question is always raised concerning admission criteria, transfer to
other facilities in the case of violence, or overzealous use of medication.
It seems reasonable to the APA task force that if a hospital elects to treat
violent and/or self-destructive mental patients, it will need to have
available to it restraint appliances and seclusion rooms. Generally
speaking, of the two modalities, restraint is the more hazardous inter-
vention because it involves subjugation of the patient during which in-
jury or death can occur. Cases of litigation have arisen from suffocation
induced by holding an obese patient to the ground, face down, without
proper observation. Occasionally, a restrained patient may be hurt by
another patient who takes advantage of the former’s immobility. How-
ever, failure to restrain can also arise as a cause of litigation, as when a
patient who is not properly controlled becomes violent toward others
or injures him- or herself. The patient who is unrestrained in seclusion
may be capable of self-injury as well.
Restraint and Seclusion Committee

A new recommendation made by the APA task force is the formation of
a local monitoring agency for restraint and seclusion use. In a specific
local area, such a group of administrators or clinicians could oversee the
extent of practices, review complicated protocols such as those needed
for long-term restraint, and review any injuries that might occur. The
documentation of injuries is already part of JCAHO policy.
Key Points
■ Seclusion and physical restraint are techniques to manage violent
patients in hospitals that are regulated by governmental agencies
and the psychiatric profession.
■ Seclusion and restraint are used in emergencies to prevent immi-
nent harm to other persons, as well as to the patient, and to pre-
vent serious disruption of the treatment environment.
■ Seclusion or restraint may be contraindicated on the basis of the
patient's clinical condition and should not be used as a punitive
response or for the convenience of the staff.
■ Seclusion and restraint must be ordered by a physician or, in some
situations, by another licensed independent practitioner who
must see the patient face to face within one hour after initiation
of seclusion or restraint.
■ The duration of seclusion or restraint is limited on the basis of the
patient’s age.
■ Clinicians must know and adhere to the policies of the hospital in
which they practice in regard to other parameters such as tech-
niques of using and renewing seclusion and restraint.
■ The patient in seclusion or restraint must be observed properly
and receive adequate nursing care and medical care, including the
use of medication.
■ Patients may be released from seclusion or restraint when the
patient's behavior is under control and no longer poses a threat to
self or others or a further disruption of the treatment environ-
ment.
References
Ahmed MB, Lepnurm M: Seclusion practice in a Canadian forensic psychiatric
hospital. J Am Acad Psychiatry Law 29:303–309, 2001
American Psychiatric Association: Seclusion and Restraint: Report No 22 of the
American Psychiatric Association Task Force on Seclusion and Restraint.
Washington, DC, American Psychiatric Association, 1985
American Psychiatric Association Task Force on Seclusion and Restraint: Seclu-
sion and Restraint: Report of the American Psychiatric Association Task
Force on Seclusion and Restraint (unpublished). Submitted to the Ameri-
can Psychiatric Association, 2006
Binder RL: The use of seclusion on an inpatient crisis intervention unit. Hosp
Community Psychiatry 30:266–269, 1979
Busch A, Shore MF: Seclusion and restraint: a review of the recent literature.
Harv Rev Psychiatry 8:261–270, 2000
Convertino K, Pinto RP, Fiester AR: Use of inpatient seclusion at a community

mental health center. Hosp Community Psychiatry 31:848–850, 1980
Crenshaw WB, Francis PS: A national survey on seclusion and restraint in state
psychiatric hospitals. Psychiatr Serv 46:1026–1031, 1995
Crenshaw WB, Cain KA, Francis PS: An updated national survey on seclusion
and restraint. Psychiatr Serv 48:395–397, 1997
Dubin W, Lion JR (eds): Clinician Safety: American Psychiatric Association Task
Force Report No. 33. Washington, DC, American Psychiatric Press, 1993
Joint Commission on Accreditation of Healthcare Organizations: The Official
Handbook. Oakbrook Terrace, IL, Joint Commission on the Accreditation
of Healthcare Organizations, 2002
Kaltiala-Heino R, Tuohimaki C, Korkeila J, et al: Reasons for using seclusion
and restraint in psychiatric inpatient care. Int J Law Psychiatry 26:139–149,
2003
Lavoie FW: Consent, involuntary treatment, and the use of force in an urban
emergency department. Ann Emerg Med 21:25–40, 1992
Lion JR, Danto BL: The hardware of violence containment, in Creating a Secure
Workplace. Edited by Lion JR, Dubin WR, Futrell DE. Chicago, IL, Ameri-
can Hospital Publishing, 1996, pp 195–208
Mattson MR, Sacks MH: Seclusion: uses and complications. Am J Psychiatry
135:1210–1213, 1978
Oldham JM, Russakoff LM, Prusnofsky L: Seclusion: patterns and milieu.
J Nerv Ment Dis 171:645–650, 1983
Plutchik R, Karasu TB, Conte HR, et al: Toward a rationale for the seclusion.
J Nerv Ment Dis 166:571–579, 1978
Ray KN, Rappoport ME: Use of restraint and seclusion in psychiatric settings in
New York State. Psychiatr Serv 46:1032–1037, 1995
Sailas E, Fenton M: Seclusion and restraint for people with serious mental ill-
nesses. Cochrane Database Syst Rev(2):CD001163, 2000
Schwab PJ, Lahmeyer RN: Uses of seclusion on a general hospital psychiatric
unit. J Clin Psychiatry 40:228–231, 1979
Soloff PH, Turner SM: Patterns of seclusion: a prospective study. J Nerv Ment
Dis 169:37–44, 1981
Swett C: Inpatient seclusion: description and causes. Bull Am Acad Psychiatry
Law 22:421–430, 1994
Tardiff K: Emergency measures for psychiatric inpatients. J Nerv Ment Dis
169:614–618, 1981
Tardiff K (ed): The Psychiatric Uses of Seclusion and Restraint. Washington, DC,
American Psychiatric Press, 1984
Tardiff K: Assessment and Management of Violent Patients. Washington, DC,
U.S. Department of Health and Human Services: 42 CFR Part 482. Medicare and
Medicaid programs; hospital conditions of participation; final rule. Fed
Regist 64:36069–36089, 1999
U.S. Department of Health and Human Services: 42 CFR Part 483.358. Orders
for the use of restraint or seclusion. Fed Regist 66:71477164, 2001
Van Rybroek GJ, Kuhlman TL, Maier GJ, et al: Preventive aggression devices
(PADS): ambulatory restraints as an alternative to seclusion. J Clin Psychi-
atry 48:401–405, 1987
Wadeson H, Carpenter WT: Impact of the seclusion room experience. J Nerv

Ment Dis 163:318–328, 1976
Way BB: The use of restraint and seclusion in New York State psychiatric cen-
ters. Int J Law Psychiatry 8:383–393, 1986
Wells DA: The use of seclusion on a university hospital psychiatric floor. Arch
Gen Psychiatry 26:410–413, 1972
P A R T V
Special Populations
C H A P T E R 1 8
Children and Adolescents

Peter Ash, M.D.
Violence is surprisingly common among children and adolescents:

four longitudinal studies in the United States using youth self-reports
have shown that by age 17, 30%–40% of boys and 16%–32% of girls have
committed a serious violent offense, defined as an aggravated assault,
robbery, gang fight, or rape (U.S. Department of Health and Human Ser-
vices 2001). Only a small fraction of these offenses resulted in arrest. De-
spite the dramatic drop in youth homicide rates since 1993, homicide re-
mains the second leading cause of death in 15- to 19-year-olds, after
accidents and ahead of suicide, accounting for approximately 1,900
deaths in the United States per year between 1999 and 2004—a rate of 9.3
per 100,000 (Centers for Disease Control and Prevention 2007). Violence
in youth appears in many forms, ranging from the benign, relatively
friendly wrestling on the schoolyard playground through such varieties
as bullying and dating violence to the more extreme gang-related kill-
ings and school shootings with multiple victims. Developmentally, the
onset of violence is a phenomenon of childhood and adolescence: if a
person has not committed a serious violent offense by his or her early
20s, the likelihood that he or she will ever do so is quite low.
Although mental health clinicians tend to look at violence as a men-
tal health problem or symptom, it is not at all clear that youth violence
is best thought of as caused by mental health problems or that the most
efficacious interventions are traditional mental health interventions.
Youth violence is a major public health concern and a focus of the juve-
nile justice system in addition to being a problem facing mental health
clinicians. It is therefore important for clinicians dealing with violent
359
youth to keep other perspectives—and other types of intervention—in

mind. Few now look at adult criminals and expect the mental health sys-
tem to prevent their recidivism, whereas delinquent youth are seen as
more amenable to mental health intervention, and a central mission of
the juvenile court is to rehabilitate them. Those whose violence is a
product of a psychotic illness make up only a small minority of youth
whose violence is a focus of attention. Therefore, although many of the
general principles pertinent to the assessment and management of
adults detailed elsewhere in this volume are relevant to the assessment
and treatment of violent youth, because of youths’ developmental dif-
ferences, different living circumstances, different precipitants, and dif-
ferent legal status, approaches to younger patients are often different
from those utilized with adults. Key differences are shown in Table 18–1.
Epidemiology
Aggression is a common behavior in a child’s development. A high per-
centage of an 18-month-old’s peer interactions involve aggression, of-
ten in reaction to frustration or wanting something another child has.
By age 2½, after the child has developed more social skills and lan-
guage, the frequency of physical peer aggression drops significantly,
and it continues to decrease until age 6 as most children shift to verbal
types of aggression. Most of the preschool child’s aggression is directed
at peers. Much of how a child learns to handle aggression is mediated
by parenting, so children who deviate from normal development and
are identified early can often be helped by parental interventions.
Aggression remains common in elementary school children. Data from
a large-scale longitudinal survey of Canadian children indicated that par-
ents rated as “sometimes” or “often true” that more than one-third of boys
and about 30% of girls ages 4–11 get into many fights and about 20% of
boys and 10% of girls physically attack people (Offord et al. 2001). Of the
15 DSM-IV-TR criteria for conduct disorder, seven code for physical ag-
gression (American Psychiatric Association 2000), so rates of conduct dis-
order give some indication of the frequency of rates of maladaptive aggres-
sion in elementary school–aged children. Epidemiological studies report
rates of conduct disorder in elementary school–aged boys as approxi-
mately 3%–7% (Loeber et al. 2000), with considerably lower rates in girls.
Violence is common throughout adolescence: in the United States,
about 30% of 12-year-old boys and 25% of 17-year-old boys surveyed in
a large-scale study in 2005 reported having gotten into a serious fight in
the past year (U.S. Department of Health and Human Services 2006). For
girls, the rates were only about one-third lower. In the same study, about
Children and Adolescents ❘ 361
TABLE 18–1. Key differences between violent behavior in adults

and adolescents
Category Compared with adults, for adolescents:
Epidemiology Violence is much more common.

Homicide accounts for a higher proportion of all
deaths.
Violent careers are shorter.
The first episode of serious violence most
commonly occurs in adolescence, sometimes in
childhood, and rarely in adulthood.
Diagnostic differences Conduct disorder is specific to children and
adolescents and is diagnosed on Axis I.
Antisocial personality disorder cannot be
diagnosed in those younger than age 18 and is
diagnosed on Axis II.
Psychotic disorder is much less common.
Behavior patterns Violent behavior occurs more in groups.
Treatment Peer group considerations are key.
Family involvement in treatment is more important.
Legal status Confidentiality and consent issues are more
complex because minors typically cannot consent,
control record release, or waive rights against self-
incrimination.
Legal consent for treatment needs to be provided by
someone other than patient.
Hospitalization over the patient’s objection can
often be accomplished without resorting to civil
commitment.
Patient’s responsibility for treatment compliance is
reduced.
Much criminal behavior is adjudicated in juvenile
court.
10% of adolescent boys and 3%–4% of girls reported that in the past year
they attacked someone with intent to seriously hurt the victim. Bullying
is a common middle school variant of violent behavior, practiced by
about 13% of sixth to tenth graders (Nansel et al. 2001). Adolescent dat-
ing violence also occurs with high frequency. In a nationally representa-
tive sample of high school students, about 9% of both girls and boys re-
ported being physically hit by a boyfriend or girlfriend in the previous
year (Centers for Disease Control and Prevention 2006). Interestingly, the
rates of dating violence were not significantly different for boys and
girls, unlike most other forms of violent behavior. Dating violence was
most strongly associated with the risk factors of being sexually active
and having attempted suicide. The cumulative prevalence of committing
a serious violent offense by age 17 is estimated at 30%–40% for boys and
16%–32% for girls. Although African American youth are arrested at
much higher rates than white youth, the self-report data cited above
show much smaller racial differences. The peak age for the onset of vio-
lent behavior occurs in adolescence, around age 16 for boys (Elliott 1994).
These rates of violence appear to have been fairly stable over the
past several decades (U.S. Department Health and Human Services
2001). However, adolescent homicide rates have been quite variable:
rates for white males tripled from 1964 to 1991 and then over the ensu-
ing 10 years fell back to the rates of the 1970s (National Center for
Health Statistics 2004). Thus, although the frequency of violence has re-
mained fairly constant, the lethality of that violence has varied consider-
ably. Both the increase and decrease of adolescent homicide rates were
linked to changing rates of using firearms by adolescents (Snyder and
Sickmund 2006). The involvement of youth in the crack trade and in-
creased gang activity led to an increase in youth homicide. Despite the
fact that possession of a handgun by an adolescent is illegal, fear on the
street led more youth to carry handguns for protection, which led to
more homicides and a spiraling cycle of yet more fear (Blumstein 2002).
In the mid-1990s, one study showed almost all incarcerated male delin-
quents owned a handgun (Ash et al. 1996). Possession of a handgun
markedly raises the potential lethality of a violent confrontation. After
the mid-1990s youth (and, to a lesser extent, adult) violent crime rates
dropped markedly. The reasons for the crime drop remain controversial
but appear related to increases in the prison population, increases in the
number of police, the decline of crack, and legalized abortion (Levitt
2004). The cycle of fear went into reverse, and firearm carrying by youth
decreased. The central role of guns in the lethality of youth violence ob-
viously has major implications for intervention.
Developmental Trajectories
Much of what we know about the development of violence has been
learned from longitudinal studies of youth. The majority of researchers
recognize at least two main patterns: an early-onset trajectory in which
the youth engages in serious violence before puberty, and a late-onset
group who do not engage in serious violence until adolescence (Moffitt
1993; National Institutes of Health 2004; U.S. Department of Health and
Human Services 2001). Significant differences between these two trajec-

tories are shown in Table 18–2. Those with early onset have more severe
and longer courses and are more difficult to treat. With research cur-
rently available, the late-onset group cannot be identified prospectively
from preadolescent symptoms, although in retrospect they experienced
many childhood risk factors.
Children first learn to manage their aggression from their parents in
toddlerhood, and poor parenting in this period sets the stage for later
problems (Tremblay et al. 2004). Poor parenting may involve abusive
parental behavior, neglect, coercive parenting, parenting by antisocial
parents, poor limit setting, or general family dysfunction. Oppositional
defiant disorder (ODD) is a frequent precursor of more serious aggres-
sive behavior, and about 30% of those with early-onset ODD progress
to conduct disorder (Connor 2002; Loeber et al. 2000). Of those with
conduct disorder, about 40% will progress to antisocial personality dis-
order (Zoccolillo et al. 1992). The most potent risk factors for preadoles-
cent violence are general, nonviolent criminal offenses and preadoles-
cent substance abuse (Hawkins et al. 2000), whereas peer effects become
the most potent risk factor in adolescent-onset violence. For both early-
onset and adolescent-onset types, there appears to be a developmental
progression of offenses, beginning with minor crime such as vandalism
and shoplifting, then progressing to aggravated assault, then robbery,
and then rape (Elliott 1994). That robbery precedes rape in more than
70% of cases is some of the strongest evidence that rape is a crime of vi-
olence, not a crime of sex. Longitudinal studies suggest that most seri-
ous violent crime—in fact, most youth crime of all types—is committed
TABLE 18–2. Comparison of developmental trajectories toward

violence
Characteristic Early onset Late onset
Onset of offending Before puberty After puberty

Serious violent
offenders, % 30±15 70±15
Violent career longer
than 2 years, % 13 2
Strongest risk factors General offenses Weak social ties
(effect size r>0.30) Substance use Antisocial delinquent
peers
Gang membership
Source. Data excerpted from Youth Violence: A Report of the Surgeon General (U.S. De-
partment of Health and Human Services 2001).
by a relatively small minority of offenders. Whereas more than one-

third of adolescents have committed a serious violent offense, about
5%–10% of youth are committing more than 75% of the violent crimes
(U.S. Department of Health and Human Services 2001).
Substance abuse, especially alcohol and marijuana, and mental dis-
order are common among incarcerated delinquents. Excluding conduct
disorder, about two-thirds of incarcerated delinquents meet diagnostic
criteria for an Axis I mental disorder (Marsteller et al. 1997; Teplin et al.
2002) and exhibit rates of disorder about triple that of the normal pop-
ulation. Axis II personality disorders are also more common among ad-
olescent offenders (Johnson et al. 2000). However, whether there is a
causal link between mental disorder and violence in adolescence re-
mains unclear.
The good news is that for most youth, violence is limited to adoles-
cence: even in the early-onset type, fewer than one in seven continue as
serious violent offenders into adulthood. The fact that so much violence
is limited to adolescence has important implications for social policy.
Zimring (2005) suggested that we consider adolescents as having a
“learner’s permit” to experiment, recognizing that experimentation
will bring with it mistakes. Juvenile justice policy, in his view, should
aim to minimize the harm of those mistakes and help those who have
trouble learning from them, rather than focusing on punishment.
Risk Factors
The high rates of violence in adolescents, compared with the general pop-
ulation, indicate that adolescence itself is a risk factor. The considerable
literature on risk factors for youth violence demonstrates numerous risk
factors at the levels of individual, family, and community (Connor 2002;
Hann 2002; Hawkins et al. 2000). The risk factor literature is complex for
several reasons. First, violence is a heterogeneous group of behaviors, and
risk factors differ for different types of violence. Second, not only are there
numerous risk factors in different domains, but given the dynamic nature
of development, different risk factors become salient at different ages. For
example, having a delinquent peer group is a potent risk factor for ado-
lescents but not for preadolescents. Third, risk factors may interact: for ex-
ample, there is considerable evidence from twin and adoption studies
that some genetic risk factors, such as having an antisocial biological par-
ent or having the low–monoamine oxidase A allele, are much more likely
to be expressed in violent behavior when an adopted child is raised in an
adverse home environment (Caspi et al. 2002; Foley et al. 2004). Finally, as
with suicidality, no combination of risk factors can predict with much
confidence whether a particular individual will become violent. From a

public health perspective, knowledge of risk factors guides prevention ef-
forts; from a clinical perspective, risk factors provide a structure for ob-
taining information and may point toward areas needing intervention.
Some of the many risk factors for violence noted in the literature are
listed in Table 18–3.
Case Examples
Case 1: Early-Onset Course
Bruce, age 13, was referred for treatment as a condition of probation for
carrying a handgun while “on duty” as a lookout for a drug seller. He
presented as an irritable teenager who initially resented having to come,
but he was quite talkative in the initial evaluation session. He had been
in foster care for 3 years beginning at age 4 when his mother was sent to
prison on a drug charge, but he was returned to her care when he was 7.
His father was unknown. His mother reported oppositional behavior at
home after age 7 and theft from other youths at school. Despite this his-
tory, he had obtained a C average in school. When he was 9, he got mad
and killed a dog with a baseball bat, and a year later, he got angry dur-
ing a baseball game and hit another player with a bat. The school re-
ported he was a bully and hung out with a peer group that harassed
other students. He had recently joined a gang and proudly showed the
evaluator the gang tattoo on his shoulder.
Case 2: Possible School Shooter

Jeremy, age 13, was suspended from school pending “psychiatric clear-
ance” when a teacher found him doodling pictures of guns on a piece of
paper that was entitled “Hit List” and listed six students in his class. Jer-
emy had no known history of violence, but he did have a long history of
not fitting in with peers. A previous therapist had diagnosed him with
pervasive developmental disorder not otherwise specified. Academi-
cally he had obtained average grades. He had complained to his parents
that “lots of kids make fun of me” and that he had been bullied at school
on numerous occasions. At the request of the evaluator, his parents
checked his computer for recent sites visited and found that he had vis-
ited a number of sites dedicated to the Columbine and Virginia Tech
school shootings. His father liked to hunt and had four rifles in the home.
Assessment
Violence, both prospective and completed, encompasses a wide range
of behaviors that call for differing approaches to assessment and inter-
vention. Violent youth are involved in multiple systems, and depend-
ing on the referral, a clinician may take one of a variety of roles, such as
TABLE 18–3. Risk factors for violence

History of prior criminal acts, including nonviolent offenses
Individual factors
Biological factors
Physiological under-arousal, including lowered heart rate
Impairments in frontal lobe functioning
Abnormal serotonin levels
Temperament
Antisocial biological parent
Psychopathology
Psychopathy
Oppositional defiant disorder, conduct disorder
Attention-deficit/hyperactivity disorder, substance abuse, mood disorder
Poor social skills
Poor school performance
Learning disabilities
Low IQ
Family factors
Poor parenting, including abuse and neglect
Antisocial parent
High family dysfunction
Negative peer relations
Delinquent peers
Gang membership
Community factors
Neighborhood crime
School tolerance of bullying and antisocial behavior
Disadvantaged neighborhoods
Availability of drugs
primary therapist, medication manager, or forensic evaluator, each of

which will call for a different type of assessment. Table 18–4 highlights
some of the dimensions in assessment that provide important informa-
tion for assessing risk and developing a treatment plan. In a full assess-
ment, it is important to obtain information from collateral sources, in-
cluding parents, schools, and often peers.
The assessment should take place in an environment where both the
clinician and patient can feel safe. For high-risk youth, this requires a
setting where the youth can be screened for weapons, where no objects
TABLE 18–4. Dimensions to consider in assessing youth violence

Clinical component Example issues
History of past violence Developmental trajectory, age at onset, recent

behaviors
Social setting Individual versus group offending
Nature of relationship to victim
(intrafamilial–stranger)
Gang involvement
Psychiatric diagnosis Comorbid conditions such as attention-deficit/
hyperactivity disorder, posttraumatic stress
disorder, mood disorders, pervasive
developmental disorder, or psychopathic
personality traits
Risk factors See Table 18–3
Protective factors Intolerant attitude toward delinquent behaviors,
high IQ, commitment to school
Intent Impulsive versus predatory
Potential lethality Carrying weapons
Imminence of risk Near future, long-range risk
are present that can be used as weapons, and where others are rapidly
available in the case of an impending assault from the patient.
Consent and Confidentiality

In discussing past violence with a youth, the interviewer may be hear-
ing about criminal acts, and because such information could potentially
be used to further criminal prosecution, issues regarding informed con-
sent and confidentiality need to be thought through carefully. Consent
issues are more complex with minors for a number of reasons. First, mi-
nors typically are not deemed competent to provide legal consent and
do not control access to their medical records. Second, minors are less
able to understand the implications of material that could constitute a
confession and are more likely than adults to defer to the wishes of
authority to provide incriminating information. Third, because of the
rehabilitative mission of juvenile courts, juvenile courts have looser
standards for admissibility, and juvenile judges have considerable dis-
cretion in how they utilize mental health information in apportioning
rehabilitative services and punishment. Finally, even when information
is obtained in a relatively confidential treatment context, if the youth
later enters the custody of the juvenile justice system, such information
may be released. The evaluator therefore balances the need to obtain

relevant information, the ability of the youth to understand the confi-
dentiality and self-incrimination parameters of the assessment, and
how information is presented in written records and reports. This judg-
ment will vary depending on the nature of the assessment: an eval-
uation for outpatient treatment will be quite different from a court-
ordered assessment of whether a delinquent youth is dangerous and
should be transferred to adult criminal court jurisdiction. At the outset
of the evaluation, the nature of the evaluation and how the information
may be used should be explained in terms developmentally appropriate
to the youth, and information in written records should be worded in a
way that does not provide evidence for prosecution (e.g., “gave a his-
tory of shooting at a person,” rather than “shot Mr. Jones on March 13 of
last year”).
When treatment is mandated by the juvenile justice system, such as
by a probation requirement or in a detention facility, confidentiality
constraints need to be clear. Will the therapist be involved in making
dispositional decisions? Will the outpatient therapist be in communica-
tion with other care providers? How much information will be given to
law enforcement and correctional personnel? Given that effective inter-
vention usually involves a multimodal approach, communication with
other care providers is usually essential, but the clinician should be
clear with the patient as to what sort of information will be shared and
what will be kept confidential.
History of Violence
Overall, the best predictor of whether a behavior will occur in the future
is whether it is occurring in the present or has occurred in the recent
past (Tremblay and LeMarquand 2001). Therefore, a history of violence
is key. The clinician needs to obtain both chronological detail (such as
when violent behavior began and with what frequency it continued)
and detailed knowledge of violent events (e.g., precipitants, emotional
state during the assault, nature of the assault, feelings after). Less struc-
tured interviewing may obtain details missed by structured question-
ing. One useful approach for discussing a violent event with a child or
adolescent is to say, “Let’s suppose I was going to make a movie of what
happened. Could you describe what happened in enough detail so I
could do that?” Follow up with questions about the event and what led
up to it, and then, once the external nature of the event is clear, go back
and ask about feeling states at key points, for example, “Tell me what
was in your mind when he said [or did]….”
In addition to obtaining history of violent episodes from the child, it

is important to obtain collateral history from other sources, such as par-
ents, school, police reports, and in some cases, peers.
Diagnosis
The most common psychiatric diagnosis applied to youth with histories
of violence is conduct disorder, the main criterion of which is “a repeti-
tive and persistent pattern of behavior in which the basic rights of oth-
ers or major age-appropriate societal norms or rules are violated, as
manifested by the presence of three (or more) [of the listed behavioral
criteria that include bullying, getting into fights, using weapons, and
robbery]” (American Psychiatric Association 2000). Conduct disorder is
thus a phenomenological diagnosis encompassing a wide range of an-
tisocial behaviors. Children with the disorder typically have a history of
previous ODD, a diagnosis characterized by a pattern of negativistic,
hostile, and defiant behavior, but ODD does not have aggressive behav-
ior as a criterion. The American Academy of Child and Adolescent Psy-
chiatry has published practice parameters for the evaluation and treat-
ment of conduct disorder (American Academy of Child and Adolescent
Psychiatry 1997) and ODD (American Academy of Child and Adoles-
cent Psychiatry 2007). Antisocial personality disorder can only be diag-
nosed in adults and has as one criterion that there was evidence of con-
duct disorder prior to age 15. ODD and conduct disorder are Axis I
disorders, but when they progress to antisocial personality disorder, the
condition is classified as an Axis II disorder. For intervention purposes,
the construct of psychopathy, which is not included in DSM, may be
useful. Psychopathy encompasses the lack of remorse and the lack of
empathy components of antisocial personality but does not include the
more behavioral components. The most common metric for psycho-
pathy, the Hare Psychopathy Checklist, does have an adolescent ver-
sion, the Psychopathy Checklist–Youth Version (PCL-YV; Forth et al.
2003; see discussion of rating scales later). Personality disorder has been
associated with recidivism in delinquents (Steiner et al. 1999).
A comprehensive psychiatric diagnostic assessment is useful to
delineate disorders that may be contributing to violence risk. Violence
can be a symptom of many diagnoses in addition to conduct disorder,
including pervasive developmental disorder and bipolar disorder.
Conduct disorder has a very high comorbidity with attention-deficit/
hyperactivity disorder (ADHD). Many violent youth give histories of
exposure to violence, either as a victim or as a witness, and may meet
criteria for posttraumatic stress disorder. Substance abuse is a signifi-
cant risk factor, especially for preadolescents, and participation in a

drug culture is likely to expose the youth to violence. Treatment of un-
derlying conditions likely lowers violence risk.
Risk Factors and Risk Assessment Scales

Risk factors listed in Table 18–3 can provide a structure for obtaining
important information. Which risk factors are especially relevant de-
pends on the age and clinical situation. For example, biological factors
appear most potent in the context of adverse parenting and are most rel-
evant in young children; a history of bullying by a latency-age child
should spur an inquiry into school attitudes and policies toward such
behavior; and questions about gang membership and peer activities are
especially relevant for adolescents.
There has been rather limited work on protective factors, which are
thought of not simply as the absence of risk factors but as factors that
independently reduce the effect of risk factors. Proposed protective factors
include intolerant attitude toward delinquent behaviors, high IQ, and
commitment to school (U.S. Department of Health and Human Services
2001), but more research needs to be done in this area.
Following work on adult actuarial risk assessment scales, efforts
have been made to modify those scales to apply to adolescents (Vincent
2006). The two scales that have the most psychometric support are the
PCL-YV (Forth et al. 2003) and the Structured Assessment of Violence
Risk in Youth (SAVRY; Borum et al. 2005). The PCL-YV utilizes a 60- to
90-minute expert interview and provides a score but does not have cut-
off values for categorical diagnosis or risk of violence. The SAVRY
guides trained evaluators in a systematic assessment of risk factors as-
sociated with violence. Evaluators then make structured professional
judgments in considering the applicability of each risk factor to the ad-
olescent being evaluated. This leads to a final determination of risk as
low, medium, or high. Thus far, prospective validity of these scales has
not been demonstrated, but they do provide a structure for assessment.
There is much weaker empirical support for structured risk assessment
in girls (Odgers et al. 2005), and even assessments for case management
of girls are more problematic.
Predatory Violence
It is clinically useful to distinguish between aggression that is impul-
sive, reactive, hostile, and affective and aggression that is predatory, in-
strumental, proactive, and controlled (Jensen et al. 2007; Vitiello and
Stoff 1997), although many youth exhibit both. There is some evidence
that different neural pathways are involved (Blair 2004). The assess-
ment of the child in the first case example, in which there is a clear his-
tory of past impulsive participation in group violence, will be different
from the assessment of the child in the second case, for whom the key
issue is the risk of an individual’s acting alone in a cold-blooded, pred-
atory manner. A youth planning predatory violence is more likely to
conceal his thinking than is a youth who acts impulsively. Therefore,
more indirect information is necessary. Although psychiatrists who
work with youth are experienced in obtaining collateral information
from parents, they are less likely to be experienced in obtaining infor-
mation from peers. Yet the evaluee’s friends are the most likely—more
so than parents—to have heard the youth express threats, even if the
friends did not take the threat seriously. One commonality in the mass
school shootings by adolescents is that in each case the shooter had ex-
pressed threats toward others prior to the event (Verlinden et al. 2000).
Depending on the level of risk suggested by other indicators, a youth’s
friends can be telephoned (with the permission of the patient) or,
in higher-risk situations, questioned by law enforcement personnel.
Whenever risk of predatory violence by an adolescent is a serious con-
sideration, if at all possible some friends should be talked to.
The second key principle in assessing risk of predatory violence is to
think in terms of a pathway toward violence (Borum et al. 1999). This
threat assessment approach, first developed for the U.S. Secret Service
(Fein and Vossekuil 1998) and later adapted to school threat assess-
ments (O’Toole 1999; Vossekuil et al. 2002), advocates focusing less on
the profile of the subject and more on whether the subject is taking steps
toward targeted violence. The path begins with fantasizing about kill-
ing, progresses to beginning planning, which might involve increased
interest in weapons or learning about how others have conducted mass
shootings by reading on the Internet, and then moves on to more de-
tailed preparation, such as obtaining weapons, scouting out sites, and
following potential victims. The farther along this path a person is, the
more risk he or she poses. It is not necessary for a person to make a threat
in order to be a threat. Because an interviewee may deny intent to harm,
when interviewing a potential attacker, one also looks for “leakage,”
such as interest in weapons and interest in other attacks, that may indi-
cate moving on a path toward violence. It is also important to explore
the motivation for the behavior that brought the subject to attention. In
the case of the potential school shooter described earlier, it would be im-
portant to explore what he had in mind when he wrote the “Hit List.”
For cases that seem to pose medium to high threat, a team of investiga-
tors may be necessary to search for possible physical evidence or inter-
view corroborative sources. It should be remembered, however, that the

base rate of mass shootings is so low that the efficacy of this approach
has not been empirically tested.
Weapons
Because of the close link between weapon carrying and the lethality of
violence, a weapon assessment should be part of the evaluation of any
youth being assessed for violence. In one study, the rate of firearm own-
ership by boys who have been in detention approaches 100%, and for
girls it is about half that (Ash et al. 1996). The assessment should in-
clude a history of how and when the youth first obtained a gun, subse-
quently obtained weapons, and has access to non-owned guns in the
home or from peers. For impulsive aggression, the issue is less one of
access, because most youths can obtain a gun if they really want one,
than of how frequently, for what reasons, and under what conditions
the youth carries a weapon and how often and under what conditions
he or she has fired at a person and demonstrates an intent to use (Ash
2002; Pittel 1998).
Formulating a Risk Assessment

Clinicians are often asked to formulate a risk assessment, as in the sec-
ond example, in which the risk to the school was the referral question.
The clinician should recognize that there is less research on the accuracy
of predictions of dangerousness of adolescents than there is for adults.
No combination of risk factors has been shown to predict with accuracy
in an individual case. Therefore, the clinician should acknowledge in
his or her report the limitations in prediction and limit the opinion to a
risk estimate, noting which risk factors are present. It is often helpful to
couch one’s opinion in terms of a comparison to some group, such as
youth of the same age and gender, youth in the same detention center,
and so on.
Management
We have come a long way from the 1970s, when the predominant think-
ing was that “nothing works” in dealing with violent youth, although
violent behavior remains a challenge to treat. Because violence is the
product of multiple factors, the most effective treatments utilize several
modalities aimed at different sources of dysfunction. These modalities
vary widely depending on the nature of the clinical situation: a 4-year-
old who was expelled from preschool for hitting other children will re-
ceive different services than the adolescent in the first case example
who has a long history of antisocial behavior.
Acute Management of High-Risk Youth

The first priority is protecting others from harm. In some cases this will
involve hospitalization. In others, removal from the social situation in
which the threat level is high, such as keeping a youth away from
school by enrolling him or her in a day treatment program, will suffice.
It is important to reduce access to weapons. Brent et al. (2000) found
that only one-quarter of parents were compliant with recommendations
to remove guns from the home when their child was suicidal. The clini-
cian can promote a weighing of the risks and benefits involved in
carrying a handgun; highlight the penalties if a minor is caught with a
handgun; and follow up to ascertain whether the advice was acted
upon. Most youths justify carrying guns for protection and safety, and
alternative methods of remaining safe can be discussed. Youths who
carry guns and demonstrate intent to use may need civil commitment,
or if control is not possible, the clinician may have a Tarasoff duty to pro-
tect others, depending on his or her jurisdiction.
On an inpatient unit, acute highly aggressive behavior may need to
be controlled. The American Academy of Child and Adolescent Psychi-
atry (2002) has developed practice parameters for these difficult sit-
uations that emphasize first utilizing measures to promote a violent
youth’s self-control and other, less restrictive means whenever possible.
When physical restraint is used on children, special attention must be
paid to maintaining an unobstructed airway and ensuring that the pa-
tient’s lungs are not restricted in the prone position by excess pressure
on the patient’s back. Staff training is a crucial factor in ensuring that se-
clusion and restraint will be applied in a reasonable manner. On mental
health units, aggressive outbursts are usually seen as a manifestation of
psychiatric problems. In juvenile detention facilities, however, such
outbursts are more typically seen as volitional behavior requiring cor-
rectional action under the institution’s punishment and use of force pol-
icies. In some cases, youths are receiving medication for their aggres-
sive outbursts and may receive as-needed medications ordered for
outbursts. It is important for psychiatrists working in such institutions
to ensure that such discretionary use is carefully monitored.
Outpatient Psychosocial Treatment

A wide variety of treatment modalities have been tried, and a signifi-
cant number are supported by some outcome studies. Most have a
strong family and/or parent training component, based on the view

that conduct problems and maladaptive aggression are developed and
sustained by maladaptive interactions. Programs that are well sup-
ported by outcome research are listed on the Web site Blueprints for Vi-
olence Prevention (Center for the Study and Prevention of Violence
2007) and discussed in several reviews (American Academy of Child
and Adolescent Psychiatry 1997; Burke et al. 2002; Cadoret et al. 1997;
Connor et al. 2006).
Two programs that have demonstrated efficacy with delinquent ad-
olescents in randomized, controlled trials are functional family therapy, a
short-term (typically 8–15 sessions) prevention and intervention pro-
gram that utilizes two-person teams to meet with the youth, families,
and schools (Alexander and University of Colorado Boulder Center for
the Study and Prevention of Violence 1998), and multisystemic therapy,
in which always-on-call therapists with low case loads provide commu-
nity-based multimodal treatment that addresses multiple risk factors
and work to empower parents and delinquent adolescents with more
adaptive coping skills (Henggeler 1998).
Medication
There is growing consensus that medication should first be used to treat
any underlying disorder, such as ADHD, depression, or bipolar disor-
der (Connor et al. 2006; Pappadopulos et al. 2003; Schur et al. 2003). One
area in which practice varies widely is the extent to which irritability in
adolescents is perceived as justifying a diagnosis of possible bipolar dis-
order and thus the utilization of a mood stabilizer. After treating any
underlying disorder, the second step is to use psychosocial approaches
to manage aggressive behavior, such as cognitive-behavioral treat-
ments, parent management training, and increasing environmental
structure. Only after those approaches have failed should medication
be considered for the target symptom of aggressive behavior.
In 2006, the U.S. Food and Drug Administration approved an indi-
cation for risperidone for the symptomatic treatment of irritability in
autistic children and adolescents. No medications have demonstrated
consistent efficacy in reducing aggression in other conditions. The most
widely utilized medications are mood stabilizers and atypical antipsy-
chotics, which appear more effective for impulsive/reactive aggression
than for predatory aggression (Connor et al. 2006). Among the mood
stabilizers, lithium and divalproex sodium have received the most re-
search support. Among the atypical antipsychotics, risperidone is the
best studied, and other antipsychotics have not yet been studied in ran-
domized, placebo-controlled studies, although they are often utilized,

especially in juvenile detention settings.
Environmental Interventions
Because association with delinquent peers and gangs is so central in ad-
olescent violence, interventions that reduce peer effects or utilize them
proactively have proved useful. For example, from 1991 to 1995, Boston,
Massachusetts, averaged 44 street homicides of youth per year. After a
community intervention beginning in 1996, that number was reduced
by 63% (Kennedy et al. 2001), the so-called “Boston Miracle.” Although
the intervention was multipronged, the basic idea was that although in
most cases the police did not know the shooter, they did know to which
gang the shooter belonged, and law enforcement came down hard on
all that gang’s members. First, there was a community outreach effort
educating gangs to the fact that following a shooting, all gang members
of the presumptive shooter’s gang would be prosecuted for any offense
to the fullest extent possible. Police resources from the city were then
concentrated on the area in which the shooter’s gang operated. Law en-
forcement and the judiciary bought into the program, and maximum
penalties were then given to that gang’s members for any offense, from
public drinking to assault. Those who violated probation in any man-
ner, including such probation requirements as going to school, had their
probation revoked. Because all of the gang suffered for a shooting, peer
pressure rapidly began discouraging shootings.
Other interventions that strive for deterrence have been less success-
ful. For example, after the crime wave of the early 1990s, concern for
public safety led to more punitive approaches toward youth. Following
the “adult crime, adult time” mantra, almost all states expanded their
criteria for waiving juveniles to adult court (Sickmund 2004). The weight
of the evidence now suggests that punishing juveniles as adults in-
creases recidivism (Fagan 1996; McGowan et al. 2007). The American
Psychiatric Association (2005) has called for reform of policies that pun-
ish large numbers of adolescents as adults.
Consultation
Aggression and violence in children and adolescents are among the
most difficult conditions to assess and treat in child and adolescent psy-
chiatry. Working with such youth also raises strong countertransference
issues, and the imprecision of risk assessment in the context of others’
lives being potentially at stake can generate considerable anxiety in the
clinician. Many child psychiatrists have little experience with this pop-
ulation, and given the national shortage of child psychiatrists, much

care is provided to adolescents by general psychiatrists and other men-
tal health professionals. In difficult situations in which one is uncertain
of what to do, it is clinically useful and prudent risk management to re-
member Jonas Rappeport’s advice, “When in doubt, shout!” Obtain
consultation from another clinician and document it.
Key Points
■ The onset of serious violence is typically an adolescent phenome-
non. Those whose violence begins in preadolescence have a signif-
icantly worse prognosis.
■ Serious violent offending is common in high school students,
but most do not continue their violent careers into adulthood.
■ Many risk factors for violence have been identified, but no
constellation of risk factors allows for accurate predictions of
future dangerousness.
■ Effective treatments for violent youth are multimodal and inter-
vene at multiple levels. Most effective treatments include parent
interventions. For adolescents, also intervening to change the
patient’s relationship to a delinquent peer group is important.
■ The best-established use of psychopharmacology is to treat
comorbid psychopathology such as ADHD or a mood disorder. No
medications specifically target aggression, but mood stabilizers
and atypical antipsychotics are sometimes utilized when available
psychosocial treatments have not proved effective.
References
Alexander J, University of Colorado Boulder Center for the Study and Preven-
tion of Violence: Functional Family Therapy. Blueprints for Violence Pre-
vention. Boulder, CO, Center for the Study and Prevention of Violence,
Institute of Behavioral Science, University of Colorado at Boulder, 1998
American Academy of Child and Adolescent Psychiatry: Practice parameters
for the assessment and treatment of children and adolescents with conduct
disorder. J Am Acad Child Adolesc Psychiatry 36:122S–139S, 1997
American Academy of Child and Adolescent Psychiatry: Practice parameter for
the prevention and management of aggressive behavior in child and ado-
lescent psychiatric institutions, with special reference to seclusion and re-
straint. J Am Acad Child Adolesc Psychiatry 41:4S–25S, 2002
American Academy of Child and Adolescent Psychiatry: Practice parameter for
the assessment and treatment of children and adolescents with oppositional
defiant disorder. J Am Acad Child Adolesc Psychiatry 46:126–141, 2007

American Psychiatric Association: Adjudication of youth in the criminal justice
system: position statement. 2005. Available at http://www.psych.org/edu/
other_res/lib_archives/archives/200507.pdf. Accessed February 26, 2006
Ash P: Children’s access to weapons, in Principles and Practice of Child and
Adolescent Forensic Psychiatry. Edited by Schetky DH, Benedek EP. Wash-
ington, DC, American Psychiatric Publishing, 2002, pp 225–230
Ash P, Kellermann AL, Fuqua-Whitley D, et al: Gun acquisition and use by ju-
venile offenders. JAMA 275:1754–1758, 1996
Blair RJR: The roles of orbital frontal cortex in the modulation of antisocial be-
havior. Brain Cogn 55:198–208, 2004
Blumstein A: Youth, guns, and violent crime. Future Child 12:38–53, 2002
Borum R, Fein R, Vossekuil B, et al: Threat assessment: defining an approach for
evaluating risk of targeted violence. Behav Sci Law 17:323–337, 1999
Borum R, Bartel PA, Forth AE (eds): Structured Assessment of Violence Risk in
Youth. New York, Guilford, 2005
Brent DA, Baugher M, Birmaher B, et al: Compliance with recommendations to
remove firearms in families participating in a clinical trial for adolescent
depression. J Am Acad Child Adolesc Psychiatry 39:1220–1226, 2000
Burke JD, Loeber R, Birmaher B: Oppositional defiant disorder and conduct dis-
order: a review of the past 10 years, part II. J Am Acad Child Adolesc Psy-
chiatry 41:1275–1293, 2002
Cadoret RJ, Leve LD, Devor E: Genetics of aggressive and violent behavior. Psy-
chiatr Clin North Am 20:301–322, 1997
Caspi A, McClay J, Moffitt TE, et al: Role of genotype in the cycle of violence in
maltreated children. Science 297:851–854, 2002
Center for the Study and Prevention of Violence: Blueprints for Violence Pre-
vention. 2007. Available at http://www.colorado.edu/cspv/blueprints/in-
dex.html. Accessed March 15, 2007
Centers for Disease Control and Prevention: Physical dating violence among
high school students—United States, 2003. MMWR Morb Mortal Wkly Rep
55:532–535, 2006
Centers for Disease Control and Prevention: CDC Wonder: Compressed Mortal-
ity File [author analysis]. 2007. Available at http://wonder.cdc.gov/
mortSQL.html. Accessed March 30, 2007
Connor DF: Aggression and Antisocial Behavior in Children and Adolescents:
Research and Treatment. New York, Guilford, 2002
Connor DF, Carlson GA, Chang KD, et al: Juvenile maladaptive aggression: a re-
view of prevention, treatment, and service configuration and a proposed
research agenda. J Clin Psychiatry 67:808–820, 2006
Elliott DS: Serious violent offenders: onset, developmental course, and termina-
tion: the American Society of Criminology 1993 Presidential Address.
Criminology 32:1–21, 1994
Fagan J: The comparative advantage of juvenile versus criminal court sanctions
on recidivism among adolescent felony offenders. Law Policy 18:77–114, 1996
Fein RA, Vossekuil B: Protective Intelligence and Threat Investigations: A Guide

for State and Local Law Enforcement Officials (Publ. NCJ 170612). Wash-
ington, DC, U.S. Department of Justice, 1998
Foley DL, Eaves LJ, Wormley B, et al: Childhood adversity, monoamine oxidase
a genotype, and risk for conduct disorder. Arch Gen Psychiatry 61:738–744,
2004
Forth A, Kosson D, Hare R: Hare Psychopathy Checklist: Youth Version. Tor-
onto, ON, Canada, Multihealth Systems, 2003
Hann DM (ed): Taking Stock of Risk Factors for Child/Youth Externalizing Be-
havior Problems. Bethesda, MD, National Institute of Mental Health, 2002.
Available at http://eric.ed.gov/ERICDocs/data/ericdocs2sql/content_
storage_01/0000019b/80/1a/ce/6a.pdf. Accessed January 22, 2008
Hawkins JD, Herrenkohl TI, Farrington DP, et al: Predictors of Youth Violence
(NCJ 179065). Washington, DC, Office of Juvenile Justice and Delinquency
Prevention, 2000
Henggeler SW: Multisystemic Treatment of Antisocial Behavior in Children and
Adolescents: Treatment Manuals for Practitioners. New York, Guilford, 1998
Jensen PS, Youngstrom EA, Steiner H, et al: Consensus report on impulsive ag-
gression as a symptom across diagnostic categories in child psychiatry: im-
plications for medication studies. J Am Acad Child Adolesc Psychiatry
46:309–322, 2007
Johnson JG, Cohen P, Smailes E, et al: Adolescent personality disorders associ-
ated with violence and criminal behavior during adolescence and early
adulthood. Am J Psychiatry 157:1406–1412, 2000
Kennedy DM, Braga AA, Piehl AM, et al: Reducing Gun Violence: The Boston
Gun Project’s Operation Ceasefire (NCJ 188741). Washington, DC, National
Institute of Justice, 2001
Levitt SD: Understanding why crime fell in the 1990s: four factors that explain
the decline and six that do not. J Econ Perspect 18:163–190, 2004
Loeber R, Burke JD, Lahey BB, et al: Oppositional defiant and conduct disorder:
a review of the past 10 years, part I. J Am Acad Child Adolesc Psychiatry
39:1468–1484, 2000
Marsteller FA, Brogan D, Smith I, et al: The prevalence of psychiatric disorders
among juveniles admitted to DCYS Regional Youth Detention Centers: tech-
nical report. Atlanta, GA, Department of Children and Youth Services, 1997
McGowan A, Hahn R, Liberman A, et al: Effects on violence of laws and policies
facilitating the transfer of juveniles from the juvenile justice system to the
adult justice system: a systematic review. Am J Prev Med 32:S7–S28, 2007
Moffitt TE: Adolescence-limited and life-course-persistent antisocial behavior:
a developmental taxonomy. Psychol Rev 100:674–701, 1993
Nansel TR, Overpeck M, Pilla RS, et al: Bullying behaviors among US youth:
prevalence and association with psychosocial adjustment. JAMA 285:2094–
2100, 2001
National Center for Health Statistics: Health, United States, 2004. DHHS Publi-
cation No. 2005–0152. Hyattsville, MD, National Center for Health Statis-
tics, 2004
National Institutes of Health: Preventing violence and related health-risking so-
cial behaviors in adolescents: an NIH State-of-the-Science Conference. Pre-
sented at National Institutes of Health, Bethesda, MD, October 13–15, 2004
Odgers CL, Moretti MM, Reppucci ND: Examining the science and practice of
violence risk assessment with female adolescents. Law Hum Behav 29:7–
27, 2005
Offord DR, Lipman EL, Duku EK: Epidemiology of problem behavior up to age
12 years, in Child Delinquents: Development, Intervention, and Service
Needs. Edited by Loeber R, Farrington DP. Thousand Oaks, CA, Sage, 2001,
pp 95–116
O’Toole ME: The School Shooter: A Threat Assessment Perspective. Washing-
ton, DC, U.S. Department of Justice, 1999
Pappadopulos E, Macintyre JC, Crismon ML, et al: Treatment recommendations
for the use of antipsychotics for aggressive youth (TRAAY), part II. J Am
Acad Child Adolesc Psychiatry 42:145–161, 2003
Pittel EM: How to take a weapons history: interviewing children at risk for vio-
lence at school. J Am Acad Child Adolesc Psychiatry 37:1100–1102, 1998
Schur SB, Sikich L, Findling RL, et al: Treatment recommendations for the use
of antipsychotics for aggressive youth (TRAAY), part I: a review. J Am
Acad Child Adolesc Psychiatry 42:132–144, 2003
Sickmund M: Juveniles in Corrections (NCJ 202885). Washington, DC, Office of
Juvenile Justice and Delinquency Prevention, 2004
Snyder HN, Sickmund M: Juvenile Offenders and Victims: 2006 National Re-
port. Washington, DC, Office of Juvenile Justice and Delinquency Preven-
tion, 2006
Steiner H, Cauffman E, Duxbury E: Personality traits in juvenile delinquents: re-
lation to criminal behavior and recidivism. J Am Acad Child Adolesc Psy-
chiatry 38:256–262, 1999
Teplin LA, Abram KM, McClelland GM, et al: Psychiatric disorders in youth in
juvenile detention. Arch Gen Psychiatry 59:1133–1143, 2002
Tremblay RE, LeMarquand D: Individual risk and protective factors, in Child
Delinquents: Development, Intervention, and Service Needs. Edited by
Loeber R, Farrington DP. Thousand Oaks, CA, Sage, 2001, pp 137–164
Tremblay RE, Nagin DS, Seguin JR, et al: Physical aggression during early child-
hood: trajectories and predictors. Pediatrics 114:43–50, 2004
U.S. Department of Health and Human Services: Youth Violence: A Report of
the Surgeon General. Rockville, MD, U.S. Department of Health and Hu-
man Services, 2001
U.S. Department of Health and Human Services, Substance Abuse and Mental
Health Services Administration, Office of Applied Studies: National Survey
on Drug Use and Health, 2005 [computer file]. ICPSR04596-v1. Research Tri-
angle Park, NC, Research Triangle Institute [producer], 2006. Ann Arbor, MI,
Inter-University Consortium for Political and Social Research [distributor],
2006-11-16. Available at http://www.icpsr.umich.edu/cocoon/SAMHDA/
STUDY/04596.xml. Accessed March 26, 2006
Verlinden S, Hersen M, Thomas J: Risk factors in school shootings. Clin Psychol
Rev 20:3–56, 2000
Vincent GM: Psychopathy and violence risk assessment in youth. Child Adolesc
Psychiatr Clin North Am 15:407–428, 2006
Vitiello B, Stoff DM: Subtypes of aggression and their relevance to child psychi-
atry. J Am Acad Child Adolesc Psychiatry 36:307–315, 1997
Vossekuil B, Fein R, Reddy M, et al: The Final Report and Findings of the Safe
School Initiative: Implications for the Prevention of School Attacks in the
United States. Washington, DC, U.S. Department of Education, Office of El-
ementary and Secondary Education, Safe and Drug-Free Schools Program
and U.S. Secret Service, National Threat Assessment Center, 2002
Zimring FE: American Juvenile Justice. New York, Oxford University Press,
2005
Zoccolillo M, Pickles A, Quinton D, et al: The outcome of childhood conduct
disorder: implications for defining adult personality disorder and conduct
disorder. Psychol Med 22:971–986, 1992
C H A P T E R 1 9
The Elderly
Stephen Read, M.D.
Violence in geriatric patients involves many of the same factors and

considerations as in younger persons. Although less frequently than
younger individuals, elderly people can commit crimes, react to para-
noid fears, abuse drugs, and act in revenge. Any cause of violence in a
younger individual can also be a cause for violence in the geriatric pop-
ulation. These situations in older people generally need to be treated in
similar ways, with minor modifications to account for special geriatric
issues. The difference is the lower frequency of such violence as individ-
uals mature and become less impulsive. During clinical interventions,
physical and physiological changes concomitant with advancing age,
such as increasing frailness and reduced metabolism and clearance,
need to be taken into consideration.
However, there are specific problems that are more common in the
elderly, and this chapter focuses in particular on violence as related to
those factors. Elderly individuals are more likely to have memory and
other cognitive problems, including dementia. Memory problems can
lead to agitation when elderly individuals forget where they put things
and think their possessions are stolen or forget who specific individuals
in their home might be. Disorientation can lead to suspiciousness and to
lashing out at “strangers” in what the patient perceives as self-defense.
381
Some dementias respond to clinical intervention with an emphasis

on careful diagnostic assessment. Even if a diagnosis of an untreatable
condition, such as progressive dementia, is found, family and caregiv-
ers can develop more realistic expectations and plans. Also, current
treatments can decrease the rate of decline and diminish agitation.
As patients lose the ability to care for themselves, the development
of feelings of frustration, humiliation, and helplessness can be frighten-
ing. In response to these feelings, the older patient can lash out at care-
givers, which in turn can give rise in the caregivers to impatience, burn-
out, and other negative responses. Intervention at this juncture can
make a big difference to ease the caregivers’ burden.
Long-term care facilities that house the elderly may have special
challenges. Patients can correctly or incorrectly feel abandoned by their
loved ones when placed in such facilities. Independence is very impor-
tant to such people. Well-meaning social workers sometimes can be too
ready to remove patients from their homes because of relatively minor
self-care deficiencies. For some such patients, independence may be
worth some relatively small physical safety risk. The safest solution
may not always be the best one. The values of the patient and family
must be considered; otherwise, the resulting frustration can lead to vio-
lent actions.
Sexual aggression in the elderly can present problems in nursing
homes and elsewhere. Individuals with Alzheimer’s disease or other
dementias, stroke, or other mental illnesses can develop mania or con-
fusion that can lead to inappropriate sexual touching, paraphilias, or
other unwelcome behaviors that can be perceived as aggressive. Delu-
sional jealousy or misidentification brought on by cognitive problems
can be associated with violence and inappropriate aggression.
The elderly often take multiple medications with additive side ef-
fects, drug–drug interactions, and/or resultant confusion. Reevaluat-
ing the need for each medication may be crucial because some may no
longer be needed or may lead to confusion. Discontinuing some of these
medications can be important, as can lowering dosages that may be
excessive. Because of pharmacokinetic differences, lower dosages of
many medications may be necessary. Treatment compliance remains a
significant problem across all age groups but may pose a special prob-
lem among the elderly. Memory difficulties can lead to forgetting doses
or taking extra ones, with resultant confusion. Organizing medications,
such as with daily medication organizers, can be important and may fix
the problem.
There has been an upsurge in people over the age of 60 in the prison
population in recent years, but the data suggest that this is not due to
The Elderly ❘ 383
an increase in offenses in that group but rather a reflection of the elderly

receiving harsher sentences (Yorston and Taylor 2006). The most likely
cause is mandatory sentencing guidelines, but for some crimes, such as
sex offenses, elderly individuals may be punished more harshly than
younger individuals. In a study of elderly forensic evaluees, most had
alcohol dependence problems and nearly half had dementia (Lewis et
al. 2006).
Assessment
Violent acts attract attention. One common setting in which this be-
comes relevant is in a general hospital. A patient can be profoundly de-
pressed or even confused without attracting attention until the person
threatens suicide—or throws a bedpan in frustration at a nurse’s aide.
The psychiatric attention that was being left until discharge suddenly
becomes an emergency. Implicit in this scenario are two principles: that
actual or threatened violence remains one of the more reliable routes to
psychiatric evaluation, and that the act itself commonly occurs—and is
interpreted as “crazy”—in its perceived inappropriateness in context.
A consequence of this scenario is that the referral for evaluation is
likely to emphasize the act and its consequences—that is, how upset the
victim (or potential target) is or the danger to others—more than the
context. In the communicated need to “do something” urgently, the
evaluator may have difficulty gaining information about the contextual
issues, information that would substantially guide the “prescription”—
the course of recommended response and treatment. In many circum-
stances the psychiatric consultant in a general hospital may not be able
to perform a complete assessment (Devanand 2005; Silver and Herr-
mann 2004), especially in urgent circumstances. Many similar problems
arise in the outpatient setting. Assessment of the problem is essential to
provide appropriate management and treatment. Because any of the el-
ements of a complete evaluation may provide the critical key to the
problem, we first review the relevant portions of a complete assessment.
Chief Complaint
The patient may or may not have a “complaint”—the incident that led
to the referral may have been forgotten—or the patient’s complaint may
be the result of the response of others: “Why is everyone mad at [afraid
of] me?” Eliciting the patient’s recall provides a valuable key to mem-
ory, insight, and lability of mood and may actually reveal the precipi-
tant for the patient’s action. In addition, in this inquiry, the consultant
should note appearance, evidence for neurological or other impair-

ments, and level of arousal and agitation.
It is necessary to understand both the violent act and the context
from the point of view of others as well as the patient. The chief com-
plaint per se will commonly originate from a caregiver, or perhaps a
family member, rather than the patient. The sympathies of the care-
giver, therefore, are not only important in and of themselves, but their
continued commitment to the patient is also important to maintain. For
example, care of dementia patients at home becomes increasingly more
challenging as the dementia progresses, and in such circumstances
home care generally requires increasing recourse to other services
(Read 1990). It has long been recognized that the decision to place a pa-
tient elsewhere is most closely related to behavioral issues (Rabins et al.
1982). Caregivers, therefore, are likely to be the critical factors in suc-
cessfully maintaining the patient’s present situation or hastening the
transfer to a more restrictive environment. They may also be the agents
on whom one must rely for interventions to be carried out. For both pa-
tient and caregivers, it is recommended that the examiner maintain
nonjudgmental sympathy to avoid prejudging, choosing sides, or acti-
vating the ever-present risk of feelings of guilt that will inhibit disclo-
sure.
Case Example 1
Mr. M was a 74-year-old man who had retired from the U.S. Postal Ser-
vice after 43 years of delivering the mail in the small town where he
lived. He had married, relatively late in life, a widow he had met on his
rounds whose mother had urged her daughter to pursue him. Mr. M
therefore became stepfather to the widow’s two children as well as a
“solid citizen” who knew everyone by name and was most proud of be-
ing the greeter every Sunday prior to church services.
The police were startled one morning by his phone call: “I think I
have just killed my wife.” They arrived within a few minutes to find Mr.
M, hammer in hand, standing by the kitchen table where his wife was
slumped, clearly dead. Mr. M said, “I don’t know why I did it.” After his
arrest and jailing, he was given a psychiatric evaluation. His psychiatric
history, which he corroborated, included being hospitalized three times
for severe depression. Each time, his depression had responded to anti-
depressant medicine, principally sertraline, with modest dosages of
(different) antipsychotic medicines. His first hospitalization occurred
after age 50, and the most recent had been 18 months prior to his lethal
assault. This last hospitalization was precipitated by Mr. M’s having ap-
peared at the door of the police station with an axe, shouting threats and
challenging the police to shoot him. Observers had agreed that Mr. M
had returned to his quiet, well-behaved self after discharge. However,
The Elderly ❘ 385
no psychiatric follow-up was available in his small rural community.

His primary care doctor had elected to taper his medicines, and they
were discontinued 2 months before the attack on his wife.
Mr. M was a modest-sized man with psychomotor retardation and a
soft voice. He said at first that the events of the day of the assault were
“foggy”; in fact, cognitive testing revealed modest short-term memory
deficits, poor visuospatial (copying) skills, and difficulty with sequen-
tial tasks, but preserved language. He reported his mood as, “It doesn’t
matter” and “depressed”; to observation, he spoke mechanically, with-
out emotional modulation, and his facial and body expressions were
also very constrained.
Mr. M generated the following account of the fatal day: He had, over
the 2 months since discontinuing his medicine, become increasingly
withdrawn—“like I was before.” His wife had begun discussions with
her children about Mr. and Mrs. M moving so that the children would
be able to assist with their needs and eventually, perhaps, their care. It
was in this context, in fact, that Mr. M revealed that he had been or-
phaned early in life and that he and his sister had lived under constant
fear and threat of the orphanage. Mrs. M had in fact been discussing the
potential move by telephone with her daughter before Mr. M’s attack.
The entire narration was marked by an almost complete absence of ob-
servable or reported emotional reaction.
The psychiatric examiner’s evaluation emphasized the interaction
of three factors: 1) severe melancholic depression, at least approaching
psychotic proportions, with relapse likely attributable to discontinued
treatment; 2) cognitive disorder not otherwise specified, classifiable ei-
ther as mild cognitive impairment or as mild dementia, most likely due
to subcortical white matter ischemic vascular disease, consistent with
findings reported on magnetic resonance imaging scan; and 3) reactive
social and situational factors that, in context, had reactivated deep-
seated fears of abandonment based on childhood experiences. Note that
all three factors made critical contributions to the action: depression pre-
disposed Mr. M to the most pejorative interpretation of his wife’s rea-
sonable exploration of how their future could be more secure, and it also
limited his own capacity to envision this turning out well. His prior life
experience may have played a critical role (the fear of abandonment is
certainly fundamental) but was in fact brought to Mr. M’s mind in the
interview by his reflection on his homicide. Cognitive impairment (fron-
tal systems brain impairment) was judged to have eroded impulse con-
trol and, in this situation, also to have compromised Mr. M’s compe-
tence to assist with his defense in a trial.
In this example, of course, the examiner was aided by his evalua-

tion’s being at some distance in time and space from the events, but the
elicitation of the content of Mrs. M’s telephone call and the connection
to Mr. M’s deep-seated fear of abandonment based on having been
orphaned—counterpoised by Mr. M’s own inability to acknowledge,
express, or experience the associated emotional reactions—opened an
understanding of the events (sadly, of course, much too late). Note that
this information, although in some sense applicable to the “chief com-
plaint,” appeared only late in the interview, after rapport had been es-
tablished and following a gradual approach to the fraught circum-
stances of the event.
Past History
Medical/Surgical
Confusion is the term most commonly used in referrals when the diag-
nosis of delirium eventuates. In the elderly, in fact, delirium may com-
monly herald the onset of an illness such as myocardial infarction that
in younger people is announced by more specific symptoms. Because of
the high frequency with which mental functions abruptly decline in the
elderly due to delirium related to some recent physiological challenge,
the medical history is a high priority in the evaluation of the violent ge-
riatric patient. In addition, knowledge about medical status will be vital
to the choice of treatment.
The consultant should seek recent laboratory studies. Asking about
recent acute illnesses or changes in the patient’s status may identify, for
example, a bladder infection that in an older person is associated with
“confusion” more than with identifiable physical discomfort. Recent
medication changes and their effects, such as the anticholinergic effects
of amitriptyline commonly given to alleviate neuropathic pain, may
precipitate delirium, especially in individuals with dementia. Identify-
ing a physiological abnormality not only may lead to a specific and
sometimes prompt resolution of mental derangement but also can be vi-
tal to preserving brain function.
Case Example 2
Mrs. J was a high-functioning 72-year-old woman with no personal or
family history of major mental disorder who slowly became increas-
ingly agitated over a several-week period, to the dismay of her family.
Eventually, psychiatric consultation was sought, which led to voluntary
psychiatric hospitalization. However, her hospital stay became involun-
tary after she assaulted a staff member when her husband could not be
reached by phone. She was diagnosed with acute mania and started on
quetiapine and lithium. She became alternately sedated and agitated as
well as somewhat confused; in response to needed care interactions, she
frequently responded by hitting attendants, and more than once struck
other patients in a fit of irritability. At these times, she would typically
be given intramuscular lorazepam and confined to the “quiet room” or
restrained in her bed for several hours. She was noted to sweat pro-
The Elderly ❘ 387
fusely and had low-grade temperature elevation. Medications were

switched to risperidone and valproic acid, but without any significant
change in her clinical condition.
After 5 weeks, Mrs. J was transferred to an inpatient geropsychiatry
unit; the referring doctor implied that she had a rapid-onset dementing
condition. She had lost almost 30 pounds and gave the impression of hy-
peralertness while being sedated and of parkinsonian features of rigidity,
drooling, and tremor. Neuroleptic malignant syndrome was ruled out by
normal creatine phosphokinase and lactate dehydrogenase levels, but
with comprehensive laboratory studies her thyroxine level was found to
be 17, with a high reference level of 12.5. Vigorous treatment of her now-
diagnosed Graves’ disease was initiated on the geropsychiatry service by
her consulting internist, with progressive calming and reduced agitation
over a 3-week period. Valproic acid was discontinued and risperidone
was tapered, and thus her rigidity was reduced. Temperature normal-
ized and weight stabilized as her euthyroid state was reestablished.
This case illustrates the need for careful medical analysis and collab-
oration. Late-onset mania is rare. Hyperthyroidism cannot be expected
to respond to “psychiatric” treatment, even when it presents “psychiat-
ric” symptoms. A more disturbing feature of this case is that in this
same time frame, after the return of euthyroid state, Mrs. J was noted to
have memory impairment of which there had been no sign prior to her
hospitalization, and she went on to have full-blown dementia due to
Alzheimer’s disease.
Alternatively, the inquiry into the mental and physical condition
prior to the violent incident may lead to the identification of an emer-
gent, not-yet-recognized dementia. A brief example is a consultation in
a lovely elderly woman who lived alone. Her family stopped by to pick
her up for church and found all the drawers in her kitchen had been
emptied on the floor and everything scattered. After it was ascertained
that this was not a break-in or robbery, she was evaluated for delirium
or dementia. She presented very articulately—she was in a book club,
for example—but oddly could not recall the titles of any recent books.
This observation led to the demonstration of impaired short-term mem-
ory, and a history emerged of episodes of accumulating impairment
during a 2-year period—and a diagnosis of Alzheimer’s disease.
Case Example 3
Mr. T was a 67-year-old veteran admitted to the acute psychiatric unit
from an emergency department. He had a history of depression that had
been treated with fluoxetine and of very mild early dementia. He had
abused crack cocaine in the past, with episodes of delusions and paranoia
treated with olanzapine. His most recent hospitalization had been for pul-
monary insufficiency. He was a longtime smoker and had emphysema.
He presented to the general hospital emergency department with

severe agitation. He would struggle with staff when they attempted to
take his vital signs or draw blood. The emergency physician decided Mr.
T’s problems were psychiatric and attributable to his being off medica-
tion. On admission, it could not be determined whether the patient had
any family. He was medically cleared by “eyeballing” him and was sent
to an acute psychiatric ward.
Because he was combative, Mr. T was placed in four-point restraints
on the psychiatric ward. He was struggling with nursing staff. The med-
ical consultant recommended giving him an intramuscular injection of
haloperidol, lorazepam, and benztropine to calm him down so he could
be examined. Before this was done, his blood pressure was taken and
found to be 65/40, and his pulse oxygen was 78. He was sent to the in-
tensive care unit, where he was found to be badly dehydrated with pul-
monary insufficiency. Hydration cured his “psychiatric” problems. For-
tunately, he had not been given the haloperidol and lorazepam, because
the resultant blood pressure drop could have killed him.
This case illustrates the danger of physicians misdiagnosing an el-

derly patient with medical problems as having a psychiatric disorder
when there is any history of psychiatric problems or dementia. Such
misdiagnosis could lead to a patient’s death. Fortunately, the psychia-
trist and psychiatric nursing staff were alert to this possibility. Also,
medical clearance by eyeballing a patient can be very risky.
Psychiatric
Psychiatric history information may be most elusive in the elderly. For
example, the patient may have suppressed personal recall related to the
stigma of such experiences, and family members (children) may never
have been told. In addition, the patient may be delirious or demented,
and hospital or nursing home staff may have no real information about
prior history. Misidentification delusions in the elderly can result in vi-
olence. Substance abuse histories may also be vital but are subject to de-
liberate concealment or ignorance about the extent of alcohol intake, for
example, which may contribute to the clinician’s overlooking delirium
tremens or drug-induced psychosis or affective disorder. Another cau-
tion is that a condition that today would be recognized as a major de-
pressive episode may have been diagnosed in the early adult life of an
85-year-old as “neurosis,” “schizophrenia” or other psychosis, or sim-
ply a “nervous breakdown” in older nomenclature. Elicitation of psy-
chiatric history is usefully done in conjunction with the medical history.
However, a history of prior psychiatric disorder can be critical for iden-
tifying the precipitating episode of a reemergence. As discussed above,
in emergency department and medical settings, confusion and agitation
The Elderly ❘ 389
can often be dismissed as psychiatric when the cause may be purely

medical and physiological. A misdiagnosis or a tendency to decide that
any confusion in a patient with a psychiatric history must be psychiatric
can be life-threatening.
Psychosocial
Inquiry into psychosocial factors provides background information
that is valuable for rapport and for identifying sources of stress (and for
observing the patient’s resilience and capacity to understand and ap-
preciate such forces). Such inquiry may also reveal clues to a family ill-
ness or to an event relevant to the patient’s actions. A serious issue
involving a close family member, a recent bereavement, or a major fi-
nancial or social decision may dominate the patient’s thinking and have
a greater meaning than one would expect. In addition to the direct
physical and physiological effects of the patient’s condition, his or her
awareness of it may be a critical stressor. Whether the concern is “heart
attack,” “cancer,” or care consequences (e.g., a lap belt used on a patient
with dementia to prevent forgetful attempts to stand after hip replace-
ment surgery), it may be experienced directly by the patient as an exac-
erbation of vulnerability—a perception of threat that can lead to a vio-
lent reaction to a stimulus that otherwise would not be threatening.
Mental Status Examination

Different aspects of the mental status examination will emphasize se-
lected features likely to be especially relevant to the evaluation of a vi-
olent elderly patient. Ideally, the evaluator will be able to observe at a
distance prior to direct encounter. In addition, observation at a distance
may reveal the sign of a relevant general medical condition—for ex-
ample, asymmetric facial or limb movements in a stroke or other focal
brain injury, proptosis in Graves’ disease, movement disorder, or the
moon face of Cushing’s syndrome. Familiarity with the expression of
cognitive, mood, and behavior disorders associated with various ab-
normalities of brain function will also usefully inform the examination
(Bogousslavsky and Cummings 2000; Strub and Black 1988).
Attentional Problems
Attentional problems are the core symptoms of delirium. The first issue
is whether the patient is paying attention—that is, is able to listen to and
respond appropriately to questions or requests from caregivers. Is the
patient easily distracted? Can the patient reorient to the previous ques-
tion after an interruption? On the other hand, can he or she then turn
attention to another topic or matter on request? How much effort does

this require? Is it onerous (irritating, inducing hopelessness, etc.)?
Conditions that can lead to false positives in this realm are signifi-
cant hearing loss, fluency only in another language, or severe dementia.
The most basic test of attention is to ask the patient to repeat single
words and then phrases up to longer sentences. Copying line drawings
may also reveal problems with sustaining effort and attention, and the
preservation of this or other skills may suggest the presence of a lan-
guage disorder, such as fluent aphasia, that can disrupt communication
in a way that incorrectly suggests inattention.
It can be very helpful to describe the patient’s performance over the
course of the delirium. Among the several available tools, we rely on
the orientation, mental control, and registration and recall sections of
the Mini-Mental State Examination (MMSE) of Folstein et al. (1975). The
MMSE was in fact developed as a tool to identify and give a ballpark
estimation of cognitive impairment in patients seen in consultation in a
general hospital setting. These items, rated over time, document perfor-
mance on attentional measures more than adequately. In addition, the
MMSE has gained widespread acceptance as an overall rating of the
cognitive severity of dementia, and its questions tap a broader range of
functions (including visuospatial skills) than many other short rating
scales. Use of the MMSE therefore serves double duty in terms of its rec-
ognizability as a proxy for the overall cognitive impairments that may
predate the acute illness phase or persist afterward.
Cognitive Impairment With Impaired Memory

Cognitive impairment with poor short-term memory is the core symp-
tom complex of dementia. Impaired memory is required for a diagnosis
of dementia in DSM-IV-TR (American Psychiatric Association 2000),
and if the patient can repeat three words (suggesting at least a modicum
of attentional capacity), his or her inability to recall those same words
after several minutes signals the presence of this hallmark symptom.
From the history, the consultant should have some information, at least
about recent medical events, that can serve as a basis for assessing long-
term memory. Other cognitive areas warrant attention, even in the ini-
tial urgent examination (for more detailed resources, see, e.g., Strub and
Black 1985):
• Language impairments can be associated with high levels of frustra-

tion and distress and, as a practical issue, must be accounted for in
developing a plan for management.
The Elderly ❘ 391
• Visuospatial tasks, such as copying simple drawings, can help evalu-

ate the patient’s ability to engage and focus. Abnormalities in the
face of good effort are supportive evidence for dementia. Grossly
disproportionate abnormalities may be associated with deficits in
brain regions that organize one’s sense of orientation in the most ba-
sic sense.
• Reasoning tasks (e.g., interpreting proverbs and idioms) may be
poorly tolerated but are indicative of the level of language skill, at-
tention, and the willingness to tolerate some degree of annoyance or
perceived irrelevance. Intact responses at an abstract level are coun-
tersuggestive of dementia but may be compatible with significant
encephalopathy and resulting delirium.
• Executive functioning as assessed by, for example, the Executive Inter-
view (Royall et al. 1992) overlaps with and depends on attentional
mechanisms but may be independently disrupted. Deficits may indi-
cate the presence of impaired frontal lobe functions, with associated
loss of empathy, judgment, and impulse control, functions that may
reduce the threshold for violent, apparently impulsive reactions.
Case Example 4
Dr. F, a former surgeon and community leader, had been living at home
with his wife and attending day care. He had support from his out-of-
town children as well. He had been functioning as a surgeon until his
memory problems became apparent. He had been cared for at home for
several years after diagnosis of dementia due to Alzheimer’s disease
(which itself followed several years of declining short-term memory
and risky financial decisions), and this home stay was supplemented by
his attending adult day care. He was evaluated for nursing home place-
ment after an emergency call about his being “agitated.” Some restless-
ness and “agitation” had been mitigated with low-dose risperidone.
However, near the end of the day care schedule, in fact within several
minutes of 2:30 P.M. every day, Dr. F would suddenly transition from a
relatively affable, cooperative, good-humored man into a restless, agi-
tated, active and impatient man. He would typically claim he “needed
to go [somewhere],” and he would try to push his way out toward the
parking lot. Day care staff had worked out somewhat elaborate at-
tempts to divert him, but on the day in question, he had been unde-
terred and had struck a staff person who finally tried to block his way
physically. His wife had arrived shortly thereafter, and she also (for the
first time) feared that he would be “violent.”
The 2:30 P.M. transition time endured in the nursing setting. Dr. F
was bright, confident, helpful, and cooperative in the morning but be-
came impatient, demanding, and urgent from before 3:00 P.M. until past
dinnertime. If staff were unable to distract or redirect him, he could be-
come very forceful. Although this was generally limited to pushing or
shoving obstructions out of his way (including staff persons or, occa-
sionally, another patient who was unable to recognize the situation), at
least the threat of more focused aggression existed. Containing his ag-
gressive impulses was difficult—doors were “secure,” but Dr. F’s athlet-
icism enabled him to get through several barriers and even to climb over
the six-foot-high perimeter wall. Once out, he was also able to move
very quickly.
As staff came to know him, their interventions became more effec-
tive. Organizationally, because his behavioral change was timed close to
change of shift, administrative team intervention was required to ensure
a continued high level of observation (instead of the distraction of com-
pleting mandated charting) and to manage the relative confusion atten-
dant on staff comings and goings. These efforts alone remained insuffi-
cient. His behavior worsened at every attempt to reduce antipsychotic
dosage, and he required doses of risperidone 2–3 mg daily, most given
after lunch to maximize mid-afternoon effect and thus mute the impulse
and contain the risk of aggression.
Dr. F’s dementia was moderately severe at nursing home admission
(e.g., MMSE score was 14/30) and continued to progress over the next
5 years. He was started on donepezil with some mild improvement. Al-
though the diurnal pattern was unchanged, management became easier
as his capabilities diminished. After surgery to repair an intertrochan-
teric (hip) fracture, he experienced delirium and, after recovery, was not
only less mobile but also had a substantial incremental worsening of his
dementia. He was also more passive, and at that point the diurnal agi-
tation became much less marked, allowing for the tapering and discon-
tinuation of risperidone.
Violence is common in a significant minority of Alzheimer’s pa-

tients. Management of these patients requires consideration of a combi-
nation of both psychosocial milieu factors and medication.
Thought Processing
The presence of hallucinations, delusions, preoccupations, or other ab-
normalities may be indicative of impaired brain function and may also
directly contribute to the aggressive impulse.
Disorders of Mood and Affective Regulation

Many depressive patients are irritable, which can lead to reactive hos-
tility and, especially in the face of other damage to frontal lobe struc-
tures, to disinhibition and violence. There is evidence that aggression in
the substantial minority of patients with dementia is strongly linked to
the presence of depressive symptoms (Lyketsos et al. 1999). Most de-
mentia patients in the community are not violent. If depression is fac-
tored out, dementia patients may be no more violent than other individ-
The Elderly ❘ 393
uals. The hyperactive, driven, grandiose manic patient may also react
violently to what would otherwise be mild provocation or frustration.
Attentional and cognitive deficits, as well as delusions, can be associ-
ated with both poles of mood disorder and further compromise the
patient’s impulse control. Because apathy can be an early sign of de-
mentia, sometimes it is difficult to distinguish from depression in the
elderly.
Case Example 5
Mrs. L was a 78-year-old woman who had been a social leader of her
community and who had been married nearly 50 years to a man who
continued to adore her. His devotion led him to seek assistance from ev-
ery source he could find when she developed a persistent, savagely anx-
ious depressive syndrome—and to seek an alternative if success were
not promptly forthcoming.
Mrs. L had acted at nearly hysterical levels for a substantial part of
every day for more than 3 years. She would throw herself into walls at
times, although there was no clear attempt to commit suicide. She had
torrents of tears and refused to be comforted and would at times strike
at or claw at anyone who presumed to get close, including her husband
and the caregivers he hired. Sleep, appetite, energy, and mental focus
were all grossly deficient, and she had lost over 40 pounds in the year
preceding evaluation. Although a diagnosis of depression had been
made (several times), treatment had never been consistently pursued
because of her own hopelessness and resistance and her husband’s great
concern every time someone mentioned the possibility of side effects.
Evaluation confirmed a diagnosis of severe recurrent major depres-
sion. Consultation with an internist and laboratory studies revealed no
medical cause; in fact, Mrs. L remained in good health. Mr. L was in-
volved in all phases of the assessment—with an emphasis on his accept-
ing that the primary diagnosis was psychiatric. Once that was estab-
lished, it was possible to provide realistic information about treatment
options and the course of improvement that could be expected. Accord-
ingly, Mrs. L was started on venlafaxine. Treatment response began at a
dosage of 75 mg daily, with additional improvement and real symptom-
atic relief evident at a dosage of 150 mg daily. Each improvement, how-
ever small, was strongly reinforced both with Mrs. L and her husband.
Some level of distress, agitation, and anxiety persisted with optimal
doses of venlafaxine, but further improvement was achieved with the
addition of small doses of risperidone and then antianxiety medication.
Mrs. L eventually returned to her primary care doctor for follow-up.
Twice in 4 years she was reevaluated when symptoms returned—both
times after an attempt to “stop medicine because I was better.” The third
relapse differed. At this point, overt difficulty with memory was seen at
evaluation, and this proved to be the early finding in her developing
dementia due to Alzheimer’s disease. Donepezil was started, but with
little effect. As her dementia progressed, she became more impaired.
In addition to increasing cognitive impairment, at every attempt to re-

duce mood medications, agitation and depression returned, associated
at times with paranoia and agitation that could lead to her hitting care-
givers or anyone else in the vicinity.
Agitation can be part of depression and can lead to violence. Depres-

sion can be an early sign of Alzheimer’s but usually manifests more as
apathy in such cases. Considering the persistence of the depression and
the lack of response to donepezil, there most likely were two indepen-
dent problems in this patient.
Treatment
Consider Acute/Emergent Treatment
At times the consultant must consider initiating treatment of violent be-
havior before the full evaluation process can be completed. A rapid
medical response, such as restraining a patient in a bed, may be neces-
sary to ensure the safety of the patient and others or to limit the morbid-
ity of other responses. Two groups of medications remain the mainstays
of acute calming of aggressive and violent behavior: benzodiazepines
and neuroleptics. Both require careful titration and close observation
but have a high safety profile in short-term use, especially when com-
pared with the potential for injury from agitated or aggressive behavior.
“Close observation” should include frequent visualization, prefera-
bly with the patient in constant line of sight. In addition to monitoring
level of agitation and aggression (i.e., the response of target symptoms),
observations should include vital signs, color, any appearance of phys-
ical distress, urine output, and level of awareness. Regular notes should
be made (and retained) and a supervisor and/or physician should be
notified promptly of any deterioration in any parameter.
Despite the Clinical Antipsychotic Trials in Intervention Effective-
ness–Alzheimer’s Disease (CATIE-AD; Schneider et al. 2006) studies and
recent concerns about neuroleptic medications in the elderly, most au-
thorities continue to prefer the use of these agents in the elderly due to
the greater likelihood of aggravating cognitive impairment and the risks
of unsteady gait with benzodiazepines. Benzodiazepines often can also
be disinhibiting. Neuroleptics may be given orally, intramuscularly, or
intravenously (e.g., in an intensive care setting with established venous
access), but the availability of liquid (elixir) and fast-dissolving oral prep-
arations has greatly reduced the need for parenteral administration.
At the time of this writing there is controversy about the meaning of
the CATIE-AD studies regarding the use of atypical antipsychotics in
The Elderly ❘ 395
the elderly to treat agitation and aggression. It appears that these med-
ications are effective, but this effectiveness can be negated by side ef-
fects (Schneider et al. 2006). The U.S. Food and Drug Administration
(FDA) has not approved these medications for the treatment of demen-
tia-related psychosis because of increased danger of death (Karlawich
2006). There may be a small increased risk of death (Schneider et al.
2005), but there are contradictory findings and interpretations, and
many think these agents have a place and that the danger has been
overblown (Barak et al. 2007; Raivio et al. 2007).
Ensure Safety of Patient and Others

Implicit in the referral will be concerns about safety, not only for the pa-
tient, but also for others. Besides understanding the context and precip-
itating factors for the violent act(s) that led to consultation, questioning
of family and other caregivers should also gauge the commitment,
skills, and resources within the existing caregiving matrix—with the
consideration of whether the patient requires more intense care, either
immediately or in the long run, to ensure safety. Our experience is that
for the large majority, family members and caregivers remain commit-
ted to the care effort—sometimes unwisely so (e.g., failing to recognize
the risk of allowing an 83-year-old woman to continue caring for her
larger, more vigorous, and pathologically paranoid husband of more
than 50 years—who no longer recognizes her consistently). Therefore,
in this basic area, the consultant’s recommendations may be the most
fundamental with regard to emergency referral, acute psychiatric hos-
pitalization, and the degree of ongoing support and structure required,
whether at home or in an institutional setting (Read 1990). Use of re-
straints in the elderly also requires special caution. The frailness of such
patients can be a risk. Confused patients can panic in restraints and
might even develop a myocardial infarction. Restraints are sometimes
used as an easy way to keep patients in bed or to prevent them from
pulling out intravenous lines and other medical equipment. Restraints
can be used also as in younger patients for out-of-control aggression.
Identify and Treat Causes of Medical Decompensation,

Especially Sources of Delirium
Delirium most commonly develops over the course of hours to days. Be-
haviors are varied and may fluctuate over the course of the day (Amer-
ican Psychiatric Association 2000). Recognition of early symptoms, in-
cluding memory impairment, incoherence, disorientation, disrupted
sleep cycle, hallucinations, and irritability and other mood changes (de
Jonghe et al. 2007) allows for mitigating their effects. These importantly
include diminished immediate safety, negative impact on the patient,
and the fact that prolonged delirium is associated with poor outcomes,
including death and permanent brain damage. For example, in case ex-
ample 2, Mrs. J’s prolonged thyrotoxicosis left her with permanent mem-
ory loss, and she subsequently developed progressive dementia.
Because brain function is a sensitive marker of decompensation in
the function of any major organ system—cardiac, pulmonary, hepatic,
or renal—identifying the presence and cause of such dysfunction is crit-
ical. Asthma, incipient pneumonia, congestive heart failure, and elec-
trolyte imbalance are other common illnesses for which the typical
symptoms in the elderly may be “confusion” rather than specific text-
book medical symptoms. An especially common cause is a bladder in-
fection, which reliably causes confusion and abrupt decompensation in
the patient with moderate to severe dementia. Delirium tremens may
appear in the covertly alcoholic person. For these reasons, the consult-
ant is urged not to dismiss the observation “she’s just not herself” from
a credible caregiver who knows the patient well (whether a family
member, home health aide, or certified nursing assistant or nurse), be-
cause those who work closely with the patient over time will be the
most sensitive to these changes.
Treat/Manage Identified Precipitants to the Violence

A major goal of the assessment process is to identify precipitants or pre-
monitory symptoms for the violent act. In retrospect, the patient may
have had pain or hunger or other discomfort; factors such as time of day
or loud noises may figure in; or the actions of a caregiver or another pa-
tient, or some other clear factor, may emerge from a careful evaluation.
In the obvious situation, identification by the consultant may be suffi-
cient, and the health or social/family system will respond. In other in-
stances, the consultant’s expertise may be needed to assist with a plan
or to ratify and approve a clinical approach.
The settings in which the precipitating incident occurs (e.g., skilled
nursing, acute general medical floor, inpatient psychiatry, home, as-
sisted living, day care) will bear on the range of responses available, be-
cause these locations are staffed by persons with differing levels of
training and experience and are governed by differing sets of regula-
tions. For example, in a skilled nursing facility, one can assume 24/7
availability of licensed nursing personnel, supervision of prescriptions,
and a high standard of recordkeeping compared with, for example, the
The Elderly ❘ 397
variability of home and assisted-living settings. In a skilled nursing fa-

cility, assessments documented by means of the Minimum Data Set—
a federally mandated data system that includes functional and behav-
ioral observations and is reevaluated on at least a quarterly basis—will
often assist the evaluation of context. However, in many facilities, also
as a consequence of federal regulations, the consultant may find that the
regulations are interpreted so as to restrict treatment and management
options. The consultant who works in long-term care settings will rec-
ognize the high prevalence of mental and behavior disorders (Rovner et
al. 1990) and is advised to learn the capabilities and limitations imposed
by the different levels of licensure (which also vary among the different
states). Useful resources are available for the specifics of behavioral
management in long-term care (Katz et al. 2005; Reichman and Katz
1996; Szwabo and Grossberg 1993; see also materials available from the
American Medical Directors Association [n.d.]). Although we empha-
size the necessity of identifying the particularities of each case, some ex-
amples are offered.
• Any unrecognized source of physical discomfort can cause irritability

and lead to “violent” reactions. This is perhaps especially important
in assessing a demented patient who may no longer be able con-
sciously to recognize sources of discomfort, although the “stoical”
patient may not connect his or her irritability with a pain he or she
was trying to ignore. The source of pain may be mechanical, such as
a lumpy object left in the seat of the wheelchair, or it may be an un-
recognized injury, such as an occult vertebral compression fracture
or a broken hip from even a minor fall. Other “physical” sources of
discomfort may be poor temperature regulation (elderly do not de-
fend their body temperatures as effectively), hunger, or a blocked in-
dwelling catheter. Careful systematic palpation and physical exam-
ination may be necessary to exclude a source of pain in a very de-
mented person or someone with severe aphasia.
• Irritating environmental stimuli may provoke violence. Noise, bright
lights, smells, a person approaching too close, or simply a harsh tone
of voice can be jarring. At times the substrate will be a neurological
deficit (for example, a patient [and caregivers] may not have recog-
nized a homonymous hemianopsia, in which someone approaching
is not seen until he or she suddenly appears in front of the patient—
seeming, to the uninsightful patient, to have “snuck up” on him).
Patients with advancing Alzheimer’s disease can be observed to
look straight ahead and appear to lose reactivity to peripheral vi-
sion. Patients with thalamic or other subcortical lesions may have
decreased “gating” protection from sounds or other stimuli. Recog-

nition of these interactions guides management (one warns the
hemianopic patient of an approach by talking), and the staff are pre-
pared for a defensive “striking out” if they must tread on the pa-
tient’s sensibilities. In addition, it is common for agitation to develop
in the late afternoon in patients with dementia due to Alzheimer’s
disease (Cohen-Mansfield 2007), at which times they are likely to be
more reactive to stimuli they would ignore earlier in the day, as in
the case of Dr. F (case example 4).
• Specific care needs may also be a precipitant. Caring for a dependent
elder, whether the cause is mental (dementia) or physical (paralysis
after a stroke), requires the most intimate contact, and at close quar-
ters. Family and caregivers must also be aware of other demands
and are subject to fatigue themselves, and thus may become impa-
tient or seem gruff at times. Patients may also be angry about their
disability, be embarrassed, or have other causes of diminished im-
pulse control, or there may be no specific identifiable provocation
for a patient’s striking out. Bathing is an especially sensitive activity.
Many patients with dementia become averse to water (sometimes
evoking the tired simile of “second childhood”) and may become
agitated while naked and slippery. The bather may lower his or her
guard while attending to some sensitive detail (e.g., cleaning the
perineal area), and a frightened or hurt patient at close quarters may
react with a slap or punch. Adjusting bathing expectations (patients
do not always require a daily bath), providing a second assistant, or
finding a more congenial time of day may solve the problem. Some
patients seem to manage for long periods with sponge baths in their
beds only. At other times, judicious use of a short-acting antianxiety
medication such as alprazolam, 0.25–0.5 mg given half an hour be-
fore commencing the activity, may be critical for allowing this kind
of care to proceed. Communication problems around personal care
tasks are responsible for a substantial proportion of violent incidents
(Almvik et al. 2006).
Diagnose and Treat Underlying Psychiatric Disorders

Psychiatric disorders may account for a substantial part of the propen-
sity to violent actions, either alone or by potentiating other factors. For
example, the reaction to being bathed by a spouse or child is altered
when the patient no longer recognizes that person, who understand-
ably has reason to expect familiarity. Consideration of the following
specific comments is important.
The Elderly ❘ 399
• Coordination of team efforts, situational adjustments, and activities all

have roles in mitigating difficulties in many patients (Katz et al.
2005). Group and individual psychotherapy may be of benefit in
some patients. In our opinion, the application of such efforts war-
rants the same critical analysis and reassessment expected for phar-
macological therapy.
• Pharmacological treatments for Alzheimer’s disease include cholines-
terase inhibitors (including donepezil, rivastigmine, and galan-
tamine) and now the NMDA (N-methyl- D -aspartate) receptor
blocker memantine. Cholinesterase inhibitors partially compensate
for the deficit in acetylcholine, the major neurotransmitter deficit
demonstrated in Alzheimer’s disease, with modest improvements
in memory, attention, and clarity of thinking; benefits vary from pa-
tient to patient, and these agents do not alter the progressive course
of the disease process. However, behavioral improvements are sus-
tained in a significant percentage of patients on continuous cho-
linesterase inhibitor treatment, including reduced irritability, anxi-
ety, aberrant motor behavior, delusions, and disinhibition (Aupperle
et al. 2004; Gorman et al. 1993). Memantine has been shown to slow
the course of progression, but short-term responses may include re-
duced agitation/aggression and irritability/lability (Cummings et
al. 2006).
• Alcohol and nicotine habits can be associated with agitation and even
violence in relation to the sense of urgency (craving) and confusion
induced by use (presumably relative hypoxia for smokers). Naltrex-
one for alcohol and bupropion (sometimes with nicotine patches) for
smoking have been consistently effective and reduced this nexus of
distress (with the added benefit of health and safety improvement).
Smoking is generally forbidden or limited in care facilities, and this
can be a basis for violent attacks. As it is in younger patients, sub-
stance abuse treatment can be very important (Aradt et al. 2002).
• Psychosis and mood disorders may be important contributors to the di-
athesis leading to violence, and their treatment may be important to
the control of violence. The interaction of these factors is demon-
strated in the following case.
Case Example 6
Mrs. A was a wealthy, divorced 84-year-old woman originally seen in
the context of a bitter and complex family struggle over property issues.
In fact, this complicated situation was largely explained by Mrs. A’s
early dementia, which had been unrecognized. Resolution of the de-
mentia led to conservatorship, with a commitment that she be cared for
at home. Full evaluation, including single-photon emission computed

tomography studies, supported a diagnosis of mixed dementia—Alz-
heimer’s disease plus ischemic-vascular disease (Read et al. 1995).
Mrs. A was considered “spoiled,” and indeed she was used to hav-
ing her own way. Her major recreational activity was shopping fol-
lowed by expensive lunches, but her caregivers noted problematic dis-
inhibition: Mrs. A would yell out the car window and make vulgar
gestures to others whom she perceived to be “in her way,” and she was
not mindful of the potential for dangerous retaliation. On other occa-
sions she was very rude to people in restaurants. She had no apprecia-
tion for the effects of these behaviors, despite numerous discussions and
illustrations; besides stigmatization of Mrs. A, the caregivers worried
about her provoking physical response from an offended stranger.
Mrs. A’s behaviors were mitigated with the use of donepezil and
quetiapine without any evident side effects, and she was able to con-
tinue her “recreation.” She was followed up at home, and it was noted
that when she was “nervous,” she began chain-smoking. After three cig-
arettes, she became slightly bluish and much more confused. She of
course did not notice, experiencing her distress as the result of “com-
pany.” However, with the use of nicotine patches and an agreement by
conservator and caregivers, cigarettes were removed and her distress
“smoothed out.”
An additional piece to the puzzle was the observation that she had
several-week episodes of increased motor activity, demanding be-
haviors, and agitation that included threats to people perceived by her
as frustrating. At these times, the rapport that had been established
through regular supportive discussions at monthly visits broke down.
This pattern provoked review and the realization that Mrs. A was ex-
pressing the strong family history of bipolar disorder (which had also
made major contributions to the original family strife). Because of her
demonstrably good response to quetiapine, these periods were man-
aged by increasing the dosage of neuroleptic rather than exposing her to
possible complications of other mood-stabilizing agents.
The final challenge was how to handle a crisis—the suicide of a son,
probably himself bipolar, who was beloved by Mrs. A but who had also
precipitated the original crisis by defrauding her out of a large and valu-
able property. By this time the psychiatric consultant was familiar to all
the family members, as well as caregivers and conservator, and this en-
abled a unified plan of management.
Empirically Evaluate Continuing Use of Medications

for the Violent Patient
After treatment of specific factors as described earlier, violent behavior
may continue, mandating consideration of ongoing pharmacotherapy.
Unfortunately, in these days when we are exhorted to practice “evi-
dence-based medicine,” treatment of the violent elderly patient must be
undertaken in the face of a paucity of data. Causes for this conundrum
The Elderly ❘ 401
include the complexity of defining and monitoring treatment (includ-

ing assessing outcomes); the protean heterogeneity of conditions un-
derlying violence in the elderly; and the general neglect of studies of
treatment in elderly persons. In addition, the issues evoke strong and
contrary feelings, and the target patient’s capacity (necessary at least,
for example, for informed consent) may well be challenged by the very
condition that underlies the behavior itself. Ambivalence about phar-
macological treatment has also been heightened in the past 2 years by
the description of safety issues related to risks that have appeared with
the study of large (but heterogeneous) databases of medication use.
This section emphasizes principles of treatment of the violent elderly
patient, building on the assessment process described earlier.
Although not sanctioned by FDA-approved studies, neuroleptic
medications (approved for the treatment of psychosis) have been the
mainstay of treating agitation, aggression, and violent behaviors in the
elderly in clinical practice—with an emphasis on the use of “atypical”
antipsychotics (Sink et al. 2005). These medications displaced the previ-
ously widespread use of antianxiety agents, especially benzodiaze-
pines, because of problems with unsteady gait and risk of falling, over-
sedation, and deteriorated cognition with benzodiazepines. Following
clinical practice, we outline below an approach to use in practice.
• Side effect profile largely determines choice of the particular agent:

olanzapine, for example, if inducing sleep and appetite are seen as
beneficial side effects (in the insomniac anorectic patient); ziprasi-
done for the reverse situation; quetiapine or aripiprazole or risperi-
done if neither of those dimensions is important. An additional con-
sideration is the route of administration and the availability of drug
in elixir, intramuscular, intravenous, or rapidly dissolving forms.
More rarely, depot injectable medication may be indicated.
• Dosage is “titrated” to optimal effect, monitoring response and side
effects. Titration follows the geriatric adage “start low, go slow”
(sometimes difficult to adhere to in the initial urgency created by a
violent elderly patient). In addition, it is also true that some elderly
patients are relatively resistant to medications—for us, the operant
principle is that variability is what increases with age—and require
dosages fully equivalent to, or higher than, those in younger adults.
Although the biological basis for this dose–response variability is
poorly understood (and in fact has been little investigated), the ex-
istence of this variability complicates the possibility of firm guide-
lines for dosage in complex situations such as treating violence—
much less the underlying psychiatric syndromes that may have
been identified. This dose–response variability, combined with fre-

quently medical complexity in the geriatric patient, mandates that
pharmacological intervention be undertaken in a setting with
knowledgeable and experienced personnel who have sufficient
equipment to observe response and potential adverse effects, espe-
cially when assessing the need for further dosage increments.
• Monitoring the response mandates targeting symptoms and keeping
records. Because the violent acts may be relatively intermittent, it is
best also to monitor premonitory symptoms (e.g., irritability) in
terms of timing, frequency, and intensity—and this enhances staff
alertness to the overall situation.
• The prominent side effects of concern are extrapyramidal motor symp-
toms (drug-induced Parkinsonism) and autonomic instabilities, es-
pecially orthostatic hypotension. A major hazard for both of these is
the risk of falling and sustaining injury. Motor symptoms should be
monitored clinically, and blood pressure responses require the avail-
ability of a device and someone who knows how to use it.
Application of these principles is illustrated in the following case.
Case Example 7
An 86-year-old widow was seen for psychosis and agitation associated
with the episodic delusion that the house where she lived was “not my
home.” This delusion appeared reliably every afternoon at 4 P.M., at
which time this sweet, docile woman who enjoyed many activities be-
came angry, anxious, and distressed and would strike out at her care-
giver or anyone else who tried to reassure or correct her misperception.
In fact she had run away from home twice, and consideration was being
given to placing her in a care facility.
Evaluation revealed an articulate and opinionated woman with
moderately severe dementia (MMSE score 14/30) fully consistent with
Alzheimer’s disease, as confirmed by positron emission tomography
scan. The initial intervention was family education, informing them that
the patient was unable to understand what was happening and did not
intend her actions and that this was fully typical for midstage Alz-
heimer’s disease, including the afternoon emergence. As a result, they
arranged for a home health aide (who fortunately established good rap-
port with the patient). Cholinesterase inhibitor therapy was started
(donepezil, titrated to 10 mg daily), together with memantine.
On this program, the patient’s cognition improved modestly
(MMSE rose to 17/30), and, although the afternoon psychosis contin-
ued, she was less adamant and responded better to attempts to demon-
strate that the house really was her home. However, after several
months, cognition again declined, and the distress and agitation and at-
tempts to elope continued. She even threatened one evening to hit her
The Elderly ❘ 403
caregiver, whom at other times she had come to call her “new daugh-
ter.” At this point it was decided to add low-dose atypical antipsychotic
medication, specifically quetiapine. Dosing was started at 25 mg daily
at 2 P.M. and 6 P.M. She was seen 1 month later, and review of daily logs
indicated there were no physical changes and that although the inten-
sity of symptoms was less, distress still emerged in late afternoon. Dos-
age was slowly increased in 25-mg increments. At 100 mg twice daily,
there were no reported incidents and the patient remained free of par-
kinsonian or metabolic side effects. She continued on this dosage with-
out apparent adverse effects.
The use of neuroleptic medication has become more controversial in

the past 2 years due to two major side effect issues: 1) treatment with
antipsychotic agents potentiates (at least) the emergence of the “meta-
bolic syndrome,” which may expose patients to a higher risk of diabe-
tes, lipid abnormalities, and cardiovascular disease; and 2) there is an
increased risk of stroke. In addition, such studies as have been under-
taken have provided little support for the long-term efficacy of neuro-
leptic treatment (in striking contrast to the perceptions of “frontline” cli-
nicians). These concerns have culminated in adverse advisories from
the FDA (U.S. Food and Drug Administration 2005), which specifically
emphasizes that these medications were approved for use in schizo-
phrenia but not for geriatric patients with dementia. These statements
have been followed by “black box” warnings on the package inserts of
these medications, although there is a dispute as to whether all atypical
antipsychotics (or older neuroleptic medications such as haloperidol)
warrant equal concern.
These difficulties have highlighted the widespread “off-label” use of
these drugs in the elderly, despite a paucity of controlled studies. For
these reasons, as well as poor efficacy in at least some patients, there has
been interest in a variety of other medications and a corresponding
large anecdotal literature (and word-of-mouth practice). Positive re-
ports have appeared for trazodone, other serotonergic antidepressants,
anticonvulsants (gabapentin, carbamazepine, and valproate), bus-
pirone, diphenhydramine, benzodiazepines, transdermal nicotine, and
estrogen (in men) (Sink et al. 2005).
At this writing, these issues remain unsettled—and unsettling. The
importance of effective management and treatment of agitation and vi-
olence in elderly patients is recognized by caregivers, facilities, and the
patients and families themselves in most cases. Nonpharmacological
interventions that address behavioral issues and unmet needs, includ-
ing those of caretakers, may be efficacious (Avalon et al. 2006). In cases
where nonpsychopharmacological interventions have been insuffi-
cient, medications may enable these patients to be maintained in less

restrictive settings, such as the home or an institution or assisted living
facility with amenities, as opposed to a secure nursing facility.
The best approach to violence in the elderly is multifaceted. That is
true also in patients with Alzheimer’s disease (Lavretsky and Nguyen
2006). The clinician must, however, continually consider new evidence
as it becomes available, reassess the patient’s ongoing need for medica-
tion, and rely on careful informed consent.
Key Points
■ Geriatric patients can be violent for the same reasons as younger
individuals.
■ Violence resulting from dementia and delirium with resultant
confusion is more common in the elderly.
■ Treatment of the underlying condition may be essential to control
of violence.
■ Frailness in many elderly persons may require special caution and
considerations.
■ Nonetheless, violence in the elderly can be serious and requires
active intervention.
References
Almvik R, Rasmussen K, Woods P: Challenging behavior in the elderly moni-
toring violent incidents. Int J Geriatr Psychiatry 21:368–374, 2006
American Medical Directors Association: Management Tools. n.d. Available at
http://www.amda.com/managementtools/index.cfm.
Aradt S, Turvey CL, Flaum M: Older offenders, substance abuse, and treatment.
Am J Geriatr Psychiatry 10:733–739, 2002
Aupperle PM, Koumaras B, Chen M, et al: Long-term effects of rivastigmine
treatment on neuropsychiatric and behavioral disturbances in nursing
home residents with moderate to severe Alzheimer’s disease: results of a
52-week open-label study. Curr Med Res Opin 20:1605–1612, 2004
Avalon L, Gum AM, Feliciano L, et al: Effectiveness of nonpharmacological in-
terventions for the management of neuropsychiatric symptoms in patients
with dementia. Arch Intern Med 166:2182–2188, 2006
Bogousslavsky J, Cummings JL: Behavior and Mood Disorders in Focal Brain
Lesions. New York, Cambridge University Press, 2000
The Elderly ❘ 405
Barak Y, Baruch Y, Mazeh D: Cardiac and cerebrovascular morbidity and mor-

tality associated with antipsychotic medications in elderly psychiatric in-
patients. Am J Geriatr Psychiatry 15:354–356, 2007
Cohen-Mansfield J: Temporal patterns of agitation in dementia. Am J Geriatr
Psychiatry 15:395–405, 2007
Cummings JL, Schneider E, Tariot PN, et al: Behavioral effects of memantine in
Alzheimer disease patients receiving donepezil treatment. Neurology
67:57–63, 2006
de Jonghe JFM, Kalisvaart KJ, Dijkstra M, et al: Early symptoms in the prodro-
mal phase of delirium: a prospective cohort study in elderly patients un-
dergoing hip surgery. Am J Geriatr Psychiatry 15:112–121, 2007
Devanand DP: Psychiatric assessment of the older patient, in Kaplan and Sa-
dock’s Comprehensive Textbook of Psychiatry, 8th Edition. Edited by Sa-
dock BJ, Sadock VA. Philadelphia, PA, Lippincott Williams & Wilkins, 2005,
pp 3603–3610
Folstein MF, Folstein SE, McHugh PR: “Mini-Mental State”: a practical method
for grading the cognitive state of patients for the clinician. J Psychiatr Res
12:189–198, 1975
Gorman DG, Read S, Cummings JL: Cholinergic therapy of behavioral distur-
bances in Alzheimer ’s disease. Neuropsychiatry Neuropsychol Behav
Neurol 6:229–234, 1993
Karlawich J: Alzheimer’s disease: clinical trials and the logic of clinical purpose.
N Engl J Med 355:1604–1605, 2006
Katz IR, Streim JE, Datto CJ: Psychiatric aspects of long-term care, in Kaplan
and Sadock’s Comprehensive Textbook of Psychiatry, 8th Edition. Edited
by Sadock BJ, Sadock VA. Philadelphia, PA, Lippincott Williams & Wilkins,
2005, pp 3793–3797
Lavretsky H, Nguyen LH: Diagnosis and treatment of neuropsychiatric symp-
toms in Alzheimer’s disease. Psychiatr Serv 57:617–619, 2006
Lewis CF, Fields C, Rainey E: A study of geriatric forensic evaluees: who are the
violent elderly? J Am Acad Psychiatry Law 34:324–332, 2006
Lyketsos CG, Steele C, Galik E: Physical aggression in dementia patients and its
relationship to depression. Am J Psychiatry 156:66–71, 1999
Rabins PV, Mace NL, Lucas MJ: The impact of dementia on the family. JAMA
248:333–335, 1982
Raivio MM, Laurila JV, Strandberg TE, et al: Neither atypical nor conventional
antipsychotics increase mortality or hospital admissions among elderly pa-
tients with dementia: a two-year prospective study. Am J Geriatr Psychia-
try 15:416–424, 2007
Read SL: Community resources, in Alzheimer’s Disease: Long-Term Manage-
ment. Edited by Cummings JL, Miller BL. New York, Marcel Dekker, 1990,
pp 235–244
Read SL, Miller B, Mena I, et al: SPECT in dementia: clinical and pathological
correlation. J Am Geriatr Soc 43:1243–1247, 1995
Reichman WE, Katz PR: Psychiatric Care in the Nursing Home. New York, Ox-
ford University Press, 1996
Rovner BW, German PS, Broadhead J, et al: The prevalence and management of
dementia and other psychiatric disorders in nursing homes. Int Psychoge-
riatr 2:13, 1990
Royall DR, Mahurin RK, Gray KF: Bedside assessment of executive cognitive
impairment: the executive interview. J Am Geriatr Soc 40:1221–1226, 1992
Schneider LS, Dagerman KS, Insel P: Risk of death with atypical antipsychotic
drug treatment for dementia: meta-analysis of randomized placebo-
controlled trials. JAMA 294:1934–1943, 2005
Schneider LS, Tariot PN, Dagerman KS, et al: Effectiveness of atypical antipsy-
chotic drugs in patients with Alzheimer’s disease. N Engl J Med 355:1525–
1538, 2006
Silver IL, Herrmann N: Comprehensive psychiatric evaluation, in Comprehen-
sive Textbook of Geriatric Psychiatry, 3rd Edition. Edited by Sadovoy J, Jar-
vik LF, Grossberg GT, et al. New York, WW Norton, 2004, pp 253–279
Sink KM, Holden KF, Yaffe K: Pharmacological treatment of neuropsychiatric
symptoms of dementia: a review of the evidence. JAMA 293:596–608, 2005
Strub RL, Black FW: The Mental Status Examination in Neurology. Philadel-
phia, PA, FA Davis, 1985
Strub RL, Black FW: Neurobehavior Disorders: A Clinical Approach. Philadel-
phia, PA, FA Davis, 1988
Szwabo PA, Grossberg GT: Problem Behaviors in Long-Term Care: Recognition,
Diagnosis, and Treatment. New York, Springer, 1993
U.S. Food and Drug Administration: FDA public health advisory: deaths with
antipsychotics in elderly patient s with behavioral disturbances. 2005.
Available at http://www.fda.gov/cder/drug/advisory/antipsychotics.htm.
Accessed June 1, 2007
Yorston GA, Taylor PJ: Commentary: older offenders—no place to go? J Am
Acad Psychiatry Law 34:333–337, 2006
P A R T V I
Special Topics
C H A P T E R 2 0
Forensic Issues
F orensic expertise is often requested in situations involving a poten-

tially dangerous person. But what does forensic actually mean, and how
does this evaluation process differ from the provision of clinical care?
The American Academy of Psychiatry and the Law (2005) provided the
following definition of forensic psychiatry in its ethics guidelines: “Foren-
sic Psychiatry is a subspecialty in which scientific and clinical expertise
is applied in legal contexts involving civil, criminal, correctional, regu-
latory matters, and in specialized clinical consultation in areas such as
risk assessment or employment” (p. 1).
This chapter reviews situations involving known, unknown, and de-
ceased offenders in which a forensic evaluation may be helpful. Although
clinical assessment skills are important in conducting a forensic examina-
tion, providers must be aware that having clinical expertise is vastly dif-
ferent from having the requisite skills to perform a forensic examination.
When conducting a forensic examination, the evaluator must understand
the relevant legal standard, the skills to evaluate the person in relationship
to this standard, the capacity to apply information to the legal construct,
and the capability to effectively translate and communicate his or her
findings in the context of the legal system (Grisso 1998). Table 20–1 high-
lights important areas to consider when a forensic evaluation is requested.
409
TABLE 20–1. Forensic evaluation checklist

What is the specific referral question?
Is the evaluation for treatment or for legal purposes?
Who is the party requesting the evaluation?
Have appropriate parties been notified in advance of the evaluation?
Has the evaluee or legally governing authority consented to the

evaluation?
Have the parameters of confidentiality been explained to the evaluee?
Have appropriate collateral records been reviewed?
Have appropriate third parties to interview been identified?
Forensic Evaluations of Known Offenders

Forensic evaluations are often requested of defendants involved in the
criminal justice system, and the evaluator must know if any legal stan-
dard governs the particular evaluation requested. “Competency to stand
trial” evaluation requests are the most common referrals for criminal fo-
rensic examinations (Rogers et al. 2001; Warren et al. 1991). The legal
standard for assessing a defendant’s competency was articulated in
Dusky v. U.S. In this 1960 landmark case, the U.S. Supreme Court an-
nounced that the “test must be whether he has sufficient present ability
to consult with his lawyer with a reasonable degree of rational under-
standing and whether he has a rational as well as a factual understanding
of the proceedings against him” (Dusky v. U.S. 1960). Although this is less
commonly requested, the forensic evaluator may also be asked to evalu-
ate the defendant’s sanity at the time of the offense and to comment on
the relationship of mental illness to violent behavior. In general, the in-
sanity defense excuses a mentally ill defendant from legal responsibility
for his or her criminal behavior. The exact standard for determining a de-
fendant’s sanity varies according to jurisdiction. The majority of states in
the United States use some variation of a cognitive test of insanity that de-
termines whether the defendant, as a result of a mental disease or defect,
knows or understands the nature and quality of his or her actions and/
or is able to distinguish right from wrong at the time of the offense
(Giorgi-Guarnieri et al. 2002). Finally, the court may also ask the psychia-
trist to outline risk factors regarding a criminal defendant’s risk of future
Forensic Issues ❘ 411
violence. Specific forensic expertise is required in assessing the risk of fu-

ture dangerousness in two particular offender types: stalkers and rapists.
Case Example 1
Jill is a 47-year-old woman who has recently separated from her hus-
band Jack after a volatile 10-year marriage. Jack has been arrested in the
past for driving under the influence and has physically assaulted Jill
during the course of their marriage. After Jill files for divorce, she dis-
covers notes on her car from Jack that say she will “always be his,” and
he leaves numerous threatening phone calls that include both pleas to
reunite and veiled threats to kill her. Jill becomes afraid and seeks out
mental health counseling for advice.
Stalking
Stalking is a serious problem in the United States. All 50 states, the fed-
eral government, and the District of Columbia classify stalking as a
crime. Although precise statutory definitions vary, most stalking stat-
utes incorporate the following elements:
• A course of conduct is specified

• The conduct is directed at a specific person
• The conduct results in a reasonable person experiencing fear
Approximately 1 in 12 women and 1 in 45 men will be stalked at

some point in their lifetime. Nearly 90% of stalkers are men, and the ma-
jority of female and male victims know their stalker. Women are more
likely than men (59% vs. 30%) to be stalked by an intimate partner. Al-
though the average duration of stalking is 1.8 years, the duration in-
creases to 2.2 years when the stalking relationship involves an intimate
partner. More than 70% of current or former intimate partners verbally
threaten their victims with violence; 81% of women stalked by a current
or prior partner are eventually physically assaulted, and more than 30%
are sexually assaulted (Tjaden and Thoennes 1998).
Numerous typologies have attempted to classify stalking behavior.
One of the most commonly referenced typologies was developed by
Michael Zona, who initially divided stalkers into the following three
categories (Zona et al. 1993):
1. Simple obsessional: These individuals usually have a prior relation-

ship with the victim and are motivated by a desire to enact revenge
on or to force reestablishment of the lost relationship. This group
poses the greatest risk of harm to their victim.
2. Love obsessional: In contrast to the simple obsessional, the vast ma-

jority of these stalkers have had no prior relationship to their victim.
These perpetrators may become focused on their victim after seeing
him or her in the media or another public forum. They are com-
monly viewed by others as obsessed fans. A significant number of
these individuals have a mental disorder such as schizophrenia or
bipolar disorder.
3. Erotomanic: Stalkers in this category delusionally believe that their
love object also loves them. The typical perpetrator is a female who is
convinced that an older male, usually of higher status, returns her af-
fection despite the lack of any rational evidence to support this belief.
Mullen et al. (1999) expanded the Zona typology of stalking to in-

clude five categories of stalkers, described by their primary motivation,
the context in which the stalking developed, and the function of the
stalker’s behavior. The primary types described include the rejected,
the intimacy seekers, the resentful, the predatory, and the incompetent.
Characteristics of each stalker category are noted in Table 20–2.
A more recent typology classifies stalkers according to their rela-
tionship, if any, with the victim and the private versus public-figure
context of their pursuit. The acronym RECON was selected for this
scheme because it is both relationship (RE) and context (CON) based
(Mohandie et al. 2006). The four categories of stalkers, described by the
type of victim selected, are labeled “intimate,” “acquaintance,” “public
figure,” and “private stranger.” An outline of this categorization
scheme and the associated features are shown in Table 20–3.
The examiner should consider the possibility that a victim may
make a false allegation that he or she has been stalked. Five contexts
involving false claims include 1) stalkers who claim to be victims, 2) in-
dividuals who have delusions of being stalked, 3) persons who have
been previously stalked and then misperceive benign acts of others,
4) persons with factitious disorder attempting to achieve the sick role,
and 5) malingerers who fabricate claims for external reward such as
money or to avoid criminal prosecution (Mullen et al. 2000).
Stalking can occur in a variety of circumstances and may include at-
tempts to contact the victim directly or indirectly through the phone,
mail, faxes, or personal notes left at a particular location. With the ad-
vent of electronic communication, stalkers may employ cyberspace
technology and the Internet to maintain contact with their victim, either
through e-mails or through gathering information about the victim by
using common search engines (McGrath and Casey 2002). Text messag-
ing or short message service via a mobile phone represents yet another
TABLE 20–2. Mullen stalker typology

Type Characteristics
Rejected Predominantly males who pursue an ex-intimate
Goal is reconciliation or revenge
Usually personality disordered rather than psychotic
Frequently persistent and intrusive
Intimacy seekers Desired attachment is usually romantic but can be to
parent, child, or close relationship
Believes target loves him/her and intimate relationship
will occur
Persists with pursuit despite responses from victim
May have underlying psychotic disorder
Resentful Targets person who stalker feels has wronged them
Stalking behaviors are intended to cause fear
Sense of power and control gained from stalking
Feels justified as acts are retribution for misjustice
Predatory Predominantly men who target unsuspecting women
Stalking behaviors are preparation for sexual assault
Pursue multiple victims over time
Incompetent Feels entitled to relationship with person of interest
Indifferent to target’s preferences
No insight regarding target’s lack of reciprocity
Persistent inept attempts
Source. Mullen et al. 1999.
developing method for the stalker to maintain communication with the

victim without actual physical contact (Eytan and Borras 2005).
The degree of danger posed by a stalker depends on a variety of fac-
tors. Intervention plans to curb or stop stalking behavior should be tai-
lored to each specific case. General recommendations noted to reduce
the impact of stalking include the following (Mullen et al. 2000):
• Communicating early and clearly that any contact or attention is un-

wanted
• Carefully protecting personal information, to include limiting distri-
bution of home address, telephone numbers, and cyberspace infor-
mation
• Informing trusted others at home and work to prevent inadvertent
disclosure of information and to protect their safety
• Contacting appropriate helping agencies such as police, victim sup-
port organizations, mental health clinics, and domestic and sexual
violence programs when applicable
414
TABLE 20–3. RECON (relationship and context) stalker typology
Stalker type Relationship category Characteristics Risk management
❘
Intimate Previous relationship: Most dangerous group, with history of violence Intense probation/parole
marriage, cohabiting, Quickly escalates supervision
dating/sexual Abuses alcohol and stimulants Intervene to decrease risk of
> 50% physically assault victim domestic violence before and
One-third use or threaten use of weapon after separation

> One-third have suicidal ideation or behavior
Acquaintance Previous relationship: Pursuit is sporadic but relentless Careful diagnostic assessment
employment related, Strong desire to initiate relationship Work with law enforcement and
affiliative/friendship, One-third will assault victim or damage mental health
customer/client property
Public figure No previous relationship; Greater proportion of female stalkers and male Professional protection of target
pursuit of public-figure victims Psychiatric treatment
victim Older, with less violence history Prosecution, with forensic
Increased likelihood of psychosis hospitalization as option
Unlikely to threaten and low violence risk
Private stranger No previous relationship; Often mentally ill men Psychiatric treatment
pursuit of private-figure 12% suicidal Aggressive prosecution
victim Communicate directly
One-third are violent toward person or
property
Source. Mohandie et al. 2006.
• Documenting and preserving all stalker contacts

• Recording all phone calls on an answering machine and keeping a
separate private line for personal calls
• Obtaining self-defense training
• Avoiding all contact and confrontations
The decision to obtain a restraining order against the stalker is one

that requires careful consideration, and obtaining an order may be in-
effective or actually inflammatory in certain situations. In particular,
Orion (1997) emphasized that restraining orders are likely to be ineffec-
tive against ex-intimates, who are heavily invested in the relationship,
and erotomanic or delusional stalkers, who view legal orders as not ap-
plicable to their situation. De Becker (1997) noted that restraining orders
are most likely to be effective in those situations that involve a casual ac-
quaintance with limited emotional investment and no prior history of
violence. If a decision is made to obtain a restraining order, the victim
should be aware that stalkers are at higher risk to act violently during
the time frame immediately following the issuance of the order, so that
added precautions can be taken. A protection order should be viewed
as only one component of a comprehensive plan designed to minimize
risk to the victim, and such an order may not be appropriate for every
case.
Rapists
Although the specific definition of rape varies according to jurisdiction,
common legal elements of rape include the penetration of a human or-
ifice by another person’s body part or object. According to the National
Crime Victimization Survey, there were more than 190,000 victims of
rape or sexual assault in the United States during 1995 (Catalano 2006).
The majority of rapes and sexual assaults are committed by men against
women.
Groth and Birnbaum (1979) classified rapists into four main types.
Two of the categories emphasize the use of sexual aggression to satisfy
the rapist’s need for power and the other two categories highlight the
use of sexual aggression to express anger. Characteristics of each rapist
subtype in this typology are outlined in Table 20–4.
A more recent typology developed by Knight and Prentky (1990),
classifying rapists according to their primary motivation, includes the
following four types:
TABLE 20–4. Groth rapist typology

Rapist type Motivating factors
Power-Reassurance Alleviate feelings of sexual inadequacy
Power-Assertive Express potency, mastery, and dominance
Anger-Retaliation Express rage toward women; seek revenge by
degrading women
Anger-Excitation Obtain sexual gratification from victim’s suffering
Source. Groth and Birnbaum 1979.
• Opportunistic rapists: Offenders who commit impulsive, unplanned

predatory acts to achieve immediate sexual gratification
• Pervasively angry rapists: Offenders who are angry in general and
who seek out targets as recipients of their anger rather than to meet
a sexual need
• Sexual rapists: Offenders who may have recurrent intrusive rape fan-
tasies and who assault to gratify sexual needs
• Vindictive rapists: Offenders who are primarily angry at women and
who attack to degrade and humiliate
When assessing a known rapist’s risk for future dangerousness, the

forensic examiner must conduct an extremely thorough interview, gen-
erally combined with structured assessments and review of collateral
records. Key assessment components include the following:
• Detailed clinical interview recording the individual’s account of his

or her actions
• Review of key collateral records to compare victim’s account and
police account to alleged perpetrator’s report
• Evaluation of any associated mental and/or substance use disorder
• Assessment of psychopathy and/or associated personality disorder
• Administration of standardized personality tests such as the Minne-
sota Multiphasic Personality Inventory
• Use of standardized questionnaires and sexual inventories
• Incorporation of actuarial risk assessment instruments and struc-
tured clinical interviews designed for the risk assessment of sexual
offenders
Although physiological measures such as the penile plethysmo-

graph or polygraph have been used to monitor treatment progress of
sex offenders, they are generally not permissible in court and may not
be allowed as part of the forensic examination. The recidivism risk for

rapists varies according to the study conducted and the length of fol-
low-up time measured. In their follow-up of 136 rapists, Prentky et al.
(1997) found that 39% reoffended over the 25-year follow-up period.
Factors associated with reoffense for rapists include the following
(Prentky and Burgess 2000):
• Impulsive, antisocial behavior

• Psychopathy
• Sexual drive strength
• Sexual coercion and rape fantasies
• Number of prior sexual offenses
• Offense planning
• Attitudes (global/pervasive anger, hypermasculine/macho, seeks
ways to con others, criminal)
Forensic Evaluations of Unknown Offenders

Forensic evaluators may be asked to assist law enforcement in identify-
ing violence risk factors regarding an offender whose exact identity is
yet unknown. Perhaps the most dangerous and terrifying criminal is
that individual who murders multiple people yet remains undetected.
One of the most famous approaches to profiling unknown perpetra-
tors was developed by the Behavioral Sciences Unit of the FBI Academy.
FBI profilers attempt to identify a suspect by searching for specific phys-
ical and behavioral clues at the crime scene. Investigators divide crime
scenes into two broad categories: organized and disorganized. Organized
crimes are characterized by planning in advance of the offense, targeting
a stranger victim, using restraints on a victim, hiding the dead body, re-
moving the weapon and/or evidence, and general control of the crime
scene. In contrast, disorganized crimes are described as unplanned, ran-
dom, and sloppy, with minimal use of restraints, sudden violence to the
victim with subsequent sexual acts, and a failure by the killer to remove
the weapon or body from the crime scene. This information theoretically
serves as a personality fingerprint to assist law enforcement in narrowing
the field of potential perpetrators. According to the FBI profiling system,
organized and disorganized crime scenes should match murderers with
organized and disorganized characteristic or traits. The original data set
underlying the FBI profiling system was developed by FBI agents who
interviewed 36 convicted sexual murderers, 29 of whom were serial sex-
ual killers. Profile characteristics of organized versus disorganized mur-
derers as defined by the FBI are outlined in Table 20–5 (Ressler et al. 1988).
TABLE 20–5. Profile characteristics of organized and disorganized

murderers
Organized Disorganized
Average to above-average intelligence Below-average intelligence
Socially competent Socially inadequate
Skilled work preferred Unskilled work
Sexually competent Sexually incompetent
High birth-order status Low birth-order status
Father’s work stable Father’s work unstable
Inconsistent childhood discipline Harsh discipline as a child
Controlled mood during crime Anxious mood during crime
Use of alcohol with crime Minimal use of alcohol
Precipitating situational stress Minimal situational stress
Living with partner Living alone
Mobility with car in good condition Lives/works near crime scene
Follows crime in news media Minimal interest in news media
May change jobs or leave town Significant behavior change (e.g.,
drug/alcohol abuse, religiosity)
Source. Ressler et al. 1988.
Criticisms of the FBI profiling methodology have included observa-

tions that crime scene characteristics do not clearly separate organized
and disorganized crimes or offenders and the concern that profile char-
acteristics of serial murderers may not generalize to nonsexual offend-
ers (Canter et al. 2004).
Forensic Evaluations of Deceased Persons

The circumstances surrounding a person’s death are sometimes hazy,
leaving an air of mystery as to why and how that person died. Consider
the situation in which a wife kills her husband with a single gunshot
through the heart. Was the deceased shot by a dangerous or deranged
woman, was she defending herself against a brutal attack, or did her
husband purposely provoke her to kill him in a veiled suicide attempt?
Or consider that for many, Marilyn Monroe’s death in 1962 continues to
remain shrouded in mystery, with theories about her cause of death
ranging from an accidental overdose to homicide committed by agents
of the U.S. government.
E.S. Shneidman, co-founder of the LA Suicide Prevention Center,
coined the term psychological autopsy to describe a posthumous evalua-
tion process in which the examiner conducts a thorough retrospective

investigation to determine the decedent’s intentions and other possible
causes of his or her death. Dr. Shneidman and his colleagues envisioned
the use of the psychological autopsy in circumstances where the mode
of death was equivocal in nature, such as drug overdoses, “suicide by
cop,” Russian roulette, vehicular accidents, murder-suicide, and auto-
erotic asphyxia (Shneidman 1981).
The evaluator conducting a psychological autopsy should carefully
consider the following five concepts during retrospective investigation:
1) cause, 2) mode, 3) motive, 4) intent, and 5) lethality.
Cause explains how the person actually died. Examples of potential
causes of death include a single gunshot wound to the head, a crush in-
jury from a car accident, or a massive heart attack. Although the cause
of death may be clear to the coroner, the mode of death is often more
ambiguous.
Mode refers to the circumstances that led to the cause of death. When
determining mode, the evaluator may find it helpful to classify the death
according to the acronym NASH, which stands for Natural, Accidental,
Suicide, or Homicide. In 5%–20% of death cases reviewed by the medical
examiner (coroner), the mode of death is unclear (Shneidman 1981). If the
mode of death is determined to be suicide, motive addresses why the de-
cedent committed suicide. When determining a person’s motive, the ex-
aminer attempts to understand the reasons and/or events that prompted
the individual to act. In many suicides, the motive is unknown and must
be inferred from the available evidence (Biffl 1996; Massello 1986). To as-
sist in the examiner’s investigation of the deceased’s motive, Shneidman
(1981) recommends careful review of the reasons why an individual com-
mitted suicide and why he or she chose that specific time to do it.
In contrast to motive, intent represents the resolve of an individual,
either consciously or unconsciously, in carrying out his or her death.
Understanding a person’s intent is facilitated by reviewing the degree
of lethality, or risk-taking, involved in the suicidal behavior (Peck and
Warner 1995). Shneidman (1981) noted that lethality represents the
probability that an individual will successfully kill him- or herself in the
immediate future. He divided the degree of lethality into high, me-
dium, low, and absent, although he did not provide precise classifica-
tion criteria for these categories.
A review of the literature indicates that a variety of techniques are
used to conduct psychological autopsies. Common characteristics of all
techniques include a careful review of collateral records (such as the au-
topsy, toxicology reports, medical and mental health records, and per-
sonal diaries of the victim), interviews of survivors with a focus on the
time frame immediately preceding the death, and a review of specific

mental health factors such as any prior psychiatric history, prior suicide
attempts, and behaviors suggesting that the person was planning to die.
Suicide by cop and murder-suicide are two situations in which the fo-
rensic evaluator helps unravel the relationship of violence toward oth-
ers to the offender’s own death.
Case Example 2
Joe is a 48-year-old man with a long-standing history of impulsive, an-
gry outbursts. He is extremely narcissistic and becomes aggressive
whenever criticized. Because of Joe’s personality, he has been unable to
sustain any long-term relationships and is socially isolated. He spends
a great deal of his time reading weapons magazines and fantasizes
about dying while shooting others in a “blaze of glory.” Because of his
temper problems, he was terminated from work. He blamed his super-
visor for all of his problems and decided to go to his workplace and
“take out anyone and everyone.”
Suicide By Cop
The phrase suicide by cop refers to behaviors by an individual intended
to provoke a law enforcement officer to use lethal force that will result
in the person’s death. In a study of more than 430 exchanges of fire be-
tween police and a suspect over a 10-year period in Los Angeles
County, California, researchers classified 10.5% of the cases as suicide-
by-cop situations (Hutson et al. 1998). In their review of 15 deaths of sui-
cidal persons who provoked law enforcement officers into killing them,
researchers (Wilson et al. 1998) described 10 characteristics of these in-
dividuals, which are summarized in Table 20–6.
In their review of the literature, Mohandie and Meloy (2000) out-
lined both verbal and behavioral clues indicating risk for suicide-by-
cop that may be helpful when conducting a psychological autopsy.
Twelve verbal clues associated with a suicide-by-cop situation included
suspects demanding authorities kill them, setting a deadline for author-
ities to kill them, threatening to kill or harm others, wanting to “go out
in a blaze of glory,” giving a verbal will, telling hostages or others they
want to die, looking for a “macho” way out, offering to surrender to
person in charge, indicating elaborate plans of their own death, express-
ing feelings of hopelessness/helplessness, emphasizing that jail is not
an option, and making biblical references, particularly to resurrection
and to the Book of Revelations.
Behavioral clues to suicide-by-cop risk include being demonstrative
with a weapon, pointing a weapon at police, clearing a threshold in a
TABLE 20–6. Characteristics of “suicide by cop” in 15 incidents

• Incidents were perceived as life-threatening to law officers and witnesses
• 14 of the victims (i.e., suicides) were male, 13 were Caucasian, and the
mean age was 32 years
• All verbally threatened homicide and resisted arrest
• Two-thirds of the victims took hostages
• All victims possessed a handgun or other weapon
• All victims posed or used their weapon during the incident
• 60% used the weapon with the intent to harm others
• 40% were intoxicated with alcohol
• 40% had documented psychiatric diagnoses, and 60% had evidence of
psychiatric illness
• Depression and substance abuse were the most common psychiatric
diagnoses
Source. Wilson et al. 1998.
barricade situation in order to fire a weapon, shooting at police, reach-

ing for a weapon with police present, attaching a weapon to one’s own
body, giving a countdown to kill hostages with police present, assault-
ing or harming hostages with police present, forcing confrontation with
police, advancing on police when told to stop, calling police to report a
crime in progress, continuing hopeless acts of aggression even after in-
capacitation by gunfire, self-mutilating with police present, pointing a
weapon at oneself with police present, refusing to negotiate, not mak-
ing any escape demands, and getting intoxicated (Mohandie and Meloy
2000).
Murder-Suicide
Murder-suicide occurs when an individual commits suicide after taking
the life of another person. In the National Violent Death Reporting Sys-
tem (Bossarte et al. 2006), only suicides that occur within 24 hours after
a murder qualify the deaths as murder-suicide, whereas other authors
extend this period to up to one week (Marzuk et al. 1992). Various labels
have been used to describe the phenomenon of a murderer who subse-
quently takes his or her own life, including “homicide-suicide,” “dyadic
death,” “doubly violent aggression,” and “despondent killers.”
Because there is no national surveillance system for murder-suicide
in the United States, the exact prevalence is difficult to determine. In the
majority of studies, murder-suicide rates have been reported to range
from 0.2 to 0.3 per 100,000 persons (Coid 1983; Marzuk et al. 1992;
Milroy 1995), although rates as high as 0.4 to 0.5 per 100,000 persons
have also been noted (Hannah et al. 1998; Hanzlick and Koponen 1994).
Hanzlick and Koponen (1994) identified common precipitants for mur-
der-suicide, as outlined in Table 20–7.
In addition to the motivators just noted, Bossarte et al. (2006), in
their study of 65 murder-suicide incidents, found that a legal problem
was the most common associated circumstance, experienced by one of
every four perpetrators.
Marzuk et al. (1992) proposed a murder-suicide typology based on
the relationship between the perpetrator and the victim. The proposed
categories of murder-suicide are 1) spousal/consortial, 2) familial, and
3) extrafamilial.
Spousal/Consortial Murder-Suicides
Numerous studies indicate that the majority of murder-suicides involve
male perpetrators who kill spouses or intimates (Aderibigbe 1997; Felt-
hous and Hempel 1995; Malphurs and Cohen 2002; Marzuk et al. 1992;
Milroy et al. 1997; Palermo et al. 1997) with a handgun (Malphurs and Co-
hen 2002). Nearly one-third of men who kill their spouse or partner will
commit suicide, a statistical phenomenon not matched by females who kill
intimate partners (Bossarte et al. 2006). Common psychiatric diagnoses in
perpetrators of couple murder-suicides include depression (Rosenbaum
1990) and alcohol intoxication or abuse (Comstock et al. 2005).
TABLE 20–7. Thirteen suggested motivators for murder-suicide

1. Impending divorce
2. Previous divorce
3. Release or perceived loss of a nonmarital partner
4. Jealousy
5. Retaliation
6. Mercy killing
7. Altruism
8. Financial stressors
9. Family stress or dysfunction
10. Alcohol
11. Drugs other than alcohol
12. Psychiatric illness
13. Unspecified or unknown factors
Source. Hanzlick and Koponen 1994.
Marzuk et al. (1992) divided spousal/consortial murder-suicides

into two subtypes: 1) amorous-jealous and 2) declining health. The
amorous-jealous subtype is the most common, representing between
50% and 75% of all spousal/consortial murder-suicides. In the amo-
rous-jealous subtype, the perpetrator is commonly a young man who
kills his spouse or girlfriend with a firearm in a jealous rage during a pe-
riod of actual or impending separation (Marzuk et al. 1992). More recent
studies of murder-suicide in older persons also note that interpersonal
conflict remains a potential trigger for these deaths, particularly in an
older man married to a younger woman (Cohen et al. 1998).
In the declining-health subtype, the murderer is typically an older
man (potentially in poor health) caring for his ailing wife. The perpetra-
tor may believe his actions are altruistic and serve as a mercy suicide.
Both parties may view their deaths as a dual suicide pact in which the
perpetrator’s actions are part of an assisted suicide.
Murder-Suicide of Other Family Members

Murder-suicides may involve a perpetrator who kills one or more fam-
ily members other than a spouse or intimate partner. In an Australian
study examining murder-suicides of children over a 29-year period, re-
searchers found that when fathers killed their children, they were more
likely to also kill their spouse, in contrast to mothers, who killed only
their children. Furthermore, compared with men, women tended to use
less violent methods to commit murder and suicide (Byard et al. 1999).
Filicide is broadly defined as the murder of a child. Three types of
filicide include
1. Neonaticide: Murder of a child less than 1 day old

2. Infancticide: Murder of a child older than 1 day and less than 1 year
3. Pedicide: Murder of a child older than 1 year and younger than
16 years
High rates of suicide after a filicide have been noted: 16%–29% of

mothers and 40%–60% of fathers take their life after murdering their
child (Hatters Friedman et al. 2005; Marzuk et al. 1992; Rodenburg
1971). In a study of 30 family filicide-suicide files, the most common
motive involved an attempt by the perpetrator to relieve a real or imag-
ined suffering of the child, an action known as an “altruistic filicide.”
Eighty percent of the parents in this study had evidence of a past or
current psychiatric history, nearly 60% having depression, 27% having
psychosis, and 20% experiencing delusional beliefs (Hatters Friedman
et al. 2005).
Familicide is defined as the murder of an entire family. These family

annihilators are usually men with depression, intoxication, or both
(Dietz 1986). Other risk factors associated with family annihilators in-
clude ongoing marital conflict and anger over separation, illness in a
child, and financial stress (Hatters Friedman et al. 2005; Morton et al.
1998). In certain cases, the perpetrator believes that murdering the fam-
ily members will alleviate future suffering and views his or her action
as altruistic. Rare cases of depressed or psychotic adolescents have also
been described in which the child kills his or her entire family prior to
taking his or her own life (Malmquist 2006).
Because of the high reported rates of mental illness in parents who
kill their children, evaluators should carefully consider the possibility
that their depressed, suicidal, or psychotic patients who are parents
may represent a potential risk of harm to their child. In addition to a
standard suicide risk assessment, the clinician should explore other ar-
eas that may assist in preventing a tragic death (Hatters Friedman et al.
2005). Sample questions include
• What do you believe will happen to your child if you die or commit
suicide?
• Do you have any fears or concerns that your child may be harmed
by others?
• Do you have any worries regarding your child’s health or unneces-
sary suffering?
• Are you having any thoughts about harming your child?
• Have you taken any steps to harm your child?
• If you have had thoughts of harming your child, what has kept you
from doing so thus far?
Extrafamilial Murder-Suicides
Suicides after the murder of a person who is not a family member or in-
timate partner are relatively rare. Murder-suicides outside the family
have occurred in the workplace, school settings, and public environ-
ments such as shopping malls or tourist locations. The perpetrators also
have been referred to as “mass killers” or “rampage killers.” Mullen
(2004) proposed a classification scheme for separating types of mass
killers that is defined by the relationship between the killers’ intentions
and their victims. This typology is described in Table 20–8.
The perpetrator of an autogenic (i.e., self-initiated) mass murder
typically involves a heavily armed male who randomly shoots individ-
uals before turning the gun on himself. The murderer may target for his
TABLE 20–8. Mass killing categories

Victim-specific mass killings: deaths of the particular victims are the
intended outcome.
Family slaying
Revenge killings
Cult killings
Gang killings
Instrumental mass killings: murder is a means to an end and perpetrator
intends to advance a particular objective.
Terrorist killings
Killings incidental to other criminal activity
Massacres: murders are indiscriminate and killing people is the goal.
Social conflict between different groups or classes resulting in civil
massacres
Autogenic massacres: individual driven by personal agenda and
psychopathology
Source. Mullen 2004.
first killing a person against whom he has a grudge and subsequently

expand his rampage to random victims. Such perpetrators are likely de-
pressed and may frequently have obsessional traits with marked hyper-
sensitivity and paranoia. Mullen (2004) found the following seven
characteristics of perpetrators of autogenic massacres who survived de-
spite their intent to commit suicide:
1. Male
2. Younger than 40 years of age
3. Socially isolated without close relationships
4. Unemployed or minimally employed
5. Bullied and/or isolated as a child
6. Fascinated with weapons
7. Collector of weapons
These murderers may also provoke law enforcement personnel to

kill them after their murders—again, “suicide by cop.” One study of 98
lone rampage killers in the United States found that those who were ul-
timately killed by police officers had the largest number of victims
when compared with those who committed suicide or who were ulti-
mately captured (Lester et al. 2005).
Key Points
Forensic evaluations of dangerousness are performed in a wide variety
of situations that may involve a known, unknown, or even deceased
offender. Regardless of the circumstance, forensic examiners should:
■ Conduct a detailed psychiatric examination to search for the pres-
ence of mental illness, substance use disorder, and/or personality
disorder
■ Carefully review collateral records
■ Interview individuals familiar with the offender
■ Understand unique characteristics of both the offender and the
potential victims when organizing a violence-prevention plan
■ Be familiar with key typologies to understand underlying motiva-
tions and risk factors
References
Aderibigbe YA: Violence in America: a survey of suicide linked to homicide.
J Forensic Sci 42:662–665, 1997
American Academy of Psychiatry and the Law: Ethical guidelines for the prac-
tice of forensic psychiatry, adopted May 2005. Available at http://aapl.org/
pdf/ETHICSGDLNS.pdf. Accessed March 20, 2007
Biffl E: Psychological autopsies: do they belong in the courtroom? Am J Crim
Law 1:123–146, 1996
Bossarte RM, Simon TR, Barker L: Homicide-suicide: characteristics of homi-
cide followed by suicide incidents in multiple states, 2003–04. Inj Prev
12(suppl):33–38, 2006
Byard RW, Knight D, James RA, et al: Murder-suicides involving children: a 29-
year study. Am J Forensic Med Pathol 20:323–327, 1999
Canter DV, Alison LJ, Alison E, et al: The organized/disorganized typology of
serial murder: myth or model? Psychol Public Policy Law 10:293–320, 2004
Catalano SM: National Crime Victimization Survey: Criminal Victimization
2005 (BJS Bulletin, NCJ 214644). Washington, DC, Office of Justice Statistics,
U.S. Department of Justice, 2006
Cohen D, Llorente M, Eisdorfer C: Homicide-suicide in older persons. Am J
Psychiatry 155:390–396, 1998
Coid J: The epidemiology of abnormal homicide and murder followed by sui-
cide. Psychol Med 13:855–860, 1983
Comstock RD, Mallonee S, Kruger E, et al: Epidemiology of homicide-suicide
events, Oklahoma, 1994–2001. Am J Forensic Med Pathol 26:229–235, 2005
De Becker G: The Gift of Fear: Survival Signals That Protect us from Violence.
London, Bloomsbury, 1997, pp 200–214
Dietz PE: Mass, serial, and sensational homicides. Bull NY Acad Med 62:477–
491, 1986
Dusky v. U.S., 362 U.S. 402 (1960)

Eytan A, Borras L: Stalking through SMS: a new tool for an old behavior? Aust
N Z J Psychiatry 39:204, 2005
Felthous AR, Hempel A: Combined homicide-suicides: a review. J Forensic Sci
40:846–857, 1995
Giorgi-Guarnieri D, Janofsky J, Keram E, et al: AAPL practice guideline for fo-
rensic psychiatric evaluation of defendants raising the insanity defense.
J Am Acad Psychiatry Law 30(suppl):S3–S40, 2002
Grisso T: Forensic Evaluation of Juveniles. Sarasota, FL, Professional Resource
Press, 1998
Groth AN, Birnbaum J: Men Who Rape: The Psychology of the Offender. New
York, Plenum Press, 1979, pp 12–60
Hannah SG, Turf EE, Fierro MF: Murder-suicide in central Virginia: a descrip-
tive epidemiologic study and empiric validation of the Hanzlick-Koponen
typology. Am J Forensic Med Pathol 19:275–283, 1998
Hanzlick R, Koponen M: Murder-suicide in Fulton County, Georgia 1988–1991:
comparison with a recent report and proposed typology. Am J Forensic
Med Pathol 15:168–173, 1994
Hatters Friedman S, Huorda DR, Holden CE, et al: Filicide-suicide: common
factors in parents who kill their children and themselves. J Am Acad Psy-
chiatry Law 33:496–504, 2005
Hutson HR, Anglin D, Yarbrough J, et al: Suicide by cop. Ann Emerg Med
32:665–669, 1998
Knight RA, Prentky RA: Classifying sexual offenders: the development and cor-
roboration of taxonomic models, in The Handbook of Sexual Assault. Ed-
ited by Marshall WL, Barbaree HE. New York, Plenum, 1990, pp 23–52
Lester D, Stack S, Schmidtke A, et al: Mass homicide and suicide deadliness and
outcome. Crisis 26:184–187, 2005
Malmquist CP: Combined murder-suicide, in Textbook of Suicide Assessment
and Management. Edited by Simon RI, Hales RE. Washington, DC, Amer-
ican Psychiatric Publishing, 2006, pp 495–509
Malphurs JE, Cohen D: A newspaper surveillance study of homicide-suicide in
the United States. Am J Forensic Med Pathol 23:142–148, 2002
Marzuk P, Tardiff K, Hirsch CS: The epidemiology of murder-suicide. JAMA
267:3179–3183, 1992
Massello W: The proof in law of suicide. J Forensic Sci 31:1000–1008, 1986
McGrath MG, Casey E: Forensic psychiatry and the Internet: practical perspec-
tives on sexual predators and obsessional harassers in cyberspace. J Am
Milroy CM: The epidemiology of homicide-suicide (dyadic death). Forensic Sci
Int 71:117–122, 1995
Milroy CM, Dratsas M, Ranson DL: Homicide-suicide in Victoria, Australia.
Am J Forensic Med Pathol 18:369–373, 1997
Mohandie K, Meloy JR: Clinical and forensic indicators of “suicide by cop.”
J Forensic Sci 45:384–389, 2000
Mohandie K, Meloy JR, McGowan MG, et al: The RECON typology of stalking:
reliability and validity based upon a large sample of North American stalk-
ers. J Forensic Sci 51:147–155, 2006
Morton E, Runyan CW, Moracco KE, et al: Partner homicide-suicide involving

female homicide victims: a population-based study in North Carolina,
1988–1992. Violence Vict 13:91–106, 1998
Mullen PE: The autogenic (self-generated) massacre. Behav Sci Law 22:311–323,
2004
Mullen PE, Pathe M, Purcell R, et al: A study of stalkers. Am J Psychiatry
156:1244–1249, 1999
Mullen PE, Pathe M, Purcell R: Stalkers and Their Victims. New York, Cam-
bridge University Press, 2000, pp 187–237
Orion D: I Know You Really Love Me: A Psychiatrist’s Journal of Erotomania,
Stalking, and Obsessive Love. New York, Macmillan, 1997, pp 150–160
Palermo GB, Smith MB, Jenzten JM, et al: Murder-suicide of the jealous para-
noia type: a multicenter statistical pilot study. Am J Forensic Med Pathol
18:374–383, 1997
Peck DL, Warner K: Accident or suicide? Single-vehicle car accidents and the in-
tent hypothesis. Adolescence 30:463–472, 1995
Prentky RA, Burgess AW (eds): Prediction, in Forensic Management of Sexual
Offenders. New York, Springer, 2000, pp 99–142
Prentky RA, Lee AFS, Knight RA: Recidivism rates among child molesters and
rapists. Law Hum Behav 21:635–669, 1997
Ressler RK, Burgess AW, Douglas JE: Sexual Homicide: Patterns and Motives.
New York, Lexington Books, 1988
Rodenburg M: Child murder by depressed parents. Can Psychiatr Assoc J
16:41–48, 1971
Rogers R, Grandjean N, Tillbrook CD, et al: Recent interview-based measures of
competency to stand trial: a critical review augmented with research data.
Behav Sci Law 19:503–518, 2001
Rosenbaum ML: The role of depression in couples involved in murder-suicide
and homicide. Am J Psychiatry 147:1036–1039, 1990
Shneidman ES: The psychological autopsy. Suicide Life Threat Behav 11:325–
340, 1981
Tjaden P, Thoennes N: Stalking in America: findings from the National Violence
Against Women Survey. NCJ 169592. Washington, DC, Office of Justice Pro-
grams, National Institute of Justice Centers for Disease Control and Pre-
vention, 1998
Warren JI, Fitch L, Dietz PE, et al: Criminal offense, psychiatric diagnosis, and
psycholegal opinion: an analysis of 894 pretrial referrals. Bull Am Acad
Psychiatry Law 19:63–69, 1991
Wilson EF, Davis JH, Bloom JD, et al: Homicide or suicide: the killing of suicidal
persons by law enforcement officers. J Forensic Sci 43:46–52, 1998
Zona MA, Sharma KK, Lane J: A comparative study of erotomanic and obses-
sional subjects in a forensic sample. J Forensic Sci 38:894–903, 1993
C H A P T E R 2 1
Legal Issues of Prediction,

Protection, and Expertise
L ike the gift from the gods in numerous Greek myths (e.g., Midas, Pan-
dora, Icarus), psychiatry and psychology’s acquisition of the capacity to
assess the risk of violence comes with burdens that may overshadow
the gift. One burden of acumen in assessing risk of violence, with con-
sequences as tragic as any visited by the mythological Greek gods, is
that when violence occurs, psychiatrists and psychologists are often
transformed, with the aid of the legal process, from heroes to villains for
not having used their gift to prevent the violence from happening. An-
other burden is that when violence is legally relevant, the law expects
to receive accurate prophecies from psychiatrists and psychologists,
even at the expense of confidentiality and cautious science. Accord-
ingly, a comprehensive treatment of violence necessitates a hard look at
legal rules that bear on the assessment of violence. This chapter intro-
duces those contexts and the issues they raise.
The first section of this chapter addresses an all too common legal
dilemma for psychiatrists and psychologists: the impact of acumen in
violence assessment and management on the rules governing thera-
pist–patient confidentiality. Most states either recognize a statutory
duty of confidentiality for psychotherapists or incorporate a profes-
sional ethics code containing a duty of confidentiality into state licens-
ing law. If psychiatrists or psychologists have unique insights about a
429
patient’s desire to commit a violent act, should they be required to pro-

tect the victim, violate this duty of confidentiality, and reveal informa-
tion about the risk of violence (acquired in a confidential relationship)
to the police or an identifiable person at risk? Consider the irony of con-
cluding that a psychiatrist may not reveal even patients’ names to con-
cerned family members or friends but may owe a duty to protect a
third-party stranger who might be harmed as a result of the patient’s vi-
olent behavior, requiring a breach of confidentiality to that person or to
the police. How can the protection of confidentiality and the duty to
protect coexist in a psychiatrist or psychologist’s daily practice?
The second part of the chapter examines how the courts have re-
ceived expert evidence assessing the risk of violence. If the law con-
cludes that psychiatry and psychology’s ability to assess violence is
good enough to recognize a special relationship and impose a “duty to
protect” the non-patient victim, does it follow that courts will invari-
ably allow psychiatrists and psychologists to testify about violence as-
sessment? Once again, there is irony in the legal recognition of a duty to
protect while at the same time the legal system questions psychiatry’s
or psychology’s acumen in cases involving violence.
Finally, no legal inquiry into violence assessment would be com-
plete without discussing how dangerousness has shaped the nation’s
criminal and civil justice systems. We may question whether psychia-
trists and psychologists can accurately assess the risk of violence, but a
host of extant legal procedures rest on the assumption that they can
(e.g., civil commitment). The decision to detain criminal defendants
prior to trial or to impose capital punishment, for example, often turns
on assumptions regarding the ability to assess the risk of violence.
Tarasoff’s Legacy: Predicting Violence in

Mental Health Patients
There are many places one could begin a legal examination of the as-
sessment of the risk of violence. Chronologically, Tarasoff v. Regents of
the University of California (1976) finds the debate in the courts about the
ability of psychiatrists and psychologists to predict violence in full
stride, with a wry twist. In civil commitment proceedings, psychiatrists,
among other mental health professionals, had been touting their predic-
tive abilities to justify preventive confinement. Rather than touting their
wisdom in this case, however, the psychiatrists’ and psychologists’
briefs in Tarasoff tried to convince the courts that they are simply not
that good at prediction and therefore should not be liable for getting it
wrong.
Legal Issues of Prediction, Protection, and Expertise ❘ 431
In Tarasoff, the parents of Tatiana Tarasoff, a U.C. Berkeley coed who

was murdered by an international student who was a patient of a uni-
versity psychologist, sued the psychologist, the university, and the cam-
pus police for failing to use reasonable care to protect their daughter
from that patient’s acts. During individual therapy sessions, the patient,
Prosenjit Poddar, a spurned suitor of the victim, revealed that he was
going to kill Ms. Tarasoff when she returned from vacation. The psy-
chologist notified the campus police, advising them that his patient
should be committed. However, after speaking to him, the campus po-
lice decided that the young man was rational and not dangerous and re-
fused to confine him. When the victim returned from vacation, the pa-
tient murdered her. He was convicted of second-degree murder, which
was later reversed for an erroneous jury instruction and not retried. The
civil claim for negligence was filed in California state trial court, which
promptly granted the defendants’ motion to dismiss because the pros-
ecution had failed to state a valid legal claim.
The trial court dismissed the claim because Tarasoff was not a patient
of the psychologist and therefore the psychologist did not have a duty to
protect her from his patient. An accepted principle of the common law
is that one person does not have the duty to control the conduct of an-
other, warn another of danger, or rescue another unless he undertakes
that duty voluntarily or a special relationship exists that gives rise to a
right of protection or a duty to control. After an appeal from the original
dismissal, as well as a withdrawal of its initial opinion, the Supreme
Court of California decided that either the psychologist’s relationship to
his patient or to his patient’s victim satisfied the special relationship:
Although, as we have stated above, under the common law, as a general

rule, one person owed no duty to control the conduct of another…the
courts have carved out an exception to this rule in cases in which the de-
fendant stands in some special relationship to either the person whose
conduct needs to be controlled or in a relationship to the foreseeable vic-
tim of that conduct…. Applying this exception to the present case, we
note that a relationship of defendant therapists to either Tatiana or Pod-
dar will suffice to establish a duty of care; as explained in section 315 of
the Restatement Second of Torts, a duty of care may arise from either
“(a) a special relation…between the actor and the third person which
imposes a duty upon the actor to control the third person’s conduct, or
(b) a special relation…between the actor and the other which gives to
the other a right of protection.” (Tarasoff v. Regents of the University of Cal-
ifornia 1976)
The Supreme Court of California based this reasoning on a belief in

the psychologist’s predictive abilities. Accordingly, the court reasoned,
when a psychotherapist determines, or should determine, that “his pa-

tient presents a serious danger of violence to another, he [the therapist]
incurs an obligation to use reasonable care to protect the intended vic-
tim against such danger.” Although the California Supreme Court obli-
gated psychotherapists to safeguard third parties endangered by their
patients, it did not address whether the defendant psychotherapist sat-
isfied this obligation by informing the campus police and requesting
confinement of his patient. Instead, the trial court dismissed the case be-
fore any evidence was presented, and the case was settled before it was
tried after remand. If this same set of circumstances reached the Califor-
nia trial courts today, the correct answer regarding the steps to be taken
by the psychotherapist would be clear. Designed to clarify when a duty
arises and what fulfills it, a layer of legislation now answers these ques-
tions: “There shall be no monetary liability on the part of, and no cause
of action shall arise against, a psychotherapist who, under the limited
circumstances specified above, discharges his or her duty to warn and
protect by making reasonable efforts to communicate the threat to the
victim or victims and to a law enforcement agency” (Cal. Civ. Code §
43.92[b] [West 2007]). Whatever the Tarasoff court might have thought
30 years ago, the psychologist in Tarasoff apparently met this new statu-
tory requirement by notifying the campus police.
Although Tarasoff changed California law, in our system of federal-
ism with semiautonomous, coequal states, it did not by itself change the
laws in other states. Free to choose about most elements of tort law and
criminal law, however, many states have accepted Tarasoff in whole or in
part. Many have also implemented statutes to clarify and limit the re-
sponsibility of psychiatrists and psychologists to report potentially vio-
lent patients and limit claims for breach of that duty. The vast majority
of these statutes attempt to limit the situations in which reporting is re-
quired: typically, they state that the danger or threatened act of violent
behavior must be imminent, although this cautionary measure is not
scientifically grounded or meaningful; the potential victim must be eas-
ily identifiable from the patient’s threats; and the harm to the threatened
potential victim must be serious and reasonably foreseeable. For exam-
ple, in New Jersey (N.J. Stat. § 2A:62A–16 [2007]) and Delaware (16 Del.
C. § 5402 [2007]), a duty to protect arises only when 1) the patient has
communicated to that practitioner a threat of imminent, serious physi-
cal violence against a readily identifiable individual or against him- or
herself and the circumstances are such that a reasonable professional in
the practitioner’s area of expertise would believe the patient intended to
carry out the threat; or 2) the circumstances are such that a reasonable
professional in the practitioner’s area of expertise would believe the pa-
tient intended to carry out an act of imminent, serious physical violence

against a readily identifiable individual or against him- or herself.
For example, if the patient revealed to the therapist that next month
he or she was going to vaporize everyone in the world who had brown
hair and blue eyes, this would not invoke the therapist’s duty to warn
or protect. Whatever imminent means, this is not imminent. Millions of
people have brown hair and blue eyes; the patient’s description of the
potential victims does not make them easily identifiable. Finally, his
threat to “vaporize” is not realistic or reasonably foreseeable. On the
other hand, if the patient had told the therapist that tonight, with a deer
hunting rifle, he or she was going to kill his neighbor with brown hair
and blue eyes, then the answer would be different. The time frame is
now closer and therefore less accommodating of exploring alternatives,
and the victim is easily identifiable. If the threat is credible, then it
would be reasonably foreseeable that the patient could use a rifle to kill
the victim. Although these examples are extreme, this is the type of
analysis that the statutes and courts encourage therapists to engage in
when deciding how to weigh their confidential obligations to their pa-
tients against the duty to protect potential victims.
In an additional attempt to limit the duty to protect and the thera-
pist’s revelation of confidential information, some states define the in-
dividuals who may and/or must be warned of the possibility of future
violent behavior. For example, if the patient’s threat to kill his or her
neighbor tonight is credible, the therapist must report this information
to the proper authorities and the victim (but should not call the local
news station or newspaper to report this threat). In California and
Washington, for example, the therapist discharges his or her duty to
warn and protect by making reasonable efforts to communicate the
threat to the victim or victims and to a law enforcement agency (Cal.
Civ. Code § 43.92[b] [West 2006]; Rev. Code Wash. §71.05.120 [2007]).
A few states, such as Texas, have rejected Tarasoff outright. Texas
psychiatrists and psychologists are permitted to reveal patient threats of
violence but are not required to do so. The Texas statute governing con-
fidentiality provides: “A professional may disclose confidential infor-
mation only…to medical or law enforcement personnel if the profes-
sional determines that there is a probability of imminent physical injury
by the patient to the patient or others or there is a probability of imme-
diate mental or emotional injury to the patient” (Texas Health and
Safety Code 2006, emphasis added).
In Thapar v. Zezulka (1999), the Texas Supreme Court interpreted this
statutory discretionary disclosure to be controlling in its determination
or recognition of the duty to protect/warn. The surviving spouse of a
person murdered by a psychiatrist’s patient sued the psychiatrist, alleg-

ing negligence in diagnosing the patient’s condition and in failing to
warn family members or law enforcement officers of the patient’s
threats against his eventual victim. The Texas Supreme Court inter-
preted the state confidentiality statute’s authorization of permissive
disclosure of confidential information as a rejection of compelled dis-
closure, from which the court inferred a legislative rejection of Tarasoff.
Of course, the legislature could have mentioned Tarasoff by name if they
intended to reject it, but then again if they disagreed with Thapar they
could have amended the law. Neither occurred.
Texas is not alone in allowing the therapist to use discretion when de-
ciding whether to reveal the patient’s violent threats. In Mississippi a
therapist “may communicate the threat only to the potential victim or vic-
tims, a law enforcement agency, or the parent or guardian of a minor who
is identified as a potential victim” (Miss. Code Ann. § 41–21–97 [2006],
emphasis added). Florida law also grants the therapist substantial discre-
tion in choosing to disclose (Fla. Stat. § 491.0147 [2006]). In both of these
states, the courts have agreed that the permissive disclosure language of
the statutes is inconsistent with recognition of a Tarasoff duty (Boynton v.
Burglass 1991; Evans v. United States 1995; Green v. Ross 1997).
What do these differing state statutes and court interpretations de-
mand of the best practices of psychiatrists and psychologists? They de-
mand attention at the beginning of the relationship and the process of
informed consent. Patients must be informed that confidentiality is not
absolute and told what limits exist. When a psychiatrist or psychologist
determines that a patient poses a risk to a third party, best practice de-
mands attention in a consultation with colleagues and/or legal counsel
to consider the risk of violence as well as the risks and benefits of alter-
native responses to the duty imposed. The best practice also demands
that in dealing with Tarasoff, which is, at bottom, a negligence claim for
failing to act as a reasonable psychologist would act under the circum-
stances, clinicians remain current in the research on assessment of the
risk of violence. Of course, best practice standards also demand that
psychiatrists and psychologists know the relevant laws for their states
regarding what gives rise to a duty to protect third-party victims and
the proper entities to notify with this information.
Daubert’s Legacy: Violence and Expert Testimony

Assessment of the risk of violence is certainly not the only topic on
which psychiatric or psychological expert testimony is offered. Yet it
has played an important role in establishing rules for admissibility of
expert testimony. Thus it is worthwhile to examine it for its own sake as

well as for the lessons it offers about expert testimony more generally.
The approach of courts to accommodating the need for expert assis-
tance and scientific orthodoxy has varied over time. Until the twentieth
century, even on matters of science, the courts rarely demanded more
than a qualified expert as a condition of admissibility, leaving the jury to
determine the reliability of the method or process on which an expert’s
opinion rests through cross-examination and the testimony of opposing
experts. Thus, for witnesses presented as experts, the sole inquiry in ad-
missibility focused on the witnesses’ education, training, or experience.
Any formulas, devices, or techniques that the witness relied upon were
matters of believability for the jury to decide, not matters of admissibil-
ity. That began to change with the D.C. Court of Appeals decision in Frye
v. United States (1923). In Frye, a defendant charged with murder offered
exculpatory evidence from an early polygraph relying on systolic blood
pressure. Reasoning that the qualifications of the expert witness pre-
senting the test results were not sufficient to justify admission without
examining the accuracy of the machine itself, the decision articulated
the famous “general acceptance” test to assess the admissibility of novel
scientific evidence.
Frye was a refreshing recognition of the need to look at both the sci-
ence and the scientist as a condition of admissibility of scientific expert
testimony. However, it had its problems. It equated popularity (“general
acceptance”) with scientific accuracy, it left unresolved whose acceptance
is required (polygraph examiners, psychologists, psychiatrists, or the
National Academy of Sciences) and what signaled general acceptance
(50%, 75%, or 90%), and it insulated the judge from scientific disputes by
directing the judge to defer to scientific majority. Although the debate
raged about Frye’s merits as well as its survival under the Federal Rules
of Evidence, which were intended to liberalize the admissibility of evi-
dence, another evidentiary debate loomed over the admissibility, in cap-
ital sentencing, of predictions of violence based on a brief clinical exam-
ination. Texas was one of the states at the center of this debate, because
of both its capital punishment criteria and the presence of a very persua-
sive psychiatric expert witness, Dr. James Grigson, who most often
found himself assisting the prosecution. A critical question in Texas for
a capital jury choosing whether to impose the death penalty is the risk of
violence the defendant poses if not executed. In Barefoot v. Estelle (1983),
the U.S. Supreme Court heard a constitutional challenge to imposing the
death penalty based on the argument that Dr. Grigson’s clinical-opinion
testimony was not generally accepted as a reliable basis for a prediction
of future violence, although he was very successful in persuading juries.
The Supreme Court rejected the prisoner’s argument that the U.S. Con-
stitution prohibits imposition of the death penalty based on clinical pre-
dictions regarded as unreliable by the American Psychiatric Association,
among others, which filed an amicus brief in the case. Acknowledging
problems with the accuracy of clinically based predictions, the Court re-
fused to permit a private organization to frustrate imposition of the
death penalty it had approved.
Barefoot v. Estelle (1983) presented a question of constitutional mini-
mums—does a death sentence that rests on a clinical prediction violate
due process, thus requiring a fundamental change in the criteria for the
imposition of capital punishment? Because it was a state court trial and
conviction, it did not specify what the Federal Rules of Evidence might
demand of an expert in these circumstances above the constitutional
minimums. That part of the puzzle was revealed a decade later, when
the Supreme Court decided Daubert v. Merrell Dow Pharmaceuticals, Inc.
(1993). Daubert was a toxic tort case, filed in California state court and
removed to federal court on diversity of citizenship grounds. The plain-
tiff alleged that Bendectin, an anti-nausea drug manufactured by Mer-
rell Dow Pharmaceuticals, caused limb reduction birth defects. The de-
fendant’s Motion for Summary Judgment asserted that epidemiology
was the generally accepted standard for evaluating a drug’s toxicity
and that no published epidemiological study found a significant rela-
tionship between Bendectin and limb reduction birth defects. When the
plaintiff’s experts, all well-qualified research scientists with impressive
credentials, offered another approach to analyzing the data, the court
rejected their expert testimony, relying on Frye, which it assumed to be
the standard applied under the Federal Rules of Evidence, and granted
the defendant’s motion for summary judgment.
The case made its way to the Ninth Circuit and eventually to the Su-
preme Court. The Court granted review to resolve the test for expert ev-
idence under the Federal Rules of Evidence, unresolved since the rules
were enacted in 1975. Examining the text of the rules, the court con-
cluded that because nowhere in the rules was Frye mentioned, it could
not have been intended to be the test for scientific evidence under the
Federal Rules of Evidence. Instead, relying on falsifiability as the hall-
mark of science, the court adopted a pragmatic approach to relevance
and reliability, taking into account whether the underlying methods
and procedures were testable and had been tested; whether they had
been subjected to peer review and publication, and if so what was the
error rate and could it be controlled; and finally a rebirth of Frye’s gen-
eral acceptance test. Two other related decisions followed (General Elec-
tric v. Joiner [1997] and Kumho Tire Co. v. Carmichael [1999]) that made
clear that these considerations were committed to the discretion of the

trial court, who might apply some but not others as appropriate.
After much prognosticating about Daubert’s likely impact on psychi-
atric and psychological expert testimony, the post-Daubert world for
psychiatric and psychological experts for the most part has not differed
significantly. Daubert was intended to ask, as a matter of admissibility,
why we should believe an expert’s assertion. That scrutiny has rarely
occurred. For the most part, scrutiny of the methods and procedures
used by psychiatric expert witnesses testifying on issues such as com-
petence to stand trial, the insanity defense and punishment, and mental
or emotional loss in personal injury claims has not changed. When the
Daubert question is raised in the context of violence assessment in death
penalty cases, courts often assume incorrectly that Barefoot, which ad-
dressed constitutional minimums, also disposed of the evidentiary
question under Daubert (Johnson v. Cockrell 2002). It did not. Ironically,
clinically based testimony is rarely challenged; rather, it is testimony
based on written tests that claim an actuarial basis that more often in-
curs a challenge.
Where the issue is addressed and courts do not view the evidence
issue as resolved by Barefoot, many conclude that risk assessment is not
novel science to which Frye or Daubert apply (In re detention of Thorell
[2004]). When they do apply Frye or Daubert, the written tests used in
risk assessment (Static 99, Rapid Risk Assessment for Sexual Offense
Recidivism, Structured Anchored Clinical Judgment–Minimum) al-
most always survive the scrutiny applied (In re commitment of R.S.
[2004]). As contrasted with Daubert’s application in toxic tort cases, its
application in violence assessments and sexually violent predator com-
mitments seems to result in less demanding scrutiny.
Violence Assessment and the Dilemma of Daubert

The assumption that reliable violence risk assessment is available is an
unstated but central premise of numerous legal processes and proce-
dures. As far as public safety or incapacitation is concerned, criminal
punishment without reliable individual violence assessments is a blunt
instrument, as reflected in judicial dissatisfaction with one-size-fits-all
sentencing guidelines. Unreliable violence assessments in domestic vi-
olence cases pose frightening consequences. If the assessment of future
dangerousness as required by the statutes that govern civil commit-
ment of the mentally ill and sexually violent predators is unreliable, the
resulting decision to commit is random. Necessary treatment would not
reach its intended targets, and the public would be lulled into a false
sense of security. The judicial system has historically relied on what it

assumes to be reasonably accurate violence assessment to decide whom
to incarcerate and commit.
Although criminal procedure varies from state to state, for purposes
of illustration we highlight the different stages at which Texas criminal
procedure calls for violence assessment. Both formal diversion (e.g., ac-
ceptance of defendant in mental health courts rather than criminal pros-
ecution) and informal diversion (e.g., acceptance of prisoner in psychi-
atric emergency department rather than jail booking) turn on a violence
assessment, among other things. If the defendant has been detained for
a crime before guilt or innocence is decided, the court is required to ad-
dress the prisoner’s potential for violence to determine bail. One of the
factors to be considered in setting bail is “the future safety of a victim of
the alleged offense and the community” (Tex. Code Crim. Proc. Art.
17.15[5] [2006]). If society cannot otherwise be protected, bail can be de-
nied or increased; but to operate effectively, this decision requires accu-
rate assessment of the risk of violence.
During sentencing, evidence of “dangerousness” can be introduced
as either aggravating or mitigating evidence. In capital murder sentenc-
ing, violence assessment is critical in choosing between life imprison-
ment and the death penalty. One of the factors that the jury must consider
when determining whether to impose the death sentence in Texas for a
defendant convicted of capital murder is “whether there is a probability
that the defendant would commit criminal acts of violence that would
constitute a continuing threat to society” (Nenno v. State 1998; Tex. Code
Crim. Proc. Art. 7.071[2][b][1] [2006]).
As noted, predicting future violent behavior is also a foundation for
civil commitment. Civil commitment is a preventative measure to pro-
vide protection from the risks some mentally ill individuals pose to
themselves and society. In O’Connor v. Donaldson (1975), the Supreme
Court held that “a State cannot constitutionally confine without more
a nondangerous individual who is capable of surviving safely in free-
dom.” If violence could not be predicted, then civil commitment could
not be justified unless the state offered something “more,” a justifica-
tion that would presumably focus on the availability and the effective-
ness of treatment, a vastly more demanding and expensive yardstick. In
addition to leaving unanswered the “something more” question, the
Supreme Court did not explain how “dangerousness” should be as-
sessed, although the decision rests on the assumption that dangerous-
ness can be reliably assessed.
Consistent with the assumption of expertise, most states have en-
acted statutes requiring specific mental health experts to testify in civil
commitment cases. For example, in Illinois, “no respondent may be

found subject to involuntary admission unless at least one psychiatrist,
clinical social worker, or clinical psychologist who has examined him
testifies in person at the hearing” (405 Ill. Comp. Stat. 5/3–807 [2007]).
Rhode Island requires that “in determining whether there exists a like-
lihood of serious harm the physician and the court may consider previ-
ous acts, diagnosis, words or thoughts of the patient” (R.I. Gen. Laws §
40.1–5–2[7][4][2007], emphasis added). In Texas “the proof [for civil
commitment] must include expert [psychiatric] testimony and, unless
waived, must include evidence of either a recent overt act or a continu-
ing pattern of behavior in either case tending to confirm the likelihood
of serious harm to the person or others or the person’s distress and de-
terioration of ability to function” (State for Interest of P.W. 1990).
If we are to take Daubert seriously and ask why we should believe
what the expert is offering, how many of these assessments would sur-
vive rigorous scrutiny? Do the data support that reliable assessment
techniques exist for each legally relevant context? When reliable risk as-
sessment methods or techniques exist, how widely are they used? Here
a lesson might be learned from clinical practice and its use of evidence-
based medicine to encourage use of the most reliable techniques. A crit-
ical analysis of existing forensic practice resulting in research-based fo-
rensic guidelines offers the potential for greater accuracy and consis-
tency of forensic assessments of the risk of violence.
To build these best techniques on a solid foundation and reinforce
the procedures they inform, it would also be useful first to engage law-
yers, judges, psychiatrists, and psychologists in an interdisciplinary di-
alogue to identify all of the contexts in which legal determinations turn
on the risk of violence, and then to determine what the research reveals
about what psychiatrists and psychologists have to contribute to the de-
bate. In any of those contexts in which the analysis leaves us confident
in the reliability of the methods used, it might be possible to recognize
that concurrence so that judicial notice could be taken of the acceptance
of these methods, thereby resolving the Daubert/Frye question. In any of
these contexts in which the analysis leaves us uncomfortable with the
reliability of the methods used to assess risk in this context, identifica-
tion of the context should result in consideration of new assessment ap-
proaches as well as an exploration of alternative legal rules or standards
(e.g., examining alternative bases for imposing capital punishment) to
improve the foundation for legal decision making and avoid the temp-
tation for psychiatrists and psychologists to answer without a scientific
foundation.
Key Points
■ Psychiatrists are expected by their patients and their profession
to make accurate violence assessments in their clinical practice.
■ Psychiatrists are expected by their patients and their profession
to know what is required of them to meet their obligations to
patients, those whom the patient may endanger, and society.
■ Psychiatrists will be held accountable by their patients, those
whom the patients harm, and society for harm that results from
the failure to engage in state-of-the-art, evidence-based assess-
ments of violence.
References
Barefoot v. Estelle, 403 U.S. 880 (1983)
Boynton v. Burglass, 590 So. 2d 446 (Fla. App 1991)
Cal. Civ. Code § 43.92(b) (West 2006)
Cal. Civ. Code § 43.92(b) (West 2007)
Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993)
16 Del. C. § 5402 (2007)
Evans v. United States, 883 F.Supp. 124 (S.D. Miss. 1995)
Fla. Stat. § 491.0147 (2006)
Frye v. United States, 293 F. 1013 (D.C. Cir. 1923)
General Electric v. Joiner, 522 U.S. 136 (1997)
Green v. Ross, 691 So. 2d 542 (Fla. App. 1997)
405 Ill. Comp. Stat. 5/3–807 (2007)
In re commitment of R.S., 801 A.2d 219 (N.J. 2004)
In re detention of Thorell, 72 P.3d 708 (Wash. 2000), cert. denied 541 U.S. 990 (2004)
Johnson v. Cockrell, 306 F.3d 249 (5th Cir. 2002)
Kumho Tire Co. v. Carmichael, 526 U.S. 137 (1999)
Miss. Code Ann. § 41–21–97 (2006)
Nenno v. State, 970 S.W.2d 549 (Tex. Crim. App. 1998)
N.J. Stat. § 2A:62A–16 (2007)
O’Connor v. Donaldson, 422 U.S. 563 (1975)
Rev. Code Wash. (ARCW) § 71.05.120 (2007)
R.I. Gen. Laws § 40.1–5–2(7)(4) (2007)
State for Interest of P.W., 801 S.W.2d 1 (Tex. App. 1990)
Tarasoff v. Regents of the University of California, 17 Cal. 3d 425, 131 Cal. Rptr. 14,
551 P.2d 334 (1976)
Tex. Code Crim. Proc. Art. 17.15(5) (2006)
Tex. Code Crim. Proc. Art. 7.071(2)(b)(1) (2006)
Texas Health and Safety Code § 611.004(a)(2) (2006)
Thapar v. Zezulka, 494 S.W. 2d 635 (Tex. 1999)
C H A P T E R 2 2
Sexual Violence and

the Clinician
Paul Fedoroff, M.D.
Sexual Violence: A Review

Sexual violence is a multidimensional concept that has various defini-
tions, depending on which professional group is using the term or the
context in which it is used. Violent sexual behavior, such as a sexually
motivated homicide or violent serial sexual offenses against women
and children, clearly fits the definition of sexual violence. “Any sexual
behavior against a nonconsenting partner” represents a broader defini-
tion of sexual violence and the one most commonly used. This broad
definition of sexual violence would cover most sexually deviant behav-
ior and most sexual offending behavior (Bradford 2006). It would in-
clude most of the “hands-on” paraphilias or sexual deviations (e.g., sex-
ual sadism, pedophilia) and would exclude most of the “hands-off”
paraphilias (e.g., voyeurism, fetishism, exhibitionism) (Bradford et al.
1992). Our review of sexual violence in this chapter follows this broad
definition. Sexual aggression is a term more commonly used by mental
health professionals and sexologists to cover the broad categorization
of sexual violence. It is also broadly broken down into sexual aggression
against women and sexual aggression against children, following the
common classification of sexual offenders and sexual crimes.
441
Sexual violence thus includes any sexual act with nonconsenting

partners. It also includes physical violence associated with a paraphilia
or sexual deviation. Sexual violence includes extreme physical violence
associated with coercive sexual activity. This type of violence usually in-
volves the sexual assault and murder of adults or children as part of sa-
distic, sexually motivated homicides. Sexually motivated homicides are
in turn classified as sadistic or nonsadistic (Gratzer and Bradford 1995).
Because acts of sexual violence with nonconsenting partners are against
the law, most of the perpetrators of these acts would also be classified as
sexual offenders. It is important to note that the converse is not neces-
sarily true and that not all sexual offenders have a paraphilia or sexual
deviation as defined in DSM-IV-TR (American Psychiatric Association
2000). It should also be noted that sex and violence are of universal in-
terest—as demonstrated by an Internet search using the search engine
Google in which the keyword “sex” generates more than 680 million
“hits” (Fedoroff, in press). A search of the major databases such as the
National Library of Medicine, also using the keyword “sex,” resulted in
more than 4,000 journal article citations in the past 10 years, and a search
on “sadomasochism” generated 148 journal articles (Fedoroff, in press).
Studies of sexual violence are increasing, and interest in the subject has
also been increasing in various professional groups, particularly in psy-
chology. Psychiatrists, including forensic psychiatrists, have sadly ne-
glected this important field even though they have many of the skill sets
required to work in this area. Expertise in both the psychological and bi-
ological bases to sexual violence as well as medical and pharmacologi-
cal skills for assessing and treating these individuals are critical. Most
psychiatrists have these skill sets and should be involved in research as
well as the assessment and treatment of sexual violence.
The consequences of sexual violence are very serious both for the
perpetrators and for the victims (Fedoroff 1996). In fact, about one-third
of men who molest children were sexually abused themselves and are
therefore victims as well as perpetrators (Fedoroff 1996). The relation-
ship between perpetrators and victims has led to the theoretical as-
sumption that abuse victims may become perpetrators themselves (Fe-
doroff 1996). If this theoretical assumption is true, population studies
on the incidence and prevalence of sexual abuse are extremely impor-
tant if preventive strategies are to be put in place to reduce victimiza-
tion. Various studies have shown that between 10% and 16% of males
and between 20% and 27% of females have been victims of childhood
sexual abuse (Finkelhor 1984; Finkelhor and Lewis 1988; Finkelhor et al.
1990). These data are critical to understanding not only the nature of pe-
dophilia but also the degree or level of risk within the population for the
Sexual Violence and the Clinician ❘ 443
sexual abuse of children. A Scandinavian study on population statistics

reported the incidence and prevalence of sexual abuse in Denmark
(Helweg-Larsen and Larsen 2005). These researchers found the average
annual incidence of child sexual abuse to be 0.06 per 1,000 children un-
der 15 years of age, using the National Patient Register as a database.
When other data sets were used, for example the National Criminal
Register, the incidence was found to be 0.5 per 1,000. When police re-
ports containing comprehensive information from victims were used,
the reported incidence of child sexual abuse was found to be 1.0 per
1,000 children younger than 12 years of age and 0.6 per 1,000 when ex-
hibitionism was excluded. About 50% of intrafamilial child sexual
abuse cases and about 40% of extrafamilial cases resulted in a convic-
tion (Helweg-Larsen and Larsen 2005). Translated into North American
terms, the incidence in the United States would be at least 300,000 chil-
dren sexually abused in any given year and 30,000 children in Canada.
Another study on childhood sexual abuse using computer-based self-
administered questionnaires submitted by a national representative
sample of 15- and 16-year-olds in Denmark resulted in close to 6,000 re-
spondents. Of these, 11% reported unlawful sexual experiences (7% of
boys and 16% of girls). Interestingly, these young people’s interpre-
tation of the experience was that only 1% of boys and 4% of girls felt
that they had been “definitely” or “maybe” sexually abused (Helweg-
Larsen and Boving Larsen 2006). This type of study is critical to under-
standing child sexual abuse. It also illustrates the complexity of the
sexual interaction and sexual experiences of young people in recent de-
cades. It should also focus attention on the sexual abuse of children,
which should be regarded as a public health problem.
In many Western countries, programs to prevent child sexual abuse
by educating children and sensitizing them to the issue have been put
in place. Various programs have been developed to educate children
about “good touch” and “bad touch.” In addition, Western countries
have instituted laws requiring the mandatory reporting of child sexual
abuse as tools to prevent victimization. The actual impact of these mea-
sures on the incidence and prevalence of child sexual abuse is still a
matter of debate.
In more recent years, the dramatic growth of the Internet has
brought a whole new dimension to child sexual abuse. The Internet has
become a new vehicle for the sexual abuse of children, requiring that
new strategies be put in place to deal with this threat. The Internet has
clearly become a focus of deviant sexual behavior, as demonstrated by
the dramatic increase in individuals charged with the possession of
child pornography. Very serious concerns have arisen about the Inter-
net being used as a tool by pedophiles to target children. This is a world-

wide phenomenon, and the concern has led to recent studies on how the
Internet is being used for sexual purposes and to classify Internet of-
fenders (Alexy et al. 2005). As the natures of perpetrators on the Internet
and of the interactions that are taking place become better understood,
the Internet has also become a focus for the protection of children
(Mitchell et al. 2001, 2005a, 2005b, 2005c). There clearly is significant un-
wanted exposure to sexual solicitation and harassment of young per-
sons (Mitchell et al. 2001). Young persons also experience considerable
unsolicited exposure to pornography on the Internet (Wolak et al. 2007).
The exact nature and the impact of these developments in the future is
impossible to determine at this time. Studies of Internet perpetrators
have shown they are quite frequently sexual offender recidivists and
also have engaged in other deviant sexual offenses (Seto and Eke 2005).
Individuals’ use of the Internet to lure children into meeting them in
order to seduce them is relatively infrequent but is of serious concern
when it does occur. These perpetrators are most likely long-standing
pedophilic sexual offenders using the Internet as a strategy to access
children for sexual purposes. Clearly, the Internet is a new frontier for
exposure to pornography by young persons as well as a vehicle for ex-
posure to sexual solicitation by pedophiles and other Internet offenders.
Case Example: The Internet Perpetrator

Mr. S is a 52-year-old married scientist with daughters who are teen-
agers. He was arrested at his place of work by local law enforcement of-
ficers as part of an FBI investigation of Internet child pornography. His
laptop computer at work had approximately 1,000 images of child por-
nography on its hard drive. Further investigation through a search war-
rant found 2,000 images of child pornography on his home computer,
including images backed up onto compact discs. The images were care-
fully classified according to age groups and were all of girls 13 years of
age or younger. A wide spectrum of ethnic groups was represented.
Most of the images depicted young girls in sexual poses without engag-
ing in sexual activity. About 10% of the images depicted sexual activity
with an adult male, mostly performing fellatio but also including sexual
intercourse in a small number of pictures. No videos were included in
this child pornography collection.
Mr. S was regarded as an excellent employee; he had no criminal
record, and an intensive investigation by the local child protection
agency produced no evidence of his having sexually abused his own
children. There was no evidence of any marital dysfunction, including
no history of sexual difficulties. Mr. S was referred to a specialized sex-
ual behaviors clinic for a pre-sentence forensic psychiatric evaluation.
This included the usual psychiatric history and diagnosis but also a sex
hormone profile, various sexual questionnaires (including the Bradford

Sexual History Inventory; Sexual Fantasy Checklist; Derogatis Sexual
Functioning Inventory; and various scales measuring pedophilic cogni-
tive distortions, hostility, and impulsivity, as well as the Michigan Alco-
hol Screening Test and the Drug Screening Test), along with objective
measures of sexual interest measured through penile plethysmography.
The assessment revealed that Mr. S had experienced sexual fantasies
of prepubertal children during adolescence and over the years had mas-
turbated to these fantasies but had never sexually acted out with chil-
dren. On the Sexual Fantasy Checklist he endorsed moderate ongoing
levels of heterosexual pedophilic fantasy. He showed mild pedophilic
cognitive distortions. Objective measures of sexual interest showed an
increased sexual preference toward prepubertal female children, with
a Pedophile Index of 2.4. This meant that the ratio of his pedophilic
arousal was 2.4 times higher than his measured arousal to consenting
sex with adult females. The sensitivity of this examination is 85% and
the specificity is 90%, meaning that based on a study of 100 clinical ad-
mitted pedophile subjects and 100 community control subjects, 85% of
the pedophiles and 90% of the community control subjects were cor-
rectly classified by using various measures of sexual arousal. There was
no evidence of any sadistic sexual preference. Mr. S immediately en-
tered into treatment for pedophilia. His formal DSM-IV-TR diagnosis
was pedophilia, opposite-sex children, nonexclusive type, mild. No
other Axis I diagnoses were present. There was no evidence of signifi-
cant personality disorder on Axis II. A Hare Psychopathy Checklist
score of 8 was obtained on risk assessment evaluation. Formal sexual of-
fender risk assessment using the Static-99 was not valid because he had
no previous convictions for sexual offenses. His risk of future sexual of-
fense recidivism was very low.
His treatment program consisted of psychological and pharmaco-
logical components. He was treated individually for the pedophilic cog-
nitive distortions and attended an ongoing relapse prevention group.
He was treated with selective serotonin reuptake inhibitors (SSRIs),
specifically sertraline, 150–200 mg/day, to suppress the pedophilic fan-
tasies. The pharmacological regimen was successful in eliminating the
pedophilic fantasies without affecting his sexual performance. Because
he was fired from his place of work, vocational counseling and rehabil-
itation through occupational therapy assisted him in finding alternative
employment. Marital and family therapy were successful in maintain-
ing his family unit. He received a suspended sentence and probation
with a condition of treatment for 3 years. At 5-year follow-up Mr. S
showed no evidence of sexual offense recidivism.
Sexual Sadism and Sexually Motivated Homicide

The operational criteria for sexual sadism in DSM-IV-TR are found un-
der the sexual and gender identity disorders and specifically under the
paraphilias. Questions have arisen as to the adequacy of the criteria and
how they have been applied. At least one study of experienced forensic
psychiatrists found that sexual sadism as a diagnosis was not being ap-
plied in Canadian correctional facilities as defined in DSM-IV-TR. The
kappa coefficient for reliability across diagnoses was only 0.14, which is
extremely poor (Marshall et al. 2002). Without reviewing all the details
of this particular study, it is clear that confusion about this very signifi-
cant diagnosis is evident. Confusion about the clinical aspects of sexual
violence occurs even among experienced forensic psychiatrists, in part
due to difficulties related to the “coercive paraphilic disorders.” There
was a recommendation by the Subcommittee on Sexual Disorders to in-
clude this group of paraphilias in the DSM-III-R diagnostic schema
(Abel 1989; American Psychiatric Association 1987). Although there was
strong empirical evidence for a diagnostic grouping of the “coercive
paraphilic disorders,” there was also considerable concern that the in-
clusion could lead to forensic psychiatric misuse of such a diagnostic
classification (Abel 1989). This debate has left a gap in the diagnostic
classification for males who meet the criteria for the paraphilias as well
as established clinical criteria for the paraphilias (e.g., typical natural
history of the paraphilias) and who are offenders against adult females.
Some of these men commit rape and show a sexual preference for rape
over consensual sex with adult females. The lack of an official diagnosis
has led to some of these men being diagnosed as having sexual sadism
and others as having paraphilia not otherwise specified. Neither of these
diagnostic classifications accommodates this type of paraphilic male
very well, leading to diagnostic confusion as well as forensic misuse in
sexually violent predator cases in the United States.
Clinical studies of sexually sadistic homicide perpetrators are rare,
mostly because this is a small group of individuals infrequently seen
even in forensic psychiatric settings (Swigert et al. 1976). Two studies
have looked at the offender and offense characteristics, including crime
scene behavior and other factors related to sexually sadistic homicide
(Dietz et al. 1990; Gratzer and Bradford 1995). Both of these studies had
to review a large number of homicide perpetrators and sexual offenders
before finding a relatively small number of males that met the diagnos-
tic criteria. Dietz et al. (1990) completed an uncontrolled descriptive
study of 30 sexually sadistic homicide or attempted homicide perpetra-
tors. Intentional torture for sexual arousal was the common characteris-
tic of this group of males. It was also clear that careful planning, includ-
ing stranger victim selection, most commonly occurred. Bondage,
beating, and homicide by strangulation were common offense charac-
teristics. Gratzer and Bradford (1995) compared the Dietz et al. sample
with another sample of sexually sadistic homicide and attempted homi-
cide perpetrators and a third sample of nonsadistic sexually motivated

homicide and attempted homicide perpetrators. There were many com-
mon characteristics in the sadistic homicide samples (Gratzer and Brad-
ford 1995). Previous studies of sexually sadistic homicide perpetrators
had shown that strangulation was the most common method used to
perpetrate the homicide (Dietz et al. 1990; Gratzer and Bradford 1995).
A well-controlled study from Finland looked at all homicides over a
7-year period in which strangulation was the cause of death (Hakkanen
2005). In contrast to previous studies, they found that there was no as-
sociation between sexually sadistic homicides and strangulation as the
cause of death. Although this does not mean that strangulation should
be disregarded as a signature of a sadistic sexually motivated homicide,
it is clear that further research is needed. The contrast between the
North American studies and the Finnish study may be a cultural varia-
tion, because the chosen method of homicide does vary across different
cultures. Although the North American studies point to a possible com-
mon psychopathology in sexually sadistic homicide perpetrators, there
may be considerable cultural differences in this group of perpetrators.
Various studies have shown some association between sexual sa-
dism and brain abnormality, commonly temporal lobe damage in the
brain (Hucker and Stermac 1992; Langevin et al. 1988). This includes a
study of sexually sadistic homicide perpetrators in which 50% of the
sample was noted to have some type of neurological abnormality
(Gratzer and Bradford 1995). These neuroanatomical abnormalities and
their association to deviant sexuality are also supported by studies on
brain activation in relation to sexual stimuli (Mouras et al. 2003; Red-
oute et al. 2000; Stoleru et al. 1999). More recently, a study by Briken et
al. (2006b) of sexually motivated homicide perpetrators found that
about one-third had obvious brain abnormalities. The sexually moti-
vated murderers with brain abnormalities differed significantly from
the group in having a higher incidence of early behavioral problems; a
higher total number of paraphilias; younger victims, particularly those
6 years of age or younger; and a higher incidence of transvestic fetish-
ism and paraphilias not otherwise specified (Briken et al. 2006b). The
study strongly emphasized the need for detailed neuropsychiatric eval-
uation of sexually motivated homicide perpetrators. In another well-
controlled study by Briken et al. (2006a) of male sexually motivated ho-
micide perpetrators, the prevalence of XYY syndrome was examined.
The prevalence was 1.8%, which is considerably higher than the preva-
lence in male offenders or in the general population (which would be
0.01%). In addition to the chromosome abnormality, these individuals
were also diagnosed as having sexual sadism and scored in the psycho-
pathic range on the Hare Psychopathy Checklist (Briken et al. 2006a).

The higher incidence of a chromosome abnormality, particularly one in-
volving the Y chromosome, possibly implicates androgens in the psy-
chopathology of sexually motivated homicide. This again emphasizes
the need for detailed neuropsychiatric and neurobiological evaluations
as part of the forensic psychiatric workup of sexually motivated homi-
cide perpetrators.
Although this type of homicide is a relatively rare event, the level of
media coverage and the degree of psychological trauma it causes in the
general public are significant, and anything that might allow us to have
a better understanding of this behavior would be extremely helpful. It
would also give us a stronger basis to understand and treat the condi-
tion early on in its development and therefore provide some degree of
secondary prevention and protection to the public. When sexually mo-
tivated homicides are perpetrated against children, the impact on the
general public is even more traumatic than if adults were involved. The
degree of shock, abhorrence, and horror that occurs in any community
if a child from that community is abducted, sexually assaulted, and
murdered is enormous. This has often led to the misconception that pe-
dophiles are responsible for the murder of children. To some extent this
impression is based on literary work by the Marquis de Sade in the eigh-
teenth century and factual accounts of Giles de Rais. Giles de Rais was
tried for the sexually motivated homicides of 40 young boys (Brown-
miller 1975; Sade et al. 1966). In actual fact, the most common perpetra-
tors of homicides against children are the caretakers of children, and
specifically the parents of children, as opposed to pedophiles. A well-
controlled study completed in England confirmed this finding (Dolan et
al. 2003). In this study, fathers or surrogate fathers were responsible in
nearly two-thirds of the cases. Children younger than 6 months of age
were the most vulnerable, and victim behavior and relationship difficul-
ties appear to be the precipitants in more than two-thirds of the cases. By
contrast, sexually motivated homicide occurred in 18.7% of the cases.
The impetus for “sexually motivated homicide” would appear to be
obvious; however, there has in fact been scientific debate and contro-
versy as to what motivates these individuals to commit their crimes.
This question arguably may apply more to serial sexual homicide, defined
as sexual homicide by perpetrators who kill three or more victims in a
noncontinuous fashion using a predatory form of violence (Meloy and
Felthous 2004). One hypothesis is that serial sexual homicide perpetra-
tors are motivated by a need to achieve power and control over their
victims and that sexual gratification is completely secondary. This
comes from a now-dated theory of motivation for rape involving power
and control (Groth et al. 1977). Crime scene examinations usually show
evidence of a sexual motivation. Myers and colleagues (2006) believe
that serial sexual murderers should be considered sexual offenders.
They also suggest the modification of DSM criteria to accommodate
this. They believe that these individuals mostly have paraphilic disor-
ders in the sadistic spectrum and that a diagnostic classification of “sex-
ual sadism, homicidal type” should be included as a subtype of sexual
sadism (Myers 2002; Myers et al. 2006).
If sexually sadistic homicide perpetrators do have a sexual motiva-
tion, then it should be measurable by physiological methods. Sexual
arousal can be measured by penile tumescence techniques known as
phallometry. There is evidence both for and against the capability of
this technique to discriminate between nonoffender and offender
populations or between different types of offenders (Marshall and
Fernandez 2000). Nonetheless, it is useful in differentiating groups of
men convicted of child molesting offenses, particularly extrafamilial
child sexual abuse, and in a meta-analysis on sexual offender recidi-
vism, phallometry was found to be one of the most reliable predictors
of recidivism for child molesters (Hanson and Bussiere 1998). There are
several ways to measure sexual arousal, although the one favored by
the lead author of this chapter (J.M.W.B.) is the use of indices reflecting
relative sexual arousal or sexual preference. A calculation is made
based on responses to audiotape descriptions of sex with children
where the arousal to children (the numerator) is divided by the arousal
to mutually consenting sex with adults (the denominator) and an index
is calculated by the ratio. This means an index greater than 1 means a
sexual preference in the direction of pedophilia. Indices can also be
used for sexual preference in a sadistic direction. In a study of 27 child
molesters who had committed or who had attempted a sexually moti-
vated homicide, 189 nonhomicidal child molesters, and 47 community
control subjects, there were clear differences between the three groups
based on phallometric measures (Firestone et al. 2000a). Significantly
more homicidal and nonhomicidal child molesters had pedophile indi-
ces greater than 1 compared with nonoffenders. The homicidal and
nonhomicidal child molesters did not differ from each other on pedo-
phile indices. This is not surprising, because both groups were pedo-
philes. However, when it came to assault indices, which measured
arousal to nonsexual violence (a measurement of sadistic sexual prefer-
ence), significantly more homicidal child molesters had assault indices
of 1 or greater or a sexual preference for nonsexual violence, compared
with the other two groups. The nonoffenders and nonhomicidal child
molesters did not differ from each other on this measure (Firestone et
al. 1999). This physiological evaluation was able to differentiate homi-

cidal child molesters from nonoffenders and nonhomicidal child mo-
lesters on the basis of sexual preference. This strongly supports a sexual
motivation to sexually motivated sadistic homicides.
The average psychiatrist is not likely to be asked to evaluate or treat
these extremely high-risk individuals. What may occur, however, in
general psychiatric practice is that individuals may present with homi-
cidal sexual fantasies against women or children. If this presentation
has been associated with any stalking of potential victims or even the
urge to do this, an extreme psychiatric emergency situation exists and a
referral to a forensic psychiatric subspecialist in sexology needs to occur
immediately. Involuntary civil commitment to a psychiatric facility to
protect the public is also indicated.
Recidivism
Understanding recidivism is fundamental to understanding sexual vio-
lence. There is a large body of scientific literature on sexual offender re-
cidivism that covers different types of sexual offenders, and there is con-
siderable consistency in the research findings even in studies from
different countries. Recidivism studies provide information to predict
subsequent risk of reoffense as well as information about treatment out-
come. There are well-established and significant differences in the recid-
ivism rates of different types of sexual offenders. The results of a task
force report from the American Psychiatric Association (1999) showed
that sexual assaulters of adult females (“rapists”) have the highest recid-
ivism rates, followed by extrafamilial child molesters; intrafamilial child
molesters (“incest” perpetrators) have the lowest rates of recidivism.
There is still evidence that recidivism rates are reflections of sexual of-
fenses that are underreported. In order to compensate for underreport-
ing, most recent studies include conviction rates and rearrest rates.
Arguably the most accurate reflection would be a combination of con-
viction rates, rearrest rates, and self-reported rates of sexual offenses.
Most recent studies involve a survival analysis technique. In general
terms, the longer the follow-up period, the higher the reported rates of
recidivism. In general, a follow-up period of at least 5 years while of-
fenders are at risk in the community is necessary for a valid study.
A number of meta-analyses have been completed on sexual offender
recidivism studies, the most significant being a study by Furby et al.
(1989), followed by two by Hanson (Hanson and Broom 2005; Hanson
and Thornton 1999) and a treatment outcome study meta-analysis by
Alexander (1999; Furby et al. 1989; Hanson and Broom 2005; Hanson
and Bussiere 1998). The Furby et al. (1989) study was generally re-
garded as a pessimistic one that was highly critical of the methodology
in existing recidivism studies, but at the same time it played an impor-
tant role in ensuring that future recidivism studies had improved meth-
odology. The Hanson and Bussiere (1998) meta-analysis comprised
more than 28,000 sexual offenders with a median follow-up period of
approximately 4 years and included 87 studies from six different coun-
tries. The meta-analysis documented the specific factors associated
with a high risk of sexual offense recidivism. The strongest predictors
of sexual offender recidivism were related to sexual deviance. Phallo-
metric measures of pedophilic sexual preference were highly correlated
with the risk of future sexual offense recidivism (r=0.32). Prior sexual
offenses (r= 0.19); age (r= 0.13); early onset of sexual offending (r= 0.12);
any prior offenses (r= 0.13); and never having been married (r= 0.11)
were also strongly correlated with sexual offender recidivism. The over-
all recidivism rate for sexual offenses was 13.4% (Hanson and Bussiere
1998). Alexander (1999) reviewed 79 treatment outcome studies includ-
ing more than 11,000 subjects. She showed that all the psychological
treatments included in the study resulted in lower recidivism rates
compared with an untreated group for both adults and adolescents. She
also showed that mandatory treatment appeared to have a positive ef-
fect on treatment outcome compared with voluntary treatment of sex-
ual offenders. Hanson and Broom (2005), in a follow-up meta-analysis,
used different analytical techniques to examine the trends in recidivism.
Sexual offender recidivism studies have been a focus of our own re-
search. We have completed recidivism studies of rapists, extrafamilial
child molesters, incest perpetrators, and exhibitionists (Firestone et al.
1998, 1999, 2000c; Greenberg et al. 2002). The mean follow-up period for
the studies was approximately 7 years, and all exceeded 5 years. The
studies included phallometric data and Hare Psychopathy Checklist
(PCL) scores as well as many other variables and used rearrest rates and
conviction rates as a measure of recidivism taken from the Canadian Po-
lice Information Computer, a national database of arrest and conviction
rates. Phallometric measures of deviant sexual preference and scores
were important predictors of sexual offense recidivism, and the PCL
scores were also strong predictors of violent and general recidivism. We
have completed other recidivism studies as well that looked more
closely at other factors related to recidivism, such as hostility (Firestone
et al. 2000b, 2005b, 2006; Greenberg et al. 2000). Incest perpetrators have
generally been regarded as a homogeneous group having the lowest risk
of recidivism for child molesters. Some differences in recidivism risk
based on victim age had been reported, and this was felt to be an impor-
tant factor for future risk assessment. A study of 119 incest perpetrators
consisted of a group of men (n = 48) who had victims younger than 6
years of age compared with another group of men (n=71) whose victims
were adolescents and more typical of incest perpetrators. Both groups
showed deviant sexual preference; however, the group with the young
victims had significantly more psychopathology, including substance
abuse (Firestone et al. 2005a). There are also important differences in the
recidivism risk for incest perpetrators whose victims are their biological
daughters compared with those whose victims are stepdaughters
(Greenberg et al. 2005). Deviant sexual preference was significantly
lower in men who molest their biological daughters, and this crime
therefore carries a lower risk of future sexual offense recidivism.
Sexual offense recidivism risk can be easily estimated by using the
Static-99, developed by the Research Division of the Solicitor General of
Canada (Hanson and Bussiere 1998). The ability to identify high-risk
sexual offenders provides the criminal justice system with a mechanism
to prevent further harm to the general public as well as a basis for sen-
tencing of these individuals or dealing with them through civil commit-
ment. These statistical instruments are considerable improvements
over the use of unstructured clinical judgment (Hanson and Bussiere
1998). The Static-99 is scored using easily obtained information with
limited training and yet provides significant accuracy in the prediction
of sexual offense recidivism (Hanson and Thornton 2000; Nunes et al.
2002). It also classifies sexual offender risk levels based on scoring of the
instrument as low, medium-low, medium-high, and high for sexual of-
fense recidivism based on the potential risk. It also gives the percentage
risk of future violent recidivism. This useful instrument can be utilized
by the forensic or general psychiatrist not trained in specialized forensic
sexology, thus providing a valid measure of future risk for sexual vio-
lence or violence in general that is considerably more accurate than un-
structured clinical evaluation.
Treatment
In recent years the focus of psychological treatments has been on the
cognitive-behavioral treatment spectrum. These treatments help the
person with a paraphilia block or reduce thoughts of or fantasies about
deviant behaviors such as child molesting. The treatments can be given
on an individual basis, but in most specialized treatment programs it is
performed in a group treatment setting. This not only improves the
cost-effectiveness of treatment but also has a therapeutic advantage be-
cause other individuals in the group with a similar paraphilia both
recognize and challenge cognitive distortions used by various members

in the group. Cognitive distortions are the rationalizations used by
paraphilic individuals, typically pedophiles, to justify their behavior.
Covert sensitization teaches the patient to imagine the negative social
consequences resulting from the deviant sexual urges to engage in
paraphilic behavior. Nearly all paraphilic behavior is preceded by devi-
ant thoughts or fantasies and urges. Individuals are taught to recognize
this cycle and to interrupt it. Olfactory aversion, desensitization, satia-
tion (masturbatory), and developing nonparaphilic interests are all
techniques that are used. Without exception, psychological treatments
are self-administered. Other techniques for training prosocial behaviors
include social skills training, anger management training, victim empa-
thy training, and relapse prevention. Relapse prevention in various
forms is the most commonly used psychological intervention. It is an
extensive framework of techniques for avoiding high-risk situations
and the risk of relapse. It is based on the assumption that sex offenders
make a decision to engage in treatment to stop committing deviant sex-
ual acts but find themselves in high-risk situations frequently involving
stress, interpersonal conflicts, or negative emotional states. Relapse pre-
vention helps the patient develop strategies to prevent this sequence of
events. It usually involves relapse rehearsal and a maintenance plan to
avoid high-risk situations (Abel et al. 1988; American Psychiatric Asso-
ciation 1999).
Pharmacological treatments consist of anti-androgens, hormonal
agents, and specific serotonin reuptake inhibitors. The pharmacological
approach to treatment is based on the suppression of deviant sexual
fantasies and urges and a reduction in sexual drive. In addition, there is
evidence that deviant sexual arousal patterns measured by phallometry
can also be suppressed (American Psychiatric Association 1999; Brad-
ford 2000; Bradford and Pawlak 1993a, 1993b). The anti-androgen used
in Canada is cyproterone acetate (not available in the United States). It
can be given orally or intramuscularly and is a powerful blocker of in-
tracellular androgen receptors throughout the body, including intra-
cerebral androgen receptors in the hypothalamus. Medroxyprogester-
one acetate has been widely used in the United States, and although not
a true anti-androgen, it does bring about significant reductions in sex-
ual drive by reducing circulating testosterone. Its mode of action is to
induce liver enzymes to clear plasma testosterone, and it also has an
antigonadotropic effect (Bradford 2000, 2001). More recently, luteiniz-
ing hormone–releasing hormone (LHRH) agonists such as leuprolide
acetate have been used as pharmacological castration agents (Bradford
2000, 2001). As well, some SSRIs are used to treat paraphilic behavior
by reducing sexual interest, sexual fantasies, and sexual drive. The

agents most commonly used are fluoxetine hydrochloride and sertraline
hydrochloride (Bradford 2001). Many of the studies of these pharmaco-
logical treatments are uncontrolled treatment outcome studies, al-
though some double-blind studies and recidivism studies have been
completed (Bradford 2000, 2001).
The lead author of this chapter has published an algorithm for the
pharmacological treatment of sexual deviation (Bradford 2000, 2001).
This algorithm first classifies the deviant sexual behavior into mild,
moderate, severe, or catastrophic, based on a modification of DSM-III-R
criteria. In addition, there is a six-level algorithm of treatment. Level 1
involves psychological treatments; level 2, the introduction of SSRIs;
level 3, the combination of an SSRI and a low dose of anti-androgen
treatment, either cyproterone acetate or medroxyprogesterone acetate;
level 4, full oral doses of cyproterone acetate or medroxyprogesterone
acetate; level 5, intramuscular cyproterone acetate or medroxyproges-
terone acetate; and level 6, pharmacological castration using intramus-
cular cyproterone acetate or an LHRH agonist such as leuprolide ace-
tate. Mild paraphilias would be treated at the stage one and two level;
moderate at stages two and three; severe at stages four and five; and cat-
astrophic at stage six (Bradford 2000, 2001). The introduction of SSRIs
in level 2 of treatment requires a dosage level of sertraline, 150–250 mg/
day, or fluoxetine, 40–60 mg/day. These SSRIs are the treatment of
choice because they suppress sexual drive without causing significant
problems of sexual dysfunction, particularly sexual performance. The
low-dose oral anti-androgen treatment used in association with SSRIs is
50–100 mg/day of either cyproterone acetate or medroxyprogesterone
acetate. Full oral anti-androgen treatment with either of these agents
would have an oral dosage range of 200–400 mg/day. Intramuscular
dosages of cyproterone acetate would be 100 mg every 2 weeks and of
medroxyprogesterone acetate would be 400 mg every 2–4 weeks. Phar-
macological castration would require leuprolide acetate, 7.6 mg intra-
muscularly, given monthly. Because the first 4–6 weeks of LHRH ago-
nist treatment actually increases testosterone levels by an outpouring of
LHRH from the hypothalamus, significant risk of deviant sexual acting
out occurs during this time frame. This risk must be managed with the
addition of an oral anti-androgen such as flutamide to cover this initial
period of increased risk. The longer-term management of anti-androgen
treatment requires ongoing sex hormone profile evaluations, partic-
ularly monitoring prolactin levels to avoid gynecomastia. In addition,
increased risk for osteopenia and osteoporosis with long-term treat-
ment needs to be monitored with yearly bone scans. Treatment with
vitamin D and calcium supplements may be helpful to offset this poten-

tial risk. For a complete review of pharmacological sexual offender
treatment, reference to review articles by the lead author of this chapter
is recommended (Bradford 1998, 2000).
Sex is a basic biological drive, and there is considerable scientific
information both from animal and human research about the neuro-
biological aspects of sexual behavior. Furthermore, there is considerable
research showing that the actions of various pharmacological agents on
hormones and neurotransmitters affect sexual behavior. Physicians,
and particularly psychiatrists, are in a strong position by virtue of their
training to be involved in the assessment and treatment of sexual devi-
ation. It is unfortunate that mainstream psychiatry, including medical
school training, has neglected this important area of psychiatric and
medical treatment. If a major impact is to be made on child sexual
abuse, psychiatrists at all levels should be more engaged in the assess-
ment and treatment of sexual deviation. Psychology and other mental
health disciplines are already strongly engaged in the assessment and
treatment of sexual offenders and sexual deviation.
Key Points
■ Sexual violence involves any sexual act with a nonconsenting
partner and also includes physical violence associated with a
paraphilia or sexual deviation.
■ Various studies have shown that between 10% and 16% of males
and between 20% and 27% of females have been victims of
childhood sexual abuse.
■ In recent years the Internet has become a vehicle for sexual
offenses, usually child pornography. Studies have defined
different types of Internet offenders.
■ Clinical studies of sexually sadistic homicide perpetrators are rare,
mostly because this condition only rarely occurs. Clinical features
have been defined and an association has been found between
sexual sadism and brain abnormalities.
■ Sexual arousal has been shown to be one of the most reliable
predictors of sexual offense recidivism. It has also been shown
to discriminate between sexual offenders and nonoffenders,
particularly in relation to pedophilia, and to discriminate
between pedophilic homicidal perpetrators and nonhomicidal
perpetrators.
■ There is a large body of scientific evidence on sexual offender

recidivism that shows considerable consistency even in studies
conducted in different countries. Those who sexually assault
adult females have the highest recidivism rates, followed by
extrafamilial child molesters and lastly by intrafamilial child
molesters.
■ The Static-99 provides an easy way of estimating sexual offense
recidivism with easily obtainable information.
■ The focus of psychological treatments has been the cognitive-
behavioral treatment spectrum, with a specific focus on relapse
prevention.
■ Pharmacological treatments include SSRIs, anti-androgens, and
LHRH agonists as well as some other hormonal agents.
■ An algorithm for the pharmacological treatment of sexual
offenders has been developed.
References
Abel GG: Paraphilias, in Comprehensive Textbook of Psychiatry, 5th Edition.
Edited by Kaplan HI, Sadock BJ. Baltimore, MD, Williams & Wilkins, 1989,
pp 1069–1085
Abel GG, Mittelman M, Becker JV, et al: Predicting child molesters’ response to
treatment. Ann N Y Acad Sci 528:223–234, 1988
Alexander MA: Sexual offender treatment efficacy revisited. Sex Abuse 11:101–
116, 1999
Alexy EM, Burgess AW, Baker T: Internet offenders: traders, travelers, and com-
bination trader-travelers. J Interpers Violence 20:804–812, 2005
Disorders, 3rd Edition, Revised. Washington, DC, American Psychiatric
Association, 1987
American Psychiatric Association: Dangerous Sex Offenders: A Task Force Re-
port of the American Psychiatric Association. Washington, DC, American
Psychiatric Association, 1999
Bradford JM: Treatment of men with paraphilia. N Engl J Med 338:464–465, 1998
Bradford JM: The treatment of sexual deviation using a pharmacological ap-
proach. J Sex Res 37:248–257, 2000
Bradford JM: The neurobiology, neuropharmacology, and pharmacological
treatment of the paraphilias and compulsive sexual behaviour. Can J Psy-
chiatry 46:26–34, 2001
Bradford JM: On sexual violence. Curr Opin Psychiatry 19:527–532, 2006
Bradford JM, Pawlak A: Double-blind placebo crossover study of cyproterone

acetate in the treatment of the paraphilias. Arch Sex Behav 22:383–402,
1993a
Bradford JM, Pawlak A: Effects of cyproterone acetate on sexual arousal pat-
terns of pedophiles. Arch Sex Behav 22:629–641, 1993b
Bradford JM, Boulet J, Pawlak A: The paraphilias: a multiplicity of deviant be-
haviours. Can J Psychiatry 37:104–108, 1992
Briken P, Habermann N, Berner W, et al: XYY chromosome abnormality in sex-
ual homicide perpetrators. Am J Med Genet B Neuropsychiatr Genet
141:198–200, 2006a
Briken P, Habermann N, Kafka MP, et al: The paraphilia-related disorders: an
investigation of the relevance of the concept in sexual murderers. J Forensic
Sci 51:683–688, 2006b
Brownmiller S: Against Our Will: Men, Women and Rape. New York, Simon
and Schuster, 1975
Dietz PE, Hazelwood RR, Warren J: The sexually sadistic criminal and his of-
fenses. Bull Am Acad Psychiatry Law 18:163–178, 1990
Dolan M, Guly O, Woods P, et al: Child homicide. Med Sci Law 43:153–169, 2003
Fedoroff J: The genesis of pedophilia: testing the “abuse to abuser” hypothesis.
The Journal of Offender Rehabilitation 24:85–101, 1996
Fedoroff J: Sadism, sadomasochism, sex and violence. Can J Psychiatry (in press)
Finkelhor D: How widespread is child sexual abuse? Child Today 13:18–20,
1984
Finkelhor D, Lewis IA: An epidemiologic approach to the study of child moles-
tation. Ann N Y Acad Sci 528:64–78, 1988
Finkelhor D, Hotaling G, Lewis IA, et al: Sexual abuse in a national survey of
adult men and women: prevalence, characteristics, and risk factors. Child
Abuse Negl 14:19–28, 1990
Firestone P, Bradford JM, McCoy M, et al: Recidivism in convicted rapists. J Am
Firestone P, Bradford JM, McCoy M, et al: Prediction of recidivism in incest of-
fenders. J Interpers Violence 14:511–531, 1999
Firestone P, Bradford JM, Greenberg DM, et al: Differentiation of homicidal
child molesters, nonhomicidal child molesters, and nonoffenders by phal-
lometry. Am J Psychiatry 157:1847–1850, 2000a
Firestone P, Bradford JM, Greenberg DM, et al: The relationship of deviant sex-
ual arousal and psychopathy in incest offenders, extrafamilial child molest-
ers, and rapists. J Am Acad Psychiatry Law 28:303–308, 2000b
Firestone P, Bradford JM, McCoy M, et al: Prediction of recidivism in extrafamil-
ial child molesters based on court-related assessments. Sex Abuse 12:203–
221, 2000c
Firestone P, Dixon KL, Nunes KL, et al: A comparison of incest offenders based
on victim age. J Am Acad Psychiatry Law 33:223–232, 2005a
Firestone P, Nunes KL, Moulden H, et al: Hostility and recidivism in sexual of-
fenders. Arch Sex Behav 34:277–283, 2005b
Firestone P, Kingston DA, Wexler A, et al: Long-term follow-up of exhibition-
ists: psychological, phallometric, and offense characteristics. J Am Acad
Psychiatry Law 34:349–359, 2006
Furby L, Weinrott MR, Blackshaw L: Sex offender recidivism: a review. Psychol

Bull 105:3–30, 1989
Gratzer T, Bradford JM: Offender and offense characteristics of sexual sadists: a
comparative study. J Forensic Sci 40:450–455, 1995
Greenberg D, Bradford J, Firestone P, et al: Recidivism of child molesters: a
study of victim relationship with the perpetrator. Child Abuse Negl
24:1485–1494, 2000
Greenberg D, Firestone P, Nunes KL, et al: Biological fathers and stepfathers
who molest their daughters: psychological, phallometric, and criminal fea-
tures. Sex Abuse 17:39–46, 2005
Greenberg SR, Firestone P, Bradford JM, et al: Prediction of recidivism in exhi-
bitionists: psychological, phallometric, and offense factors. Sex Abuse
14:329–347, 2002
Groth AN, Burgess W, Holmstrom LL: Rape: power, anger, and sexuality. Am J
Psychiatry 134:1239–1243, 1977
Hakkanen H: Homicide by ligature strangulation in Finland: offence and of-
fender characteristics. Forensic Sci Int 152:61–64, 2005
Hanson RK, Broom I: The utility of cumulative meta-analysis: application to
programs for reducing sexual violence. Sex Abuse 17:357–373, 2005
Hanson R, Bussiere MT: Predicting relapse: a meta-analysis of sexual offender
recidivism studies. J Consult Clin Psychol 66:348–362, 1998
Hanson RK, Thornton D: Improving risk assessments for sex offenders: a com-
parison of three actuarial scales. Law Hum Behav 24:119–136, 2000
Helweg-Larsen K, Boving Larsen H: The prevalence of unwanted and unlawful
sexual experiences reported by Danish adolescents: results from a national
youth survey in 2002. Acta Paediatr 95:1270–1276, 2006
Helweg-Larsen K, Larsen HB: A critical review of available data on sexual
abuse of children in Denmark. Child Abuse Negl 29:715–724, 2005
Hucker SJ, Stermac L: The evaluation and treatment of sexual violence, necro-
philia, and asphyxiophilia. Psychiatr Clin North Am 15:703–719, 1992
Langevin R, Bain J, Wortzman G, et al: Sexual sadism: brain, blood, and behav-
ior. Ann N Y Acad Sci 528:163–171, 1988
Marshall WL, Fernandez YM: Phallometric testing with sexual offenders: limits
to its value. Clin Psychol Rev 20:807–822, 2000
Marshall WL, Kennedy P, Yates P: Issues concerning the reliability and validity
of the diagnosis of sexual sadism applied in prison settings. Sex Abuse
14:301–311, 2002
Meloy J, Felthous A: Introduction to this issue: serial and mass murder. Behav
Sci Law 22:289–290, 2004
Mitchell KJ, Finkelhor D, Wolak J: Risk factors for and impact of online sexual
solicitation of youth. JAMA 285:3011–3014, 2001
Mitchell KJ, Finkelhor D, Wolak J: The Internet and family and acquaintance
sexual abuse. Child Maltreat 10:49–60, 2005a
Mitchell KJ, Finkelhor D, Wolak J: Protecting youth online: family use of filter-
ing and blocking software. Child Abuse Negl 29:753–765, 2005b
Mitchell KJ, Wolak J, Finkelhor D: Police posing as juveniles online to catch sex
offenders: is it working? Sex Abuse 17:241–267, 2005c
Mouras H, Stoleru S, Bittoun J, et al: Brain processing of visual sexual stimuli in

healthy men: a functional magnetic resonance imaging study. Neuroimage
20:855–869, 2003
Myers WC: Juvenile Sexual Homicide. San Diego, CA, Academic Press, 2002
Myers WC, Husted DS, Safarik ME, et al: The motivation behind serial sexual
homicide: is it sex, power, and control, or anger? J Forensic Sci 51:900–907,
2006
Nunes KL, Firestone P, Bradford JM, et al: A comparison of modified versions
of the Static-99 and the Sex Offender Risk Appraisal Guide. Sex Abuse
14:253–269, 2002
Redoute J, Stoleru S, Gregoire MC, et al: Brain processing of visual sexual stim-
uli in human males. Hum Brain Mapp 11:162–177, 2000
Sade DAF, Wainhouse A, Seaver R: The Marquis de Sade: The 120 Days of So-
dom and Other Writings. New York, Grove Press, 1966
Seto MC, Eke AW: The criminal histories and later offending of child pornogra-
phy offenders. Sex Abuse 17:201–210, 2005
Stoleru S, Gregoire MC, Gerard D, et al: Neuroanatomical correlates of visually
evoked sexual arousal in human males. Arch Sex Behav 28:1–21, 1999
Swigert VL, Farrell RA, Yoels WC: Sexual homicide: social, psychological, and
legal aspects. Arch Sex Behav 5:391–401, 1976
Wolak J, Mitchell K, Finkelhor D: Unwanted and wanted exposure to online
pornography in a national sample of youth Internet users. Pediatrics
119:247–257, 2007
C H A P T E R 2 3
Violence Toward Mental

Health Professionals
Autumn Ning, M.D.
T he idea that a clinician can be the target of violence by a patient is an

inconceivable one to most caregivers. Yet clinicians are at significant
risk for being assaulted. According to the Department of Justice’s Crime
Victimization Survey for 1993–1999 (Duhart 2001), the annual rate for
nonfatal violent crime (rape and sexual assault, robbery, and aggra-
vated and simple assault) for all occupations was 12.6 per 1,000 work-
ers. For physicians, the rate was 16.2, and for nurses, it was 21.9. How-
ever, for psychiatrists, the rate was 68.2 per 1,000; for mental health
custodial staff, it was 69.0; and for other mental health workers, it was
40.7. Of psychiatrists who have responded to surveys, the rate of assault
ranges from 3% to 40%, with an average of 40% (American Psychiatric
Association 1993; Davies 2001; O’Sullivan and Meagher 1998). Among
psychiatric residents, the percentage of respondents reporting being
assaulted ranges from 19% to 64% (American Psychiatric Association
1993; Black et al. 1994; Coverdale et al. 2001; Schwartz and Park 1999),
with a high rate of repeat assaults that ranges from 10% to 31% (Chaim-
owitz and Moscovitch 1991; Fink et al. 1991; Milstein 1987).
Nonpsychiatric residents also experience a high rate of assaults.
Twenty percent of surgical residents have reported being assaulted
(Barlow and Rizzo 1997); other studies report that 16%–40% of internal
461
medicine residents who responded to surveys were assaulted (Cover-

dale et al. 2001; Milstein 1987; Paola et al. 1994; VanIneveld et al. 1996).
The emergency department may be the most dangerous place to
work in a hospital. Emergency departments are open to the public and
are accessed by an unscreened patient population. Police bring poten-
tially violent patients to the emergency department on a daily basis, and
the number of drug-abusing patients who present has also increased.
Studies have documented the alarming incidence of weapons brought
into the emergency department (Goetz et al. 1991; Thompson et al. 1988)
and the high incidence of violence against staff (Foust and Rhee 1993;
Jenkins et al. 1998; Lavoie et al. 1988; Pane et al. 1991; Wyatt and Watt
1995). Of psychiatric patients seen in psychiatric emergency services or
an emergency department, 4%–17% have been reported to have weap-
ons (Anderson et al. 1989; Goetz et al. 1991; McNeil and Binder 1987).
Statistics regarding the homicide rate within the medical profession
are not available from either the American Medical Association or the
American Psychiatric Association. There are no aggregate studies ad-
dressing homicide against physicians. Data of such events are either in
case or media reports (American Psychiatric Association 1993; Dubin
and Lion 1996; Ladds and Lion 1996). Fatal assaults occur in a range of
settings, including private outpatient offices, community mental health
centers, academic centers, hospitals, and even military installations,
perpetrated by patients with a variety of psychiatric diagnoses. Homi-
cides have been committed by patients using guns, knives, blunt instru-
ments, and physical assault.
Case Examples
Case Example 1:
Violence in the Psychiatric Emergency Service
Mr. A, a 30-year-old man, was brought to the psychiatric emergency ser-
vice by police at his own request after he told them he was depressed
and suicidal and wanted to go to the hospital. He was using $200 worth
of cocaine per day and reported symptoms of major depression. He
stated, “I am hopeless and feel like I might hurt myself or others.” Mr.
A had presented a month earlier with a similar complaint and had been
referred to outpatient treatment. The managed care company reported
the patient had a long history of drug and psychiatric treatment related
to his drug use. He had never followed up with outpatient treatment,
even with a case manager.
In the psychiatric emergency service the patient was labile and eas-
ily agitated and could only with effort be redirected. He was tempo-
rarily placed in an open seclusion room in view of the nurses’ station
Violence Toward Mental Health Professionals ❘ 463
while waiting to be seen by the psychiatric resident. He was observed to

pace in the room. When the resident approached the patient, the patient
became calmer and cooperative. He denied the history reported by the
managed care company, stating he did not know what they were talking
about, that he had not had treatment recently and had not sought help
in a year. The physician confronted the patient with the fact that he had
a record of the patient presenting in the past month. The patient became
agitated and shouted, “Who the f*** cares? I’m suicidal NOW.” The phy-
sician began to explain to Mr. A in a quiet tone that there was a note in
the patient’s records from the insurance company regarding the many
presentations he had had. The patient then kicked the physician twice
in the abdomen before staff was able to restrain him.
This patient had a known history of drug use, was labile, easily agi-
tated, and was seen pacing in the seclusion room. The resident chose to
confront the patient and, in essence, called the patient a liar, violating
the dynamic of violence by humiliating this already labile patient.
Rather than focusing on affect management, the resident chose to give
a rational explanation to the patient about the documented history of
drug treatment and his history of noncompliance. The resident instead
should have addressed the affect, considered medication, and, most im-
portantly, had additional staff present with him during the interview.
Case Example 2:
Failure of a Risk Assessment in an Outpatient Office
Mr. B, in his early 30s, held his psychiatrist hostage for 90 minutes,
threatening to beat him up and to destroy the property in the psychia-
trist’s private home office. When the patient had initially called for an
appointment, the psychiatrist learned that Mr. B had a history of violent
episodes and paranoid responses to psychiatrists. Mr. B stated on the
phone that he was an impossible case but that his initial response to this
psychiatrist was positive. During the first four treatment sessions, Mr. B
had continually pressed the psychiatrist to demonstrate an interest in
him. After the fourth session he had called the psychiatrist at 11 P.M. and
asked to meet with him to discuss a crisis. The psychiatrist responded
that he could not meet him and that he would see him at the next sched-
uled time. At the next appointment Mr. B walked into the office, locked
the door, and began his threatening behavior. Mr. B screamed and ver-
bally abused the psychiatrist for 90 minutes, pushing him around and
knocking diplomas off the wall. Although angry and fearful, the psychi-
atrist talked to the patient in a calming manner. The psychiatrist did not
challenge Mr. B but calmly pointed out possible repercussions. The pa-
tient finally left the office and never returned or contacted the psychia-
trist. The psychiatrist acknowledged his own sense of grandiosity and
vanity had been enhanced when the patient made positive comments
about him on the phone and that this further strengthened his denial of
the risk of aggression that this patient posed.
This patient had a risk profile that suggested he was at risk for vio-
lence—that is, a history of violence and conflicts with previous psychia-
trists. The psychological defense of denial resulted in the psychiatrist’s
minimizing the risk and treating the patient in his home office. The psy-
chiatrist should have either treated this patient in a more secure setting or
declined to take this patient into treatment at the initial phone interview.
Case Example 3: Threats of Violence Toward a Surgeon

Mr. C was a 30-year-old man with no formal psychiatric diagnosis who
was seeing a resident plastic surgeon in the clinic to be evaluated for a rhi-
noplasty. He had a history of stalking behavior and threatening with a
gun, which was not known until after the incident. After multiple elective
rhinoplasties, the patient became violent in the clinic and was removed
and told never to return. The resident surgeon on the patient’s first rhino-
plasty completed his residency and moved to another state, where the pa-
tient found him through the unwitting participation of the residency de-
partment. Mr. C began to write letters to the surgeon, calling her a
“butcher” and stating “you won’t be able to operate again.” She notified
the police, the FBI, and the postal authorities, who told the patient that it
was unlawful to write threatening letters. The patient continued to write
threats on the outside of envelopes. The doctor was informed that be-
cause these weren’t actual letters, there was no way to press charges. The
threats continued, but at a lessened frequency, and eventually ceased.
Unknown in this case is the degree to which the department physi-

cians and staff were attuned to the risks attendant in the specialty of
plastic surgery and whether they had a methodology for psychologi-
cally assessing patients and risks. Also unknown is how they handled
requests by patients regarding information about residents who gradu-
ated. The physician in this case acted quickly and decisively and may
have deterred more aggressive action by the patient by enlisting law en-
forcement officers to speak with the patient.
Case Example 4: Violence as a Result of Psychotic

Transference in an Outpatient Setting
Ms. D, a woman in her late 30s, walked into her psychiatrist’s private of-
fice in his home and pulled out a knife. According to the psychiatrist,
Ms. D was having a positive transference reaction and believed that the
psychiatrist was the object of her desires, which she could not control.
The psychiatrist saw the threat as her way of destroying him in order to
free herself from this predicament. The psychiatrist noted that he
had compounded the problem: because Ms. D was a physically small
woman, he had failed to interpret a previous acting-out episode in
which she had thrown an ashtray at him. He had never told her that
such behavior was dangerous and unacceptable.
The psychiatrist responded to Ms. D in a calm, clear voice, saying

that it was not acceptable to threaten him with the knife, that she did not
have to do this to relieve her pain, and that he would not hurt her. As he
talked, he grabbed her wrist and bent it forward, and she released the
knife. Changes in treatment included active interpretation of the pa-
tient’s fears of closeness, as well as stricter limit setting. The psychia-
trist’s belief that he could overpower the patient led to his denial of the
discomforting anxiety one usually feels when a psychotic transference
develops in a therapeutic relationship.
Because of the patient’s size and possibly her gender, this psychia-
trist used denial and minimized the risk by failing to set limits with the
patient after the first aggressive incident. After that first episode the
psychiatrist also should have reevaluated the safety of treating this pa-
tient in his home office. Grabbing the patient’s wrist to take away the
knife was a questionable strategy and risked possible injury to the psy-
chiatrist or the patient. It would have been preferable to continue the
talk-down strategy.
Facilitating Clinician Safety

Risk Assessment
The most effective strategy for enhancing clinician safety is to anticipate
potential aggression. Tardiff (1996) suggested certain clinical, psycho-
logical, and historical variables that increase a patient’s potential for vi-
olence (Table 23–1). Although there is no specific combination or num-
ber of these risk factors that can predict violence, their presence alerts
the clinician that the patient poses a risk. The clinician who is aware of
these risk factors has the opportunity to develop treatment strategies to
minimize the potential for violence. Several important risk factors for
violence are outlined in Table 23–1.
For psychiatrists with offices in their homes or in office buildings, all
new patients should have at least a 15- to 20-minute interview on the
phone that includes a risk assessment for violence. A risk assessment
evaluation should include intent to harm self or others, possession or ac-
cess to a weapon, recent violence, formulation of a definitive plan of vi-
olence, drug and alcohol use, adherence with aftercare and medication
management, and associated psychiatric or medical conditions (Petit
2005). Other components of a risk assessment include present illness,
past psychiatric history, military history, legal history, and a mental sta-
tus examination (Buckley et al. 2003). Patients with a history of violence
or paranoia, or who have borderline personality disorder with little im-
pulse control, should not be interviewed initially in a private office (Berg
TABLE 23–1. Risk factors for violence

• A past history of repetitive violence
• Agitation, anger, disorganized behavior
• Poor compliance during the interview
• A detailed or planned threat of violence
• Available means for inflicting injury, such as ownership of a weapon
• History of childhood physical or sexual abuse
• Presence of organic disorder
• Presence of psychotic psychopathology, especially delusions or command
hallucinations
• Presence of borderline or antisocial personality disorder
• Presence of alcohol or drug use
• Belonging to a demographic group with an increased prevalence of
violence: young, male, lower socioeconomic group
Source. Adapted from Tardiff 1996.
2000). If possible, such patients should be interviewed the first time in a

more secure setting such as an outpatient department, a crisis service, or
an emergency department. For unknown patients who are making their
first visit, appointments should be scheduled for the middle of the day
when numerous staff are present as opposed to early morning or late
evening. Sessions may be scheduled so that other staff is immediately
available in the vicinity of the office or in the interview room during the
appointment. For high-risk patients in hospital settings, the clinician
may choose to leave the office door open during the interview, with staff
or security present, or interview the patient in a conference room where
a large table can provide a barrier between the clinician and patient.
A major impediment to the effective management and treatment of
a violent patient is the psychological defense of denial. Denial allays the
clinician’s anxiety by disavowing thoughts, feelings, or external reality
factors that are consciously intolerable. Therefore, clinicians ignore clin-
ical data or behavior that suggests a patient may become violent. For ex-
ample, rather than acknowledging anxiety and fear, the psychiatrist
may project an image of false machismo, fearlessness, and confidence.
Other manifestations of denial are failure to obtain pertinent data re-
garding a patient’s previous history of violence or arrests and failure to
question a patient about current aggressive behavior. A risk assessment
begins to neutralize the potent psychological defense of denial in the
clinician. Anticipation of violence leads to preventive treatment plan-
ning and can significantly enhance safety.
Dynamics of Violence
The successful management of violence is predicated on an understand-
ing of its dynamics. Violence is a reaction to feelings of passivity and
helplessness. A patient’s threatening behavior is commonly an overre-
action to feelings of impotence, helplessness, and perceived or actual
feelings of humiliation. A clinician who encounters a threatening pa-
tient should avoid becoming verbally or physically aggressive toward
the patient. Psychiatrists who respond to threatening patients with
physical or verbal aggression are significantly more likely to be injured
or have property destroyed than those who acknowledge their fear but
also express a desire to help the patient (Dubin et al. 1988). The strategy
of a nonthreatening offer to provide help is reassuring to the patient and
is the centerpiece for intervention with a potentially violent patient.
The Prodrome of Violence

Aggression rarely occurs suddenly and unexpectedly. Generally there is
a prodromal syndrome consisting of increasing tension and anxiety, es-
calating verbal stridency and abuse, and increasing motor activity usu-
ally characterized by pacing behavior. Intervention using talk-down
strategies during this period of escalation will frequently avert violent
behavior. In such an escalating situation, the clinician must be sure that
the patient can hear and respond. A patient who is under the influence
of alcohol or drugs is not a good candidate for talk-down techniques. By
using a soft, assertive voice and short sentences, the clinician can rap-
idly determine if the patient if paying attention (Maier 1996). Volume,
tone, and rate of speech should be lower than the patient’s, although if
too low, the patient may perceive it as a threat (Berg et al. 2000). The cli-
nician should talk down a patient by agreeing with him or her and not
arguing (Maier 1996). It is important not to respond to the content of the
patient’s speech. The patient should be “overdosed” with agreement
(Maier 1996). An escalating patient should be approached from the
front or side, because an approach from behind is extremely threaten-
ing. The clinician should also never turn his or her back to the agitated
or threatening patient (Berg et al. 2000).
Affect Management
The main strategy for de-escalating a potentially violent patient is affect
management. Patients who are affectively aroused will need to ventilate
their history, and the clinician should not overly intrude into the in-
terview (Eichelman 1995). Often the patient who is overwhelmed with
angry affect intimidates the clinician, who then responds with logical and
rational explanations. This type of intervention only inflames the patient.
Affect management involves acknowledging the patient’s affect, validat-
ing the affect when appropriate, and encouraging the patient to talk about
his or her feelings. For instance, the clinician might say, “I can see how an-
gry this makes you. If I were given medication against my will I would be
as angry as you are. Let’s talk more about your feelings.” Phrases such as
“ventilate,” “talking it out,” “getting it off your chest,” or “catharsis” are
colloquialisms that refer to the process of allowing a person to discharge
his or her affect. Addressing the affect serves several purposes. It teaches
the patient to reduce his or her internal state of tension by verbalizing feel-
ings and teaches that it is not necessary to hit someone or destroy furni-
ture to feel better. Giving the patient the opportunity to ventilate affect of-
ten defuses an escalating patient and averts a more violent confrontation.
Additional Management Techniques

Emotionally distraught patients require an active response from a clini-
cian. Active eye contact and body language that signal attentiveness
and connectedness to the patient will reduce the probability that the pa-
tient will need to explode or assault to get his or her point across (Eich-
elman 1995). However, prolonged or intense direct eye contact can be
perceived as menacing by the patient (Petit 2005). Eichelman (1995) de-
scribed interventions that are effective in aggression management. The
use of active listening techniques, such as paraphrasing to the patient in
brief, encapsulated form the content of his or her statements, helps to
convey that the clinician understands what the patient is experiencing.
It is important to be honest and precise when responding to patients.
Dishonesty may set the clinician up for either retribution or a tenuous
therapeutic relationship. Eichelman (1995) further recommended that
in all situations the clinician keep a proper physical distance from the
patient. Assaultive patients have a larger body buffer zone, and a rule
of thumb is to keep two quick steps or at least an arm’s distance from
the patient. A personal space can be visualized as an oval zone extend-
ing 4–6 feet all around (Berg et al. 2000). If the patient is standing, the
clinician should stand. If the patient is sitting, the clinician should also
sit down and not stand over the patient during the interview. If the pa-
tient is pacing, the clinician can model for the patient by walking with
the patient but at a much slower pace. Berg et al. (2000) recommended
that the clinician take a posture that makes him or her appear small and
thus less threatening. This can be done by holding the hands at waist
level with palms up and open or assuming the Thinker stance (one fore-
arm crosses the chest, the opposite elbow rests on it with the index fin-
ger touching the cheek or chin).
Limit Setting
At times clinicians react to escalating or agitated behavior with punitive
threats, in an attempt to set limits. A threatening intervention, however,
is contrary to the dynamics of violence because it evokes feeling of im-
potence or humiliation in the patient and increases the risk of violence.
Overt anger or hostility should never be expressed toward an agitated
patient (Petit 2005). Limit setting can be therapeutic and avert violent
behavior. Green et al. (1988) described the basic philosophy behind
limit setting, which is to contain and counteract maladaptive behavior
that interferes with therapy and threatens the safety of the clinician.
Green et al. (1988) further note that effective limit setting involves clear
identification of the specific behaviors that need to be altered and pre-
cise articulation of the consequences that will follow if the inappropri-
ate behavior persists. If the therapist lacks clarity in his or her thinking
or communications to the patient concerning inappropriate behavior,
the intervention may confuse and disorganize the patient. Whenever
possible, interpretive interventions should precede the imposition of
limits, because this affords the patient greater flexibility in exercising
his or her own autonomy and discretion.
Successful limit setting is most effective when this sequence is fol-
lowed (Green et al. 1988):
1. The patient is told of the behavior that is unacceptable.

2. The patient is told why the behavior is unacceptable.
3. The patient is offered several alternative treatment interventions.
For example, a clinician can say to a patient, “You cannot yell, curse,
or threaten other patients in the day room. They are afraid of you, and
they think that you will harm them. Therefore, you can go to your room
and listen to the radio until you feel calmer, or we can walk to the se-
clusion room and I will give you some medication.” Given several op-
tions, the patient will usually accept whichever is preferred. If given a
choice, the patient will pause to consider the options, and each pause
decreases the amount of energy behind the anger. As this process con-
tinues, the patient will slowly regain self-control (Maier 1996). Offering
only one option invites the patient to argue and negotiate, which leads
to further escalation and frequently culminates in an assault against the
clinician or restraint of the patient.
Thackrey (1987) described several important clinical caveats regard-

ing limit setting. Alternatives regarding both expected and prohibited
behaviors must be stated concretely and in terms of actions that can be
performed immediately. Whenever possible, directives or alternatives
should be expressed in positive terms (“Do this,” which describes ac-
ceptable behavior) rather than negative terms (“Don’t do that,” which
describes no acceptable alternatives). The best limits are absolute rather
than relative (e.g., “Don’t bang on the windows” rather than “Don’t
bang on the windows so hard”). An essential part of limit setting is for
the clinician to determine whether the patient is capable of responding.
In general, the greater the degree of cognitive impairment, the less able
the patient is to understand or respond to limit setting. In these in-
stances, and depending on the location of the threat, the clinician
should call for help or leave the interview office immediately if no one
else is available to assist in the management of the patient.
Safety in the Psychiatric Emergency Service and

Inpatient Unit
There are certain environmental variables that can be modified to de-
crease the potential for violence, especially in emergency departments
or inpatient units. They include shortening the waiting time, decreasing
stimuli by offering the patient a comfortable chair in an office or an op-
tion to lie down, and offering the patient a cup of water or juice or some
food (Petit 2005).
When interviewing patients who have been violent or who have the
potential for violence, especially in an emergency department or inpa-
tient unit, the clinician should remove his or her glasses, if possible
(Tardiff 1996). Neckties should be removed or tucked in, and jewelry
such as necklaces and earrings should be removed (Tardiff 1996). The
physician may want to consider tightly securing long hair. Clinicians
should always consider that running away from a patient may become
necessary and should always wear shoes that will make running easier.
Clinicians should leave when a situation seems totally uncontrolla-
ble (Bowie 1989). Before leaving, the clinician should consider what
must be done to escape and identify the nearest safe place, how far it is,
and the best way to get there (Bowie 1989). The clinician should not run
as a panic reaction but should leave as a positive action. The clinician
should run toward a place of safety and not just away from danger.
Once beginning the escape, the clinician should not hesitate or stop un-
til he or she is free and clear.
Threat Management
Overview
A neglected area in clinician safety is threat management. Unfortunately,
there is a paucity of research in this area. There are no data detailing the
clinical context in which most threats occur or of the outcome of threats.
Threats can take many forms. They can be verbal, written, by phone, or re-
layed by a third party. Patients can threaten the clinician in an impulsive,
emotional outburst; by a calm, serious statement; in a joking, flippant
manner; or through vague innuendos (Tardiff 1996). Threats can be in the
form of property damage, visits to a therapist’s home, or loitering around
a therapist’s office on days when there is no appointment (Jenkins 1989).
The sending of love letters, pornographic materials, or vacation pictures
can represent threats (Jenkins 1989). Threats can also take the form of
veiled comments that show a patient is involved in the clinician’s personal
life (Maier 1996). Such comments might include knowledge of the clini-
cian’s car or home address or the names of the clinician’s children. These
statements are made as a way of showing interest in the clinician but are
usually out of proportion to the therapeutic relationship (Maier 1996).
Threats to clinicians can occur in a variety of settings, including clin-
ical settings such as the emergency department, inpatient unit, outpa-
tient clinic, or private office. They can occur in custody hearings, dis-
ability evaluations and hearings, forensic evaluations and hearings,
competency hearings, and in nonpsychiatric medical settings, or they
may come from spouses of patients.
Dynamics of a Threat
Threats are a means by which a patient tries to gain control of others
through manipulation (Maier 1996). When patients make manipulative
comments, they are often of such a nature that the clinician is not en-
couraged to share them with his or her peers. For instance, a patient
may ask a female clinician if she is pregnant or has her period. He may
tell a male clinician that he looks hung over or make some comment
about the clinician’s sexual identity. The interplay between positive
comments and personal judgments can provide for effective manipula-
tion, resulting in the clinician’s behavior becoming predictable and thus
usable by the patient against the clinician at some future time. In this
process, the patient establishes a secret relationship with the clinician,
binding the clinician to the patient, governing the clinician’s conduct,
and distorting his or her judgment (Maier 1996). Sharing the secret with
a colleague is the first step in managing this process.
Types of Threats
Brown et al. (1996) described two types of threat situations: situational
and transferential threats. A situational threat occurs when a psychiatrist
acts as an administrator, usually on an inpatient unit or emergency de-
partment. For example, a psychiatrist frustrated a patient’s wish by de-
nying a request to go out on a pass to get more medication. The patient
threatened the psychiatrist, an emergency code was called, the patient
was restrained or escorted from the hospital, and the threat situation
ended with no psychiatrist being injured.
A transference threat occurs within the context of ongoing psycho-
therapy. The threats to therapists described by Brown et al. (1996) were
often insidious and in several instances continued for many years. Al-
though no psychiatrists who were the targets of transference threats
were physically injured, the threats were very disruptive, both to the
psychiatrists and to their families. The treating psychiatrist who ini-
tially viewed the threat as an issue to be resolved in therapy often toler-
ated transference threats. The threat situation often continued for many
months before the psychiatrist recognized the inherent danger. Rather
than diluting the transference by disengaging from the patient, many of
the therapists intensified the transference by increasing the frequency of
treatment sessions. Therapists often have difficulty disengaging from a
patient. Such situations may be compared with the problem of marital
separation when two parties are locked in a pathological relationship
(Lion 1995). By the time the treating psychiatrist in this example sought
consultation from a colleague, the clinician was so enmeshed in the pa-
tient’s distorted or psychotic transference that the resolution of the
threat situation was complicated, if not impossible. When a clinician is
threatened or perceives a threat, he or she should initially seek expert
consultation.
Monitoring Threats
Psychiatrists must pay close attention to any changes in either transfer-
ence or countertransference feelings as they arise in the therapy situa-
tion. Any changes in behavior or affect, either by the therapist or the pa-
tient, should alert the clinician to a potential change in the therapeutic
relationship. Such behavior might include patient requests for more
therapy sessions, frequent phone calls or messages on the answering
machine, notes or gifts between sessions, increased and frequent flat-
tery, or increased anger, hostility, or withdrawal. Therapists who treat
more primitive character disorders or paranoid or psychotic patients
run a certain risk as closeness develops (Lion 1995). This is a risk that
must be continuously assessed and dealt with. Often, subtle actions,
such as the patient’s moving back his or her chair, defensive posturing,
or tardiness late in the course of therapy, should alert a clinician to a
problem in the transference (Lion 1995). In such situations, immediately
reviewing the case with a colleague is a prudent first step toward un-
derstanding the change in the therapeutic relationship, assessing
whether a threat exists, and if one does exist, the level of risk. Similarly,
a clinician should also monitor his or her countertransference. Feelings
of increasing attraction, dread, anxiety, or anger toward a patient might
signal the beginning of a distorted transference.
Managing Threat Situations

When a threat is made, a clinician should act decisively and immedi-
ately. Threats are messages and require comment (Lion 1995). To ignore
them is to indicate that the clinician is indifferent to suffering and does
not care. When a clinician is threatened, direct confrontation such as
“You’re scaring me with your threat” or “Why do you have to go around
scaring me and others with your threats? Is this the only way that you
can relate to people, to be scary?” is often effective (Lion 1995). If the
clinician has any alliance with the patient, interpretive statements can
facilitate understanding and resolution of the threat (Lion 1995). Ex-
amples of interpretive comments include, “Why do you go around
threatening and alienating people?” or “Do you have any positive feel-
ing about our therapy?” The therapist may try to clarify the meaning of
the threat, but a patient’s failing to respond to reasonable interventions
and continuing to make threats are considerations for termination of
therapy. Rather than intensifying the transference by continuing to en-
gage the patient in therapy, the therapist should focus on diluting the
transference by establishing distance and separation from the patient.
A written threat must be preserved with a minimal disruption of the
physical evidence. Envelopes and all packaging materials must be
saved, only minimally handled (preferably with cotton gloves touching
the extreme edges), and stored in a plastic bag (Dietz 1990). Telephone
threats that are on tape should be saved, and under no circumstances
should the tape be erased. If a threat is by telephone or in person, an at-
tempt should be made to reconstruct the conversation verbatim and to
immediately record as much detail as possible (Dietz 1990).
For clinicians who work in institutional settings, diluting the trans-
ference can be done by involving the chief clinical administrator. In de-
veloping a response to a threat, it can be very useful and reassuring to
meet with departmental representatives from security, legal affairs, ad-

ministration, employee health, human affairs, the director of clinical
services, and a psychiatric consultant familiar with violence manage-
ment techniques (Tardiff 1996). This group can provide various per-
spectives on evaluating different options, from legal to therapeutic, in
responding to threats.
A designated administrator should notify the patient that the ad-
ministrator is aware of the threat situation and that it will not be toler-
ated. Furthermore, the patient should be informed that the threatened
clinician will not accept phone call or letters or have any further inter-
action with the patient. The administrator should offer to help the pa-
tient find another therapist should he or she wish to continue therapy.
Legal assistance should be concomitantly obtained. The hospital legal
staff should unambiguously convey to the patient that threatening a
staff clinician is behavior that will not be tolerated and that he or she
will be prosecuted if the threats continue.
For psychiatrists in private practice, involving the district attorney
is more effective than obtaining help from the family lawyer. Consulta-
tion with a colleague, especially one with experience in dealing with
such situations, is extremely valuable. Threats must be dealt with deci-
sively and without ambivalence.
When threats persist despite the interventions discussed, additional
steps should be considered. If the patient is told not to return to the ther-
apist’s office, security guards or doormen should be alerted not to allow
the patient entrance and to notify the therapist if the patient tries to en-
ter the building. A description of the patient should be given to the se-
curity staff. If security staff or doormen are not available, the local po-
lice precinct should be notified immediately if the patient is seen on the
premises. While the threat is ongoing, the therapist should alter his or
her schedule and not leave the office at night alone or come in early in
the morning alone. The therapist should park near the office, in as pub-
lic a location as possible. The therapist should avoid parking in isolated,
dark areas where there is little traffic and where there are places for a
patient to hide. If threats are severe and ongoing, the therapist should
consider varying daily routines and travel routes. Although this can be
disruptive to the therapist’s life, it also increases his or her safety.
Stalking
There are various definitions of stalking (Meloy 1998). From a clinical
perspective, stalking is obsessional pursuit, harassment, and intimida-
tion by a person who has a significant personal relationship (or believes
one to exist) with the object of the unwanted attention (Miller 2001).
Stalking of clinicians is a behavior representative of a pathological
attachment or deranged transference, or it may be the result of a dissat-
isfying outcome (Lion and Herschler 1998). Eight percent of adult
American women and 2% of adult American men have been stalked
sometime in their lives (Meloy 1998). At least half of stalkers may ex-
plicitly threaten their victims, and the frequency of violence toward
their objects ranges from 25% to 35% (Meloy 1998). The homicide rate
among victims of stalking is less than 2% (Meloy 1998). Physicians and
mental healthcare staff are at even greater risk of being stalked than the
general population, particularly by their patients. Recent studies have
found that 11% of mental health professionals responding to a survey
had been stalked, with psychologists and psychiatrists more likely to
experience longer periods of stalking (Galeazzi et al. 2005). Sandburg et
al. (2002) reported that 53% of inpatient clinical staff were stalked,
threatened, or harassed at some point during their career. Gentile et al.
(2002) found that 10% of psychologists had been stalked at least one
time, and Ashmore et al. (2006) found that 50% of responding mental
health nurses had been stalked.
Although violence is obviously the most disturbing potential out-
come of stalking, this phenomenon also deserves close attention be-
cause most victims of stalking experience major life disruptions and
psychological disturbance, including anxiety, depression, or symptoms
of trauma (Meloy 1998). Lion and Herschler (1998) presented cases of
psychiatrists who have been threatened or stalked and spent enormous
amounts of time and money on protective measures and legal fees;
some had to relocate their families and practices to other parts of the
country.
Immediate management of stalking is imperative and does not dif-
fer significantly from the management of threats. Meloy (1997) recom-
mended 10 guidelines for the clinical management of stalking, covering
the following points: a team approach, personal responsibility for
safety, documentation and recording, no initiated contact, protection
orders, law enforcement and prosecution, treatment if indicated, segre-
gation and incarceration, periodic violence risk assessment, and the
importance of dramatic moments, which are events that shame or hu-
miliate the perpetrator. Lion and Herschler (1998) further suggested be-
ing attuned to early inappropriate behaviors and boundary violations
and considering the risk that such violations will escalate over time to
the point of physical danger. Clinicians should seek legal and forensic
consultation early—preferably prior to an intervention.
The Armed Patient

If a patient appears in a treatment setting with a weapon, as few staff as
possible should be exposed to the risk of injury (Tardiff 1996). Staff
should retreat to an office. The clinician should acknowledge the obvi-
ous—“I see you have a gun” (Tardiff 1996). He or she should be calm
and not become counteraggressive or threatening. The clinician should
encourage the patient to talk during the initial phases and repeat the pa-
tient’s concerns. The firearm is almost invariably an expression of feel-
ings of inadequacy and fear (Dubin 1995). The clinician should try to
speak to the underlying psychological issues. It is important to identify
areas in which the patient’s viewpoint is correct rather than initially try-
ing to demonstrate the areas in which the patient’s viewpoint is wrong.
The alliance can be enhanced if the clinician can identify similarities be-
tween him- or herself and the patient.
If a short time passes without the patient’s actually firing the gun,
the likelihood of its eventual use is diminished. Initially, however, the
clinician should comply with whatever demand the patient may make
and take special care to avoid further upsetting the patient. There
should be no attempt to take the weapon from the patient. A suggestion
should be made to have the patient put the weapon down gently (Tar-
diff 1996). One should not reach for the gun or tell the patient to drop the
gun, because it might discharge (Tardiff 1996).
Office Safety
The most problematic issue is the individual clinician who practices
alone. Office safety requires planning and persistence, and clinicians
should be cognizant of safety issues. Ideally, offices should have two
doors, one into the reception room and another locked door that leads
into the actual treatment office. This second door should have a peep-
hole so that the clinician can see who is in the outer office. Ideally, a cli-
nician may wish to have two entrances to his or her office so that if a
threatening patient comes into the reception area, there is another
means of exit. If a clinician does not have a receptionist, he or she
should consider a panic alarm or buzzer system to notify either build-
ing security or police in the event of a threatening or aggressive patient.
Institutional outpatient sites should be constructed so that there is a
physical and personal buffer for the clinician. Offices in outpatient clin-
ics should have panic buttons or an organized strategy to notify a recep-
tionist, other staff, or even the police of a threatening situation. A pro-
tocol should be developed to train office staff to recognize patients who
are at risk for violence. Strategies should be put in place so that the of-
fice staff can notify the clinical staff or, if necessary, the police without
alarming the patient when he or she begins to escalate or presents to the
office in a threatening manner. For instance, a simple code such as “Dr.
Smith, can you see Mr. Jones immediately?” may be a signal that Mr.
Jones is demonstrating behavior that puts him at risk for violence. This
nonthreatening phone call then allows the clinician to implement other
strategies, such as calling the police or mobilizing other clinic staff to
help contain the patient. A different code should be developed to alert
staff that an armed patient is in the waiting area. Such an alarm should
be simple and not threatening, for example, “Is room 22 available?” Di-
rect alarm systems to the police should also be considered, especially
for the situation in which a patient has a weapon or is suspected of hav-
ing a weapon. Such a system should be inconspicuous so as not to alert
the patient that an alarm is being sounded.
There are certain architectural features that can further enhance
safety. An office should be decorated in a manner consistent with the
type of patient that the clinician is treating. If a clinician is treating psy-
chotic patients or patients with a history of aggression, or if he or she
frequently evaluates new patients, there are specific office safety issues
that should be considered. Safe offices will have heavy furniture that
cannot be lifted or used as a weapon. Offices, especially in emergency
departments or inpatient units, should not have hard, sharp objects
such as small ashtrays, artwork, lamps, or other decorations that can be
thrown or used as weapons. All office doors should swing out into a
hall and not into an office. This prevents a patient inside the office from
blocking the therapist’s egress by leaning against the door. Berg et al.
(2000) recommended several strategies for enhancing safety in emer-
gency departments and clinics. Having windows in the doors of the ex-
amining rooms allows privacy while lending a sense of the possibility
of being monitored for unacceptable behavior. Security cameras also
provide a sense that behavior is being monitored, and posting rules
makes it clear that violence will not be tolerated and has consequences.
Key Points
Managing aggressive, violent patients is a clinical challenge. However,
if the clinician periodically reviews the key points of clinical manage-
ment, most encounters with a violent patient can have a satisfactory
outcome. The fundamental management strategies include:
■ Performing a risk assessment on all new patients.
■ Not evaluating or treating patients at risk for violence alone or in
an isolated office.
■ Remembering that violence is a response to feelings of helpless-
ness, passivity, and perceived or actual feelings of humiliation.
■ Using nonthreatening talk-down strategies and using affect
management as the centerpiece of the intervention.
■ Setting limits by offering the patient two options, with one being
the preferred option.
■ Being cognizant of the different manifestations of a threat,
because threats take many forms.
■ Monitoring transference and countertransference and evaluating
any change in the context of a developing threat situation.
■ Immediately seeking consultation if a threat is perceived or if the
clinician questions whether a patient’s behavior is a threat.
■ Immediately initiating threat management strategies without
hesitation or ambivalence.
■ Anticipating the potential for stalking and immediate and early
consultation with a forensic expert.
■ Responding in a nonthreatening manner to an armed patient and
offering help and understanding.
■ Periodically evaluating offices and patient areas and implement-
ing changes that will enhance safety.
References
American Psychiatric Association: Clinician Safety: Report of the American
Psychiatric Association Task Force on Clinician Safety. Report No. 33.
Washington, DC, American Psychiatric Association, 1993
Anderson AA, Ghali AY, Bansil RK: Weapon carrying among patients in a psy-
chiatric emergency room. Hosp Community Psychiatry 40:845–847, 1989
Ashmore R, Jones J, Jackson A, et al: A survey of mental health nurses’ experi-
ences of stalking. J Psychiatr Ment Health Nurs 13:562–569, 2006
Barlow CB, Rizzo AG: Violence against surgical residents. West J Med 167:74–
78, 1997
Berg AZ, Bell, CC, Tupin J: Clinician safety: assessing and managing violent pa-
tients. New Dir Ment Health Serv 86:9–29, 2000
Black KJ, Compton WM, Wetzel M, et al: Assaults by patients on psychiatric res-
idents at three training sites. Hosp Community Psychiatry 45:706–710, 1994
Bowie V: Coping With Violence: A Guide for the Human Services. Sydney, Aus-
tralia, Karibuni Press, 1989
Brown GP, Dubin WR, Lion JR, et al: A preliminary classification of threats
against clinicians. Bull Am Acad Psychiatry Law 24:367–376, 1996
Buckley PF, Noffsinger SG, Smith DA, et al: Treatment of the psychotic patient
who is violent. Psychiatr Clin North Am 26:231–272, 2003
Chaimowitz GA, Moscovitch A: Patient assaults against psychiatric residents:
the Canadian experience. Can J Psychiatry 36:107–111, 1991
Coverdale J, Gale C, Weeks S, et al: A survey of threats and violent acts by pa-
tients against training physicians. Med Educ 35:154–159, 2001
Davies S: Assaults and threats on psychiatrists. Psychiatr Bull 25:89–91, 2001
Dietz PE: Defenses against dangerous people when arrest and commitment fail,
in Review of Clinical Psychiatry and the Law, Vol 1. Edited by Simon RI.
Dubin WR: Assaults with weapons, in Patient Violence and the Clinician. Ed-
ited by Eichelman BS, Hartwig AC. Washington, DC, American Psychiatric
Press, 1995, pp 53–72
Dubin WR, Lion JR: Violence against the medical profession, in Creating a Se-
cure Workplace: Effective Policies and Practices in Health Care. Edited by
Lion JR, Dubin WR, Futrell DE. Chicago, IL, American Hospital Publishing,
1996, pp 3–14
Dubin WR, Wilson SJ, Mercer C: Assaults against psychiatrists in outpatient set-
tings. J Clin Psychiatry 49:338–345, 1988
Duhart DT: Violence in the Workplace, 1993–1999. Bureau of Justice Statistics
Special Report, NCJ 190076. Washington, DC, U.S. Department of Justice,
2001
Eichelman BS: Strategies for clinician safety, in Patient Violence and the Clini-
cian. Edited by Eichelman BS, Hartwig AC. Washington, DC, American
Psychiatric Press, 1995, pp 139–154
Fink DL, Shoyer B, Dubin WR: A study of assaults against psychiatric residents.
Acad Psychiatry 15:94–99, 1991
Foust D, Rhee KJ: The incidence of battery in an urban emergency department.
Ann Emerg Med 22:583–585, 1993
Galeazzi GM, Elkins K, Curci P: The stalking of mental health professionals by
patients. Psychiatr Serv 56:137–138, 2005
Gentile SR, Asamen JK, Harmell PH, et al: The stalking of psychologists by their
clients. Prof Psychol Res Pr 33:490–494, 2002
Goetz RR, Bloom JD, Chenell SL, et al: Weapons possession by patients in a uni-
versity emergency department. Ann Emerg Med 20:8–10, 1991
Green SA, Goldberg RL, Goldstein DM, et al: The art of limit setting, in Limit
Setting in Clinical Practice. Edited by Green SA, Goldberg RL, Goldstein
DM, et al. Washington, DC, American Psychiatric Press, 1988, pp 1–13
Jenkins MG, Rocke LG, McNicholl BP, et al: Violence and verbal abuse against
staff in accident and emergency departments: a survey of consultants in the
UK and the Republic of Ireland. J Accid Emerg Med 15:262–265, 1998
Jenkins PL: Harassment, in Hazards of Psychiatric Practice: A Clinical and Le-
gal Guide for the Therapist. Chicago, IL, Year Book Medical Publishers,
1989, pp 14–25
Ladds B, Lion JR: Severe assaults and homicide within medical institutions: ep-
idemiologic issues, in Creating a Secure Workplace: Effective Policies and
Practices in Health Care. Edited by Lion JR, Dubin WR, Futrell DE. Chi-
cago, IL, American Hospital Publishing, 1996, pp 15–24
Lavoie FW, Carter GL, Danzi, DF, et al: Emergency department violence in
United States teaching hospitals. Ann Emerg Med 17:1227–1233, 1988
Lion JR: Verbal threats against clinicians, in Patient Violence and the Clinician.
Edited by Eichelman BS, Hartwig AC. Washington, DC, American Psychi-
atric Press, 1995, pp 43–52
Lion JR, Herschler JA: The stalking of clinicians by their patients, in The Psy-
chology of Stalking: Clinical and Forensic Perspectives. Edited by Meloy
JR. New York, Academic Press, 1998, pp 163–173
Maier GJ: Managing threatening behavior: The role of talk down and talk up.
J Psychosoc Nurs Ment Health Serv 34:25–30, 1996
McNeil DE, Binder RL: Patients who bring weapons to the psychiatric emer-
gency room. J Clin Psychiatry 48:230–233, 1987
Meloy JR: The clinical risk management of stalking: “Someone is watching over
me…” Am J Psychother 51:174–184, 1997
Meloy JR: The psychology of stalking, in The Psychology of Stalking: Clinical
and Forensic Perspectives. Edited by Meloy JR. New York, Academic Press,
1998, pp 1–23
Miller MC: Stalking. Harv Ment Health Lett 17:5–6, 2001
Milstein V: Patient assaults on residents. Indiana Med 80:753–755, 1987
O’Sullivan M, Meagher D: Assaults on psychiatrists: a three-year retrospective
study. Ir J Psychol Med 15:54–57, 1998
Pane GA, Winiarski AM, Salness KA: Aggression directed toward emergency
department staff at a university teaching hospital. Ann Emerg Med 20:283–
286, 1991
Paola F, Malik T, Qureshi A: Violence against physicians. J Gen Intern Med
9:503–506, 1994
Petit JR: Management of the acutely violent patient. Psychiatr Clin North Am
28:701–711, 2005
Sandberg DA, McNiel DE, Binder RL: Stalking, threatening, and harassing be-
havior by psychiatric patients toward clinicians. J Am Acad Psychiatry
Law 30:221–229, 2002
Schwartz TL, Park TI: Assaults by patients on psychiatric residents: a survey
and training recommendations. Psychiatr Serv 50:381–383, 1999
Tardiff K: Assessment and Management of Violent Patients, 2nd Edition. Wash-
ington, DC, American Psychiatric Press, 1996
Thackrey M: Therapeutics for Aggression. New York, Human Sciences Press,
1987
Thompson BM, Nunn J, Kramer I, et al: Disarming the department: weapon

screening and improved security to create a safer emergency department
environment. Ann Emerg Med 17:419, 1988
VanIneveld CH, Cook DJ, Kane SL, et al: Discrimination and abuse in internal
medicine residency. J Gen Intern Med 11:401–405, 1996
Wyatt JP, Watt M: Violence towards junior doctors in accident and emergency
departments. J Accid Emerg Med 12:40–42, 1995
C H A P T E R 2 4

and the Clinician
Intimate partner violence (IPV) can take any of several forms, includ-
ing emotional, physical, and/or sexual abuse. This chapter focuses spe-
cifically on the physical and sexual abuse inflicted by one individual
against his or her intimate partner, who may be a spouse, a live-in part-
ner, or a date of the opposite or same sex. It should be recognized, how-
ever, that emotional abuse and physical violence are often related. The
physical violence may take numerous forms, including battering or
beating, rape, murder, or forced suicide; may be effected through the
use of fists, feet, sexual organs, poisoning, drowning, hanging, fire, elec-
trical shocks, knives, guns, and/or other weapons (Loue 2001b); and
can occur in a variety of settings, not just the home. The violence may
be perpetrated for various proffered reasons, including economic pres-
sures and dissatisfaction with the partner’s attitudes or behavior.
Incidents of physical abuse may occur only intermittently, but an
abusive partner may use emotional abuse to gain and retain control
during the interim periods (Dutton and Golant 1995). This “cycle of
This research was supported in part by National Institute of Mental Health

grant R01 MH63016.
483
violence,” as it has been referred to, consists of three phases. The first, or
tension-building, phase is characterized by verbal, emotional, and
sometimes physical abuse of lesser severity. Often the victim will at-
tempt to alleviate the situation by placating the batterer. The second, or
acute battering, phase represents the discharge of built-up tension. Al-
though the victim may be blamed for “triggering” the abuse, the actual
cause of the violence is often a stressor external to the relationship (such
as difficulties at work) or internal to the abuser. Acute battering episodes
are often unpredictable. During the third, or “honeymoon,” phase, the
abuser attempts to apologize for his or her actions through apologies,
gifts, helpfulness, and increased emotional closeness with the victim. It
is during this phase that the bond between the abuser and his or her vic-
tim is intensified, because the victim now experiences the positive as-
pects of the relationship and comes to believe in the partner’s voiced
commitment to the relationship. This intensified emotional closeness
and commitment increases the difficulty of leaving the abusive relation-
ship (Walker 2000).
In the United States, the lifetime prevalence of physical assault by an
intimate partner against women in a population not defined by mental
illness has been found to range from 9% to 30% (Bureau of Justice Statis-
tics 1998; Tjaden and Thoennes 2000), and the prevalence of rape by an
intimate is approximately 8% (Tjaden and Thoennes 2000). The range of
estimates is due, in part, to variations across studies in the methods used
to collect data (e.g., personal interviews, telephone interviews, written
surveys); the sampling methods used (e.g., hospital patients, general
population, clinic outpatients); and definitions of partner violence.
Previous research has identified the following risk factors for part-
ner violence in the United States: younger age, urban residence, lower
levels of education, and lower income (Sorenson et al. 1996). Prior his-
tory of abuse, including childhood abuse (Friedman and Loue 2007),
may also increase risk. Female victims of IPV have been found to be
more likely to use multiple substances (alcohol, cigarettes, and illegal
drugs) than are nonvictims (Martin et al. 1996). Pregnancy has also been
established as a risk factor for IPV victimization (Miller and Finnerty
1996). Pregnant women especially at risk for battering during their
pregnancies are those who have been battered prior to pregnancy (Mc-
Farlane et al. 1992). Homicide is a leading cause of death during preg-
nancy (Frye 2001). Research indicates that homicide is more likely to oc-
cur among couples of lower socioeconomic status and those in which
the wife is significantly younger than the husband (Chimbos 1998). It
has been hypothesized that the homicide may be related to partner con-
cerns about paternity or changing role expectations.
Intimate Partner Violence and the Clinician ❘ 485
Various theories have been advanced in an attempt to understand

and explain why partner violence occurs. These theories are outlined in
Table 24–1. Some theories focus on why the violence occurs, whereas
others attempt to understand why battered women remain in the abu-
sive relationship.
Intimate Partner Violence and

Severe Mental Illness
Specific diagnoses, including schizophrenia, depression, anxiety disor-
ders, substance use disorders, and personality disorders, may elevate the
risk of victimization from IPV (Bergman and Ericsson 1996; Briere et al.
1997). Limited reality testing, impaired judgment, impaired executive
function, and difficulty with social relationships may potentially increase
a person’s vulnerability to abusive or coercive relationships (Goodman et
al. 1995). Social isolation and the stigma of mental illness, which make
some women eager to please, also may increase their risk (Gearon and
Bellack 1999). Women may be unable to distinguish between the physical
closeness that signifies emotional intimacy and the physical closeness as-
sociated with assaultiveness, and thus they may not remove themselves
from potentially dangerous situations (DeNiro 1995).
There is a well-established association between IPV and posttrau-
matic stress disorder (PTSD), depression, and substance abuse (Coker
et al. 2005; El-Bassel et al. 2003; Houry et al. 2005; Mueser et al. 1998).
Symptoms of mental illness may worsen with victimization (Campbell
2002). Victimization from IPV or other traumas may act as a stressor,
further increasing symptomatology. Among battered women in treat-
ment for depression or anxiety, the majority believed that the battering
had worsened their symptoms (Weingourt 1990).
Evidence suggests that IPV victimization may increase the already
elevated suicide risk among the mentally ill. Studies have found that
more than one-third of battered women have attempted suicide (Diene-
mann et al. 2000; Golding 1999). Similarly, up to 44% of suicide attempts
may have abuse as a precipitating factor. A study of outpatients with de-
pression or anxiety found that those who had been physically assaulted
were three times more likely to attempt suicide (Kaplan et al. 1995).
Case Example
Marta [all names used in the case example are fictitious] is a Hispanic
woman in her mid-30s who was diagnosed with major depression some
years ago. Marta moved from Puerto Rico to the mainland United States
at a relatively young age. Both her brother and an uncle repeatedly
486
TABLE 24–1. Theories of causation of intimate partner violence
Theory Description
❘
Why the violence occurs
Culture of violence Subcultures develop norms that permit the use of physical force or violence.
(Wolfgang and Ferracuti 1967)
Ecological theory (Belsky 1980) Risk of assault is a function of the interplay between factors at the macrosystemic,

exosystemic, microsystemic, and ontogenetic levels of the environment.
Evolutionary theory 1. Obedience is valued in societies with a hierarchically organized structure; violence may
(Lenski and Lenski 1970; be used as a means of securing obedience.
Rohner 1975; Wilson and Daly 1993) 2. Partner violence is evolutionary and occurs as a mechanism to ensure the male partner’s
sexual dominance and reproductive advantage.
3. Sexual proprietariness is a psychological adaptation of the human male; the jealousy and
related violence are directly tied to women’s reproductive value.
Exchange theory (Gelles 1983) People use violence to obtain their goals as long as the benefits outweigh the costs.
Theory of marital power Those who lack power will be more likely to physically abuse their partners.
(Cromwell and Olson 1975)
Patriarchy theory Wife assault is a systematic form of domination and social control of women by men; assault
(Dobash and Dobash 1979) is committed by men who believe that patriarchy is their right; the use of violence to
maintain male dominance is acceptable to society.
Resource theory Decision-making power within a family derives from the value of the resources that he or
(Blood and Wolfe 1960; Goode 1971) she brings to the relationship. The more external control one has of resources outside of the
family, the less likely he or she will be to use violence as a means of control.
Social learning theory (O’Leary 1988) Family violence arises due to a constellation of factors, including individual characteristics,
couple characteristics, and societal characteristics.
Intimate Partner Violence and the Clinician
TABLE 24–1. Theories of causation of intimate partner violence (continued)
Theory Description
Why people stay
Investment theory (Rusbult 1980) One’s willingness to remain in a relationship increases as the balance of the rewards over
costs from staying in the relationship exceeds the balance of rewards over the costs in
alternative relationships or arrangements.
Traumatic bonding theory Over time, the imbalance of power within a relationship increases as the dominant party
(Dutton and Painter 1981, 1993) develops an inflated self-image and the subordinate person feels increasingly negative
about him- or herself and becomes increasingly dependent on the abusive partner.
❘
487
sexually abused her as a young child. Marta was raised in the Roman
Catholic faith and more recently became a member of the religious com-
munity known as Mita y Aaron, founded by a Puerto Rican woman. Al-
though Marta has been involved in several relationships, she has never
married and has no children.
Marta has not used any substances for approximately 7 years, but
during her prior relationship with Jorge, she was dependent on alcohol
and heroin. Jorge was physically abusive, but Marta believed that, as a
good woman, she was responsible for her partner’s happiness. She de-
scribed his abuse thus:
One day we were walking down the street when he punched me

on my jaw. He accused me of looking at a man sitting on his
porch…. I felt dizzy, and he started yelling at me and calling me
a bitch. I could not wake up before him because he would beat
me. I had to lay there and look at the ceiling. That is the worst
feeling in the world. You stare at the walls and ceiling and think
about your life a lot…. He would put his arm around me to
make sure I would not go anywhere.
Marta explained why she tolerated the abuse, which included rape:
One day he got crazy, and he kicked me out of the apartment be-
cause his sister was telling him bad things about me. He took me
close and threw me down the steps. For a day I slept in the hall-
way. Then [he] said he called the cops and the cops told me I
needed to leave his place. I could not say anything because the
apartment was under his name only…. He would tell me what
I was thinking, and he was right. It was my job to sexually sat-
isfy him, so he continued to have anal sex with me.
One incident finally convinced Marta to leave Jorge:
He was asking me for money for drugs. I didn’t want to give it

to him…. He would buy clothes and spend all of his money, and
then he would want my half of the money. On that day he told
me to withdraw all the money, and I told him I was not going to
withdraw all the money…. After he beats me he starts to cry on
top of me and tells me he loves me. After that he tells me to get
up and tells me to go with him on Broadway so I can sell my
body. Never in my life I’ve done something like that, honest to
God…. He hits me and takes me over there. He tells me to get
dressed and he cuts my leg with a knife. He said if I didn’t give
him the money I was going to have to sell myself. Well, he takes
me and on the way there he beats me…. On Broadway he gave
me a big beating. It was a beating like you could not imagine….
…I didn’t want to sell my body so I resisted. Well he took me back
to the house and told me to get dressed because we were going
to the bank. I said to myself I can take no more beatings anymore
so I’m going to give him the money. He took me to a bridge where

there was a lady doing drugs…. She had AIDS. Then, he asked
her for the needle…. I wanted that money to leave him and then
he takes me under the bridge…. Because he didn’t clean it and
that needle automatically would give him hepatitis. He automat-
ically would get hepatitis. For sure he got it. I knew afterwards.
I got hysterical and wanted to leave but he didn’t let me leave.
Well, he started hitting me…and the girl would tell him look she
is a nice girl don’t hit her…. I started to run…he hit me. He
stabbed me with the needle on my hands and thighs. When I felt
him stabbing me I thought about the lady he shared the needle
with. I thought of her and how I would get AIDS…. I’m fucked
up now…. I lost my mind and I told God I did not deserve this.
Everything went through my mind…. I was struggling so he
would stop. I threw dirt in his eyes; I tried to defend myself. I fell
on the floor and there was glass on the floor. I started to cut my-
self. Because I said to myself I’m already sick. I had no hope. I
gave my life to him…. A lot of things went through my mind. It
was like I was going to die. I left a good man and I started to cut
my venas [veins]. Well when he saw me cutting myself he said,
“Ah, you want to die,” so he began to hit me on my head. He
didn’t cut himself; he took me back to the house we stayed at. He
makes me take a bath and get dressed.
After this incident, Marta attempted suicide with an overdose of pills

and was hospitalized for 3 months. After discharge, she had the courage
to leave Jorge and continued her treatment for depression and addic-
tion. She resides with her current partner, Humberto, and his child from
a previous marriage in a large urban city on the mainland.
This unfortunate case demonstrates multiple issues for the clinician

regarding IPV. Marta had depression and self-esteem difficulties and a
childhood history of abuse as well as IPV victimization. IPV can occur
not just at home but also in public (including some carryover to the
workplace). Her victimization did not merely occur in solitude but also
in front of others and on the street, and she was victimized with multi-
ple methods. Suicidality, such as Marta experienced, is relatively fre-
quent among IPV victims. Her thoughts and decisions about staying in
the relationship or leaving are illustrative as well.
Assessment
Assessing Abuse
Partner victimization is more common than many of the symptoms that
are routinely brought up during psychiatric visits, and yet it is infre-
quently discussed. Often women do not report victimization to their

therapists (Hilberman and Munson 1987; Jacobson and Richardson
1987) even when directly asked. However, IPV victimization should of-
ten be a consideration on the differential diagnosis.
Reasons for the reluctance of victims to discuss their abuse are nu-
merous. Victims may come to accept the victimization, may be ashamed,
or may not even conceptualize their own experiences as abuse or
assault. Male victims may be especially reluctant to reveal abuse per-
petrated on them by either a female or a same-sex partner. As a con-
sequence, many physicians do not know that their patients are victims
(Sahay et al. 2000). PTSD among those with serious mental illness is also
frequently undetected (Cusack et al. 2006; Salyers et al. 2004), although
it is a frequent consequence of victimization.
Given the frequency and consequences of IPV, physicians should
consider integrating into their practice careful, specific, routine inquir-
ies into experiences of victimization (Carlile 1991; Carlson et al. 2003).
Questioning about violence can be approached with a discussion about
decision making in relationships (Jacobson and Richardson 1987). Nor-
malizing statements regarding the frequency of victimization can intro-
duce inquiry (Friedman and Loue 2007). Women can be asked about
fears of being harmed or of their children being harmed. Abuse screen-
ing measures also exist (Soeken et al. 1998). Direct specific inquiry into
present and past traumatic experiences can be fruitful. Practitioners
should also explore the possibility of violence by same-sex partners and
by women.
Because many women seek care primarily from family practitioners
or obstetrician-gynecologists, these professionals may be the first to de-
tect IPV during routine screening. IPV can have serious psychiatric con-
sequences (exacerbating underlying conditions or triggering new ones),
so a referral to psychiatry should be considered. The psychiatrist can
then provide evaluation, monitoring, and, if necessary, treatment for co-
morbid mental illness. The psychiatrist can also obtain a victimization
history, which can be critical in medication management, discharge
planning, and safety planning.
Assessing for the Commission of Abuse

It is equally important that clinicians conduct an assessment to deter-
mine a patient’s risk of committing partner violence. In one study of 375
men screened in three family medicine clinics, 8.5% (32) had committed
acts of physical violence against their intimate partners during the pre-
ceding year (Oriel and Fleming 1998).
As in many areas of violence, the strongest predictor of IPV perpetra-

tion is past violence, especially before adulthood (Moffitt et al. 2001).
Other risk factors associated with the commission of IPV by men include
parental rejection or shaming behavior, physical assault by a mother,
fear of abandonment, trauma symptoms, the use of alcohol or drugs to
deal with pain, a tendency to blame others, frequent anger, and mood
swings. Male batterers have also been found to have antisocial person-
ality traits, to act out their hostility, to need an excessive amount of con-
trol, to experience tremendous insecurity, to have high levels of jealousy,
and/or to have low self-esteem (Dutton and Golant 1995; Jacobson and
Gottman 1998). Although it is popularly believed that men are more
likely to batter their partners than are women, more recent studies sug-
gest that the rate of violence committed by female partners against their
male intimates may equal that committed by men against their female
partners (Straus 1993). Male batterers, however, are responsible for most
acts of the more serious and egregious forms of physical violence, such
as choking, punching, and the use of weapons (Gormley 2005).
Questions to ascertain whether an individual is at risk of commit-
ting violence may initially focus on his or her relationship with a part-
ner and can become progressively more specific about techniques that
are used by the couple to handle anger. Certain disorders have been
found to be associated with the commission of violence in general: per-
sonality disorders, some types of schizophrenia, mood disorder, and
impulse control disorders (Ferris et al. 1997). A physician diagnosing
one of these disorders may, in the presence of other indicators, wish to
address the issue of perpetration of partner violence. Men with sub-
stance use disorders and women with histories of self-harm or suicide
attempts are also common among batterers (Buttell and Carney 2006).
Treatment and Management

The Victim
It is notable that Minnesota Multiphasic Personality Inventory–2 profiles
of abused women may be quite similar to those of inpatients with
schizophrenia or borderline personality disorder (Khan et al. 1993). A
patient who is an IPV victim may feel unsafe in her own home without
being “paranoid.” Similarly, the inability to sleep may be self-protective.
Psychiatrists should be careful in evaluating these symptoms. Pharma-
cotherapy choice can reflect consideration of IPV. For example, sedating
medications could decrease the ability to escape from or respond to dan-
gerous situations.
Mistrust, emotional isolation, and impaired self-esteem in IPV victims

are fodder for psychotherapy (Hilberman and Munson 1987). Role-
playing may be a useful method (Gearon and Bellack 1999). Cognitive-
behavioral therapy may address symptoms of anxiety and depression.
Psychiatrists should be able to refer their patients either directly to
shelters, services, and crisis management services or to someone famil-
iar with these resources. In addition, the development of a safety plan
and available legal options should be considered. A patient’s decision to
leave a violent relationship is not made lightly. The patient may be am-
bivalent, fear for his or her children, fear being alone, fear retaliation,
have religious beliefs about staying in the marriage, or have significant
financial difficulties and be isolated from his or her support group.
The Assaultive Partner

Male batterers are not a homogeneous group and may differ in their per-
sonality style (moody and emotional, cold and calculating, or insecure
and jealous [Hamberger and Hastings 1988]). Abusive females may be
compulsive, histrionic, and/or narcissistic (Buttell and Carney 2006).
Accordingly, treatment must consider the characteristics of the individ-
ual. Batterers’ physiological responses to their own violence (cold and
remorseless or reactive and threatened) are also important. Many batter-
ers receive treatment through court-mandated batterer intervention pro-
grams, which use multiple sessions of cognitive-behavioral therapy to
reduce batterer aggression. The effectiveness of these programs is ques-
tionable, however—possibly because they offer a single approach to
dealing with a multidimensional problem (Buttell and Carney 2006).
Couples or family therapy is another means of treating batterers. Al-
though some therapists advise a victim to leave the relationship, others
believe that advising this is unethical and encourages passivity in the
victim. In cases of severe violence where the victim is not psychologi-
cally competent, the therapist may be ethically bound to intervene. In
cases of lower-severity mutual violence, however, the therapist might
best adopt the principle of neutrality, respecting both victim and bat-
terer as autonomous individuals (Wilcoxen 1985).
Legal Issues
Protection and the Reporting of
In some instances in which the healthcare provider believes that a pa-
tient may be a danger to his or her intimate partner, the provider may
have a duty to protect the potential target, according to a line of court

cases that began in 1976 with Tarasoff v. Regents of the University of Cali-
fornia (see also Chapter 21). The court held that when a patient “pre-
sents a serious danger…[the therapist] incurs an obligation to use rea-
sonable care to protect the intended victim against such danger.” That
obligation could be satisfied by warning the intended victim of the po-
tential danger, by notifying authorities, or by taking “whatever other
steps are reasonably necessary under the circumstances.” Some version
of a Tarasoff duty, whether through case law or statute, has been adopted
in many states and may provide immunity to the psychiatrist for the
breach of confidentiality in cases of violence risk.
Most, if not all, states have laws mandating the reporting by health-
care providers of elder abuse, including violence inflicted by an inti-
mate partner (Loue 2001a). As of 2000, seven states (California, Colo-
rado, Kentucky, Mississippi, Ohio, Rhode Island, and Texas) had laws
requiring providers to report IPV injuries (Houry et al. 2002), but these
laws vary drastically. California and Colorado notably further require
that healthcare providers report instances of injury resulting from part-
ner violence to law enforcement, whereas Mississippi and Kentucky
providers report to the department of public welfare. In Ohio, physi-
cians must have documented IPV injuries in the record, whereas Rhode
Island requires injuries to be reported for data collection purposes only.
Texas requires documentation, shelter referral, and informing patients
that IPV is a crime. Twenty-three states had statutes in 2000 requiring
reporting of injuries received from crimes (and IPV is a crime in these
states) (Houry et al. 2002). However, many physicians are unaware of
these reporting statutes or may be aware yet still not report (Houry et
al. 2002). Arguments against mandatory IPV reporting, in addition to
those involving the exception to confidentiality, include that a reporting
requirement may increase risk for retaliatory abuse, may deter victims
from seeking care, and may create expectations that the system might
not be able to meet. Arguments for mandatory reporting include that it
would increase detection of abuse and potentially improve victim
safety while not deterring most victims from seeking medical care
(Houry et al. 2002; Sachs and Rodriguez 2000).
Separation and the pursuit of legal remedies may actually increase
a woman’s risk of harm (Jordan 2004). Fewer than half of IPV incidents
are reported (Tjaden and Thoennes 2000), and arrest rates vary. Intimate
partners who assault or rape are less likely to be arrested than other in-
dividuals committing similar acts. Over a quarter of those who are ar-
rested re-assault prior to trial (Jordan 2004). Civil Protection Orders
(CPOs; also known as restraining orders) may be helpful but do not
guarantee safety; CPOs are violated approximately 40% of the time and
may even elevate the woman’s risk (Spitzberg 2002). Considerations re-
garding the seeking of CPOs include offender’s employment or social
standing, severity and persistence of the violence, length of relation-
ship, presence of children, and living situation (Jordan 2004).
The Psychiatrist as Expert

“Self-defense” as an affirmative legal defense usually requires an objec-
tive, honest, reasonable belief that one’s life is in immediate danger. “Im-
perfect self-defense,” in contrast, lowers the severity of the charge and
usually requires that the belief be subjectively reasonable. A woman in
an abusive relationship may believe that she is in acute danger related to
specific circumstances and may act to kill her aggressor, whereas a “rea-
sonable man” (without a history of victimization) in the same situation
might not have perceived the same risk. “Battered women’s syndrome”
has been used to explain this type of “imperfect self-defense.”
Battered women’s syndrome is not a DSM diagnosis but rather a
term used in the legal arena. The concept derives from the theory of
learned helplessness, which holds that individuals will not attempt to
escape abusive situations where they have learned from previous expe-
riences that all such efforts will be futile (Seligman 1975). The concept
of the syndrome has been criticized because of the lack of an agreed-
upon definition, the underlying assumption of pathology in the victim,
and the application of the term to partner violence only in situations in-
volving the commission of a crime, such as the killing of the abusive in-
timate (Dutton 2006). In addition, it has been erroneously asserted that
in order to claim a battered-woman defense, a woman who kills her
male partner in self-defense must have been suffering from PTSD.
Expert testimony regarding battered women has been used in the
United States, Canada, Australia, New Zealand, and the United King-
dom (Schuller and Rzepa 2002; Tang 2003). Since the landmark Ameri-
can case of Beverly Ibn-Tamas, in 1979, expert evidence about battered
women’s syndrome may be allowed in the courtroom (Ibn-Tamas v.
United States 1979). However, some courts require what is known as a
Daubert hearing (Daubert v. Merrell Dow Pharmaceuticals 1993) to deter-
mine whether there is sufficient scientific validity to go forward with the
theory. If the defense fails to establish the scientific validity of the theory
at the time of this hearing, they will not be permitted to present this evi-
dence to the jury at the time of the trial. It is critical to note here that the
term scientific validity in the legal context carries a different meaning
than it does in the context of scientific research (see Chapter 21).
The expert can provide information regarding the context or frame-

work for the woman’s violence, the woman’s belief in the reasonableness
of her actions, why women stay in these relationships, the use of excessive
force, the impact of the abusive relationship, the learned helplessness
model, psychological effects of battering, and the cyclical pattern of vio-
lence, including the potential contrition (“honeymoon”) phase (Dutton
2006; Schuller and Rzepa 2002; Walker 2000). However, the expert must
be careful not to mislead the court into believing that learned helplessness
is the only appropriate pattern of victim response to IPV. Other women
cope differently with IPV victimization, and the expert should use cau-
tion not to create a prejudice against women who do not fit the pattern.
One study (Schuller and Rzepa 2002) asked Canadian undergraduates
(N=200) about a hypothetical case of a battered woman who killed her
abuser. They found that the student mock jurors showed greater leniency
when presented with expert testimony (primarily with a nullification
instruction) and presented with a passive response history for the
woman.
Key Points
■ Intimate partner violence (IPV) can be perpetrated by men
against women, by women against men, or by same-sex
partners.
■ Severely mentally ill women may be at increased risk of IPV.
■ IPV may exacerbate existing symptoms of mental illness.
■ Patients may not disclose IPV to their care providers out of
fear, shame, and embarrassment.
■ Responses to IPV may mimic symptoms of mental illness.
For instance, a woman may appear to be disorganized and
paranoid when she may actually be living in fear of severe
injury.
■ Psychiatrists and other health professionals may perform an
assessment to determine if their patient fears for his or her
safety because of IPV or if the patient may be likely to
perpetrate violence against a partner.
■ Healthcare providers may be required by law to warn the
potential victim if they believe that a patient is likely to
perpetrate IPV and/or to report to designated authorities
the occurrence of IPV.
References
Bergman B, Ericsson E: Family violence among psychiatric inpatients as measured
by the Conflict Tactics Scale (CTS). Acta Psychiatr Scand 94:168–174, 1996
Belsky J: Child maltreatment: an ecological integration. Am Psychol 35:320–335,
1980
Blood RO, Wolfe DM: Husbands and Wives: The Dynamics of Married Living.
Glencoe, IL, Free Press, 1960
Briere J, Woo R, McRae B, et al: Lifetime victimization history, demographics,
and clinical status in female psychiatric emergency room patients. J Nerv
Ment Dis 185:95–101, 1997
Bureau of Justice Statistics: Bureau of Justice Statistics Sourcebook of Criminal
Justice Statistics, 1997. Washington, DC, U.S. Department of Justice, 1998
Buttell FP, Carney MM: Women Who Perpetrate Relationship Violence. Bing-
hamton, NY, Haworth Press, 2006
Campbell JC: Health consequences of intimate partner violence. Lancet
359:1331–1336, 2002
Carlile JB: Spouse assault on mentally disordered wives. Can J Psychiatry
36:265–269, 1991
Carlson E, McNutt LA, Choi DY: Childhood and adult abuse among women in pri-
mary health care: effects on mental health. J Interpers Violence 18:924–941, 2003
Chimbos PD: Spousal homicides in contemporary Greece. Int J Comp Sociol
39:213–223, 1998
Coker AL, Weston R, Creson DL, et al: PTSD symptoms among men and women
survivors of intimate partner violence: the role of risk and protective factors.
Violence Vict 20:625–643, 2005
Cromwell RE, Olson DH: Power in Families. New York, Wiley, 1975
Cusack KJ, Grubaugh AL, Knapp RG, et al: Unrecognized trauma and PTSD
among public mental health consumers with chronic and severe mental ill-
ness. Community Ment Health J 42:487–500, 2006
Daubert v. Merrell Dow Pharmaceuticals, 509 U.S. 579 (1993)
DeNiro DA: Perceived alienation in individuals with residual-type schizophre-
nia. Issues Ment Health Nurs 16:185–200, 1995
Dienemann J, Boyle E, Baker D, et al: Intimate partner abuse among women di-
agnosed with depression. Issues Ment Health Nurs 21:499–513, 2000
Dobash RE, Dobash R: Violence Against Wives. New York, Free Press, 1979
Dutton DG, Painter SL: Traumatic bonding: the development of emotional at-
tachments in battered women and other relationships of intermittent
abuse. Victimology 1:139–155, 1981
Dutton DG, Painter SL: Emotional attachments in abusive relationships: a test
of traumatic bonding theory. Violence Vict 8:105–120, 1993
Dutton MA: Intimate partner violence: 25 years of expert testimony. Presenta-
tion at the annual meeting of the American Academy of Psychiatry and the
Law, Chicago, IL, October 2006
Dutton MA, Golant SK: The Batterer: A Psychological Profile. New York, Basic
Books, 1995
El-Bassel N, Gilbert L, Witte S, et al: Intimate partner violence and substance
abuse among minority women receiving care from an inner city emergency
department. Womens Health Issues 13:16–22, 2003
Ferris LE, Norton PG, Dunn EV, et al: Guidelines for managing domestic abuse
when male and female partners are patients of the same physician. JAMA
278:851–857, 1997
Friedman SH, Loue S: Incidence and prevalence of intimate partner violence by
and against women with severe mental illness: a review. J Womens Health
16:471–480, 2007
Frye V: Examining homicide’s contribution to pregnancy-associated deaths.
JAMA 285:1510–1511, 2001
Gearon JS, Bellack AS: Women with schizophrenia and co-occurring substance
use disorders: an increased risk for violent victimization and HIV. Commu-
nity Ment Health J 35:401–419, 1999
Gelles RJ: An exchange/social theory, in The Dark Side of Families: Current
Family Violence Research. Edited by Finkelhor D, Gelles RJ, Hotaling GT,
et al. Beverly Hills, CA, Sage, 1983, pp 151–165
Goode W: Force and violence in the family. J Marriage Fam 33:624–636, 1971
Goodman LA, Dutton MA, Harris M: Episodically homeless women with seri-
ous mental illness: prevalence of physical and sexual assault. Am J Ortho-
psychiatry 65:468–478, 1995
Golding JM: Intimate partner violence as a risk factor for mental disorders: a
meta-analysis. J Fam Viol 14:99–132, 1999
Gormley B: An adult attachment theoretical perspective of gender symmetry in
intimate partner violence. Sex Roles 11/12: 785–795, 2005
Hamberger LK, Hastings J: Characteristics of male spouse abusers consistent
with personality disorders. Hosp Community Psychiatry 39:763–770, 1988
Hilberman E, Munson K: Sixty battered women. Victimology 2:460–470, 1987
Houry D, Sachs CJ, Feldhaus KM, et al: Violence-inflicted injuries: reporting
laws in the fifty states. Ann Emerg Med 39:56–60, 2002
Houry D, Kaslow NJ, Thompson MP: Depressive symptoms in women experi-
encing intimate partner violence. J Interpers Viol 20:1467–1477, 2005
Ibn-Tamas v. United States, 407A.2d 626 (1979)
Jacobson A, Richardson B: Assault experiences of 100 psychiatric inpatients:
evidence of the need for further inquiry. Am J Psychiatry 144:908–913,
1987
Jacobson NS, Gottman JM: When Men Batter Women: New Insights into Ending
Abusive Relationships. New York, Simon and Schuster, 1998
Jordan CE: Intimate partner violence and the justice system. J Interpers Viol
19:1412–1434, 2004
Kaplan ML, Asnis GM, Lipschitz DS, et al: Suicidal behavior and abuse in psy-
chiatric outpatients. Comp Psychiatry 36:229–235, 1995
Khan FI, Welch TL, Zillmer EA: MMPI-2 profiles of battered women in transi-
tion. J Pers Assess 60:100–111, 1993
Lenski G, Lenski J: Human Societies: An Introduction to Macrosociology. New
York, McGraw Hill, 1970
Loue S: Elder abuse and neglect in medicine and law: the need for reform. J Leg
Med 22:159–209, 2001a
Loue S: Intimate Partner Violence: Societal, Medical, Legal and Individual Re-
sponses. New York, Plenum, 2001b
Martin SL, English KT, Clark KA, et al: Violence and substance use among
North Carolina pregnant women. Am J Public Health 86:991–998, 1996
McFarlane J, Parker B, Soeken K, et al: Assessing for abuse during pregnancy:

severity and frequency of injuries and associated entry into prenatal care.
JAMA 267:3176–3178, 1992
Miller LJ, Finnerty M: Sexuality, pregnancy, and childbearing among women
with schizophrenia-spectrum disorders. Psychiatr Serv 47:502–506, 1996
Moffitt TE, Caspi A, Rutter M, et al: Sex Differences in Antisocial Behavior: Con-
duct Disorder, Delinquency, and Violence in the Dunedin Longitudinal
Study. Cambridge, United Kingdom, Cambridge University Press, 2001
Mueser KT, Goodman LB, Trumbetta SL, et al: Trauma and posttraumatic stress
disorder in severe mental illness. J Consult Clin Psychol 66:493–499, 1998
O’Leary KD: Physical aggression between spouses: a social learning theory per-
spective, in Handbook of Family Violence. Edited by Hasselt VB, Morrison
RL, Bellack AS, et al. New York, Plenum, 1988, pp 31–55
Oriel KA, Fleming MF: Screening men for partner violence in a primary care set-
ting: a new strategy for detecting domestic violence. J Fam Pract 64:393–
398, 1998
Rohner RP: They Love Me, They Love Me Not: A Worldwide Study of the
Effects of Parental Acceptance and Rejection. New Haven, CT, HRAF Press,
1975
Rusbult CE: Commitment and satisfaction in romantic situations: a test of the
investment model. J Exp Soc Psychol 16:172–186, 1980
Sachs CJ, Rodriguez MA: Should Physicians be required to report domestic vi-
olence to the police? West J Med 173:225, 2000
Sahay S, Piran N, Maddocks S: Sexual victimization and clinical challenges in
women receiving hospital treatment for depression. Can J Commun Ment
Health 19:161–174, 2000
Salyers MP, Evans LJ, Bond GR, et al: Barriers to assessment and treatment of
posttraumatic stress disorder and other trauma-related problems in people
with severe mental illness: clinician perspectives. Community Ment Health
J 40:17–31, 2004
Schuller RA, Rzepa S: Expert testimony pertaining to battered woman syn-
drome: its impact on jurors’ decisions. Law Hum Behav 26:655–673, 2002
Seligman MEP: Helplessness: On Depression, Development, and Death. San
Francisco, CA, WH Freeman, 1975
Soeken KL, McFarlane J, Parker B, et al: The abuse assessment screen, in Em-
powering Survivors of Abuse: Health Care for Battered Women and Their
Children. Edited by Campbell JC. Thousand Oaks, CA, Sage, 1998, pp 195–
203
Sorenson SB, Upchurch DM, Shen H: Violence and injury in marital arguments:
risk patterns and gender differences. Am J Public Health 86:35–40, 1996
Spitzberg BH: The tactical topography of stalking victimization and manage-
ment. Trauma Violence Abuse 3:261–288, 2002
Straus MA: Physical assaults by wives: a major social problem, in Current Con-
troversies in Family Violence. Edited by Gelles RJ, Loseke DR. Newbury
Park, CA, Sage, 1993, pp 67–87
Tang K: Battered woman syndrome testimony in Canada: its development and
lingering issues. Int J Offender Ther Comp Criminol 47:618–629, 2003
551 P.2d 334 (1976)
Tjaden P, Thoennes N: Full report of the prevalence, incidence, and conse-

quences of violence against women: findings from the National Violence
Against Women Survey. Washington, DC, U.S. Department of Justice, 2000
Walker LEA: The Battered Woman Syndrome, 2nd Edition. New York, Springer,
2000
Weingourt R: Wife rape in a sample of psychiatric patients. Image J Nurs Sch
22:144–147, 1990
Wilcoxen M: A Handbook for Health Professionals, Education Wife Assault.
Toronto, ON, Canada, 1985
Wilson M, Daly M: An evolutionary psychological perspective on male sexual
proprietariness and violence against wives. Violence Vict 8:271–294, 1993
Wolfgang ME, Ferracuti F: The Subculture of Violence: Toward an Integrated
Theory of Criminology. London, England, Tavistock, 1967
C H A P T E R 2 5
Workplace Violence
and the Clinician
Workplace violence has been a major concern for the American work-
place and public in recent decades. Although international terrorism has
been the focus of foreign policy and the media, workplace violence has
continued to be a leading concern of employers (Pinkerton Consulting
and Investigations 2003). The term workplace violence conjures up images
of disgruntled, armed employees wreaking havoc, killing and injuring
coworkers, and in most cases killing themselves or being killed in the
process of apprehension by police. Contrary to public perception and
media portrayals, such stereotypical workplace violence episodes have
decreased in frequency since 1994 (Bureau of Labor Statistics 2006a), as
has the overall rate of violent crime in the United States. Much more com-
mon is an array of behaviors that are justifiably included under a broad
definition of workplace violence or aggression: nonfatal assaults, bully-
ing, harassment (both sexual and nonsexual), verbal abuse, threats from
both known and anonymous sources, and hoaxes. Acts of terrorism with
conventional or chemical, biological, radiological, or nuclear weapons
can also be included as incidents of workplace violence, given that work-
places are common targets of terror attacks (Schouten et al. 2004).
The need to understand and manage acts of violence in the work-
place has given rise to multiple theories, proposed methods for iden-
tifying potentially violent employees, and prevention measures. Al-
though some, but by no means all, acts of workplace violence are
501
perpetrated by individuals with Axis I disorders, all such actions repre-

sent abnormal behavior. Therefore psychiatrists and other mental health
professionals are called upon for assistance in understanding, assessing,
managing, and preventing workplace violence. This chapter explores
some of the more common areas of workplace violence, with an empha-
sis on traditional concerns relating to fatal and nonfatal acts of interper-
sonal aggression and the challenges entailed in scientifically studying
them. It examines some of the popular myths and misconceptions relat-
ing to workplace violence, reviews approaches to preventing and man-
aging these events, and discusses the roles that mental health profes-
sionals can play in this area.
Case Example
ABCD Corporation, based in the northeastern United States, is a manu-
facturer and distributor of consumer electronics. Over the past 5 years,
ABCD has been gradually outsourcing its manufacturing operations,
service centers, and distribution networks. As a result, its U.S.-based
workforce has shrunk by 30% over 2 years, with rumors of additional
cuts in the near future. Most recently, plans were announced to move the
entire customer service operation to Bangalore, India, over the next year.
The U.S. customer service representatives have been offered the option
of transferring to ABCD’s manufacturing plant in Arkansas, taking an
early retirement package, or working until such time as the U.S. opera-
tion is closed.
Six weeks after the announcement that the customer service opera-
tion would be outsourced, a large envelope bearing excess postage and
addressed to the Chief Executive Officer (CEO) arrived in the mail.
ABCD’s corporate security director took possession of the envelope and
turned it over to the local hazardous materials team for screening prior
to opening. The envelope was opened when no indication of toxic sub-
stances was found. Inside were digital photographs of the CEO’s wife
and children leaving their home on the way to school. Appropriate se-
curity measures were put in place and a full-scale investigation was
opened by law enforcement. No suspicious behavior was observed near
the CEO’s home, and the investigation turned up no useful leads that
could tie the mailing to anyone in the community or at ABCD.
Word of the mailing spread quickly among ABCD employees. Em-
ployees began wondering what might happen next and speculated who
among them might have done this. Some commented that they were
certain something else would happen, given how the company had
“screwed all of us.” Several commented that the CEO was getting what
he deserved and that although they did not want to see anyone hurt, he
had a good scare coming to him, given the terror he was causing the em-
ployees by sending their jobs overseas.
Three weeks after the mailing, graffiti began appearing in the men’s
restroom. Obscene and graphic, and written in an awkward hand, the
Workplace Violence and the Clinician ❘ 503
graffiti made direct threats of violence toward the CEO and ABCD. Po-
lice investigation, including interviews with a number of employees
who used that restroom, yielded little. Through the interviews, police
and ABCD security learned that fear among employees, as well as anger
at ABCD for the downsizing and for failing to deal with the threats, was
growing. There was discussion of installing video cameras in the men’s
room and the hall leading to it, but corporate counsel advised that these
were prohibited by state privacy laws. Employees’ concerns increased
when they began to discover pieces of office equipment inexplicably
broken.
The list of suspects who might possibly be responsible for the letter,
graffiti, and sabotage was long. No one at ABCD was happy about the
employment situation. The economy was poor, good jobs were scarce in
the region, and most of the employees had families with young children.
Among the employees, the gossip was that James Wilson was a
likely suspect. James, a programmer in the information services divi-
sion, was slated to have his job eliminated when the outsourcing began.
Thirty years of age, James was regarded as a loner, did not socialize with
coworkers, and lived with his elderly parents. He took advantage of the
company flextime policy, often coming to work late and staying until
the early morning hours. James’s sole known recreational activity was
computer fantasy games, with which he was rumored to be “obsessed.”
An employee with whom James had attended high school told col-
leagues that James had owned a handgun in high school.
Gossip about James preoccupied the plant, much to the detriment of
productivity. The gossip increased when another letter arrived with a
note threatening that “If I go down, you all go down.” James was inter-
viewed by police and corporate security and denied involvement in any
of the events. As rumors about James continued, and employees began
complaining to human resources that he seemed “odd” and they were
afraid of him, a decision was made to send him for a fitness-for-duty
evaluation, even though his job performance had remained good. The
psychiatrist retained to perform the evaluation, Dr. Anderson, was told
that James was being referred because he “fit the profile” of someone
who might commit an act of workplace violence.
Dr. Anderson found James to be an anxious young man with symp-
toms of obsessive-compulsive and avoidant personality disorders.
There was no indication of psychosis, mood disorder, or other mental
disorders, nor was there any indication of hostility toward ABCD or his
coworkers. Interviews with James’s supervisor revealed that he was
quiet and diligent, was irritable at times and avoided other employees,
but had never been threatening or violent. His personnel record was un-
remarkable; his criminal background check was clean. The expressions
of concern from coworkers were vague, unsubstantiated, and based on
their perception that he was “odd” and had a history of mental illness.
The psychiatrist concluded that James did not pose a risk of harm to
ABCD or its employees and found him fit for duty. Upon his return, a
number of coworkers complained to Human Resources that they did not
care what the evaluator said, they were still concerned. Several threat-
ened to not come to work. ABCD decided to eliminate James’s position

ahead of schedule and offered him the standard severance package.
Three weeks after the evaluation was completed, ABCD announced
the timetable for transitioning customer service operations to Bangalore.
One week later, a customer service representative visited the director of
Human Resources and hesitantly reported that one of her colleagues,
Bill Smith, had flown into a rage when he received the announcement
of the timetable by e-mail. Sitting in his cubicle, he had begun cursing
ABCD and the CEO, stating that he just was “not going to take it.” At
lunch, he told colleagues that he was “fed up with the B.S.” from ABCD
and talked about how losing his job was going to destroy his family. The
Smiths’ son had multiple medical problems, his wife stayed home to
care for the child, and he had no idea what they would do for healthcare
coverage when he lost his job. He talked about taking out a substantial
life insurance policy “just in case something happened” to him. The co-
worker also talked about how Bill muttered that “ABCD ain’t seen noth-
ing yet” and that the “fun” was just about to begin.
A background investigation of Bill revealed that he was a 48-year-
old married father of two children, ages 10 and 12, and a college gradu-
ate. Bill had no record of criminal charges, although local police had
been called to the home on two occasions related to domestic disputes.
Colleagues commented that Bill had drunk excessively at company
events over the previous year. In addition, his personnel file contained
a verbal and a written warning for verbally abusive behavior toward co-
workers.
A meeting was called of ABCD’s security director, Human Re-
sources director, and the company attorney to decide how to proceed. A
decision was made to question Bill Smith, who confessed that he was re-
sponsible for all of the incidents and explained that he had acted only
after becoming severely depressed over the upcoming changes. Bill told
the security director that he felt he had nothing to lose and that this last-
ditch effort might get ABCD to at least reconsider what it was doing. In
the meantime, James Wilson had retained an attorney, who filed a dis-
ability discrimination complaint with the appropriate agency on his be-
half as well as a civil suit alleging defamation.
This case example, a composite of actual cases, represents a range of

behaviors that fall within the parameters of workplace violence. Al-
though definitions of workplace violence vary, a commonly used defini-
tion is the one offered by the National Institute of Occupational Safety
and Health, which defines workplace violence as “violent acts, including
physical assaults and threats of assault, directed toward persons at work
or on duty” (National Institute of Occupational Safety and Health 1996).
The case example contains multiple events that would be considered
acts of workplace violence under this definition: the threat contained in
the mailing, the graffiti, and Bill’s implied threats contained in his state-
ments to coworkers. Broader definitions tend to use the term workplace
aggression and would encompass Bill’s verbal abuse of coworkers and

the sabotage (Griffin and Lopez 2005; Neuman and Baron 1998).
Inclusive definitions of workplace violence are useful in that they
convey the continuum along which acts of workplace aggression lie and
raise awareness of the impact of nonviolent aggressive behavior. How-
ever, they can also cause confusion when it comes to conducting and
reading research in the field and developing response strategies (Wad-
dington et al. 2005).
The Knowledge Base for Workplace Violence

Prevalence
The Bureau of Labor Statistics (BLS) of the U.S. Department of Labor
has maintained statistics on both fatal and nonfatal workplace injuries
for many years. BLS data going back to 1992 indicate that workplace ho-
micides peaked in 1994 and have declined overall since then (Bureau of
Labor Statistics 2006a). Overall, in 2004 and 2005, workplace homicides
dropped from third to fourth place among the most common causes
of workplace death, behind motor vehicle accidents, falls from high
places, and being struck by an object (Bureau of Labor Statistics 2006a).
It should be kept in mind that the BLS database counts individual ho-
micides rather than incidents. Thus a single event in which five people
die (four victims and the perpetrator) would be counted as five homi-
cides, meaning that the 567 workplace homicides reported by BLS for
2005 actually represent a lower number of individual incidents.
Workplace violence is divided into four types (University of Iowa
Injury Prevention Research Center 2001). Type I violence includes all in-
cidents committed by individuals who have no connection to the work-
place. These events primarily occur in the course of robberies or other
crimes, but also include acts of terrorism such as the attacks of Septem-
ber 11, 2001. Those workplace homicides were not, however, included
in the CFOI data for 2001 (Bureau of Labor Statistics 2006a) because of
their unique characteristics.
Type II violence includes acts perpetrated by customers or clients.
This includes, for example, assaults by patients on clinicians, by clients
on lawyers, and by bank customers on tellers. The events that garner
the most attention—assaults by current and former coworkers—fall
within Type III. Finally, Type IV acts are those carried out in the work-
place by relatives, current and former spouses or partners, and other ac-
quaintances. These incidents primarily include domestic violence situ-
ations in which the perpetrator seeks out the victim in the workplace.
This category has been getting increased attention as awareness of the

significance of domestic violence grows.
Since BLS began keeping its Census of Fatal Occupational Injuries
data in 1992, Type I incidents have constituted the overwhelming ma-
jority of workplace homicides. From 1997 to 2005, Type I events ac-
counted for 78% of workplace homicides. Type II events accounted for
5%, Type III for 10%, and Type IV for 7% (Bureau of Labor Statistics
2007). The Bureau of Justice Statistics has estimated that approximately
1.7 million individuals are directly affected by acts of workplace vio-
lence annually, although this is likely an underestimate (Duhart 2001).
Compared with workplace homicides, which are reported to law en-
forcement, estimates of nonfatal acts of violence or aggression in the
workplace are more difficult to obtain. According to the Bureau of Jus-
tice Statistics, as many as 27% of individuals who are physically as-
saulted at work never report it to anyone, and 52% never report the
event to law enforcement. Like homicides in the workplace, the inci-
dence of nonfatal assaults declined across all occupations from 1993 to
1999, with decreases of 51% for medical workers and 28% for mental
health workers (Duhart 2001).
Nonfatal workplace assaults appear to be distributed among the
four types in proportions somewhat different from workplace homi-
cides. A study of nonfatal workplace violence incidents reported to a
police department by Scalora et al. (2003) found that 53% of the inci-
dents were Type I, occurring at the hands of external sources. Type II
(customers, clients, or patients) accounted for 13.8% of the sample, and
Type III (coworkers) accounted for 11%. Type IV incidents (related indi-
viduals) composed 22% of the sample.
In a study of workplace violence prevention focusing on 2005, BLS
surveyed 7.4 million employers in the United States, covering 128 mil-
lion employees in private industry and state and local governments
(Bureau of Labor Statistics 2006c). The survey gathered information on
a number of aspects of workplace violence, including 1) fatal and non-
fatal workplace violence incidents of all types, as defined by physical
assaults, threats of assault, harassment, intimidation, or bullying; 2) the
impact of such incidents; 3) steps taken in response to the incidents; and
4) the existence of workplace violence policies and procedures.
The BLS study found that 5.3% of all employers reported a violent in-
cident in 2005, although almost 50% of the largest employers (with more
than 1,000 employees) reported an incident. Only 4.8% of all private em-
ployers reported an incident, whereas 32% of state government and
14.7% of local government respondents reported at least one incident
(Bureau of Labor Statistics 2006c). Approximately equal percentages
(2%) of private industry establishments reported incidents in Type I–III

categories, with a smaller percentage (0.8%) reporting a Type IV inci-
dent. Among state employers, Type III incidents were reported by 17.5%
of employers, Type II by 15.4%, Type I by 8.7%, and Type IV by 5.5.%.
Among local government employers, however, the distribution was
somewhat different: Type II, 10.3%; Type III, 4.3%; Type I, 3.7%; and
Type IV, 2.1% (Bureau of Labor Statistics 2006c). The explanation for
these disparities is open to speculation.
From a slightly different perspective, it is worth noting that of all the
nonfatal occupational injuries and illnesses resulting in lost days from
work in private industry, assaults and violent acts accounted for 1% of
the total in 2005 (Bureau of Labor Statistics 2006b). This represented a
decrease of 18% from 2004. The vast majority (96%) of these occurred in
service-providing establishments; 67% occurred in the healthcare in-
dustry and were primarily Type II (patient or client) incidents (Bureau
of Labor Statistics 2006b).
In the case example, ABCD is confronted with threats and potential
assaults that cannot be categorized when they first occur. References to
the employment situation strongly suggest that the source is internal,
that is, Type III. However, until more investigation is done, it would be
premature to rule out the possibility that the threat was coming from an
outside source, such as a member of the community who might be an-
gry about ABCD’s plans (Type I), a customer or client (Type II), or a
family member of an employee who will be affected by the changes
(Type IV).
Legal Risks Associated With Workplace Violence

Like physicians, employers are concerned with legal liability. Work-
place violence is an area fraught with legal risks, ranging from the re-
quirements of regulatory agencies to civil and criminal liability. A brief
summary of these legal risks is in order.
The General Duty Clause of the Occupational Safety and Health Act
(OSHA) requires that employers maintain a workplace that is free of
safety hazards that are known, or should be known, to the employer
(Occupational Safety and Health Act of 1970). State laws impose similar
requirements. Regulations promulgated under OSHA also require that
employers have a disaster plan in place (Occupational Safety and
Health Standards 2001). In the case example, failure to respond to any
of the potential threats could result in civil penalties if actual harm oc-
curred or if other OSHA violations, such as a lack of a disaster plan,
were detected.
Employers are responsible for providing workers’ compensation

benefits to employees who suffer injuries arising from work-related ac-
tivities. In many states, this includes stress-related injuries and ill-
nesses. If ABCD employees were to be disabled from work due to anx-
iety related to the threat situation, ABCD’s workers’ compensation
carrier would likely have to provide benefits to the affected employees.
Although this is not a liability issue per se, ABCD could find its work-
ers’ compensation insurance premiums increasing as a result.
Traditional common-law negligence claims may pose the biggest
risk to employers in this area. These could take the form of premises li-
ability (Kohler v. McCrory Stores 1992), negligent hiring (employment of
a violent individual whose potential was known or easily could have
been known), negligent retention (retaining a violent employee after
the propensity is known), negligent supervision, or vicarious liability
(Brakel 1998; Elzen 2002).
Individuals wrongly suspected or accused of threats or violent acts
may also bring suit against the employer under a variety of theories.
These can include actions for defamation (Morgan v. Bubar, et al. 2006),
violation of privacy and other civil rights (Pettus v. Cole 1996), and dis-
ability discrimination. The Americans with Disabilities Act and analo-
gous state statutes prohibit disparate treatment of individuals on the
basis of a current disability, past history of a disability, or the perception
that they have a disability (Americans With Disabilities Act 1990). In the
case example, James could allege that ABCD eliminated his position
prematurely because he was perceived as having a mental disability
and therefore posed a risk of violence (Laden and Schwartz 2000). The
Americans with Disabilities Act does not protect individuals with dis-
abilities from discipline if they violate workplace rules, even if the vio-
lation was the result of the disability (Hamilton v. Southwestern Bell 1998;
Mammone v. President and Fellows of Harvard College 2006; Palmer v. Cir-
cuit Court of Cook County 1998). Thus, Bill can be terminated for his ac-
tions, even if he could prove that his behavior was somehow linked to
a disability, such as depression. Employers may require their employees
to meet certain qualification standards, and these can include a require-
ment that the employee not pose a direct threat to the health or safety
of other individuals in the workplace (Americans With Disabilities Act
1990; Jones v. American Postal Workers Union 1999).
Risk Factors for Workplace Violence

Efforts to study the causes, perpetrators, and victims of workplace vio-
lence have been hampered by a number of methodological problems.
First, there are a limited number of cases of workplace homicides to study.

Second, nonfatal workplace assaults are underreported. Third, there is a
shortage of perpetrators to study, because they are either unwilling to
participate or unavailable due to death by suicide or in the course of ap-
prehension. Fourth, the quality of information available from individual
events is not of uniformly high quality. Much of it comes from sensation-
alized media reports that are often premature in their analysis or from
court documents that necessarily reflect biased adversarial views of the
perpetrator (e.g., opposing experts testifying on the mental state of the
perpetrator in pursuit of an insanity defense). Fifth, organizations af-
fected by acts of workplace violence strive to maintain their confidentially
and tend not to make themselves available for post hoc analysis. Finally,
prospective controlled studies of employee and organizational character-
istics, with random assignment to intervention strategies, are virtually
impossible to conduct due to ethical and legal considerations.
Considerable attention has been given to “profiles” of workplace
violence perpetrators that would allow individuals who pose a risk of
violence to be identified in advance and either excluded from the work-
place or subjected to specific interventions to prevent violence. The
problems with these profiles are well known and recognized by legiti-
mate experts in the field, yet the fact that a given individual “fits the
profile” is a common basis for concern by employers and coworkers
(American Society of Industrial Security 2005; Association of Threat As-
sessment Professionals 2005; Federal Bureau of Investigation 2002). The
fundamental problem with the profiling approach to workplace vio-
lence is the same as that encountered when mental health professionals
attempt to predict violence of any type, including suicide. Specifically,
when there is a low-incidence phenomenon, even a highly sensitive test
will result in an unacceptably high level of false positives—in effect,
many individuals will be falsely identified as being at risk (Rosen 1954).
A related problem is that none of the proffered profiles of workplace vi-
olence perpetrators has ever been empirically tested; that is, proposed
traits have not been assessed for their base rate among perpetrators as
well as in a matched control group of non-perpetrators.
In addition to the false-positive problem, profiles also create a prob-
lem with false negatives that mistakenly rule out true positives. This
problem is apparent in the case example: the traditional profile that fo-
cuses on whether an individual has a mental illness (often translated as
being “odd” in the eyes of his coworkers), is a “loner,” lives with his
parents, and may have owned a weapon made James the primary sus-
pect and object of concern. Indeed, had Bill’s coworker not come for-
ward to report his threatening behavior, James might have remained
the primary suspect. Action taken against an employee based on the

employee’s fitting a profile can provide a basis for a discrimination
claim, as in James’s case, where the action arises from the belief that the
person has a mental disability and fails to undertake an individual anal-
ysis under the direct-threat provision (Laden and Schwartz 2000).
The absence of a reliable profile that allows for prospective identifica-
tion of those who have engaged in acts of violence or are likely to do so
has been demonstrated in two excellent studies conducted under the
auspices of the U.S. Secret Service. These studies, which looked at would-
be and actual assassins (Fein and Vossekuil 1999) and school shooters
(Vossekuil et al. 2000), revealed the diversity of individuals who engaged
in such activities, amply demonstrating the absence of a set profile.
Identified Risk Factors

As outlined, there are limitations to the profiling approach and to the re-
search on workplace violence. Nevertheless, research studies of nonfa-
tal workplace aggression have identified characteristics of individuals
who engage in aggressive acts in the workplace. Similar studies have
been conducted of the characteristics of organizations that are victims
of such acts. Importantly, these studies, summarized here, do not pur-
port to identify the base rate of similar characteristics in the community.
As a result, these studies do not address the false-positive problem and
the difficulties that arise from it. A partial listing of identified risk fac-
tors is provided in the following discussion.
Individual Risk Factors

Individual and organizational risk factors for nonfatal Type III aggres-
sion have been identified in a number of studies. Greenberg and Barling
(1999) found that a past history of aggression and quantity of alcohol
consumed were positively correlated with aggression against cowork-
ers, whereas perceptions of unjust treatment and workplace surveil-
lance were related to aggression against supervisors. Their findings re-
garding the role of perceived victimization and injustice have been
confirmed by other researchers (Ambrose et al. 2002; Aquino and Brad-
field 2000; Aquino and Douglas 2003; Baron et al. 1999; Dupre and Bar-
ling 2006; Jockin et al. 2001; Skarlicki and Folger 1997). Other individual
risk factors for workplace aggression include trait anger (Chen and
Spector 1992; Douglas and Martinko 2001); threat to identity (Aquino
and Douglas 2003); hostility, low frustration tolerance, and reactivity to
stress (Chen and Spector 1992; Jockin et al. 2001; Storms and Spector
1987); negative affectivity (Penney and Spector 2005); thinking that
revenge is justified and having a tendency to blame others for personal

problems (Douglas and Martinko 2001); and a history of antisocial be-
havior (Jockin et al. 2001; Warren et al. 1999). It has also been suggested
that extremes of temperature and resultant changes in adrenaline level
may be related to workplace aggression (Simister and Cooper 2005).
Various environmental factors and acute stressors, some of which
are contained in the case example, have been associated with workplace
aggression. These include pay cuts or freezes (Baron and Neuman
1996), termination (Allen and Lucero 1998), and low level of control
over one’s job (Storms and Spector 1987). Among individuals with
mental illness, Haggard-Grann et al. (2006) found an increased risk of
criminal violence in relationship to suicidal ideation or parasuicide
within 24 hours before the violent event, hallucinations, acute conflicts
with others, and being denied psychiatric care within 24 hours before
the event. Notably, violent ideation did not appear to be associated with
increased risk, and paranoid ideation was associated with a small and
statistically nonsignificant risk.
The relationship between mental illness and violence is explored in
detail elsewhere in this volume. It is worth noting here that to date, no
empirical studies have identified severe mental illness as a risk factor
for workplace homicide specifically. Nevertheless, the presence of such
illnesses among some perpetrators of workplace violence is evidenced
by case review and my personal experience. Recent research on vio-
lence and mental illness indicates that active psychotic illness, coupled
with symptoms of paranoia and a past history of either conduct dis-
order or antisocial personality disorder, increases the risk of violent
behavior (Brennan et al. 2000; Hodgins 2006; Swanson et al. 2006). In a
community sample, examination of associations between psychotic-
like experiences and interpersonal violence in individuals without se-
vere mental illness revealed that such experiences are associated with
an increased risk of assault with intent to harm, intimate partner vio-
lence, and arrests for assault, with paranoid ideation and unusual expe-
riences such as visions serving as particular risk factors (Mojtabai 2006).
Substance abuse has also been identified as a major risk factor for vio-
lence among individuals with and without Axis I disorders (Chen and
Spector 1992; Jockin et al. 2001; Pastor 1995).
Organizational Risk Factors

Some of the organizational risk factors for workplace violence appear to
include pay cuts or freezes, use of part-time employees, changes in
management, reengineering, budget cuts, deteriorating physical work-
place environment (Baron and Neuman 1996), low work group har-
mony (Cole et al. 1997), and failure to discipline aggressive employees
(Allen and Lucero 1998). Again, some of these are found in the case ex-
ample. Karl and Hancock (1999) proposed that organizations are at in-
creased risk of workplace aggression if they conduct terminations with
more than one supervisor present or on a Monday or Tuesday.
Classification of Factors as Static or Dynamic

The risk factors for workplace violence can be broadly divided into fac-
tors that are historical or static—that is, unchangeable—and those that are
dynamic—that is, fluctuating and potentially modifiable (Douglas and
Skeem 2005; Mills 2005; Philipse et al. 2006). Historical features include in-
dividual risk factors such as trait anger and history of antisocial behavior.
Dynamic features can include organizational characteristics, life stressors,
and illness. This characterization of risk factors is of importance in the
process of assessing potential threatening situations and managing them.
Threat Assessment in Workplace Violence

Psychiatrists and other mental health professionals may be called upon
to assess the risk of violence in a specific work-related situation. These
consultations most commonly involve assessment of risk posed by an
identified current or former customer or client (Type II), an identified
current or former employee (Type III), or a person related to an em-
ployee, usually through a domestic relationship (Type IV). Less fre-
quently, the consultant may be asked to assess risk posed by an anony-
mous threatener or to assist in identifying an anonymous threatener.
As in any other consultation, the prospective consultant should first
define the exact consultation question and objectively determine
whether she or he is qualified to fulfill the request. Violence risk assess-
ment involving mental illness is a common feature of psychiatric resi-
dency and clinical psychology training, and much of mental health clin-
ical practice revolves around this activity. It is important to keep in
mind, however, that workplace violence risk assessments are not stan-
dard clinical evaluations. They differ from the clinical task in 1) their
purpose, 2) the party to whom responsibility is owed, 3) the database
that may be available, and 4) the range of options available. Each of
these elements is discussed in greater detail below.
Workplace violence risk assessments are forensic consultations, re-
quested by and for the benefit of third parties—not clinical evaluations
conducted for the benefit of the evaluee. Ideally, the evaluation will have
a beneficial impact on all parties involved, but the goal is not to diagnose
and provide treatment, or even a treatment referral, for the evaluee. The
goal is to determine the level of risk to which the referring party is ex-
posed and to assist in managing that risk. Although clinicians have a
duty to behave in an ethical manner no matter what their role, the fidu-
ciary duty that arises from the doctor–patient relationship to act only in
the best interests of the patient is owed not to the evaluee but to the
party requesting the assessment (Schouten 1993; Strasburger et al. 1997).
The evaluating clinician has a duty to the evaluee to disclose informa-
tion to the employer only with consent (American Academy of Psychi-
atry and the Law 2005), although the employer may make participation
in the evaluations and release of information a contingency of any fu-
ture employment. Even so, the evaluating clinician should only disclose
information on a need-to-know basis and should be aware of the extent
to which Health Insurance Portability and Accountability Act privacy
requirements apply to these evaluations (Gold and Metzner 2006).
The databases available in workplace violence consultations and
clinical evaluations differ, with each more limited and more complete in
certain ways. Clinical evaluations occur in person, whereas the work-
place violence consultant may never meet the subject face to face. On
the other hand, the consultant will ideally have more comprehensive in-
formation from a wider range of sources than is available to the average
clinical evaluator. In clinical settings, the primary sources of informa-
tion are generally limited to the patient him- or herself and perhaps
family members. Time and location permitting, the outpatient evalua-
tor may have access to the evaluee’s medical records. In contrast, the
workplace violence consultant will often make the risk determination
on the basis of background information, interviews with collateral
sources, review of documents and other communications, and often
comprehensive background checks, which can provide the basis for a
structured assessment based on actuarial risk factors.
It is also important to keep in mind that in workplace violence risk
assessments, only a small proportion of subjects will have identifiable
Axis I or even Axis II disorders. Thus, options available to the clinical
evaluator may not be accessible to the consultant. For example, volun-
tary or involuntary hospitalization may be available for an individual
who is engaging in threatening behavior as a result of a mental illness.
In the absence of such an illness, or sufficient risk of harm to justify in-
voluntary commitment, alternative solutions to maintaining safety
must be found. This may require a decision as to whether the matter
should be referred to law enforcement or if a legal action, such as a re-
straining order, should be pursued.
The Threat Assessment Process

Psychiatry is a solitary profession. Although clinical teams are respon-
sible for treatment in institutional settings, psychiatrists continue to be
trained in a largely dyadic model in which the primary interactions are
between doctor and patient or several patients in group therapy. Even
when the clinician is working on a team, the other team members tend
to be other mental health professionals. When called upon to render an
opinion regarding risk of violence in a workplace setting, we tend to fall
back into the familiar practice of attempting to solve the problem pre-
sented to us by relying on our individual skills and experience.
Those who would provide consultation on workplace violence con-
sultations are well advised to abandon the traditional individualistic
approach and to seek the input of team members from diverse back-
grounds (Schouten 2003, 2006). Mental health professionals are most ef-
fective in this regard when they serve as members of or consultants to
threat management teams established by organizations to handle crises
that may arise. These teams are also referred to as “threat assessment”
or “crisis management” teams, and they are common elements of work-
place violence policies—where such policies exist. BLS has reported
that 70% of establishments had no workplace violence policies as of
2005 (Bureau of Labor Statistics 2006c). Commonly composed of indi-
viduals from corporate security, human resources, and legal depart-
ments, threat management teams are ideally on call around the clock,
365 days a year, and are responsible for implementation of policies, liai-
son with law enforcement, and investigation and management of inci-
dents (American Society of Industrial Security 2005; Schouten 2003).
The team is designed in such a way that consultants such as mental
health professionals can be called in as needed (Schouten 2003).
The role for mental health professionals on the team will vary de-
pending on the sophistication of the team and the experience of the con-
sultant. In some cases, the psychiatrist or other mental health profes-
sional who possesses the necessary knowledge and experience will take
the lead; in others, the process will be led by experienced corporate se-
curity or law enforcement professionals. It has been my experience that
the best results and most satisfying experiences are the product of a
group process in which individuals from different disciplines, with mu-
tual respect for each other, share their knowledge and perspectives.
Invariably, the rest of the team looks to the mental health expert for
an assessment of the subject’s level of risk. It is up to the expert to un-
dertake an assessment involving a variety of factors, both historical and
dynamic, that may increase or mitigate risk. In this setting, as in tradi-
tional clinical settings, it is important to keep in mind that violent

behavior is the product of interactions among three sets of factors:
1) individual factors, 2) triggers, and 3) environmental factors—that is,
whether the environment encourages or dissuades potential violence
(Borum et al. 1999; Fein and Vossekuil 1995). A number of such factors
were discussed earlier in this chapter, as was the distinction between
historical and dynamic factors. Thus, in conducting the risk assessment
and advising the team, the consultant is examining a broad array of in-
formation, much of which comes from sources other than the evaluee.
The ideal threat assessment includes not only a broad range of in-
formation from background investigation and collateral sources but
also an opportunity to interview the subject of concern and to apply a
structured or semistructured approach, utilizing available instruments
where applicable. This approach, which combines actuarial and clinical
risk assessment models, is both flexible and more likely to be accurate
(Association of Threat Assessment Professionals 2005; Borum et al.
1999). The risk assessment itself will be framed in terms of relative risk,
as opposed to percentage of likelihood.
Once the risk assessment is conducted, the consultant may be called
upon to utilize skills that are more closely linked to clinical skills, such
as understanding the motivation and behavior of the subject of concern
and suggesting ways in which the situation can be managed to decrease
the likelihood of harm to all concerned. As with other situations in
which psychiatrists are applying their clinical skills in the interests of
third parties rather than the person being evaluated, there are ethical
concerns that must be considered (Arboleda-Florez 2006; Janofsky 2006;
Sen et al. 2007).
The consultant who undertakes these consultations, such as Dr.
Anderson in the case example, can avoid ethical transgressions by fol-
lowing a few basic principles. These include limiting disclosure of pre-
viously unknown information to need-to-know situations or those
where there is no expectation of privacy on the part of the subject; mod-
eling respectful behavior for all individuals; combating stereotypes of
mental illness; discouraging discrimination; and pursuing solutions
that protect the client requesting the consultation without unnecessary
harm to the individual of concern.
Assuming in the case example that a consultant finds that Bill Smith
poses a moderate risk of violence under the existing circumstances, the
challenge is to design an intervention that decreases the immediate risk
of violence, protects ABCD, and decreases the risk going forward. This
would likely include consultation with security staff on protective mea-
sures for ABCD and individual employees. Whether to discharge Bill
Smith will be a business decision that takes into consideration not only
risk issues but also the impact of retaining Mr. Smith when his behavior
is known to other employees. From a violence risk management stand-
point, however, all parties may be best served by a mandatory medical
leave, possible short-term disability, and ongoing provision of health
insurance benefits. In the event that Mr. Smith is to be discharged, a sev-
erance package that includes severance pay and ongoing health bene-
fits can decrease the stress of termination, especially in a situation like
his where family health issues constitute an ongoing stressor.
Key Points
■ Workplace violence is a subject that has captured the imagination
of the public and the media due to the human drama and tragedy
involved.
■ Because this type of violence is about abnormal behavior, psychia-
trists and other mental health professionals will continue to be
called upon to provide risk assessments, develop prevention
strategies, and suggest methods to mitigate risks.
■ As mental health professionals, we have much to offer in terms of
workplace risk assessment, prevention, and mitigation. Individu-
als who choose to pursue this rewarding professional activity can
do so effectively so long as they undertake the proper training,
understand the tasks ahead of them, and stay within the bound-
aries of knowledge, skill, and professional ethics.
References
Allen R, Lucero M: Subordinate aggression against managers: empirical analy-
ses of published arbitration abstracts. The International Journal of Conflict
Management 9:234–257, 1998
Ambrose ML, Seabright MA, Schminke M: Sabotage in the workplace: the role of
organizational injustice. Organ Behav Hum Decis Process 89:947–965, 2002
American Academy of Psychiatry and the Law: Ethics Guidelines for the Prac-
tice of Forensic Psychiatry. Bloomington, CT, American Academy of Psy-
chiatry and the Law, 2005
American Society of Industrial Security: Workplace Violence Prevention and
Response Guideline. Alexandria, VA, American Society of Industrial Secu-
rity, 2005
Americans With Disabilities Act, 42 U.S.C. § 12113(b) (1990)
Aquino K, Bradfield M: Perceived victimization in the workplace: the role of sit-
uational factors and victim characteristics. Organization Science 11:525–
537, 2000
Aquino K, Douglas S: Identity threat and antisocial behavior in organizations:

the moderating effects of individual differences, aggressive modeling, and
hierarchical status. Organ Behav Hum Decis Process 90:195–208, 2003
Arboleda-Florez JE: The ethics of forensic psychiatry. Curr Opin Psychiatry
19:544–546, 2006
Association of Threat Assessment Professionals: Violence Risk Assessment
Guideline: Standard Considerations for Assessing the Risk of Future Vio-
lent Behavior. Huntington Beach, CA, Association of Threat Assessment
Professionals, 2005
Baron R, Neuman J: Workplace violence and workplace aggression: evidence on
their relative frequency and potential causes. Aggress Behav 22:161–173, 1996
Baron RA, Neuman JH, Geddes D: Social and personal determinants of work-
place aggression: evidence for the impact of perceived injustice and the
Type A behavior pattern. Aggress Behav 25:281–296, 1999
Borum R, Fein R, Vossekuil B, et al: Threat assessment: defining an approach for
evaluating risk of targeted violence. Behav Sci Law 17:323–337, 1999
Brakel SJ: Legal liability and workplace violence. J Am Acad Psychiatry Law
26:553–562, 1998
Brennan PA, Mednick SA, Hodgins S: Major mental disorders and criminal vi-
olence in a Danish birth cohort. Arch Gen Psychiatry 57:494–500, 2000
Bureau of Labor Statistics: National Census of Fatal Occupational Injuries in
2005. Washington, DC, U.S. Department of Labor, 2006a
Bureau of Labor Statistics: Nonfatal Occupational Injuries and Illnesses Requir-
ing Days Away from Work, 2005. Washington, DC, U.S. Department of La-
bor, 2006b
Bureau of Labor Statistics: Survey of Workplace Violence Prevention, 2005.
Washington, DC, U.S. Department of Labor, 2006c
Bureau of Labor Statistics: Census of Fatal Occupational Injuries: Occupational
Homicides by Selected Characteristics, 1997–2005. Washington, DC, U.S.
Department of Labor, 2007
Chen P, Spector P: Relationships of work stressors with aggression, withdrawal,
theft and substance use: an exploratory study. Journal of Occupational and
Organizational Psychology 65:177–184, 1992
Cole L, Grubb P, Sauter S, et al: Psychosocial correlates of harassment, threats
and fear of violence in the workplace. Scand J Work Environ Health 23:450–
457, 1997
Douglas KS, Skeem JL: Violence risk assessment: getting specific about being
dynamic. Psychol Public Policy Law 11:347–383, 2005
Douglas SC, Martinko MJ: Exploring the role of individual differences in the
prediction of workplace aggression. J Appl Psychol 86:547–559, 2001
Duhart DT: Violence in the Workplace, 1993–1999 (Report No.: NCJ 190076).
Washington, DC, U.S. Department of Justice, 2001
Dupre KE, Barling J: Predicting and preventing supervisory workplace aggres-
sion. J Occup Health Psychol 11:13–26, 2006
Elzen WL: Workplace violence: vicarious liability and negligence theories as a
two- fisted approach to employer liability. Is Louisiana clinging to an out-
moded theory? La Law Rev 62:897–928, 2002
Federal Bureau of Investigation: Workplace Violence: Issues in Response. Quan-
tico, VA, U.S. Department of Justice, 2002
Fein RA, Vossekuil B: Threat Assessment: An Approach to Prevent Targeted Vi-

olence. Rockville, MD, National Institute of Justice, 1995
Fein RA, Vossekuil B: Assassination in the United States: an operational study
of recent assassins, attackers, and near-lethal approachers. J Forensic Sci
44:321–333, 1999
Gold LH, Metzner JL: Psychiatric employment evaluations and the Health In-
surance Portability and Accountability Act. Am J Psychiatry 163:1878–
1882, 2006
Greenberg L, Barling J: Predicting employee aggression against coworkers, sub-
ordinates and supervisors: the roles of person behaviors and perceived
workplace factors. Journal of Organizational Behavior 20:897–913, 1999
Griffin RW, Lopez YP: “Bad behavior” in organizations: a review and typology
for future research. J Manage 31:988–1005, 2005
Haggard-Grann U, Hallqvist J, Langstrom N, et al: Short-term effects of psychi-
atric symptoms and interpersonal stressors on criminal violence: a case-
crossover study. Soc Psychiatry Psychiatr Epidemiol 41:532–540, 2006
Hamilton v. Southwestern Bell, 136 F.3d 1047 (5th Cir. 1998)
Hodgins S: Crime and violence by persons with mental illness: new evidence re-
quires new mental health policies and practices. Neuropsychiatrie 20:7–14,
2006
Janofsky JS: Lies and coercion: why psychiatrists should not participate in po-
lice and intelligence interrogations. J Am Acad Psychiatry Law 34:472–478,
2006
Jockin V, Avery R, McGue M: Perceived victimization moderates self-reports of
workplace aggression and conflict. J Appl Psychol 86:1262–1269, 2001
Jones v. American Postal Workers Union, 192 F.3d 417 (4th Cir. 1999)
Karl K, Hancock B: Expert advice on termination practices: how expert is it?
Public Pers Manage 28:51–62, 1999
Kohler v. McCrory Stores, 615 A.2d 27 (Pa. 1992)
Laden VA, Schwartz G: Psychiatric disabilities, the Americans with Disabilities
Act, and the new Workplace Violence Account. Berkeley Journal of Em-
ployment and Labor Law 21:246–270, 2000
Mammone v. President and Fellows of Harvard College, 276 N.E.2d 847 (Mass. 2006)
Mills JF: Advances in the assessment and prediction of interpersonal violence.
J Interpers Violence 20:236–241, 2005
Mojtabai R: Psychotic-like experiences and interpersonal violence in the general
population. Soc Psychiatry Psychiatr Epidemiol 41:183–190, 2006
Morgan v. Bubar, et al. Conn. Super. 1828 (2006)
National Institute of Occupational Safety and Health: Violence in the Work-
place: Risk Factors and Prevention Strategies. DHHS (NIOSH) Publication
No. 96-100. Cincinnati, OH, National Institute of Occupational Safety and
Health, 1996. Available at http://www.cdc.gov/niosh/violcont.html. Ac-
cessed July 15, 2007
Neuman JH, Baron RA: Workplace violence and workplace aggression: evi-
dence concerning specific forms, potential causes, and preferred targets.
J Manage 24:391–419, 1998
Occupational Safety and Health Act of 1970. 29 U.S.C. § 654 (5)(a) (1970)
Occupational Safety and Health Standards, Subpart E—Means of Egress.
29 C.F.R. § 1910.38 (2001)
Palmer v. Circuit Court of Cook County, 351 F.3d 117 (7th Cir. 1997) cert denied, 522
U.S. 1096 (1998)
Pastor LH: Initial assessment and intervention strategies to reduce workplace
violence. Am Fam Physician 52:1169–1174, 1995
Penney LM, Spector PE: Job stress, incivility, and counterproductive work be-
havior (CWB): the moderating role of negative affectivity. Journal of Orga-
nizational Behavior 26:777–796, 2005
Pettus v. Cole, 57 Cal.Rptr.2d 46 (Calif. Court Appeals 1996)
Philipse MWG, Koeter MWJ, van der Staak CPF, et al: Static and dynamic pa-
tient characteristics as predictors of criminal recidivism: a prospective
study in a Dutch forensic psychiatric sample. Law Hum Behav 30:309–327,
2006
Pinkerton Consulting and Investigations: Top Security Threats and Manage-
ment Issues Facing Corporate America. Parsippany, NJ, Pinkerton Consult-
ing and Investigations, 2003
Rosen A: Detection of suicidal patients: an example of some limitations in the
prediction of infrequent events. J Consult Psychol 18:397–403, 1954
Scalora MJ, Washington DO, Casady T, et al: Nonfatal workplace violence risk
factors: data from a police contact sample. J Interpers Violence 18:310–327,
2003
Schouten R: Pitfalls of clinical practice: the treating clinician as expert witness.
Harv Rev Psychiatry 1:64–65, 1993
Schouten R: Violence in the workplace, in Mental Health and Productivity in the
Workplace. Edited by Kahn JP, Langlieb AM. San Francisco, CA, Jossey-
Bass, 2003, pp 314–328
Schouten R: Workplace violence: an overview for practicing clinicians. Psychi-
atr Ann 36:790–797, 2006
Schouten R, Callahan MV, Bryant S: Community response to disaster: the role
of the workplace. Harv Rev Psychiatry 12:229–237, 2004
Sen P, Gordon H, Adshead G, et al: Ethical dilemmas in forensic psychiatry: two
illustrative cases. J Med Ethics 33:337–341, 2007
Simister J, Cooper C: Thermal stress in the USA: effects on violence and on em-
ployee behaviour. Stress and Health 21:3–15, 2005
Skarlicki D, Folger R: Retaliation in the workplace: the roles of distributive, pro-
cedural, and interactional justice. J Appl Psychol 82:434–443, 1997
Storms P, Spector P: Relationships of organizational frustration with reported
behavioural reactions: the moderating effect of locus of control. Journal of
Occupational Psychology 60:227–234, 1987
Strasburger LH, Gutheil TG, Brodsky A: On wearing two hats: role conflict in
serving as both psychotherapist and expert witness. Am J Psychiatry
154:448–456, 1997
Swanson JW, Swartz MS, Van Dorn RA, et al: A national study of violent behav-
University of Iowa Injury Prevention Research Center: Workplace Violence: A
Report to the Nation. Iowa City, University of Iowa Injury Prevention Re-
search Center, 2001
Vossekuil B, Reddy M, Fein R: Safe School Initiative: An Interim Report on the
Prevention of Targeted Violence in School. Washington, DC, U.S. Secret
Service, 2000
Waddington PAJ, Badger D, Bull R: Appraising the inclusive definition of work-

place “violence.” Br J Criminol 45:141–164, 2005
Warren J, Brown D, Hurt S, et al: The organizational context of non-lethal work-
place violence: its interpersonal, temporal, and spatial correlates. J Occup
Environ Med 41:567–581, 1999
C H A P T E R 2 6
Vehicular Crashes
and the Role of
Mental Health Clinicians
Thomas Ala, M.D.
By virtue of their training, experience, and practice, mental health

professionals are often in a good position to identify and assess condi-
tions that may impair fitness to drive an automobile. There is much cur-
rent debate regarding whether clinicians should be expected to identify
which individuals, because of mental disorder, cannot safely drive and,
if clinicians make such an assessment, what reasonable protective mea-
sures they should be expected to take.
Through its position statement on the topic, the American Psychiat-
ric Association (1993) maintains that thorough assessment of driving
functions does not lie within the scope or abilities of psychiatrists. In
their practice, psychiatrists should not be expected to make assessments
of driving competence per se but rather to show awareness and concern
for how certain conditions may affect fitness to drive. Rather than as-
sume any protective responsibilities of a coercive or disclosing nature,
psychiatrists can foster safe driving through patient education and
monitoring. Psychiatrists can tell their patients how some symptoms of
521
their disorder can affect driving safety, especially concerning well-

researched conditions such as dementia. They should explain that cer-
tain medicines can impair a person’s alertness and coordination when
driving, especially if combined with alcohol. Finally, the American Psy-
chiatric Association encourages psychiatrists to preferentially prescribe
medications less likely to interfere with driving for their patients who
are expected to drive.
Vehicular crashes1 are responsible for untold morbidity and mortal-
ity in the United States, especially those involving teenage and young
adult drivers. Most mentally disordered individuals, like nearly every-
one else, drive and probably do so safely. Also like other drivers, the
mentally disordered are quite capable of being in a crash, even causing
it, without mental disturbance contributing in a causal sense. Then
again, sometimes a crash is related to a disturbed mental state.
Mental disorder can cause an individual to crash either deliberately
or accidentally (Felthous 2006). A mentally disordered individual may
decide to crash to commit suicide, to kill unknown victims, or both. An
accidental crash can be due to the mental disorder directly, such as
when an individual with bipolar disorder drives fast and recklessly
during a manic episode. Other neurological conditions of concern in-
clude epilepsy and other disorders of consciousness, such as narco-
lepsy. Conditions with poor regulation of emotional and behavioral
control such as intermittent explosive disorder, borderline personality
disorder, and antisocial personality disorder can all motivate risk-
taking behaviors that may elevate crash risk.
Intoxication with alcohol is an all too common cause of vehicular
crashes, even without the presence of other serious mental illness. Indi-
viduals already at risk due to mental disorder can increase their risk by
ingesting intoxicating substances. Physicians may prescribe medica-
tions that are sedating or have other properties that impair mentation
and driving competence; on the other hand, some medications and per-
formance-enhancing substances can improve driving ability. Situa-
tional distraction, distress, and especially drowsiness can compromise
safe driving in normal individuals. A cup of caffeinated coffee can help
the weary driver safely finish a long haul. Stimulants may restore nar-
coleptic individuals to safe driving capability, and the same may be true
for anticonvulsants in epileptics, antipsychotic medications in drivers
with psychotic disorders, mood stabilizers in those with bipolar disor-
1
In this discussion the term accident is used selectively, not generically, because not all
crashes are unintended.
Vehicular Crashes and the Role of Mental Health Clinicians ❘ 523
ders, and antidepressants in depressed individuals. Empirical evidence

demonstrates that when depressed patients are treated with antidepres-
sant medication, their driving competence and safety parameters im-
prove (Ärzte-Zeitung 2007). Although safe clinical practices emphasize
appropriate warnings and precautions for medications that can com-
promise safe driving, the tremendous value of medication in enhancing
driving skill and safety when prescribed appropriately must not be
overlooked.
Obviously, a great variety of illnesses, medical conditions, and phys-
ical disabilities can impair driving skill and increase the risk for a crash.
Here we focus on several mental and neurological disorders that can
adversely affect driving and that mental health clinicians are likely to
encounter. The categorical conditions included are psychosis and
schizophrenia, depression, dementia, and disturbances in conscious-
ness, particularly seizures or epilepsy and narcolepsy. Legal responsi-
bilities of treating physicians are considered by using case examples;
however, the clinician is best advised to refer to jurisdictional, espe-
cially controlling statutory, law regarding reporting requirements.
Mental Disorders
Psychosis and Schizophrenia
A literature review conducted by the National Highway Traffic Safety
Administration (Dobbs 2005) concluded that individuals with un-
treated psychotic disorders and, overall, those with schizophrenia, de-
pressive or anxiety disorders, alcoholism, and personality disorders are
at greatest risk for motor vehicular crashes. Studies of schizophrenic
drivers show a relationship to crash rates if correction is made for expo-
sure to driving. (Those with serious mental impairments spend less
time driving than their unafflicted counterparts.) There is general
agreement that one should not drive during the acute phase of a psy-
chotic illness (American Psychiatric Association 1995; Austroads 1998;
Canadian Medical Association 2000).
Little is written, however, on how to determine when a condition is
sufficiently acutely psychotic that driving should be restricted. This is
presumably because current psychotropic medications have shortened
the duration of acute psychotic exacerbations—and therefore of periods
when driving should be restricted. Moreover, the customary standard
of practice is to hospitalize the individual who is acutely psychotic, thus
minimizing for that period that individual’s exposure to driving and
potential for causing a crash.
Not everyone who is competent and willing to consent to hospi-

talization meets criteria for emergency involuntary hospitalization.
Threats, preparatory acts, and attempts to seriously harm self or others,
especially if psychotically driven, simplify the decision and the legal
justification for emergency hospitalization. The evaluator asks about
current thoughts and past acts of self-harm, suicide, and homicide.
An example of a psychotic driver who causes a crash comes from the
landmark legal case Naidu v. Laird (1988).
Case Example 1
Mr. Hilton Putney had been hospitalized numerous times, had failed to
take prescribed medication in the community, had attempted suicide on
several occasions, and had deliberately crashed his car on two occa-
sions. Immediately after he was discharged from Delaware State Hospi-
tal, Mr. Putney stopped taking medication and did not keep his outpa-
tient appointment. He deliberately crashed his car into another, killing
the other driver, Mr. George Laird. Mr. Putney was charged with man-
slaughter and found not guilty by reason of insanity. The lawsuit against
Dr. Naidu and other hospital psychiatrists resulted in a $1.4 million ver-
dict. The Supreme Court of Delaware, referring to the Tarasoff duty of
therapists to protect third persons from foreseeable harm caused by
their patients, upheld the cause of action.
Vehicular crashes by psychotically disturbed individuals cannot al-

ways be anticipated by threats. They are not necessarily preceded by an
expressed plan or intention. Worrisome signs include prior history of
crashes, especially if associated with destructive thoughts, disorga-
nized thought, or agitation. The most effective prevention is probably
hospitalization while the patient is acutely psychotic and close monitor-
ing after discharge to ensure medication compliance and mental stabil-
ity, exercising early intervention when signs of decompensation appear.
Where hospitalization is not practical but psychotic symptoms and sig-
nificantly impaired concentration cause concern, the patient should be
discouraged from driving. Family members can sometimes be helpful
in supporting the patient’s transportation needs.
Depression
Most authorities would agree that depressed individuals may drive if
their mood is stable and any medications are regulated (American Psy-
chiatric Association 1995; Austroads 1998). Austroads (1998) advised
against driving while medications are being adjusted, whereas the
American Psychiatric Association (1995) recommended that the physi-
cian warn the patient that any newly prescribed medication can affect
driving ability. There is general agreement that individuals who are pro-
foundly depressed with impaired concentration should not drive
(American Psychiatric Association 1995; Austroads 1998; Canadian
Medical Association 2000). Although controversy exists over the num-
ber of crashes caused by suicidal drivers (Dobbs 2005), practiced mental
health clinicians are familiar with individual patients who have thought
of making a suicidal crash while driving or have actually attempted
this.
Case Example 2
Mr. Tony Marconi,2 a 60-year-old retired fireman, reported at the hospi-
tal emergency department that he was thinking of committing suicide
by drug overdose or vehicular crash. Diagnoses included major depres-
sive disorder and cocaine dependence. Although he took antidepressant
medication as prescribed, he still felt depressed, so he tried to self-
augment with cocaine. Even this extra measure did not assuage his
depression. When interviewed the day after admission to the hospital
inpatient psychiatric service, Mr. Marconi denied having had a specific
method of suicide in mind: he stated that he came to the hospital only to
obtain help with depression and said he was thinking of suicide but
without a method or plan in mind. When asked specifically about drug
overdose, he acknowledged having attempted suicide by drug overdose
several years ago but denied that this was a recent consideration.
When asked about suicide by vehicular crash, Mr. Marconi said this
had occurred to him just before a prior hospitalization about 3 weeks
earlier. While he was driving on a busy interstate highway and feeling
hopeless about his unrelenting depression and insuperable cocaine ad-
diction, it suddenly occurred to him that he could at that very moment
steer his car into another vehicle and end his misery once and for all in
a fatal crash. When asked if he had concern about the other driver, Mr.
Marconi said this did occur to him as a secondary consideration. The
thought of harming another person importantly constrained him from a
disastrous turn of the steering wheel and impelled him to drive straight
to the hospital to obtain help.
The case of Mr. Marconi illustrates a general challenge in suicide as-

sessment and a specific challenge where the would-be method of sui-
cide is by crash. While in the hospital emergency department, a patient
will sometimes express a method for committing suicide and then the
following day will deny having entertained the same plan. He may
have cited the method initially in order to gain hospital admission and
2
For this and the subsequent case examples, information is altered to obscure identifica-
tion and resemblance to any particular individual.
then denied it in order to be released or to have suicide precautions

lifted. There are obviously other possible explanations for such a dis-
crepancy. In any event, description of such a method is associated with
the thought, regardless how strong or weak the intent. Without contra-
vening evidence of insincerity, such specific methods should be taken
seriously at the time, especially when availability of this patently lethal
method is established.
With further assessment, Mr. Marconi’s self-reported spontaneous
impulse to commit suicide by crash was convincing. This example illus-
trates a challenge in preventing suicide by crash. In contrast to firearms,
for example, people rely on driving cars for personal mobility, access to
services, productive employment, and living comfort, so exposure is
high. Yet the impulse to commit suicide by crash can occur impulsively
and with no advance warning. After the patient with such impulses is
no longer acutely suicidal and can be discharged, he or she should be
cautioned against driving after taking illegal drugs and when feeling
hopeless and depressed. Even before suicidal thoughts resurface, if
these conditions arise it is time to call upon mental health services.
Case Example 3
Ms. Lilian Quen is a 35-year-old woman who was admitted to the hos-
pital with major depressive disorder. She had been feeling hopeless and
having suicidal thoughts of killing her children as well as herself. Al-
though not psychotic, she could not bear the thought of her children not
being raised and cared for by herself. Two methods occurred to her:
poisoning by overdosing with medicine and driving her car with both
children inside into a river. She seriously thought of the latter method
recurrently over several years. Several times she made initial prepara-
tions by placing her children in the car and driving to a riverbank. No-
ticing that she was distraught, her children asked what was the matter,
gave solace to their mother, and in so doing interrupted Ms. Quen’s fa-
tal intentions.
This would have been a crash of another kind, driving the car into
the river in a combined homicide-filicide. It serves as another example
of the importance of assessing suicide and homicide risk by obtaining a
detailed history of prior thoughts, preparatory acts, and attempts and
by asking the details of the method(s) considered in addition to asking
about desperate thoughts. The suicide-homicide prevention plan in-
volves the child protective authorities, aggressive treatment of depres-
sion, careful discharge planning before hospital release, and close mon-
itoring afterward. Extended restriction of driving should be a part of the
prevention plan.
Dementia
For patients diagnosed with a progressive dementia such as Alzhei-
mer’s disease, it is not a matter of if retirement from driving will be
necessary but when. Alzheimer’s disease differs from other conditions
reviewed in this chapter because current treatments cannot restore driv-
ing fitness in those already impaired. Whereas patients with early, mild
forgetfulness and a safe driving record may retain sufficient ability to
drive for a time, those with more advanced impairment (i.e., deficits in
divided attention, visuospatial skills, and/or executive functioning) are
likely to pose a hazard on the road (Carr et al. 2006). The American Psy-
chiatric Association (1997) recommends that patients with moderate
dementia be required to stop driving for reasons of individual and pub-
lic safety. The American Academy of Neurology argues that even those
at the mild stage have sufficient deficits to warrant driving cessation
(Dubinsky et al. 2000).
In a landmark study by Linda Hunt and colleagues at Washington
University (Hunt et al. 1997), patients evaluated as clinically normal and
as being in the early stages of dementia were administered a detailed
on-road performance evaluation. Patients were characterized as either
very mildly demented or mildly demented by use of the Clinical De-
mentia Rating (CDR; Morris 1993), 0.5 and 1 levels, respectively. Overall
ratings of “safe,” “marginal,” or “unsafe” were assigned to each older
driver. The majority (97%) of those judged to be clinically normal (CDR
0) were found to be safe or marginal drivers. This number dropped to
81% for very mildly demented (CDR 0.5) drivers and 59% for mildly de-
mented (CDR 1) drivers. As many CDR 1 drivers were found to be un-
safe as safe—41% in both cases. Subsequent testing over time revealed
that the majority of these demented drivers moved from safe/marginal
to unsafe categories over a 2-year period, with mildly demented (CDR
1) individuals showing the steepest decline (Duchek et al. 2003).
Based on these and other findings, it is reasonable to consider the
transition from CDR 0.5 to CDR 1 stages of dementia (i.e., very mild to
mild) as the critical period for driving-related assessment, discussion of
driving retirement, and implementation of a cessation and alternative
transportation plan (Meuser et al. 2006). Primary care and specialist
physicians, including psychiatrists, can play important evaluative and
counseling roles in this process, according to the American Medical As-
sociation’s Older Drivers Project (American Medical Association 2003;
Wang and Carr 2004). In-office interview and screening procedures
are sufficient, in many cases, to “risk stratify” patients into likely safe
or unsafe categories. Advancing impairment and evidence of on-road
problems (e.g., reports of near misses, accidents, traffic tickets) would

support a recommendation to stop driving.
Few physicians have the time, confidence, or expertise to make an
independent decision on driving fitness. The American Medical Associ-
ation encourages patient referrals for on-road evaluation (which are typ-
ically provided by a certified driver rehabilitation specialist or trained
occupational therapist), mobility counseling to develop an alternative
transportation plan (often provided by a counselor or social worker),
and consultation with local support organizations (e.g., the Alzheimer’s
Association, the area Council on Aging). It often takes a community of
health and service professionals, working together with the patient and
family, to implement an effective driving retirement plan.
In contrast to other conditions reviewed in this chapter, dementia is
one for which it is recommended that the psychiatrist participate in the
driving assessment/retirement process (American Psychiatric Associa-
tion 1997). Protection of individual and public safety warrants such in-
volvement. Crash risk is doubled in persons with dementia, ranging
from an 8% to 10% crash rate per year (Brown and Ott 2004). Along with
family members, clinicians are often the first professionals to become
aware of changes in physical health and cognitive status that may affect
fitness to drive. Early intervention can reduce exposure to accidents for
the driver with dementia, his or her passengers, and others on the road.
Under ideal circumstances, the demented driver will accept medical ad-
vice to stop driving and move into driving retirement willingly. Under
less optimal circumstances, the clinician and family may need to build
a case for retirement. Often with physician input and family persis-
tence, driving retirement can be achieved even in patients who lack in-
sight. However, some patients may need to be forced via a formal report
to the state driver licensing authority—the Department of Motor Vehi-
cles in most states—to trigger a formal review and de-licensing process.
The Physician’s Guide to Assessing and Counseling Older Drivers
(American Medical Association 2003) provides step-by-step guidance
for the assessment and counseling process, culminating in state report-
ing for challenging cases where loss of awareness, denial, or stubborn-
ness obstruct prudent decision making. In dementia, fitness to drive is
not a one-time clinical concern but something that must be approached
with both sensitivity and care over a period of time (Meuser et al. 2006).
A number of recommendations for practice flow from this concept and
are applicable to psychiatric care.
• Driving retirement is appropriate for discussion soon after the diag-

nosis of progressive dementia is made. A sufficient body of research
is available on Alzheimer’s disease to indicate that retirement from

driving is an inevitable endpoint for most patients (Carr et al. 2006;
Meuser et al. 2006). Individualized assessment is necessary to deter-
mine when retirement is prudent on a per-case basis. Psychiatrists
(as well as other physicians) should encourage dementia patients
and their families to review transportation needs and actively plan
for driving cessation to occur within 1–3 years, or sooner if individ-
ual circumstances warrant more immediate action.
• When safety concerns are identified, psychiatrists should refer de-
mented patients for on-road driving evaluation and use other referral
options to support the assessment and retirement planning process.
Not all communities have such referral options, however, and referral
to the state authorities may be the only viable option in some cases.
• For those who demonstrate intact driving skills, repeat evaluations
are suggested on a yearly basis until the inevitable decision to retire
must be made (see discussion in Duchek et al. 2003).
• Driving is a complex task involving many individual and environ-
mental variables that cannot be assessed fully or controlled. The
determination of a safety risk due to medical or psychiatric health is
a clinical activity involving reasoned judgment based on available
data. This can be challenging when conflicting data exist, and the
American Medical Association recognizes that thoughtful profes-
sionals can disagree.
• In terms of legal protection, psychiatrists and other physicians who
act in the best interests of their demented patients, encourage rea-
soned decision making concerning driving safety, and document
their actions should be protected from liability if a patient causes an
accident. As a recent position statement of the American Academy
of Neurology points out, however, this is a gray area for all practic-
ing physicians today due to differences in state laws and imperfect
evaluation methods (Bacon et al. 2007). For physicians aware of a
driving fitness problem in a demented patient, the riskiest course
from legal and ethical standpoints would be to do nothing. Each
physician should be aware of the specific reporting laws in his or her
state and request legal counsel for any policy or practice in regard to
reporting clientele.
• Ultimately, it is the responsibility of state government (or other
governmental entities outside of the United States) to evaluate and
de-license drivers in response to medical fitness or other safety con-
cerns. A few states, such as California, mandate that persons diag-
nosed with Alzheimer’s disease be reported (California Department
of Motor Vehicles Health and Safety Code, Section 103900), but most
make reporting a voluntary process. Psychiatrists and other physi-

cians can do their part by facilitating reasoned decisions about driv-
ing in their demented patients and formally reporting those who
refuse to stop driving when the time has come. See the American
Medical Association’s (2003) Physician’s Guide for a review of laws in
all 50 states and suggestions for how to make a report.
Case Example 4
Mrs. Burns has very mild Alzheimer’s disease (CDR 0.5) and lives by
herself in a rural area. Her closest family member lives an hour away
and visits weekly to check on her well-being. Her family handles all of
her finances and housekeeping and sets up her weekly pillbox to assist
with medication administration. Mrs. Burns and her family report that
her driving ability remains unimpaired. Her daily routine is to drive
into town 3 miles to socialize and to have her main meal of the day. Her
friends do not drive, and visiting them requires her to operate a motor
vehicle. In addition, she is reluctant to leave the home she has enjoyed
for the past 30 years.
Case Example 5
Mr. Young has moderate Alzheimer’s disease (CDR 2) and lives with his
wife in a suburban area. Mrs. Young has never driven. They go out reg-
ularly to run errands, shop, go to restaurants, and visit friends and fam-
ily. Mrs. Young has vetoed all recommendations that Mr. Young have a
driving performance test, because they would then be isolated if he
should fail. She states that he drives well, but she does add that she has
to tell him where to turn and when to stop.
In both cases, the family believes the person with dementia to be a

safe driver. In the first case, the family is aware of Mrs. Burns’ driving
and supportive of it continuing. We are led to assume that no on-road
incidents or problems have occurred. These should be inquired about
by the clinician.
This cannot be assumed in the case of Mr. Young, however, because
his wife clearly has secondary motives for keeping him on the road. At
the moderate stage of disease, he is past the critical period for driving-
related assessment and retirement planning. The fact that she must
serve as his “copilot,” prompting him on where to go, is another con-
cern. In this case, unless Mr. Young’s driving ability can be confirmed
through an on-road evaluation, it may be necessary to submit a report
to the state Department of Motor Vehicles. A viable alternative trans-
portation plan could help ease Mrs. Young’s concerns about isolation.
Should Mr. Young refuse to stop driving, some creative efforts may be
necessary to stop his driving altogether. Such efforts may include filing
down his ignition key or disabling the car so that it is inoperable. Just
having a car in the driveway may satisfy some patients, especially those
with more advanced memory loss. The vehicle could also be sold.
Disturbances in Consciousness
Two neurological conditions involving sudden, unpredictable loss of
consciousness are seizures and narcolepsy. Other conditions with loss
of consciousness, such as syncope and sleep apnea, create similar con-
cerns about driver safety, but this discussion is limited to neuropsychi-
atric conditions. A seizure can occur once in a lifetime or it can be recur-
rent, depending on the etiology. An underlying neurological disorder as
well as electroencephalographic abnormalities predicts seizure recur-
rence (Berg and Shinnar 1991). Epilepsy by definition involves recur-
rent loss of consciousness and the function of other faculties, depending
on the nature of the disorder. Most states in the United States withhold
or withdraw the epileptic person’s license until a specific period of time
has elapsed without a seizure. The length of this seizure-free period
varies between states and ranges from 3 months to 2 years (Dobbs
2005). While the epileptic is at risk for a seizure, other activities to be
avoided include operating heavy machinery, swimming, being in the
immediate vicinity of an open fire or a body of deep water, and other
obviously perilous situations.
Falling asleep behind the wheel is thought to be a common cause of
vehicular accidents. Narcolepsy and sleep apnea are of special concern
because sudden attacks of drowsiness and sleep are not easily con-
trolled by those afflicted. Narcolepsy symptoms include catalepsy, hal-
lucinations, and sleep paralysis. Emotions can induce a spontaneous
loss of muscle strength known as catalepsy, which, in addition to “sleep
attacks,” puts the narcoleptic driver at risk. A narcoleptic individual
should not drive as long as the risk of sudden sleep remains. The Cana-
dian Medical Association (2000) recommends no vehicular driving if a
cataleptic episode has occurred within the past 12 months.
Legal Duties of Clinicians to Prevent

Vehicular Crashes
The common law principle of non-responsibility to third parties would
normally protect a clinician for liability when her or his patient harms
another in a crash. After all, clinicians should not be responsible for all
that their patients do, intentionally or accidentally. In a number of cases,
however, appellate courts have upheld causes of action against physi-
cians, even for harm inflicted on third persons, if the physician should
have taken some reasonable preventive measure (Felthous 1989a).
This liability is more likely to occur if the patient/offending driver
had been hospitalized and then released or discharged with incomplete
symptom control. In Tarasoff-like jurisprudence, such cases can be clas-
sified as involving the “foreseeable rule,” where the crash can reason-
ably be anticipated even if the other specific individual victims cannot
be identified in advance (see Felthous 1989b).
In Schuster v. Altenberg (1988) the Supreme Court of Wisconsin up-
held a claim wherein the driver, Edith Schuster, was a psychiatric out-
patient with manic-depressive illness. She was killed in the accident,
and her daughter was left with both legs paralyzed. Causes of action in-
cluded negligent diagnosis and treatment, failure to seek civil commit-
ment, and “failure to warn the patient’s family of her condition and its
dangerous implications” (p. 4). In dicta, the court equated warning the
patient with treating the patient: “Warning a patient of risks associated
with a condition and advising the patient as to appropriate conduct
constitutes treatment as to which the physician must exercise ordinary
care” (p. 6). The court’s approach to this case raises the question of
which disorders would require such a warning (Felthous 1989b). Today,
despite the Schuster decision, it is not likely that the standard of practice
would be to warn bipolar patients generally of the risk of driving once
they are deemed safe enough to be treated as outpatients.
Clinicians have a duty to inform their patients about material side
effects of the medicines they prescribe. This can include the potential
side effect of drowsiness and a warning not to drive or operate heavy
machinery until the patient becomes accustomed to the drug and aware
of its effects. If the patient is not so informed, takes the prescribed sedat-
ing or otherwise mind-altering drug, and then, because of the drug’s
side effect, loses control of his or her vehicle, resulting in an accident,
the physician can be held liable for injuries that the patient sustained
from the accident or, in some jurisdictions, injuries inflicted on others.
An oft-cited case example of this type of liability with correspond-
ing legal duty of the clinician to inform the patient is Gooden v. Tips
(1983). The plaintiff, who was struck and injured, argued that the phy-
sician who prescribed Quaalude was negligent for failing to warn her
against driving while taking this medicine. The holding for the Court of
Appeals of Texas found the petition to be sufficient. Likewise, in Kirk v.
Michael Reese Hospital and Medical Center (1985), an Illinois court found
that such a duty to warn a patient of adverse effects of a medication can
“extend to cover members of the public who may be injured as a prox-
imate cause of the failure to adequately warn” (p. 911).
If the patient already knows of the risk, however, liability may not
necessarily extend to a prescribing physician who did not give such a
warning. A psychiatrist and psychologist in Connecticut were sued
with the claim that they failed to warn a patient not to drive her vehicle
while she was taking medication that altered her sleep cycle (Weigold v.
Patel 2004). The medication caused her to fall asleep at the wheel, it was
claimed, and to strike another car, resulting in the death of the other
driver. Because the driver/patient knew that her driving was impaired,
she, not her treaters, created the proximate cause by driving anyway.
The clinicians could not control the patient’s behavior, and therefore
their failure to warn the patient not to drive was not the proximate
cause of the victim’s death. The court concluded that the psychiatrist
and psychologist had no duty to warn the patient not to drive.
Regardless of what jurisdictional law requires in the way of informing
patients of the side effects of medication and the risk of driving when se-
dated, such information is reasonable to convey. By the same token, the
benefits of a medicine should also be shared with patients, including the
likelihood that a medication or combination of medicines can improve
driving performance. Psychotically disorganized patients can be expected
to be at greater risk for an automobile accident because of their mental dis-
order. Antipsychotic medicines improve symptoms of psychosis and dis-
organized thinking, and thus driving competence should also be restored.
Anticipating the sedative potential of antidepressant medications, phar-
maceutical companies include warnings of driving risk in the package in-
serts of many such agents. Yet empirical evidence demonstrates that when
depressed patients are treated with antidepressant medication, their driv-
ing competence and safety parameters improve (Ärzte-Zeitung 2007).
Clinicians have also been held liable for failing to diagnose a condi-
tion that could risk a vehicular crash and for not informing the patient
of the risk. In Iowa, a driver lost control of a vehicle during a seizure and
ran into a pedestrian. The injured pedestrian sued the driver’s physi-
cian, claiming failure to diagnose the seizure, to determine its cause, to
advise the patient not to drive, and to warn the patient of the risk asso-
ciated with driving. Moreover, it was claimed that the physician negli-
gently assured the driver that he could drive. On appeal, the Supreme
Court of Iowa held that the plaintiff’s petition with the above claims
stated a cause of action against the physician (Freeze v. Lennon 1973).
Conclusion
It is impossible to estimate the number of vehicular crashes attributable
to a mental or neurological disorder. Crashes are generically and collo-
quially referred to as “accidents” even when the cause of apparent hu-

man error is unknown. A death from a gunshot wound will be in-
vestigated as a possible homicide, suicide, or accident, but so many
vehicular crashes appear to have been accidents that, beyond determin-
ing alcohol and drug levels, in many cases greater scrutiny and suspi-
cion are unlikely to accurately distinguish intentional from accidental
crashes after the fact. Yet mental health clinicians have, at least from ex-
perience, knowledge of suicidal or homicidal thoughts, preparation
and acts involving motor vehicles, and psychological and neurological
deficits that can result in a vehicular crash.
For psychiatrists and psychologists inexperienced in evaluating
driving competence, preventing vehicular crashes may be an unwel-
come but necessary challenge, especially where progressive dementia is
concerned. This aspect of risk management has received far less atten-
tion than, say, the clinician’s legal duty to warn or protect in regard to a
patient’s risk of harming others following a threat. Even when such a
risk is first registered by a serious verbal threat, clinicians regard them-
selves as service providers to individuals, not protectors of the public.
Yet clinicians must routinely make interventional decisions, such as
whether to hospitalize (involuntarily when this is needed and justified),
based on apparent risk of harm to self or others. The movement away
from hospital treatment toward community treatment has probably in-
creased, not decreased, the risks of mentally and neurologically disor-
dered individuals being involved in vehicular crashes. Independent of
such public policy changes, the demographically aging population,
with far more demented and otherwise disordered drivers in the future,
increases the magnitude of the challenge.
Clinicians are expected to conduct competent diagnosis and risk as-
sessment for personal violence, which should include potential for in-
tentional violence with a motor vehicle. Management of such risks can
be handled as is done for violence risk management when, for example,
a firearm is involved: with education and counseling; hospitalization if
mental disorders and risk are acute; proper treatment; alliance and coop-
eration with significant others; reduced access and driving exposure;
and selection and titration of specific risk control measures needed by
the patient. When the risk is due to a neurological condition with im-
paired driving abilities or risk of disturbed consciousness, some gross at-
tempt to rate the risk is recommended: low risk, educational measures;
high risk, interventional restrictive measures (e.g., a responsible family
member secures the car keys); intermediate risk, refer for neuropsycho-
logical and/or driving competency assessment. For specific legal report-
ing duties, clinicians should refer to controlling jurisdictional law.
Key Points
■ Medications are double-edged swords: they can impair or improve
driving competence, depending on how they are used.
■ Risk assessment interviews should be informed by the possibility
of suicide or homicide by vehicular crash.
■ For patients with Alzheimer’s disease, it is not a matter of if retire-
ment from driving should occur but when.
■ Laws or protective duties for clinicians regarding the risk of vehic-
ular crashes are variable and jurisdiction-specific.
References
American Medical Association: Physician’s Guide to Assessing and Counseling
Older Drivers. Chicago, IL, American Medical Association, 2003
American Psychiatric Association: Role of the Psychiatrist in Assessing Driving Abil-
ity: Position Statement (APA Document Reference No. 930004). Approved by
the Board of Trustees, December 1993, Approved by the Assembly, November
1993. Washington, DC, American Psychiatric Association, 1993
American Psychiatric Association: Position statement on the role of psychia-
trists in assessing driving ability. Am J Psychiatry 152:819, 1995
American Psychiatric Association: Practice guidelines in the treatment of
Alzheimer’s disease. Am J Psychiatry 154(suppl):1–39, 1997
Ärzte-Zeitung: Autofahren Klappt bei Depressiven mit Antidepressiva besser als
ohne, 2007. Available at http://www.aerztezeitung.de/docs/2006/02/03/
020a0401.asp?cat=/medizin/depression-65K-. Accessed January 24, 2006
Austroads: Assessing Fitness to Drive. Austroads Guidelines for Health Profes-
sionals and Their Legal Obligations. Sydney, Australia, Superline Printing,
1998
Bacon D, Fisher RS, Morris JC, et al: American Academy of Neurology position
statement on physician reporting of medical conditions that may affect
driving competence. Neurology 68:1174–1177, 2007
Berg AT, Shinnar S: The risk of seizure recurrence following a first unprovoked
seizure: a quantitative review. Neurology 41:965–972, 1991
Brown LB, Ott BR: Driving and dementia: a review of the literature. J Geriatr
Psychiatry Neurol 17:232–240, 2004
California Department of Motor Vehicles Health and Safety Code, Section
103900. Available at http://www.dmv.ca.gov/pubs/vctop/appndxa/hlthsaf/
hs103900.htm. Accessed January 23, 2008
Canadian Medical Association: Determining Medical Fitness to Drive: A Guide
for Physicians. Ottawa, Canada, Canadian Medical Association, 2000
Carr D, Duchek J, Meuser T, et al: The older driver with cognitive impairment.
Am Fam Physician 73:1029–1034, 2006
Dobbs BM: Medical Conditions and Driving: A Review of the Literature (1960–
2000). Washington, DC, National Highway Traffic Safety Administration,
2005
Dubinsky RM, Stein AC, Lyons K: Practice parameter: risk of driving and
Alzheimer’s disease (an evidence-based review). Report of the Quality
Standards Subcommittee of the American Academy of Neurology. Neurol-
ogy 54:2205–2211, 2000
Duchek JM, Carr DB, Hunt LA, et al: Longitudinal driving performance in early
stage dementia of the Alzheimer type. J Am Geriatr Soc 51:1342–1347, 2003
Felthous AR: The duty to warn or protect to prevent automobile accidents, in
Review of Clinical Psychiatry and the Law, Vol 1. Edited by Simon RI.
Washington, DC, American Psychiatric Press, 1989a, pp 221–238
Felthous AR: The Psychotherapist’s Duty to Warn or Protect. Springfield, IL,
Charles C Thomas, 1989b
Felthous AR: Personal violence, in Forensic Psychiatry. Edited by Simon RI,
Gold L. Washington, DC, American Psychiatric Publishing, 2006, pp 471–
500
Freeze v. Lennon, 210 NW.2d 576 (1973)
Gooden v. Tips, 651 SW.2d 364 (Tex. App. 1983)
Hunt LA, Murphy CF, Carr D, et al: Environmental cueing may effect perfor-
mance on a road test for drivers with dementia of the Alzheimer type.
Alzheimer Dis Assoc Disord 11(suppl):13–16, 1997
Kirk v. Michael Reese Hospital and Medical Center, 483 NE.2d 906 (Ill. App. 1 Dist.
1985)
Meuser TM, Carr DB, Berg-Weger M, et al: Driving and dementia in older
adults: implementation and evaluation of a continuing education project.
Gerontologist 46:680–687, 2006
Morris JC: The Clinical Dementia Rating (CDR): current version and scoring
rules. Neurology 43:2412–2414, 1993
Naidu v. Laird, 539 A.2d 1064 (1988)
Schuster v. Altenberg, Wisconsin Supreme Court, No. 87–0115, 1988
Wang C, Carr D: Older driver safety: a report from the Older Drivers Project.
J Am Geriatr Soc 52:143–149, 2004
Weigold v. Patel, 840 A.2d 19 (Conn. App. 2004)
Recommended Readings
Felthous AR: The duty to warn or protect to prevent automobile acci-
dents, in Review of Clinical Psychiatry and the Law, Vol 1. Edited
by Simon RI. Washington, DC, American Psychiatric Press, 1989, pp
221–238
Meuser TM, Carr DB, Berg-Weger M, et al: Driving and dementia in old-
er adults: implementation and evaluation of a continuing education
project. Gerontologist 46:680–687, 2006
C H A P T E R 2 7
School Violence
In dealing with school violence, one persistent impression is of confu-

sion and overlap in defining the topic. Some confine it to high schools,
whereas others include middle schools as well as violence at the college
level. A second source of confusion is that some restrict the discussion to
lethal violence, whereas others include all levels of nonfatal victimiza-
tion, including violations of school discipline. The focus of violence pre-
sents a third complication. Some focus on students as the victims of vio-
lence, ranging from those who restrict the topic to mass levels of
violence with multiple victims to those who focus on a single episode
with one perpetrator and one victim. However, faculty and nonstudents
may also be victims. A fourth difficulty in defining the topic is variations
in the scope of violence. If the violence occurs on the playground during
school hours, it would likely be included, but not necessarily if it occurs
after school hours or involves students driving away from school in cars.
A fifth source of difficulty is that the categories of individuals involved
can be a mixture: students as victims and perpetrators, nonstudents in
the school as perpetrators, ex-students, outside adults, and so on.
What these situations suggest is that the data on school violence, let
alone the reported incidence of crimes, may often be underestimated
and unreliable. This would hold not only for data from law enforcement
agencies but also that from school systems, which may not regularly re-
port such data to the public. Some school systems report “student
crime” to state departments of education, but there is a lack of consis-
tency within districts and states. Although part of this may be poorly
functioning systems, there may also be an element of trying to minimize
537
such data to protect the image of school districts. The National Research
Council and Institute of Medicine (2003) used the following criteria for
lethal school violence: taking place in or associated with schools, com-
mitted by students of the school, and resulting in multiple victimiza-
tions in a single incident. Note that this approach is narrowed to stu-
dents committing lethal acts, but it does not specify which level of
schools (such as colleges or lower grades), and the incident can be either
in or “associated with” schools.
Assessment of the Extent of School Violence

Background
Violence in school settings is not a new phenomenon. Historical evi-
dence reveals it existed in ancient civilizations back to 2000 B.C.E. There-
after, four types receive mention: rebellious clashes, focused anger with
an agenda, student protests, and random acts of violence (Midlarsky
and Klain 2005). Nor is it only recently that students brought weapons
to schools. Interestingly, school violence became more widespread in
the United States after compulsory education was adopted. By 1900,
31 states had made education compulsory, with the goal of disseminat-
ing “Americanism,” and saw an accompanying increase in violence
(Crews and Counts 1997).
During the economic depression in the 1930s, school violence was
minimal, but after 1950 the momentum accelerated. Students became
active in social movements and in various protests, such as those con-
cerned with school segregation and racial equality. In the 1960s, clashes
over civil rights, racism, and the Vietnam War were reflected in school
violence. The period from 1964 to 1968 saw the number of assaults on
teachers increase from 253 to 1,801, and weapons offenses in schools in-
creased from 396 to 1,508 (Beavan 1970). Diverse ideas arose as to why
different types of violence occurred in schools, varying from a macro
perspective on communities to a focus on individual predispositions,
and looking at issues such as the use of corporal punishment (in schools
and at home), bullying, violence in the community, weapons availa-
bility, media violence, use of psychoactive substances, and personality
difficulties.
Current Situation
In 1977 the federal government published a revealing study on school
crime (National Institute of Education 1977). It found that although
teenagers spend only 25% of their time in school, 40% of robberies and
School Violence ❘ 539
36% of physical attacks on them occurred there. A 1995 survey of stu-

dents ages 12–18 years reported 2.5 million students were victims of
some type of crime at school. Serious crimes accounted for 186,000 vic-
tims in schools (through rape, aggravated assault, sexual assault, and
robbery); 47 of these resulted in school-associated deaths, including
38 homicides (Kaufman et al. 2001).
The U.S. Department of Education and the U.S. Department of Jus-
tice conduct an annual survey reported as “Indicators of School Crime
and Safety.” They draw on a variety of sources, such as national surveys
of students, teachers, and principals, and data collected from federal de-
partments and agencies. The 2006 report indicated that in the school
year 2005–2006 for youths ages 5–18 there were 28 school-associated vi-
olent deaths—21 homicides and 7 suicides—and 48 school-associated
deaths when staff and nonstudent school associates were included—
37 homicides, 9 suicides, and 2 legal interventions (Dinkes et al. 2006).
Data from the “Indicators” report for all school-age youth during
the 2003–2004 academic year indicated 1,418 overall homicides, with 19
occurring at school, and 1,282 suicides, with 3 occurring at school. From
1992 through the 2004–2005 school years, the range was from 11 to 34
homicides at school per year. Any year with a mass school shooting el-
evates the incidence for that year. It is often stressed that homicides of
school youths are more frequent outside of school than in school. The
limitation of such a statistic is that children spend only 6–7 hours at
school, 5 days a week, for 8–9 months compared with the entire remain-
ing time period outside school in a year. Interestingly, the annual survey
for 2004 found students ages 12–18 experienced about 1.4 million “non-
fatal” crimes at school, which included 583,000 violent crimes (simple
assault and serious violent crime) and 863,000 thefts—a victimization
rate of 33 thefts and 22 violent crimes per 1,000 students.
Larger schools are more likely to have violent incidents and to report
them to police than are smaller ones. Secondary schools are also more
likely to do so than the lower grades. Ninety percent of schools with
more than 1,000 students report a violent incident. Similarly, urban dis-
tricts are more likely than rural or suburban districts to experience crime
and make reports to police. A contrast is that rampage school shootings,
although much rarer, are likely to occur in suburban or rural settings.
The Centers for Disease Control and Prevention (CDC; 2008) has an-
alyzed data from the School-Associated Violent Death (SAVD) study.
The SAVD study dealt with “school-associated” student homicides:
those occurring in public or private elementary and secondary schools,
on the campus or on the way to or from classes or school-sponsored
events. The study found 116 school-associated homicides of students,
associated with 109 homicide events, from July 1999 through June 2006.
Seventy-eight percent of the homicides occurred on school campuses. A
reported 65% included gunshot wounds, 27% included stabbings or
cuttings, and 12% included beatings, indicating some overlap. The re-
port noted that although homicide is the second leading cause of death
among those ages 5–18 years in the United States, school-associated ho-
micides represent less than 1% of all homicides of school-age children.
A major limitation of the SAVD study is that the cases were identified
from news media reports, which would result in underestimation.
Bullying
Bullying is cited so often as a link to school violence, if not shootings,
that it merits a detailed discussion. Some go so far as to argue that if bul-
lying were dealt with, school violence would be dealt with as well.
Given the American school system, this is not likely to happen. A prob-
lem is that bullying encompasses a large number of students, which in-
cludes many acquiescent observers. The “Indicators” report defined
victimization by bullying as including 1) being made fun of, called
names, or insulted; 2) being subjected to rumors; 3) being threatened
with harm; 4) being pushed, shoved, tripped, or spit on; 5) being made
to do things one does not want to do; 6) being excluded from activities
or a group on purpose; and 7) purposeful destruction of property. Some
would restrict its meaning to repeated, negative acts by a child or group
against another. Although the acts may be physical, the usage encom-
passes verbal taunts or manipulative behavior to exclude others. Im-
plicit is a power game that is played out with different scenarios.
Surveys of students reveal that 16% of children say they have been
bullied during a current school term, and about 30% of sixth- through
tenth-graders say they have been involved as a bully or a target (Nansel
et al. 2001). Adverse consequences of such behavior, reported on both
sides, include more physical and psychological problems, persistence in
the role of victim, a shift from victim to perpetrator, and problems with
self-esteem and depression (Van der Wal et al. 2003). The causal relation-
ship between school bullying and psychopathological behavior has been
debated in terms of which comes first. Psychopathological behavior,
such as social problems, aggression, and externalizing behavior prob-
lems, may be the consequence of bullying experiences rather than the
cause (Kim et al. 2006). Explanations for bullying are also being sought
from neuroscience research involving limbic activation, varying levels of
autonomic arousal, and temperamental variations from integrity of pre-
frontal/executive regulatory capacities (Sugden et al. 2006).
Newman (2004) employed a social causation model for school

shootings, utilizing multiple sources of information. Rampage shoot-
ings in schools were explained in terms of boys feeling inferior in exist-
ing school hierarchies, especially when their masculinity was threat-
ened. Newman also posited movies, television, music, and news
reports as “scripts” for how masculinity can be asserted. Widespread
availability of guns and the increasing inability of school administrators
to identify and assist disturbed or troubled youths were also noted. This
failure is parallel to the ineffectual procedures in courts and communi-
ties trying to deal with disturbed youths. There is also the problem of
bullying in those school settings where adult supervision is minimal,
such as at recess or on playgrounds.
Court decisions also limit the discipline and authority of public
school personnel. Inability to share information among schools, social
agencies, and the police due to confidentiality and privacy restrictions
is a factor. There is a catch-22 in that school administrators are legally
restricted in responding to student misbehavior, yet may be sued by
parents of victims for failing to respond to early warning signals. Of
course, the key question that remains is whether cognizance of bullying
and other disturbing behavior can predict violence and lead to inter-
vention so that serious violence or killing will not occur.
Notable Cases of School Shootings

There are a host of cases of school shootings, and each could merit a dis-
cussion or book on its own. These cases reveal a diversity that calls into
question any attempt to create a profile from factors such as being a
loner, being a bully, or being bullied. Below is a list of recent highly pub-
licized cases, followed by a discussion of three in more detail.
1. 1996, Lynville, Tennessee: A teenager fired a rifle in a school hallway,

killing a teacher and a student.
2. 1996, Moses Lake, Washington: A junior high school student on the
honor roll used a high-powered rifle to shoot and kill two students
and a math teacher.
3. 1997, Pearl, Mississippi: A 16-year-old boy killed his mother and then
went to his high school, where he shot and killed two students and
wounded seven.
4. 1997, Paducah, Kentucky: A 14-year-old opened fire with a .22-caliber
pistol on a prayer group in the lobby of his high school, killing three
students and wounding five.
5. 1998, Edinboro, Pennsylvania: At an eighth-grade graduation party,
a 14-year-old shot and killed a science teacher and wounded an-

other teacher and two students.
6. 1998, Fayetteville, Tennessee: During graduation week, an 18-year-old
honor student shot a classmate in the school parking lot because the
victim was dating the shooter’s ex-girlfriend.
7. 1998, Jonesboro, Arkansas: Two boys, ages 11 and 13 years, set off a
school fire alarm and then shot at students who were exiting the
building, killing four students and a teacher and wounding ten.
8. 1998, Springfield, Oregon: A 15-year-old killed his parents at home, then
went to his high school, where he killed two people and wounded
twenty.
9. 1999, Notus, Idaho: A tenth-grader came to school with a shotgun
and blasted it in the hallway.
10. 1999, Littleton, Colorado: Two high school students at Columbine
High School shot and killed twelve students and one teacher and
wounded twenty others before committing suicide.
11. 1999, Conyers, Georgia: A month after Columbine, a 15-year-old did
a copycat shooting, wounding six students with a .22-caliber rifle af-
ter twelve shots. He then pulled out a .357 magnum, fired three
more shots, and put the handgun in his mouth but hesitated and
was taken into custody.
12. 2000, Mount Morris Township, Michigan: A 6-year-old shot a girl in
his first grade class when they quarreled.
13. 2001, Santee, California: A boy said to be bullied shot and killed two
students in the schoolyard.
14. 2003, New Orleans, Louisiana: Four teenage gang members in a gang-
related event shot and killed a 15-year-old and wounded three other
students.
15. 2003, Cold Spring, Minnesota: A 15-year-old shot and killed one stu-
dent and wounded another at his high school.
16. 2003, San Diego, California: There were two consecutive school shoot-
ings in one year by marginalized and socially ostracized shooters
(Palinkas et al. 2003).
17. 2004, Washington, D.C.: During a confrontation between students,
one student was shot to death in their high school.
18. 2005, Red Lake, Minnesota: A 16-year-old boy on an Indian reservation
shot and killed his grandfather and the grandfather’s companion and
then proceeded to his high school, where he shot and killed a security
guard, a teacher, and five students and wounded seven other students.
19. 2006, Nickel Mines, Pennsylvania: A 32-year-old male milk delivery
truck driver killed five Amish schoolgirls and injured five others in
an execution-style killing in a one-room schoolhouse.
20. 2007, Blacksburg, Virginia: A 23-year-old college student at Virginia

Polytechnic Institute and State University shot and killed 32 stu-
dents and faculty and then committed suicide. This was classified
as the worst peacetime shooting in U.S. history.
Although such lists are revealing, they provide almost no knowl-

edge of what motivated the killings and the background scenario. Be-
cause many of the actors committed suicide, it is often an ex post facto
attempt to reconstruct the situation relying on media sources, relatives,
and past medical/psychological records if these are made available.
Appraisal of three well-known cases illustrates the diversity.
Focused Discussion of Three Diverse Cases

Columbine
The shootings at Columbine High School in Littleton, Colorado, by Eric
Harris, age 18, and Dylan Klebold, age 17, in April 1999 garnered nearly
endless publicity (Malmquist 2006). In their shooting spree, 12 students
and one teacher were killed, and 20 students were left wounded. Plan-
ning had begun up to a year beforehand. One bomb was planted a few
miles away as a diversionary tactic. Other bombs were placed in cans to
discharge while the boys were in the school, but these failed to go off.
Upon arriving at school on the day of the killings, the boys had two 20-
pound propane bombs that they planted in the cafeteria, but only one
went off.
The onset of the shooting was in the school parking lot, where two
students were killed and eight wounded. Observers reported that the
two boys were yelling “Go! Go!” upon entering the school. Later, there
was an exchange of fire with the police. A teacher who spotted the boys
was shot in the back and killed. The boys proceeded to the second floor,
shooting along the way, and then entered the library, where one student
was killed. Intended victims were first taunted. The boys then returned
to the cafeteria and tried to detonate the failed bomb, which only caught
on fire. The finale occurred back in the library, where the boys ex-
changed gunfire with law enforcement before both committed suicide
by gunshots to the head.
Investigation revealed that their goal had been to kill hundreds and
to be remembered as the greatest mass murderers of all time. An FBI
analysis did away with several myths connected to the shootings
(Cullen 2004). Some of the myths were that they were targeting athletes
and Christians, that they were part of a “trench coat Mafia” group, and
that they were outcast Goths. The fact is that they were out to have the
highest possible body count, and no one was specifically targeted.
The shootings were not impulsive, nor did they appear to be an act
of revenge against students and teachers per se, except in the sense that
the victims were now under their control, vulnerable and helpless. In-
vestigation revealed that the boys resented the possibility that they
might later be seen as “petty school shooters.” To avert this, the goal
was to amass the largest possible number of deaths, and if the bombs
had been wired correctly, about 600 deaths would have occurred. In
terms of personality characteristics, the two were quite different. Kle-
bold was hotheaded, depressed, and suicidal. Harris had more psycho-
pathic traits—“nice” on the surface but actually cold, calculating, and
homicidal. He was described as someone who took pleasure in lying
and was contemptuous of others. Without Harris, it is doubtful that
Klebold alone could have carried out such a mission.
Jonesboro, Arkansas
In a quite different type of case, Andrew Golden, age 11 and in sixth
grade, and Mitchell Johnson, age 13 and in seventh grade, carried out a
partnership shooting at Andrews Middle School on March 24, 1998.
Much of the background material on this event is taken from the case
study by the National Research Council and Institute of Medicine (2003).
The outcome was four students and one teacher dead and nine students
and one teacher wounded. Just after recess, Andrew was seen by other
students pulling the handle on the fire alarm and exiting the school. Stu-
dents responded to the alarm and marched outside to the playground.
On a hillside 100 yards away, the two boys, dressed in camouflage
shirts, opened fire. Police arriving at the top of the hill 10 minutes later
were stunned by how young the shooters were. When apprehended,
the boys had 11 guns (Remington rifles, Smith and Wesson pistols, der-
ringers, and semiautomatics) and several hundred shells that belonged
to Andrew’s father and grandfather. The shooting clearly had been
planned, because they had a van full of provisions, including sleeping
bags and pillows, a load of junk food, and a map to a remote hunting
area where they planned to hide.
The boys had driven to the school area in Mitchell’s stepfather ’s
van, which Mitchell barely knew how to handle. Nine weeks earlier
they had planned what to do if it was raining and students did not come
out to the playground if there was a fire alarm drill. Ballistics reports re-
vealed that Mitchell killed at least one but probably two people and
wounded at least three. Andrew, a more skilled shot, fired 25 shots, kill-
ing three and wounding at least two others in the course of 5 minutes.
Although all but one of the victims shot were female, it remains conjec-
tural whether anyone was specifically targeted.
Many students did not know that the two boys knew each other
beyond a casual level. Mitchell had arrived in Arkansas only 2 years
earlier after being raised in Minnesota, where his parents had gone
through a heated divorce. His mother was a correctional officer, and af-
ter the divorce she first took a job in a federal prison in Kentucky. Two
sons were born in the marriage: Mitchell and a younger brother. Prior
to the divorce Mitchell’s father was described as a hard drinker, a disci-
plinarian, mean-tempered, and explosive. In addition, when Mitchell
was 8 years old, an older boy had begun to rape him, and later his
younger brother, repeatedly. The mother married again for a third time
to a man who had served prison time on a drug charge, but they had
settled down and the family was living in a trailer camp.
Mitchell presented a mixed picture. Some saw him adjusting well as
a new student, being polite and singing in a church choir. However, his
dark side was seen in belligerent, boastful, and bullying behavior. A
few weeks before the shooting he had been disciplined for wearing a
baseball cap in school. He was both furious about this and unrepentant.
He wrote a paper stating he had some squirrels he wanted to kill, which
led the teacher to give the paper to the school principal. Although
adults saw Mitchell as a troubled boy, his peer group saw him as a
moody boy with a temper who was seeking some place in the social
pecking order.
Andrew Golden was younger, but about half of the people inter-
viewed saw him as the leader. He came from a gun-owning family, the
guns used were from his family, and he was a marksman. An only child,
he was seen by others as belligerent, although his parents saw him as
doing no wrong. School behavior was erratic, varying from class clown
to “chip off the old block,” similar to the way his father had been at the
same school as a boy.
Psychiatric reports on the two boys described Andrew as the more
troubled, and in contrast to Mitchell, he never spoke to anyone about
the incident, nor did his family. Observations in the juvenile facility
where the boys resided for several years after the trial reflect wide dif-
ferences. Mitchell was cooperative, repentant, and liked by staff; An-
drew remained silent and kept to himself. An insanity defense was
raised for Andrew, but the court ruled a juvenile was not entitled to
such a defense, a decision upheld by the Arkansas Supreme Court. Be-
cause the boys were not yet 14 years of age, they could be detained only
until they reached 21, which left many in the community incredulous.
Virginia Tech
There have been earlier notable cases of lethal violence at a college level
that are sometimes ignored in discussions of “school shootings” (Simon
1996). Charles Whitman at the University of Texas in 1966 killed his
wife and mother and then, from a tower on the campus, shot 13 people
to death and wounded 31 others before he committed suicide. Gan Lu,
an astrophysicist at the University of Iowa, shot and killed a physics
professor and a rival who won an award Lu had hoped to receive; he
then killed five other faculty members and wounded another person
before committing suicide. In 1993 Wayne Lo, an 18-year-old student at
Simon’s Rock College in Massachusetts, used a high-powered assault
rifle to kill a professor in his car and a student in the library and then
wounded four others. Perhaps it is stretching school shootings to in-
clude Ted Kaczynski, the Unabomber, but he was a one-time college
professor who sent bombs through the mail that killed 3 persons and
wounded 23 others over 17 years. The victims were either university
professors or worked in technology.
The mass killing at Blacksburg, Virginia, on April 16, 2007, by a 23-
year-old college student is a recent picture of a troubled young male
who stumbled along until the fatal day. Seung-Hui Cho murdered
32 people and wounded 17 and then committed suicide, in the worst
peacetime shooting in American history. Many factors remain elusive.
That morning he went to a dormitory and killed a male and female stu-
dent whose selection remains a puzzle. Cho then mailed off a manifesto
to NBC News with pictures of himself posing with guns and video clips
and making a rambling verbal attack on wealthy people. In an essay
found in his room, he blamed practically everyone except himself for
what he was doing—women, religion, the wealthy, debauchery, and
“deceitful charlatans” (“The Virginia Tech Massacre” 2007).
Almost a 2-hour gap occurred before Cho then appeared in a class-
room building half a mile away. He locked the doors with chains so that
those inside could not escape and proceeded into various classrooms,
trying to kill everyone in them with two semiautomatics: a Glock 9 mm
and a Walther P22. Those who survived said he was silent as he went
about shooting students and faculty at close range, putting two or three
bullets in each to make sure they died. When police burst into the build-
ing, he shot himself.
Before any name was released by officials, some classmates guessed
that the shooter was Cho. He had rarely spoken to anyone, referred to
himself as “Question Mark,” hid behind sunglasses, and was seen as in-
timidating. In a creative writing class, the themes of his papers involved
money, fury, sex, religion, and overbearing adults. In 2005, two female
students complained to police that Cho was stalking them but did not
press charges. At that time a district court found him “mentally ill and
an imminent danger to self and others,” but he never received any treat-
ment. His situation later exposed the flaws in a labyrinthine mental
health system (Schulte and Jenkins 2007). A court had committed him
for “involuntary outpatient commitment,” which exists in many states
as a category of civil commitment. The problem is in the follow-up and
the lack of clarity as to who assumes responsibility for ensuring that the
individual receives treatment. Is it the court system, some agency, or the
individual himself? In Cho’s case, no one assumed responsibility.
Discussion
There is a distinction between the pervasive problem of “school vio-
lence” and lethal school (rampage) shootings. The former is closer to the
problems of juveniles with ongoing conduct problems and is the more
classic picture of juvenile offenders in which a subset become violent.
Social and family disorganization are relevant. Lethal school shootings
present a different set of problems. Attempts are often made to create
profiles from these rampage killers’ characteristics, such as being a
loner and avoiding people, or to propose psychiatric diagnoses in retro-
spect. A major limitation is that such profiles include a great many
youths with personality characteristics and social difficulties similar to
those of the few who carry out such acts.
Consider the contrasting personalities of Golden and Johnson in
their joint act at Andrews Middle School in Arkansas, or contrast them
with that of Cho at Virginia Tech. A major obstacle to knowledge about
school killers is the lack of psychiatric and psychological data on the
perpetrators, either because they have committed suicide or because
such data are kept private. Hence, our knowledge is often based on
newspaper reports or police statements. One approach tried to system-
atize such offenders within four operant styles: adaptive, conservative,
integrative, and expressive (Fritzon and Brun 2005). The adaptive of-
fender targets specific individuals; the conservative is affected by an ex-
ternal trigger in which self-esteem is threatened; the integrative targets
others whom he identifies with his internal conflicts, and then commits
suicide; and the expressive is randomly violent.
The U.S. Secret Service developed a profile of 41 school shooters
from 37 school incidents (Vossekuil et al. 2000). The most frequent mo-
tive was revenge, with about three-fourths of the perpetrators threaten-
ing suicide before an attack. Although the report concluded there was
no evidence the shootings were the result of a mental disorder, the per-
petrators were described as feeling extremely depressed or desperate.
Two-thirds felt persecuted, and three-fourths were dealing with a major
change in a relationship or a loss of status.
This profile was consistent with a report by Meloy et al. (2001) of ad-
olescent mass murderers (not all in schools) in which a precipitating
event of personal loss or status threat had occurred. In 75% of the cases,
the shooters had communicated threats beforehand. The report empha-
sized bullying and the frequent motive of revenge. McGee and De-
Bernardo (1999) described 12 shooting incidents in middle and high
schools of “classroom avengers.” Again it was stated the shooters did
not show overt signs of a mental disorder, yet had a significantly de-
pressed mood. On reviewing these reports, the depression theme recurs
often enough to raise the question of whether those in contact with ad-
olescents are sufficiently adept at detecting depressive states.
An FBI report focused on the “myths” connected to school shootings
(O’Toole 2000). Among the myths examined were links to revenge, an-
ger about being bullied, unresolved anger about other matters, gun
availability in the homes of the perpetrators, the impact of violent video
games, and being a loner. The study argued against the idea that the
shooters shared these attributes. The debate as to whether the shooters
were mentally ill or had “mental problems” continued to intrude in all
these studies. Although the brooding ruminations of Cho suggest a se-
rious mental illness, other shooters may simply be referred to as having
a prominence of strong emotions whose ascendancy takes over their de-
cision making.
The FBI focus was on threat assessment, and they employed the con-
cept of “leakage.” The student intentionally or unintentionally “leaks”
a cry for help by way of feelings, thoughts, fantasies, or intentions about
an impending violent act. The clues may be subtle threats, boasting, in-
nuendoes, or predictions that can appear in stories, diaries, essays, let-
ters, and drawings. The leakage may involve “jokes” about violence or
destruction that are then retracted with “I was just joking.” At times
there are efforts to get friends or classmates to help with preparations
for a violent act.
A troublesome question deals with unshared information that was
available in the communities before such shootings. If such suspicions
are shared, a question arises as to how wide the sharing should be. A
retrospective study of 253 school-associated violent deaths in the
United States found more than half the perpetrators had signaled the
future event by notes, threats, or journal entries (Anderson et al. 2001).
However, unless we opt for a society with a norm of routinely inform-
ing on each other, with every suspicion being investigated within

schools or by some agency, looking for and sharing such information is
not likely to be a practical solution, especially given adolescents’ devo-
tion to secrecy.
In response to public concern about school shootings, Congress
asked the National Research Council of the Institute of Medicine to
study school shootings that had occurred in a 2-year period (National
Research Council and the Institute of Medicine 2003). Shootings in
urban and suburban communities were distinguished. Shootings in an
urban environment involved grievances between individuals known
in their communities. In the suburban milieu there was an overall low
level of crime and violence, with the boys being seen as alienated, asso-
ciating with delinquents, and having a recent change in behavior, un-
recognized mental health problems, and easy access to guns. In five of
the six communities studied, there had been rapid social change, with
parents and teachers having a poor understanding of the impact this
had on adolescents. The study report emphasized that understanding
school shootings requires recognizing a conjunction of three things:
a person with some predisposing potential for violence, a situation with
elements that create a risk of violent events, and, usually, a triggering
event. The emphasis was on using a narrative approach for an explana-
tion of particulars and societal structural factors that make violence
more likely. Focusing on situational factors, perhaps altering one link in
the chain of events could have deflected the ultimate act.
Prevention
None of the measures proposed in this section will eliminate the occa-
sional rampage homicidal shooting in schools. However, it is possible
that intervention at earlier stages might head off problems that could
eventuate in serious school violence. Such primary prevention of shoot-
ings could operate silently if intervention is successful. Two overriding
policy questions arise with increased preventive efforts. One is whether
families and children are willing to lose some liberties by greater in-
trusions into their lives when the payoff may be low. The second is
whether the public is willing to assume the additional costs that such
measures require.
A striking feature of writings on prevention of “school violence” is
the contrasting emphasis by those in clinical fields compared with those
in education. Clinically oriented writers focus on serious assaultive ep-
isodes, whereas educators are more concerned about the daily preva-
lence of various types of violence that intrude on educational processes.
Educators’ concerns are with the patterns of victimization that perme-

ate some schools, such as thefts, verbal abuse, bullying, vandalism,
fights, punching, grabbing, and sexual taunts. Prevention in this con-
text means a school milieu where students and faculty are not assaulted
and do not live in a state of fear. Tracking is difficult and often subjec-
tive, so the focus is on physical acts.
Diverse attempts to control such actions have had varying degrees
of success. Students with conduct disorders or emerging personality
disturbances may be referred for medications or taught cognitive-
behavioral techniques to foster better self-control. Alertness to detect
students with interpersonal difficulties related to various types of vio-
lence is important. The linchpin is screening for susceptible students
with whom to intervene and having the right personnel available to
connect. Even then, as in psychotherapy, some students resist participa-
tion. Similar difficulties arise when troubled students end up in court
proceedings that are plagued by loopholes and by inadequate follow-
up if therapeutic approaches are ordered, as occurred with Cho.
Various interventions are being tried, such as social skills training,
anger management, empathy training, and training in moral reasoning.
Some schools have introduced conflict resolution programs, but these
are based on the assumption that students want to be safe and free from
conflict; the reality is that in some schools, carrying a gun or being will-
ing to fight is not only seen as masculine but as a way to be safe.
Similar dilemmas arise in programs intended to control bullying.
One assumption is that asking students and faculty not to go along with
bullying will get a positive response and that those in authority will
take action in confronting such behavior. If those in authority do not re-
spond, or prefer other solutions, the student is left doubly exposed—
from the bullying itself and the failure of responsible adults (Hunter et
al. 2004). A dual approach is needed, which means intervening with the
bully and supporting the victim. Failure often reflects a lack of applica-
tion and enforcement from a majority of the participants. In some
schools, prevention may have to include guarding the school entrance.
Metal detectors have become an approach in such settings, with school
security checking bathrooms, monitoring groups loitering near the
school, and using video surveillance devices. An extension of this ap-
proach has been legislation prohibiting possession of a weapon within
so many feet of school property.
Conclusion
Those who work with adolescents and families have knowledge of
what may prevent some violence in schools. To what extent various risk
factors are dealt with so that some future act of violence is thwarted
may never be known. It is often a matter of trying to intervene with a
focus on risk factors (MacNeil 2002). An important area of potential risk
is the adolescent’s social environment, such as income level, availability
of drugs and alcohol, and family disorganization. A second area is the
psychological and neuropsychiatric aspects of an individual in terms of
impulse control, beginning conduct problems, and possible psychiatric
disorders. A third area of risk focuses on the psychological milieu of the
adolescent. This includes family conflict, inconsistent or overly harsh
punishments, and failing or changing educational performance.
The key is connecting with those adolescents most in need of inter-
vention and then taking action and involving people with sufficient ex-
pertise. In cases like Columbine, Jonesboro, and Virginia Tech, the trou-
ble was in either no one detecting the individual or too many people
ignoring what they saw and heard. An approach of trying to predict
who might engage in a mass shooting is wasteful because it bypasses
approaches that may thwart such a later outcome. It may be helpful to
zero in on high-risk students for intervention, first through implement-
ing better sharing of records and ideas within school systems and
among teachers, then by extending this to clinicians and the juvenile
justice system. As noted, this approach necessitates caution about vio-
lations of privacy and also requires a sensitive person in charge to exer-
cise wise judgment in sharing information about students.
Key Points
The following preventive measures are recommended to lessen the
likelihood of school violence:
■ Increased security measures—more security officers, metal
detectors, electronic devices, and photographic devices in and
around schools
■ Greater sensitivity to signs and symptoms that indicate a troubled
student, or one communicating harm, such as by “leakage.” This
is superior to attempts to profile “school shooters.”
■ Better training for teachers in normal and abnormal adolescent

psychology
■ Availability of competent professional personnel to assess, and
possibly treat, students seen as needing assistance
■ Greater sharing of records among schools, police, courts, and com-
munity agencies, which requires clarification of roles and confi-
dentiality issues
■ Efforts to counter the “macho” or hypermasculine milieu that per-
meates many schools, based on an athletic culture and bullying,
by offering rewards for other models and activities;
this should involve inquiry into a student culture where bullying
is tolerated by the majority.
■ Zero tolerance in schools not only for weapons but also for behav-
ior such as pushing, shoving, and intimidation. Threats raise
issues of free speech.
■ Better integration among schools, courts, and mental health facil-
ities for follow-through to avoid the type of situation that occurred
with Cho at Virginia Tech
References
Anderson M, Kaufman J, Simon TR, et al: School-associated violent deaths in
the United States, 1994–1999. JAMA 286:2695–2702, 2001
Beavan K: School violence on the increase. New York Times Educational Sup-
plement, February 12, 1970, p 16
Centers for Disease Control and Prevention: School-associated homicides—
United States, 1992–2006. MMWR Morb Mortal Wkly Rep MMWR 57(2):
33–36, January 18, 2008
Crews GA, Counts MR: The Evolution of School Violence in America. Westport,
CT, Praeger, 1997
Cullen D: The depressive and the psychopath: at last we know why the Colum-
bine killers did it. Slate, April 20, 2004. Available at http://slate.msn.com/
id/2099203. Accessed April 20, 2004
Dinkes R, Cataldi EF, Kena G, et al: Indicators of School Crime and Safety: 2006.
U.S. Departments of Education and Justice. Washington, DC, U.S. Govern-
ment Printing Office, 2006
Fritzon K, Brun A: Beyond Columbine: a faceted model of school-associated ho-
micide. Psychology, Crime and Law 11:53–71, 2005
Hunter L, MacNeil G, Elias M: School violence: prevalence, policies, and pre-
vention, in Juvenile Justice Sourcebook. Edited by Roberts AR. New York,
Oxford University Press, 2004, pp 101–125
Kaufman P, Chen X, Choy S, et al: Indicators of School Crime and Safety, 2001.
Washington, DC, U.S. Government Printing Office, 2001
Kim YS, Leventhal BL, Koh Y, et al: School bullying and youth violence. Arch
Gen Psychiatry 63:1035–1041, 2006
MacNeil G: School bullying: an overview, in Handbook of Violence. Edited by
Rapp-Paglicci I, Roberts AR, Wodarski J. New York, Wiley, 2002, pp 247–
261
Malmquist CP: Homicide: A Psychiatric Perspective. Washington, DC, Ameri-
can Psychiatric Publishing, 2006
McGee JP, DeBernardo CR: The classroom avenger: a behavioral profile of
school based shootings. The Forensic Examiner 8:16–18, 1999
Meloy JR, Hempel AG, Mohandie K, et al: Offender and offense characteristics
of a nonrandom sample of adolescent mass murderers. J Am Acad Child
Adolesc Psychiatry 40:719–728, 2001
Midlarsky E, Klain HM: A history of violence in the schools, in Violence in
Schools: Cross-National and Cross-Cultural Perspectives. Edited by Den-
mark F, Krauss HA, Wesner RC, et al. New York, Springer, 2005, pp 37–57
Nansel T, Overpeck M, Pilla R: Bullying behavior among U.S. youth: preva-
lence and association with psychosocial adjustment. JAMA 285:2094–
3100, 2001
National Institute of Education: Violent Schools, Safe Schools: The Safe Schools
Study Report to the Congress, Vol 1. Washington, DC, U.S. Government
Printing Office, 1977
National Research Council and Institute of Medicine: Deadly Lessons: Under-
standing Lethal School Violence. Edited by Moore MH, Petrie CV, Braga
AA, et al. Washington, DC, National Academies Press, 2003
Newman KS: Rampage: The Social Roots of School Shootings. New York, Basic
Books, 2004
O’Toole ME: The School Shooters: A Threat Assessment Perspective. Quantico,
VA, Critical Incident Response Group, National Center for Analysis of Vi-
olent Crime, 2000
Palinkas LA, Prussing E, Landsverk J, et al: Youth-violence prevention in the af-
termath of the San Diego East County school shootings: a qualitative as-
sessment of community explanatory models. Ambul Pediatr 3:246–252,
2003
Schulte B, Jenkins CL: Cho didn’t get court-ordered treatment. Washington
Post, May 7, 2007, p A01
Simon RI: Bad Men Do What Good Men Dream: A Forensic Psychiatrist Illumi-
nates the Darker Side of Human Behavior. Washington, DC, American Psy-
chiatric Press, 1996
Sugden SG, Kile SJ, Henderson RL: Neurodevelopmental pathways to aggres-
sion: a model to understand and target treatment in youth. J Neuropsychi-
atry Clin Neurosci 18:302–317, 2006
Van der Wal M, Cees AM, Hirasing R: Psychosocial health among young vic-
tims and offenders of direct and indirect bullying. Pediatrics 111:1312–1317,
2003
The Virginia Tech massacre. The Economist, April 21, 2007, pp 27–29
Vossekuil B, Reddy M, Fein R, et al: Safe School Initiative: An Interim Report on
the Prevention of Targeted Violence in Schools. Washington, DC, U.S. Se-
cret Service, 2000
C H A P T E R 2 8
Clinically-Based Risk
Management of Potentially
Violent Patients
The goal of violence risk management is to eliminate or decrease the

chance of another person’s injury or death resulting from actions by a
patient, as well as the potential legal liability. Risk management princi-
ples usually represent ideal or best practices, whereas the legal stan-
dard of care requires only the provision of ordinary or reasonable, pru-
dent care (Simon 2005). Risk management decisions based solely on the
clinician’s defensive desire to avoid malpractice liability or to provide a
defense against a malpractice claim can increase liability exposure by
engendering worst practices.
Clinically-based risk management principles are patient centered, sup-
porting the treatment process and the therapeutic alliance (Simon 2004).
Clinically-based risk management upholds the ethical principle of “first do
no harm.” A working knowledge of the legal regulation of psychiatry as-
sists the practitioner in managing clinical-legal dilemmas that frequently
arise in the treatment and management of potentially violent patients,
while also preventing disruption of the doctor–patient relationship.
Defensive practices can be divided into the preemptive and the
avoidant. Preemptive practices utilize procedures and treatments aimed
at preventing or limiting liability—for example, hospitalizing a patient
at low to moderate risk of violence who could be effectively treated as
555
an outpatient. Avoidant practices forego necessary procedures or treat-

ments based on the fear of being sued, even though the potentially vio-
lent patient would benefit from the interventions—for example, failing
to involuntarily hospitalize a litigious patient at high risk for violence
who is refusing voluntary hospitalization, or involuntarily hospitaliz-
ing a patient primarily as a risk management measure.
Potentially violent patients frequently confront the clinician with
complex diagnostic, treatment, and management issues, most often in
outpatient, inpatient, and emergency department settings. They can
also create judgment-numbing anxiety in the clinician. Consultation
with a colleague is always an option. The clinician should “never worry
alone” (T.G. Gutheil, personal communication, December 2002). Con-
sultation supports good clinical care while also providing a “biopsy” of
the standard of care. Clinically-based risk management puts the pa-
tient’s well-being first, avoiding defensive practices that can harm both
the patient and the clinician.
Some of the issues that potentially violent patients present to the cli-
nician are illustrated in the following hypothetical case.
Case Example
Dan, a 36-year-old married man, enters treatment with a psychiatrist for
depression, insomnia, and anxiety. His symptoms arise from long-
standing marital strife. The psychiatrist agrees to see the patient once a
week for psychotherapy and medication management. Within a month
after beginning treatment, Dan begins to suspect that his wife is having
an affair. He finds “racy” e-mails sent to his wife by a coworker. A near-
violent confrontation occurs. She adamantly denies having an affair. He
leaves the home, goes to a hotel, and calls his psychiatrist.
The patient’s psychiatric condition rapidly worsens over the next
week. He is unable to sleep. He ruminates about killing his wife. He ad-
mits to having guns at home. The psychiatrist sees the patient more fre-
quently, adjusts medications, and explores Dan’s potential for violence
toward his wife. Dan has a history of violent rages, although he has
never harmed his wife or anyone else. He intends to break into his
house, get his guns, and threaten to kill his wife.
The psychiatrist informs the patient that psychiatric hospitalization
is necessary. The patient refuses. The psychiatrist performs and docu-
ments a careful violence risk assessment, which indicates that Dan is at
acute, high risk for violence. If he does not enter the hospital voluntarily,
the psychiatrist will be forced to involuntarily hospitalize him. Dan re-
luctantly agrees to hospitalization. With his permission and in his pres-
ence, the psychiatrist calls Dan’s wife and informs her of the violent
threats by her husband. She is not surprised. She is told that Dan is go-
ing directly to the hospital. The locks on her house are changed. A secu-
rity system is installed after she speaks with the police.
Clinically-Based Risk Management ❘ 557
The psychiatrist tells Dan’s wife to remove all guns and ammunition
from the home and secure them in a place unknown to the patient. She
is asked whether guns might be kept in a car, at work, or anywhere else.
The psychiatrist asks for a callback from the patient’s wife once the guns
and ammunition are removed, which he receives within an agreed-
upon time.
Dan shows rapid improvement while on the psychiatric unit. He is
seen daily by his psychiatrist. His medications are adjusted. The treatment
team provides valuable input regarding the patient’s behaviors. No vio-
lent outbursts or threats occur. Depression and agitation moderate.
With the patient’s approval, the psychiatrist arranges for a meeting
with the patient and his wife together on the inpatient unit. With the
psychiatrist present, Dan is able to express his anger appropriately and
safely. His wife openly expresses her disappointments with the mar-
riage, stating that she has had the entire burden of caring for their three
children. The meeting is very emotional but frank and productive. The
psychiatrist recommends marital counseling, which both accept. They
continue to talk by phone.
After performing and documenting a careful violence risk assess-
ment, the psychiatrist determines that Dan’s risk of violence is now low.
As agreed upon during the inpatient meeting, the doctor and the patient
will inform the patient’s wife about the date and time of discharge. Dan
agrees to enter the hospital’s partial hospitalization program on the day
after his discharge. The couple remains separated.
Initially, the psychiatrist sees Dan three times a week to ensure sta-
bilization of his condition. Dan and his wife continue marital therapy.
By mutual agreement, the couple have no other direct contact with each
other. The patient is grateful for the psychiatrist’s care during the crisis.
The therapeutic alliance is strengthened.
Outpatients: Duty to Warn and Protect

In Tarasoff v. Regents of the University of California (1976), the California
Supreme Court recognized that a duty to protect third parties from pa-
tient violence was imposed only when a special relationship existed be-
tween a foreseeable victim, the individual whose conduct created the
danger, and the defendant. A doctor–patient relationship creates a spe-
cial relationship that supports a duty to exercise reasonable care to pro-
tect others from a patient’s acts of violence. In most states, a psychother-
apist has a duty, established by case law or statute, to act affirmatively
to protect an endangered third party from a patient’s violent or danger-
ous acts. Although some courts have declined to find a Tarasoff duty in a
specific case, a number of courts have recognized some variation of the
original Tarasoff duty. Very few courts have limited or rejected the Tara-
soff doctrine (Evans v. United States 1995; Green v. Ross 1997). Most courts
have not found a duty to warn or protect absent a foreseeable victim.
The Tarasoff court did not use the phrase “imminent danger.” “Immi-
nent” appears frequently in the mental health literature. It is common par-
lance among clinicians. It is also a legal term of art found in civil commit-
ment statutes; in duty to warn and protect statutes and case law, usually
under the rubric of dangerousness; and in seclusion and restraint policies.
“Imminence” of violence is another word for the short-term prediction of
violence, for which no standard of care exists. “Imminent” violence
should not be a substitute for comprehensive violence risk assessment. It
is a myth firmly entrenched in both psychiatry and the law (Simon 2006).
Clinicians have found that the duty to protect provides more lati-
tude for treatment interventions than the original duty-to-warn doc-
trine. Except in states with immunity statutes limiting the responsibility
of therapists for the patients’ violent acts, no hard-and-fast rules have
been created requiring clinicians to employ specific interventions to
warn and protect endangered third parties. In jurisdictions where no
duty to warn or protect currently exists, case law from other states may
be applied in deciding suits that allege such a duty.
Generally, courts have held that the therapist’s control over an out-
patient is not sufficient to establish a duty to protect without a foresee-
able victim. In treatment of an outpatient, the Tarasoff duty applies
when there is evidence, through either threats or acts, that the patient is
potentially violent to a specific, foreseeable victim. The dangers must be
substantial, involving serious bodily harm or death. If no threats or vi-
olent acts are uncovered after careful clinical evaluation, liability is un-
likely even if violence should occur.
The duty to warn does not obviate implementing other clinical in-
terventions that may be more effective. Simply warning an endangered
third party is rarely sufficient by itself. Other clinical interventions are
usually required—for example, seeing the patient more frequently, ad-
justing medication, or hospitalization.
When the clinician decides to issue a warning, the warning should
take place, if possible, in the presence of the patient and with the pa-
tient’s consent. In an emergency, however, the clinician does not need the
patient’s consent. With few exceptions, a legal problem arising in psychi-
atric treatment can be successfully addressed through good clinical prac-
tice and a clear understanding of the relevant legal requirements. Thus,
warning should be used after other clinical interventions have been tried
and failed or in conjunction with other clinical interventions.
Generally, if the clinician decides to warn, an interview may be ar-
ranged or a phone call made so that the potential victim can ask ques-
tions. Language difficulties between the clinician and the endangered
person may contribute to tragic consequences unless detected and
clarified. Sometimes, a trusted bilingual third party may act as a go-

between. The warning should be made clearly.
How a warning is given is the crucial factor (Simon and Shuman
2007). When the clinician discusses the warning with the patient prior
to giving it, the result for therapy and the alliance can be positive. Not
discussing the warning with the patient usually turns out badly for the
therapeutic alliance and the therapy. Warning a potential victim of
patient violence must be done with discretion. If the potential victim
feels that evasive action can be taken and that the therapist is genuinely
concerned and acting responsibly, the warnings are received positively.
When a potential victim sees no options for evasion and the therapist is
perceived as not behaving responsibly, a profound negative reaction
can occur. The threatened individual may take preemptive violent ac-
tion against the patient.
With the advent of e-mail and the teleconference, clinicians are treat-
ing and managing psychiatric patients at a distance. Treating a patient
by e-mail or teleconference creates a doctor–patient relationship with
an unabridged duty of care. Active case management of a potentially vi-
olent patient should not be attempted in this mode. An in-person, face-
to-face examination of the patient is necessary.
In the duty to warn and protect endangered third parties from patient
violence, a number of risk management steps should be considered (see
Table 28–1).
Inpatient Discharge
Although the Tarasoff duty was originally applied to the outpatient set-
ting, the same legal duty to protect individuals and society from harm
by mental patients arises for the release of violent inpatients. Generally,
the duty to warn is of narrower scope than the duty not to discharge a
violent patient. In outpatient cases that involve failure to warn and pro-
tect an endangered third party, the threat of violence is serious, violence
is foreseeable, and usually the victim is identifiable. The duty not to dis-
charge a violent inpatient has a broader scope because the patient may
not express specific threats toward persons or groups, thus posing a
threat to the general public. In hospital discharge cases, the clinician’s
duty extends beyond that owed to readily identifiable victims. Psychi-
atrists face greater liability exposure for the release of potentially vio-
lent hospital patients than in outpatient cases alleging a Tarasoff duty
(Simon 1992).
In inpatient release cases, the courts have held that there is a duty to
control, with or without a foreseeable victim. The duty to evaluate the
TABLE 28–1. Duty to warn and protect: risk management

• Perform and document comprehensive violence risk assessments that
inform treatment and management interventions. Evaluate and document
effectiveness of interventions.
• Perform violence risk assessments at critical junctures (e.g., on admission to
and discharge from an inpatient unit; when making outpatient decisions to
warn and protect).
• Violence risk assessment is a process, not an event. Document each risk
assessment.
• Obtain prior patient records. Courts have held clinicians responsible for not
knowing important information contained in the patient’s prior records.
• Document decision-making rationale and risk-benefit assessment (e.g.,
risks and benefits of continued outpatient treatment versus risks and
benefits of hospitalization).
• Consult with another clinician and/or attorney when confronted by
clinical-legal uncertainty in the management of a potentially violent patient.
• Issue appropriate warnings to potentially endangered third parties, if
clinical interventions fail. Warnings should be issued even when the
endangered third party is aware of the potential violence by the patient.
• Avoid reflexive warning that can harm the patient by unnecessarily
breaching confidentiality and exposing the patient to a preemptive act of
violence by the endangered third party. Also recognize that warning, by
itself, is usually insufficient.
• Hospitalize—voluntarily or involuntarily—patients at acute, high risk for
violence, whether or not an individual or the general public is endangered.
• Implement clinical interventions, including involuntary hospitalization, in
the best interest of the patient and for the safety of others, rather than as a
defensive action to avoid perceived liability.
• Consider first the duty to protect, rather than the duty to warn, allowing the
practitioner to exercise clinical options that also preserve patient
confidentiality. It may be necessary to invoke both options.
patient for the risk of violence according to usual professional stan-

dards would obviate a Tarasoff duty because the continued high risk of
violence would require further hospitalization. Litigation involving re-
lease of potentially violent patients who harm others will turn on
whether the clinician was negligent in evaluating the patient prior to
discharge. If the court releases a patient deemed dangerous by the clini-
cian, the clinician should go on record with his or her concern about the
patient’s potential for violence. The judicial decision to release insulates
the psychiatrist from liability (Simon and Shuman 2007).
Risk management principles applicable to discharge from an inpa-
tient facility are noted in Table 28–2.
TABLE 28–2. Patient discharge from inpatient facilities:

risk management
• Obtain records of previous treatments and hospitalizations. The past is
prologue.
• Document thoroughly the decision-making process regarding discharge
planning (e.g., note risks and benefits of continued hospitalization versus
risks and benefits of discharge).
• Systematically assess and document the patient’s risk of violence, including
the treatment and management interventions informed by the violence
assessment. Evaluate the effectiveness of interventions.
• Conduct violence risk assessments regularly to determine the current level
of risk. Violence risk assessment is a process, not an event. Document each
risk assessment.
• Consider obtaining a consultation from a clinician and/or attorney
regarding complex clinical-legal issues surrounding the discharge of
patients at risk for violence.
• Utilize input from the treatment team in discharge decision making.
The staff has observed and treated the patient 24 hours a day, 7 days
a week.
• Inform the person previously threatened by the patient of the patient’s
impending discharge. The warning should be made even when the clinician
is certain that the person is aware of the patient’s discharge and the
potential danger.
• Arrange a meeting with the patient’s spouse, partner, and family (if
available) to discuss issues relating to the patient’s impending discharge.
Preferably, significant others should be seen individually initially to
determine if a meeting is workable. Determine if significant others will
support or destabilize the patient.
• Structure after-care planning for maximal adherence to treatment by the
patient.
• Patients should be seen for outpatient treatment as soon as possible after
discharge. Patients should be provided written instructions regarding
postdischarge treatment and management.
• Discuss with the patient who to contact or where to go for help, if he or
she fears a loss of control over violent impulses. Provide resources and
telephone numbers. Be sure that the telephone numbers are correct.
• Do not discharge patient solely based on denial of insurance benefits or for
other financial reasons.
• Educate the patient and significant others about the patient’s mental
disorder and the necessity for continued treatment.
• Involuntarily hospitalize patients at acute, high risk for violence who refuse
voluntary hospitalization.
Patients With Access to Guns

All patients at risk for violence must be asked if there are guns at home
or easily accessible elsewhere, or if they intend to purchase a gun. Pa-
tients who have a gun at home usually have more than one gun. Per-
sons with guns in the home are at greater risk of dying from a homicide
than those without guns in the home (Dahlberg et al. 2004). Gun safety
management requires a collaborative plan between the clinician, the pa-
tient (if possible), and a designated person responsible for removing
guns from the home. The designated person should be told that all guns
must be removed, even if he or she does not believe the patient will use
a gun to harm others. Denial may doom the gun removal plan.
A callback to the clinician from the designated person is requisite to
confirm that guns and ammunition have been removed and secured ac-
cording to the agreed-upon plan. Verification is essential to gun safety
management. The verification principle in gun safety management ap-
plies to outpatients, inpatients, and patients evaluated in emergency de-
partments, although its implementation may vary according to the clin-
ical setting. Document the gun removal plan and verification (Simon
2007; see also Table 28–3).
Documentation
Documentation is an essential part of good patient care. It encourages
the clinician to sharpen clinical focus and to clarify decision-making ra-
tionale (Simon 2004). The record is an active clinical tool, not just an in-
ert document. The clinician treats the patient, not the chart. Documen-
tation is a risk management measure that also supports good clinical
care.
For patients at risk for violence, it is necessary to document clinical
interventions and the rationale for clinical decisions. Documentation
should specifically address what was done, the reason(s) for doing it,
and the rationale for rejecting alternative interventions and treatment
(Slovenko 2002). Violence risk assessment should be contemporane-
ously documented. If a malpractice claim is made against the psychia-
trist, contemporaneous documentation assists the court in considering
the many clinical complexities and ambiguities that exist in the assess-
ment, treatment, and management of patients at risk for violence.
TABLE 28–3. Gun safety management

• Ask patients at risk for violence about guns at home or elsewhere (e.g., car,
office, or workplace). Patients who have a gun at home usually have more
than one gun.
• Consider invoking the emergency exception to consent (see Chapter 21) if
a patient who is at high risk for committing violence toward others
withholds consent to contact the patient’s significant others.
• Designate a willing, responsible individual, usually a family member or
partner, to follow through with the gun removal plan as instructed by the
clinician. The patient should be included in gun safety planning, if
possible.
• Confirm that all guns and ammunition were separated and removed from
the home and safely secured in a location unknown to the patient. There is
no safe storage at home.
• Obtain a callback from the responsible, designated person confirming that
the gun(s) and ammunition have been removed and safely secured.
• Document that the gun removal plan was implemented by the designated
individual and that a callback was received from that individual
confirming the removal of guns according to the plan.
• Verification is the essence of gun safety management. Merely asking a
significant other to remove the gun(s) from the home without a confirming
callback can end tragically. Family members may not follow through with
removal of gun(s) from the home due to denial and for other reasons (see
Simon 2007).
Key Points
■ Clinically-based risk management principles are patient centered,
supporting the treatment process and the therapeutic alliance.
Unduly defensive risk management practices based on the clini-
cian’s fear of being sued or the need to provide a legal defense if
sued may subvert good patient care and invite a lawsuit.
■ Risk management principles usually represent ideal or best prac-
tices. The legal standard of care requires only the provision of ordi-
nary or reasonable care.
■ No standard of care exists for the prediction of violence. There is
no research that supports the ability of the clinician to predict who
will or will not be violent. The purpose of violence risk assessment
is to identify treatable and modifiable risk and protective factors
that will inform the clinician’s treatment and management of the
patient.
■ The clinician’s treatment and management of the potentially vio-
lent patient often requires a team approach that includes signifi-
cant others, as well as consultation with other mental health pro-
fessionals, inpatient staff, lawyers, law enforcement officials, and
the judicial system. Never worry alone.
■ The duty to protect endangered third parties from patient vio-
lence should be considered a national standard of care for all men-
tal health professionals.
■ A working knowledge of the legal regulation of psychiatry assists
the practitioner in managing the clinical-legal dilemmas that
often arise in treating and managing potentially violent patients,
while also preserving the doctor-patient relationship.
■ Active case management of a potentially violent patient should
not be attempted by e-mail or video conference.
■ Documentation is an essential aspect of good patient care and
clinically-based risk management.
References
Dahlberg LL, Ikeda RM, Kresnow MJ: Guns in the home and risk of violent
death in the home: findings from a national study. Am J Epidemiol 160:929–
936, 2004
Evans v. United States, 883 F.Supp 124 (5D Miss. 1995)
Green v. Ross, 691 502d.542 (Fla. App. 1997)
Simon RI: Clinical Psychiatry and the Law, 2nd Edition. Washington, DC,
Simon RI: Assessing and Managing Suicide Risk: Guidelines for Clinically
Based Risk Management. Washington, DC, American Psychiatric Publish-
ing, 2004
Simon RI: Standard of care testimony: best practices or reasonable care. J Am
Acad Psychiatry and Law 33:8–11, 2005
Simon RI: The myth of “imminent” violence in psychiatry and the law. Univ
Cincinnati Law Rev 75:631–644, 2006
Simon RI: Gun safety management with patients at risk for suicide. Suicide Life
Threat Behav 37:518–526, 2007
Simon RI, Shuman DW: Clinical Manual of Psychiatry and Law. Washington,
DC, American Psychiatric Publishing, 2007
Slovenko R: Psychiatry in Law/Law in Psychiatry. New York, Brunner-
Routledge, 2002
551 P.2d 334 (1976)
Index
Page numbers printed in boldface type refer to tables or figures.
AA (Alcoholics Anonymous), 155 Acute confusional state, 197

AASI (Abel Assessment of Sexual Acute dystonic reaction to lorazepam
Interest), 71–72 and haloperidol, 302, 305, 309
Abandonment, 330 Addiction, 149–155. See also Substance
fear of, among elderly persons, 382, abuse disorders
385 definition of, 149
purposeful, targeted, defensive as disease, 155
violence in reaction to, 171 identification of, 151–152
targeted, impulsive violence in media dramatization of, 151
reaction to, 172 modafinil for patients with, 150
violence related to perception or neurobiology of, 149
fear of, 177–179 treatment of, 152–155
Abel Assessment of Sexual Interest in elderly persons, 399
(AASI), 71–72 evidence-based strategies for
Acamprosate, for alcoholism, 153 relapse prevention, 152
Access of patient patient–treatment matching for,
to potential victim, 7 155
to weapons, 6, 372, 373, 557 pharmacological, 152–154
Acetaldehyde, 153 psychotherapy, 154–155
Acetaldehyde dehydrogenase, 153, 154 12-Step facilitation therapy, 155
Acetylcholine, in Alzheimer’s disease, 12-Step programs, 152, 155
399 Addiction (television series), 151
Acquaintance stalkers, 412, 414 ADHD (attention-deficit/
Acquired immunodeficiency hyperactivity disorder), 150, 374
syndrome (AIDS), 196, 489 conduct disorder and, 369
Active listening, 468 Tourette’s syndrome and, 195
Actuarial violence risk assessment, 3– Adolescents, 359–376
4, 17–31, 46, 59. See also Structured adaptive functions of aggression in,
violence risk assessment 211
compared with clinical risk antidepressant effects on suicide
assessment, 3–4, 17–21, 28 risk in, 90
on continuum of structure, 17–18 attention-deficit/hyperactivity
implications for clinical disorder in, 369
intervention, 19 bullying and, 361, 540–541, 545
in outpatient settings, 237–238 conduct disorder in, 360, 363, 369
psychological testing and, 59–60 controlling access to weapons, 372,
reliance on static risk factors, 19 373
role of clinical judgment in, 18 dating violence among, 361–362
of sexual offender evaluation, 70 depression in
time frame for validity of, 18–19 abuse or neglect and, 79–80
567
Adolescents (continued) school violence and, 79, 359, 365,

depression in (continued) 371, 537–552
bullying and, 79 self-injurious behaviors among,
deliberate self-harm and, 87 225–226
maternal depression and, 78 substance abuse among, 364, 369–
passive exposure to domestic 370
violence and, 78 Surgeon General’s report on youth
as predictor of violence, 83 violence, 37
developmental trajectories toward violence among, 11, 22, 23
violence in, 362–364, 363 violence risk assessment in, 365–372
differences between violence in consent and confidentiality for,
adults and, 360, 361 367–368
epidemiology of violence among, dimensions of, 367
360–362 formulating a risk assessment,
gangs and, 359, 362, 365, 375 372
gender and violence among, 359, history of violence, 368–369
360–361 for predatory violence, 370–372
homicide rates in, 359, 362 psychiatric diagnosis, 369–370
inhalant use by, 9 risk factors and risk assessment
intermittent explosive disorder in, scales, 364–365, 366, 370
216 setting for, 366–367
juvenile justice system for, 359, 367– weapon assessment, 372
368 Adoption studies, 364
management of violence in, 372–376 β-Adrenergic blockers, 200, 203, 316
acute management of high-risk Affect management, 463, 465, 467–468
youth, 373 Affective recognition, 330
consultation, 375–376 Affective regulation disorders, in
environmental interventions, 375 elderly persons, 392–394
outpatient psychosocial African Americans, 35
treatment, 373–374 alcohol-related intimate partner
pharmacotherapy, 374–375 violence among, 145
waiving juveniles to adult court, arrests for aggravated assaults, 37
375 culturally appropriate assessment
mental health interventions for, in psychiatric setting, 46
359–360 culturally appropriate assessment
myths about youth violence, 37 of intimate partner violence in,
passive exposure to parental 52–53
violence, 78 myths about youth violence among,
peer effects and violence in, 363, 37
364, 365, 375 overdetermination of
prevalence of violence among, 359, dangerousness, 38
360–362 risk for violence or homicide, 37
Psychopathy Checklist–Youth seclusion and restraint of, 38
Version for, 62 serial killing by, 38
rehabilitation of, 367 women
risk factors for violence in, 364–365, spiritual/religious support for,
366, 370 81
Index ❘ 569
as victims of intimate partner in dementia, 186–187

violence, 80, 81 emergency psychiatric services for,
Aftercare, 13, 14 279–280, 284–285
Age. See also Adolescents; Children; inpatient assaults and, 260
Elderly persons medication-induced, 198
at first violent episode, 35 pharmacotherapy for, 84–85, 253–
suicide risk and, 89 254, 301–318
violence risk and, 11, 22, 23 after traumatic brain injury, 187, 188
Aggression AIDS (acquired immunodeficiency
adaptive functions of, 211 syndrome), 196, 489
anger and, 212 Akathisia, 10, 198, 200, 301
antidepressant effects on, 84 lithium for, 315
biology of Alaska Natives
in mood disorders, 92–93 arrests for aggravated assaults, 37
serotonin, 92–93, 215 risk for violence or homicide, 37
bipolar disorder and, 83, 84 substance abuse and suicidality
categories of, 212 among, 144
continuum of severity of, 211–212 Alcohol Use Disorders Identification
definition of, 68, 212 Test (AUDIT), 151
dementia and, 185–187 Alcohol use/abuse, 7–9, 13, 27, 141–146
depression and, 83, 84 child abuse and, 144–145
drug-induced, 9 date rape and, 143
hostility and, 212 disinhibition induced by, 9
impulsivity and, 68, 211–228 (See early-onset, 148
also Impulsive aggression) elder abuse and, 145
instruments for assessment of, 68 fetal exposure to, 213
intent and, 212 gangs and, 146
neurochemistry of, 215–216 genetic vulnerability to, 152
pharmacotherapy for, 84–85, 253– head injury and, 149
254, 301–318 intimate partner violence and, 82–
premeditated, 212–213, 262, 312 83, 145–146
psychotherapy for, 325–337 intoxication due to, 6, 9
schizophrenia and, 10 homicide and, 142
serotonin transporter vehicular crashes and, 522, 534
polymorphisms and, 93 neurobiology of alcohol, 143
as stress management strategy in prevalence of, 141, 151
animal models, 211 psychiatric comorbidity with, 148–
theoretical models of, 212–213 149
Aggression Questionnaire, 68, 212 bipolar disorder, 85
Anger subscale of, 67 intermittent explosive disorder,
Hostility subscale of, 68 85, 217
Aggression replacement strategies, 201 posttraumatic stress disorder,
Agitation, 5 133
brain tumors and, 194 pyromania, 219
cocaine-induced, 9, 198 rape and, 143
de-escalating emotion in patients screening for, 151
exhibiting, 265 suicide and, 89, 146
Alcohol use/abuse (continued) American Academy of Psychiatry and

suicide and (continued) the Law, 409
spousal/consortial murder- American Board of Forensic
suicide, 422 Psychology, 28, 29
treatment for dependence on, 152– American Indians
155 arrests for aggravated assaults, 37
for at-risk heavy drinkers, 151– risk for violence or homicide, 37
152 substance abuse and suicidality
in elderly persons, 399 among, 144
guidelines for primary care American Medical Association’s Older
physicians, 151, 152 Drivers Project, 527–528
patient–treatment matching for, American Psychiatric Association
155 position on driving safety, 521–522,
pharmacological, 152–154 527
acamprosate, 153 practice guideline for posttraumatic
disulfiram, 153–154 stress disorder, 124
naltrexone, 152–153, 399 recognition of structural racism, 36
topiramate, 154 task force report on restraint and
psychotherapy, 154–155 seclusion, 341–342, 342, 343, 348
12-Step facilitation therapy, 155 Americans with Disabilities Act, 508
12-Step programs, 152, 155 γ-Aminobutyric acid (GABA), 215
violence and, 107, 141–146, 250 Amok, 44, 45
case examples of, 141–142 Amphetamine, 9, 198
factors associated with, 142 Amputation of extremities, 225
mechanisms of link between, Anabolic steroids, 9, 198
143 Androgens, 92, 448
withdrawal from, 133 Anger, 5, 212
benzodiazepines for, 133, 305 depression and attacks of, 252, 254
delirium during, 9 distinction from hostility, 68
pharmacotherapy for violence instruments for assessment of, 67–
during, 302 68, 72
Alcohol–disulfiram reaction, 153–154 in posttraumatic stress disorder,
Alcoholics Anonymous (AA), 155 124, 132, 134
Alkyltin exposure, 197 in schizophrenia, 10
Alprazolam, before care activities for Anger management programs, 25, 216
elderly dementia patient, 398 for youth at risk for school violence,
Alzheimer’s disease, 186–187, 382, 387, 550
391–392, 393–394, 397–398. See also Anger-excitation rapists, 416
Dementia Anger-retaliation rapists, 416
acetylcholine deficit in, 399 Anti-androgen treatment, 335, 453–455
cholinesterase inhibitors for, 392, monitoring for adverse effects of,
393, 399, 402 454–455
driving safety and, 527–531 Anticholinergic delirium, 198
American Academy of Child and Anticonvulsants, 84–85, 253, 303, 314–
Adolescent Psychiatry, 369 315, 334. See also specific drugs
American Academy of Neurology, 527, for aggressive elderly patients, 403
529 for chronic aggression, 200, 203
Index ❘ 571
combination treatment with use on 72-hour extended

antipsychotics, 314–315 observation unit, 287
driving safety and, 522 for violent psychotic inpatients, 114,
Antidepressants, 384 253–254, 267
for aggressive elderly patients, 403 for violent youth, 374–375
driving safety and, 523, 524–525, 533 Antisocial behaviors
effect on aggression, 84 in combat veterans with
effect on suicide risk, 90, 92 posttraumatic stress disorder,
for psychotic depression, 285 125–126
use on 72-hour extended in conduct disorder, 369
observation unit, 287 intellectual disability and, 220
Antipsychotics, 84, 85, 199–200, 302– intimate partner violence and, 491
314. See also specific drugs prenatal risk factors for, 213–214
for aggressive/agitated elderly self-injurious behaviors and, 226
persons, 401–403 Antisocial personality disorder, 11, 13,
acute/emergent treatment, 394– 23, 107, 181
395 conduct disorder and, 363, 369
administration routes for, 394, driving and, 522
401 purposeful, instrumental violence
case example of, 402–403 and, 168–169
choice of drug for, 401 purposeful, noninstrumental
controversy about, 403 violence and, 170
dosage of, 401–402 schizophrenia, violence and, 250–
metabolic syndrome due to, 403 251
monitoring response to, 402 substance abuse and, 148
risk of death in elderly dementia workplace violence and, 511
patients, 200–201, 395, 403 Anxiety/anxiety disorders
side effects of, 401, 402 in assault victims, 80
anti-hostility effects of, 313 intermittent explosive disorder and,
combination therapy with mood 217
stabilizers, 314–315 rage and, 334
driving safety and, 522, 533 self-injurious behaviors and, 226
electroconvulsive therapy and, 316 Anxiolytics, 198
long-term treatment with, 202, 312– Apathy, in dementia, 393, 394
314 Appearance of patient, 5–6
in psychiatric emergencies, 280, 294, Aripiprazole
295, 302–305 for aggressive elderly patients, 401
intramuscular administration of, intramuscular, 304, 305, 307, 310–311
303–311, 304, 306–307 adverse effects of, 310
suicidal patient, 295 combined with lorazepam, 310
for psychotic depression, 285 dosage of, 304, 310
racial differences in administration indications for, 310
of, 38 transition to oral administration,
second-generation, 312 311
side effects of, 401, 402 long-term treatment with, 312–314
agitation, 198 Armed patient, 476
akathisia, 301–302 Arrest reports, 5
Arsenic exposure, 197 staffing ratios, ward atmosphere

Arsonists, 218–219. See also Pyromania and, 261
ASD. See Autism spectrum disorder threatening behaviors before, 112
Asian Americans, 35 videotape recording of, 111–112,
arrests for aggravated assaults, 37 262
culturally appropriate assessment paranoia and, 107
of interpersonal violence in, psychiatric disorders in victims of,
49–50 80
barriers to, 50, 51 psychotic, 110–111, 262–263, 264,
risk for violence or homicide, 37 266–267
Assaults indicators of impending assault,
by brain-injured patients, 187 267
classification of, 109–111, 113, 264 strategies for prevention of, 267
on clinicians, 248–249, 254, 259, by psychotic patients in the
461–478 (See also Violence community, 116
toward clinicians) in schools, 539
by elderly person on day care staff, sexual (See Sexual assault/rape;
391–392 Sexual violence)
ethnicity and arrests for, 37 against stalking victims, 411
by inpatients, 109–118, 259–272 substance abuse among victims of,
on clinicians, 470 80, 82
criminal prosecution for, 269– suspiciousness and, 107–108
270, 271 of “warn” category, 112
elderly persons, 383, 386 in workplace, 506
environmental risk factors for, Assertive Community Treatment, 284
114–115, 260–261 Assessment instruments
gender and, 260 Aggression Questionnaire, 67, 68,
impulsive vs. premeditated, 262 212
instrumental vs. reactive, 109– Alcohol Use Disorders
110, 262–263 Identification Test, 151
long-term risk factors for, 261 Balanced Inventory of Desirable
management of, 113–118 Responding, 69
measures of control for, 264, 266 Barratt Impulsivity Scale, 68–69, 72,
mental states associated with, 214, 224
109 Brief Psychiatric Rating Scale, 315
motivations for, 109, 113, 261– Buss-Durkee Hostility Inventory,
270 215, 224
on nursing staff, 259, 263–265 Child Behavioral Checklist, 214
patient characteristics associated Classification of Violence Risk, 18,
with, 109, 259–260 23–26, 28, 30
prediction of, 113, 114 Clinical Dementia Rating scale, 527
prevention of, 114 Cook-Medley Hostility Inventory, 68
reactive vs. instrumental, 109– HCR-20, 21–23, 28, 30, 39, 46, 240–
110, 262–263, 264 241
research on classification of, 110– InterSept Scale for Suicidal
111 Thinking, 295
short-term risk factors for, 260 Mental Health Triage Scale, 283
Index ❘ 573
Millon Clinical Multiaxial AUDIT (Alcohol Use Disorders

Inventory–3, 63, 64, 65, 163 Identification Test), 151
Mini-Mental State Examination, Autism spectrum disorder (ASD), 221–
285, 390, 392, 402 222, 227
Minnesota Multiphasic Personality case example of, 221
Inventory–2, 28, 63–65, 67, 72, in DSM, 221
163, 416 prevalence of, 221
National Triage Scale, 283 self-injurious behaviors and, 225
Novaco Anger Scale, 67, 72 types of aggression associated with,
Overt Aggression Scale, 7, 8, 188, 222
199, 215 Autoimmune deficiency encephalitis,
Paulhus Deception Scales, 69–70 193
Personality Assessment Inventory, Available means of inflicting harm, 6–7
63, 64, 65, 163 Avoidant personality disorder
Positive and Negative Syndrome nontargeted, impulsive violence
Scale, 313, 314 incidental to emotional escape
Psychopathy Checklist, 62–63 and, 176
Psychopathy Checklist–Revised, 27, purposeful, targeted, defensive
62–63, 213 violence and, 171
Psychopathy Checklist–Youth targeted, impulsive violence and,
Version, 62, 369, 370 172, 174
Rapid Risk Assessment for Sexual
Offender Recidivism, 29, 437 Balanced Inventory of Desirable
Rorschach Inkblot test, 29, 66–67, 72 Responding, 69
Standard Family Violence Index, Baltimore Epidemiologic Catchment
127 Area Follow-Up study, 216
State/Trait Anger Expression Barbiturates, 198
Inventory, 67 Barefoot v. Estelle, 435–436, 437
Structured Assessment of Violence Barratt Impulsivity Scale, 68–69, 72,
Risk in Youth, 370 214, 224
Structured Guide for the Bathing, of patient with dementia, 398
Assessment of Violence, 285 Battered women’s syndrome, as legal
Violence Risk Appraisal Guide, 18, defense, 494. See also Intimate
26–27, 28, 30 partner violence
for youth, 369, 370 Beck Depression Inventory, 88
Athletes, anabolic steroid use by, 9 Behavioral disinhibition, 68
Ativan. See Lorazepam alcohol-induced, 9
Attentional problems, assessing benzodiazepine-induced, 317
elderly persons for, 389–390 neurocorrelates in dementia, 187
Attention-deficit/hyperactivity Behavioral disturbances
disorder (ADHD), 150, 374 brain tumors and, 194
conduct disorder and, 369 in brain-injured patients, 187
Tourette’s syndrome and, 195 of children
Attitudes parental substance abuse and,
dysfunctional, deliberate self-harm 78–79
and, 88 passive exposure to parental
racially based, of clinicians, 36 violence and, 78
Behavioral disturbances (continued) sex offenses and, 84

in dementia, 185 substance abuse and, 84, 85, 251
in systemic lupus erythematosus, suicide in, 90–91
198 lithium to reduce risk for, 91, 92
in Tourette’s syndrome, 196 after traumatic brain injury, 187
Behavioral Sciences Unit of the FBI violence prevalence in, 107, 250
Academy, 417 BLS (Bureau of Labor Statistics), 505–
Behavioral treatments for aggression, 506, 514
201, 204 Blueprints for Violence Prevention, 374
Belligerence, 9, 106. See also Borderline personality disorder, 11, 13,
Aggression; Anger 26, 166, 252
Benzodiazepines, 331 childhood abuse and, 214
abuse of, 82, 268 driving safety and, 522
for aggression, 199–200, 203, 316, 317 nontargeted, impulsive violence
for agitated/aggressive elderly incidental to emotional escape
patients, 401, 403 and, 175–176
acute/emergent management, patient dependency on therapist in,
394 331
before care activities for purposeful, instrumental violence
dementia patients, 398 and, 168–169
for alcohol withdrawal, 133, 305 self-injurious behaviors in, 226
for escalating anxiety, 334 targeted, impulsive violence and,
long-term treatment with, 316 172
in psychiatric emergencies, 302, 303 violence related to perceived/
as quenching agents, 334–335 feared loss or abandonment
during seclusion and restraint, 352 and, 178
side effects of, 198 “Boston Miracle” in assault reduction,
tolerance to, 305, 316 375
withdrawal from, 305, 316 Boufée delirante, 44, 45
Benzoylecgonine, 147 Bradford Sexual History Inventory, 445
Benztropine, 388 Bradycardia, olanzapine-induced, 308
for acute dystonic reaction, 302 Brain
Binswanger’s disease, 196 correlates of aggression in
Bipolar disorder, 25, 301 dementia, 187
in adolescents, 374 effects of prenatal substance
deliberate self-harm and, 88–89 exposure on development of,
driving safety and, 522, 532 214
ethnicity and diagnosis of, 35 encephalitis, 192–194
in genesis of violence, 83–86 minimal brain dysfunction, 196
in elderly dementia patient, 400 neurobiology of addiction, 149–150
legal problems associated with, 84 neurobiology of aggression
multiple sclerosis and, 196 in mood disorders, 92–93
pharmacotherapy for aggression/ role of serotonin, 92–93, 215
agitation in, 302–303 neurobiology of alcohol, 143
prevalence of, 90 regions of dysfunction associated
pyromania and, 219 with aggression and
self-injurious behaviors and, 225 impulsivity, 214–215
Index ❘ 575
sexual sadism and abnormalities of, Castration, 225

447 pharmacological, 453–455
traumatic injury of, 187–188, 196, Catalepsy, driving safety and, 531
204, 214–215 Catastrophic reactions, poststroke, 189
tumors of, 194, 196 Cathard, 45
white matter disorders, 196 CATIE (Clinical Antipsychotic Trials of
Brief Psychiatric Rating Scale, 315 Intervention Effectiveness), 107–
Bullying, 79, 110, 115, 173, 359, 361, 365 108
control of, 550 CATIE-AD (Clinical Antipsychotic
in schools, 540–541, 545 Trials of Intervention
Buprenorphine maintenance Effectiveness–Alzheimer’s
treatment, 154 Disease), 394–395
Bupropion, for smoking cessation, 399 Caucasians
Bureau of Justice Standards, 506 arrests for aggravated assaults, 37
Bureau of Labor Statistics (BLS), 505– risk for homicide, 37
506, 514 serial killing by, 38
Buspirone, for aggressive elderly CDR (Clinical Dementia Rating) scale,
patients, 403 527
Buss-Durkee Hostility Inventory, 215, Centers for Medicare and Medicaid
224 Services (CMS), requirements for
use of seclusion and restraint, 341,
Cafard, 45 344–345, 351
Caffeine and driving safety, 522 Central nervous system infections,
Calcium, during anti-androgen 192–194
therapy, 455 Central nervous system tumors, 194
Cambodians, 35 Cerebral glucose metabolism, 214
culturally appropriate assessment Cerebrovascular accident, 188–189,
of interpersonal violence in, 196, 389
51–52 antipsychotics and, 403
Cannabis, 86, 148, 150, 335 Chart review, 5
bipolar disorder and, 85 Chemical restraint, 351
prevalence of use, 107 Child Behavioral Checklist, 214
Capital punishment cases, 430, 435– Child custody cases, 168, 175
436, 438 Child pornography on Internet, 444–445
Carbamazepine, 203, 315 Childhood abuse or neglect, 23
for aggressive elderly patients, 403 alcohol-related, 144–145
Carbon monoxide exposure, 197 deliberate self-harm and, 87, 92, 226
Cardiovascular effects of impulsive aggression and, 214
antipsychotics mood disorders related to, 79–80
olanzapine, 308 parental substance abuse and, 78
ziprasidone, 308 physical abuse, 79
Caregivers prevention of, 82
assuring safety of, 395 sexual abuse, 79–80
behavioral strategies for, 201 verbal and emotional abuse, 79
of dementia patients, 185–186 posttraumatic stress disorder in
elderly persons lashing out at, 382 veterans and, 126
interviewing of, 384 racial reporting bias for, 38
Childhood abuse or neglect (continued) expert testimony at hearing for,

self-injurious behaviors and, 226 438–439
sexual abuse, 87, 214, 545 for imminent risk of violence, 244–
among adult perpetrators of 245
sexual violence, 442 involuntary outpatient
convictions for, 443 commitment, 547
deliberate self-harm and, 87 predicting future violence and, 438–
incest perpetrators, 450, 451–452 439
Internet as tool for, 443–445 Civil Protection Orders (CPOs), in
mood disorders related to, 79–80 cases of intimate partner violence,
prevalence of, 442–443 493–494
programs for prevention of, 443 Classification of Violence Risk (COVR),
and risk for sexual victimization 18, 23–26, 28, 30
as adult, 79, 82–83 case examples of use of, 25–26
suicide and, 80, 214, 223 classification tree methodology of,
treatment of adults with history of, 23–24
82 clinician review of, 24
Children, 359–376. See also Adolescents combining risk factors for, 23–24
aggression related to prenatal generating final risk estimate with, 24
substance exposure of, 213–214 predictive validity of, 23
bullying and, 79, 540–541 selecting and measuring risk factors
conduct disorder in, 363 for, 23
developmental trajectories toward Clinical Antipsychotic Trials of
violence in, 362–363, 363 Intervention Effectiveness
epidemiology of aggression in, 360 (CATIE), 107–108
gender and aggression in, 360 Clinical Antipsychotic Trials of
inhalant use by, 9 Intervention Effectiveness–
maladjustment of, 22, 27 Alzheimer’s Disease (CATIE-AD),
oppositional defiant disorder in, 394–395
363, 369 Clinical Dementia Rating (CDR) scale,
parental murder of, 448 527
murder-suicide, 423–424 Clinical violence risk assessment, 3–14,
passive exposure to parental 19–20, 46
violence, 78, 80 case examples of, 12–13
of substance-abusing parents, 78–79 clinicians responsible for, 3
temper tantrums in, 216 collateral sources of information for, 4
trauma exposure and later compared with actuarial or
aggression in, 214 structured risk assessment, 3–
as victims of sexually motivated 4, 17–21, 28, 30
homicide, 448 documentation of, 14
Cho, Seung-Hui, 546–547, 550 factors in, 5, 5–12, 14
Cholinesterase inhibitors, for alcohol and drug use, 7–9
Alzheimer’s disease, 392, 393, 399, appearance of patient, 5–6
402 available means, 6–7
Christians, 35 demographic characteristics, 11–
Citalopram, 316 12
Civil commitment, 430 intent, 6
Index ❘ 577
past history of violence or pathological attachment to, 332

impulsive behaviors, 7, 8 patient dependency on, 331
personality disorders, 11 physical distance and posture with
presence of violent ideation and violent patient, 468
degree of planning/ racial and cultural differences
formulation, 6 between patients and, 39
psychosis, 10 racially biased attitudes of, 36
treatment noncompliance, 11 responsibility for violence risk
implications for clinical assessment, 3
intervention, 19 risk management of potentially
in outpatient settings, 237–239 violent patients, 555–564
preventive interventions based on, safety of, 248–249, 254, 332
3, 14 threats against, 471–475
principles of, 4–5 treatment guidelines for patients
professional subjective judgment who drink too much, 151, 152
for, 19 violence toward, 461–478
questions for, 4 additional management
reassessments, 14 techniques for, 468–469
reliance on dynamic risk factors, 19 affect management for, 463, 465,
research on validity of, 19–20 467–468
time frame for validity of, 4, 18 anticipation of, 466
for violence in the short term, 3, 4 assessing risk for, 465–466, 466
Clinicians dynamics of, 467
active listening by, 468 in emergency settings, 462–463,
anxiety about therapy with suicidal 470
patients, 327–328 on inpatient unit, 470
assessment of patients’ driving limit setting and, 469–470
competence by, 527–530, 534 in outpatient settings, 248–249,
comfort in dealing with potential 332, 463–464
violence, 249 prevalence of, 461
on crisis management teams in prodrome of, 467
workplace, 514–515 psychiatrist’s denial of risk for,
duty to warn and protect potential 464, 465, 466
victims, 246, 247, 269, 333–334, as result of psychotic
373, 430–434, 524, 557–559, 560 transference in outpatient
expert testimony of, 430, 434–439 setting, 464–465
eye contact with violent patient, 468 threats toward a surgeon, 464
honesty of, 468 Clonazepam, 203, 316
inexperience with violent patients, Clozapine, 114, 253–254, 267, 303, 312–
326–327 313
irrational fear of violent patients, anti-hostility effects of, 313
332–333 for intermittent explosive disorder,
legal duties to prevent vehicular 317
crashes, 531–534 CMS (Centers for Medicare and
legal responsibility for a patient’s Medicaid Services), requirements
violent act, 247 for use of seclusion and restraint,
limit setting by, 469–470 341, 344–345, 351
Cobalamin deficiency, 196 Comprehensive Psychiatric Emergency

Cocaine, 9, 13, 39, 147, 150, 268, 525 Programs (CPEPs) (New York), 278
agitation induced by, 9, 198 Computed tomography, of aggression
crack, 9, 362, 387 in dementia, 187
disulfiram for addiction to, 154 Conduct disorder
fetal exposure to, 213, 214 antisocial personality disorder and,
topiramate for addiction to, 154 363, 369
Cognitive distortions of sexually attention-deficit/hyperactivity
violent persons, 453 disorder and, 369
Cognitive impairment oppositional defiant disorder and,
in elderly persons, 381–382 363, 369
assessment of, 390–392 schizophrenia, violence and, 250–
in persons with Intellectual 251
disability, 189–190, 219–221, 227 serious violence and, 108
Cognitive-behavioral therapy serotonin transporter
for aggression, 317 polymorphisms and, 93
for alcoholism, 155 workplace violence and, 511
for depression and posttraumatic in youth, 360, 363, 369
stress disorder resulting from Confidentiality, 429
violence exposure, 82, 132–133, breach of, 241–242
135 to protect potential victims, 246,
for sexually violent persons, 452 247, 429–434, 493 (See also
for students at risk for school Duty to warn and protect
violence, 550 potential victims)
with violent youth, 374 to seek additional sources of
“Cold” threats, 268–269 information, 241
Collateral sources of information about of interview with adolescent, 367–
patient, 4, 13 368
College students vs. safety concerns, 241
date rape among, 143 Confusion
substance abuse and suicidality amphetamine-induced, 9
among, 143 in delirium, 197
Columbine High School shootings, in elderly persons, 386
543–544 in systemic lupus erythematosus, 198
Command hallucinations, 108–109, after traumatic brain injury, 187
267, 267–268, 295 Congenital brain disorders, 189–190.
Community Mental Health Act, 277 See also Intellectual disability
Community psychiatric emergency Consultation for violence risk
services, 289–291 management, 556
crisis residence, 289 in youth, 375–376
crisis respite, 290 Cook-Medley Hostility Inventory, 68
crisis response services, 289–290 Correctional facilities, structured risk
mobile crisis intervention team, assessment for violence after
290–291 discharge from, 4
Community violence and mood Corticosteroids, 198
disorders, 81 Cortisol secretion, childhood abuse,
Competency to stand trial, 410 and trauma exposure, 78
Index ❘ 579
Cortisone, 198 Crisis respite, 290

Countertransference, 332–333 Crisis response services, 289–290
monitoring of, 472, 473 for victims of intimate partner
racism and, 40 violence, 492
in treatment of children and Crisis stabilization unit, 286–287
adolescents, 375 Culturally competent violence risk
Couples therapy, 557 assessment, 35–54. See also
for intimate partner violence, 492 Ethnicity/race
for pedophilia, 445 cross-cultural validity of
for posttraumatic stress disorder, assessment instruments,
134, 137 Psychopathy Checklist, 63
Court-mandated treatment culturally appropriate assessment,
for batterers, 492 42–43
psychotherapy, 327 culture-bound syndromes, 44, 45
Covert sensitization, for paraphilias, 453 DSM-IV-TR Outline for Cultural
COVR. See Classification of Violence Formulation, 43–44, 44
Risk in psychiatric setting, 44–46
CPEPs (Comprehensive Psychiatric denial of racism by majority
Emergency Programs) (New culture, 46
York), 278 interpersonal violence, 41–42
CPOs (Civil Protection Orders), in culturally appropriate
cases of intimate partner violence, assessment of, 47–53
493–494 problems in assessing violence risk
Crack cocaine, 9, 362, 387. See also in people of color, 37–41
Cocaine case examples of, 39–41
Criminal behavior, 5, 7 groups at greatest risk for
alcohol consumption and, 142 violence or homicide, 37–38
among combat veterans with post- overdetermination of
traumatic stress disorder, 124 dangerousness, 38–39
drop in violent crime rates, 362 youth violence, 37
among elderly persons, 381 Culture
mania and, 84 definition of, 42
motivations for, 109 effect on symptom expression, 42
prenatal tobacco exposure and, 214 protective effect on mental health,
prosecution for inpatient assaults, 43
269–270, 271 Culture of violence, intimate partner
by psychotic men, 109 violence and, 486
in schools, 538–539 Culture-bound syndromes, 44, 45
Criminal defendants, 430 Cushing’s syndrome, 389
forensic evaluations of, 409–426 Cyproterone acetate, for paraphilias,
preventive confinement of, 430 453, 454
state requirements for violence risk
assessment of, 438 Dangerousness
Crisis management teams for civil commitment and, 438–439
workplace violence risk dependence of judicial system on
assessment, 514 assessment of, 430, 437–438
Crisis residence, 289 minimization of, 4
Dangerousness (continued) Delusional disorder, 10

overdetermination in people of Delusions
color, 38–39 during alcohol withdrawal, 9
of rapist, 416 of being stalked, 412
sentencing and, 438 crimes motivated by, 109
of stalker, 413 of elderly patients, 382, 388
Date rape, 143 illicit drug–induced, 9
Dating violence, 361–362 in mania, 10
Daubert v. Merrell Down Pharmaceuti- murder-suicide and, 423
cals, Inc., 436–437, 439, 494 paranoid, 10, 12, 64, 118, 265–266,
DBH (dopamine-β-hydroxylase), 154 267, 284
de Rais, Giles, 448 persecutory, 10, 108, 266, 267
DEA (Drug Enforcement Agency), 133 in schizophrenia, 10, 12, 265–266
Death penalty, 430, 435–436, 438 violence associated with, 107–108,
“Decade of the Brain,” 151 109, 251–252
Deceased persons, forensic evaluations in dementia, 187
of, 418–425 emotional charge associated
murder-suicide, 421–425, 422, 425 with, 251–252
suicide by cop, 420–421, 421 among inpatients, 266–267, 267
Decelerative techniques, 201 substance abuse and, 149
De-escalation techniques, 265, 266 Dementia, 185–187, 381–382
affect management, 463, 465, 467–468 aggression in, 185–187
eye contact, 468 acute/emergent treatment of,
Defensiveness, assessment of, 69–70 394–395
Deinstitutionalization, 221 case example of, 391–392
Deliberate self-harm (DSH), 86–89, 92. cholinesterase inhibitors for, 392,
See also Self-injurious behaviors; 393, 399, 402
Suicide depression and, 392–394
among adolescents, 87 environmental stimuli and, 397–
childhood abuse and, 87, 92 398
definition of, 86 managing precipitants of, 396–
depression and, 86–88 398
dysfunctional attitudes and, 88 mutifaceted approach to, 404
among elderly persons, 87–88 neurological correlates of, 187
impulsive acts of, 87 nursing home placement due to,
by poisoning, 87 186
risk factors for, 86 pharmacotherapy for, 400–404
suicide and, 86, 92 physical discomfort and, 397
vs. suicide attempt, 222 predictors of, 187
Delirium, 197, 197, 388, 395–396 sexual aggression, 382
during alcohol withdrawal, 9 specific care needs and, 398
anticholinergic, 198 treating underlying psychiatric
assessment of, 390 disorders in, 398–400
early signs of, 395–396 apathy in, 393, 394
in elderly persons, 386, 395–396 atypical antipsychotics and risk of
phencyclidine-induced, 9 death in elderly patients with,
in systemic lupus erythematosus, 198 200–201
Index ❘ 581
cognitive and memory impairment serotonin level and, 93, 215

in, 381–382, 390–392 suicide and, 89–90, 223, 224
depression and, 393–394 treatment approaches for violence
driving safety and, 522, 527–531 in, 81–83
identification of, 388 violence in genesis of, 77–83
with Lewy bodies, 186, 198 community violence, war, and
prevalence of, 185 terrorism, 81
rapid-onset, 387 direct abuse or neglect, 79–81
in rheumatological diseases, 198 adult assault, 80
Demographic characteristics intimate partner violence, 41,
of suicidal patients, 89 80–81
of violent patients, 11–12 physical abuse, 79
Denial of risk by clinician, 464, 465, 466 sexual abuse, 79–80
Dependent personality disorder verbal and emotional abuse,
nontargeted, impulsive violence 79
incidental to emotional escape passive exposure, 77–79
and, 176 parental substance abuse, 78–
purposeful, targeted, defensive 79
violence and, 171 witness to domestic violence,
targeted, impulsive violence and, 77–78
172, 174 violence prevalence in, 106
Depression, 25, 284 Derogatis Sexual Functioning
in adolescents, 374 Inventory, 445
maternal depression and, 78 Desensitization, for paraphilias, 453
anger attacks in, 252, 254 Developmental disorders, 189–190,
in assault victims, 80 219–221. See also Intellectual
biology of aggression in, 92–93 disability
among combat veterans, 81 Diabetes mellitus, 106
deliberate self-harm and, 86–88 Diagnostic and Statistical Manual of
driving safety and, 523, 524–526, Mental Disorders (DSM-IV-TR)
533 autistic disorder in, 221
in elderly persons, 392–394 culture-bound syndromes in, 44, 45
in genesis of violence, 83–86 impulse-control disorders in, 218
in elderly persons, 384–385, 399 intermittent explosive disorder in,
as goal of treating violent patients, 216
328 mental retardation in, 219
intimate partner violence and, 41, Outline for Cultural Formulation,
80–81, 127, 485, 488 43–44, 44, 46, 54
mixed, 84 pyromania in, 218
murder-suicide and, 422, 423, 424 Diagnostic Interview Schedule, 107
postpsychotic, 329 Dialectical behavior therapy, 317
poststroke, 189 Dietary supplementation, during anti-
posttraumatic stress disorder and, androgen therapy, 455
128, 129 Diphenhydramine
psychotic, 285 for aggressive elderly patients, 403
public fear of violence due to, 148 for dystonic reaction, 302
self-injurious behaviors in, 226 Disability discrimination, 504, 508
Discharge from inpatient facilities, 13, Drug abuse. See Alcohol use/abuse;
14, 559–560, 561 Substance abuse disorders
Classification of Violence Risk for, Drug Enforcement Agency (DEA), 133
25–26 Drug Screening Test, 445
Discrimination, racial, 36 DSH. See Deliberate self-harm
Disorganized and organized crime DSM-IV-TR. See Diagnostic and
scenes, FBI profiling of, 417–418, Statistical Manual of Mental
418 Disorders
Disorientation, in elderly persons, 381 Dusky v. U.S., 410
Dissociation, in culture-bound Duty to warn and protect potential
syndromes, 45 victims, 246, 247, 269, 333–334,
Disturbances in consciousness, driving 373, 430–434, 524, 557–559
safety and, 531 best practices for, 434, 558
Disulfiram inpatient discharge and, 559–560
for alcoholism, 153–154 from intimate partner violence,
for cocaine dependence, 154 492–494
Divalproex. See Valproate issuing a warning, 558–559
Documentation, 562 language barriers and use of
Domestic violence, 41–42. See also translator, 558–559
Intimate partner violence risk management steps for, 559, 560
childhood mood disorders state allowances for therapist
associated with passive discretion in, 433–434
exposure to, 77–78 vs. therapist–patient confidentiality,
escalating pattern of, 7 246, 247, 429–430
parental substance abuse and, 78 Dynamics of violence, 467
screening for, 41 Dysarthria, 6
Donepezil, for Alzheimer’s disease,
392, 393, 399, 402 Eating disorders, self-injurious
Dopamine, 215 behaviors and, 224, 226
Dopamine-β-hydroxylase (DBH), 154 Ecological theory of intimate partner
Driving safety, 7, 521–535 violence, 486
dementia and, 527–531 ECT (electroconvulsive therapy), 316
depression and, 524–526 Elderly persons, 381–404
disturbances of consciousness and, alcohol-related abuse of, 145
531 assessment in, 383–394
intoxication and, 522 chief complaint, 383–386
legal duties of clinicians to prevent in inpatient setting, 383
vehicular crashes, 531–534 mental status examination, 389–
Physician’s Guide to Assessing and 394
Counseling Older Drivers, 528– attentional problems, 389–
530 390
psychosis, schizophrenia and, 523– cognitive and memory
524 impairment, 390–392
suicidal drivers, 522, 524–526 disorders of mood and
warning patients about medication affective regulation,
effects on, 522–523, 532–533 392–394
Droperidol, intramuscular, 304 thought processing, 392
Index ❘ 583
past history, 386–389 patient refusal to comply with

medical/surgical, 386–388 intake procedures of, 6
psychiatric, 388–389 pharmacotherapy for acute
psychosocial, 389 agitation/aggression in, 302,
referral for, 383 303–311, 304, 306–307
atypical antipsychotics and risk of psychiatric decision making in, 278
death in, 200–201 psychiatric emergency room, 280–
causes of violence among, 381–382 284
criminal behavior in, 381 safety in, 13, 14, 281, 282, 291, 462–
deliberate self-harm among, 87–88 463, 470
delirium in, 386 seclusion and restraint in, 13, 280,
delusions of, 382, 388 291–295, 340–341
effects of memory and cognitive violence in, 462
deficits in, 381–382 violence risk assessment in, 3
incarcerated, 382–383 violence toward clinicians in, 462–
inpatient assaults by, 383, 386 463, 470
lashing out at caretakers, 382 weapons search in, 46, 249, 462
long-term care facilities for, 382 Emotional abuse during childhood
medication-related problems in, mood disorders related to, 79
382, 386 and risk of adult sexual
physiological changes in, 381 victimization, 79, 82–83
sexual aggression in, 382 Emotional lability
treatment of violent behavior in, alcohol intoxication and, 9
394–404 in bipolar disorder, 83
acute/emergent treatment, 394– Empathy, 252
395 intrinsic derangement of, 330
for causes of medical Empathy training, for students at risk
decompensation, especially for school violence, 550
delirium, 395–396 Employee assistance programs, 134
ensuring safety of patient and Employment problems, 22, 23. See also
others, 395 Workplace violence
for identified precipitants of Encephalitis, 192–194
violence, 396–398 autoimmune deficiency, 193
pharmacotherapy, 400–404 herpes simplex, 192–193
for underlying psychiatric lethargica, 192
disorders, 398–400 other types of, 192–193
use of restraints, 395 paraneoplastic, 193
Electroconvulsive therapy (ECT), 316 β-Endorphin, alcoholism and, 152
Electroencephalogram, 214 Environmental factors and violence,
Emergency settings, 12, 277–296. See 11–12, 253
also Psychiatric emergency design of psychiatric emergency
services room, 281
extended psychiatric observation design of seclusion room, 280, 350
services, 285–289 among elderly dementia patients,
managing violent patients in, 13 397–398
with posttraumatic stress among inpatients, 114–115, 260–261
disorder, 130–131 interventions for youth, 375
Environmental factors and violence structural racism, 36

(continued) substance abuse and, 35
safety of clinician’s office, 248–249, suicide risk and, 89
332, 463–464, 465–466, 476–477 Surgeon General’s report on mental
in workplace, 511–512 health and, 36
Epidemiologic Catchment Area study, transference, countertransference
106, 148 and, 40
Epilepsy, 190–192 treatment interventions and, 38–39
driving safety and, 522, 531 violence or homicide risk and, 37
ictal aggression in, 191 intimate partner violence, 80
interictal aggression in, 191 Euphoria
postictal aggression in, 191 cocaine-induced, 9
Epinephrine secretion, childhood in intermittent explosive disorder,
abuse, and trauma exposure, 78 217
Erotomanic stalkers, 412 Evolutionary theory of intimate
Escalating pattern of violence, 7, 265 partner violence, 486
Escitalopram, for chronic aggression, 202 Exchange theory of intimate partner
Estrogen therapy, for elderly men, 403 violence, 486
Ethnicity/race. See also Culturally Executive functioning, assessing in
competent violence risk elderly persons, 391
assessment Exhibitionism, 441, 443
alcohol-related intimate partner Expert testimony, 430, 434–439
violence and, 145 Barefoot vs. Estelle, 435–436, 437
antipsychotic administration and, 38 at civil commitment hearing, 438–
arrests for aggravated assaults 439
related to, 37 Daubert v. Merrill Dow
child abuse reporting bias related Pharmaceuticals, Inc., 436–437
to, 38 Federal Rules of Evidence and, 435,
“color line” in U.S., 36 436
dearth of information in mental Frye v. United States, 435, 437
health literature on, 36 “general acceptance” test for
discrimination based on, 36 admissibility of, 435
“monocultural ethnocentrism,” 36 history of admissibility of, 435
myths about youth violence related regarding battered women, 494–495
to, 37 reliability of assessments used for,
overdetermination of 437, 439
dangerousness in people of Exposure therapy, for posttraumatic
color, 38–39 stress disorder, 132
problems in assessing violence risk Extended psychiatric observation
in people of color, 37–41 services, 285–289
psychiatric diagnosis and, 35 admission criteria for, 285–286
racial and cultural differences benefits of, 285
between patients and case example of, 287
clinicians, 39 crisis stabilization unit, 286–287
as risk marker for violence, 37, 38 length of stay in, 285
seclusion and restraint of minority 72-hour observation bed,
patients, 38 psychiatric, 288–289
Index ❘ 585
23-hour observation bed, in childhood and adolescence, 360–

psychiatric, 286 361
Extrapyramidal symptoms, 402 impulsive suicide and, 224
Eye contact with violent patient, 468 reactive, 110
Eye enucleation, 225 schizophrenia and, 10
Eyeball pressing, 225 Fight-or-flight response, 215
Filicide, 448
Factitious disorder, false allegations of Filicide-suicide, 423, 526
stalking due to, 412 Finger biting, 225
Familicide, 424 Fire setting. See Pyromania
Family. See also Caregivers Fluoxetine, 316, 387
disruption of, 11 for paraphilias, 454
domestic violence within, 41–42 for posttraumatic stress disorder, 132
incest perpetrators in, 450, 451–452 Fluphenazine decanoate, 117
informing about institutional Flutamide, 454
policies and practices on use of Forgiveness therapy, 82
seclusion and restraint, 346 Forensic evaluations, 409–426
interviewing of, 4 checklist for, 410
about past episodes of violence, 7 of deceased persons, 418–425
intimate partner violence within, murder-suicide, 421–425, 422, 425
41–42, 80–81, 483–495 (See also psychological autopsy, 418–420
Intimate partner violence) suicide by cop, 420–421, 421
murder-suicide within, 422–424 of known offenders, 410–417
removing guns from patient by, 6 for competency to stand trial, 410
Family therapy to determine risk for future
for delinquent adolescents, 374 violence, 410–411
for intimate partner violence, 492 for insanity defense, 410
for pedophilia, 445 Internet child pornographers,
Fantasies 444–445
sexual, 445 rapists, 415–417, 416
violent, 23 stalkers, 411–415, 413, 414
Federal Bureau of Investigation (FBI) skills for performance of, 409
profiling of organized and of unknown offenders, 417–418
disorganized crime scenes, FBI profiling of organized and
417–418, 418 disorganized crime scenes,
report on myths about school 417–418, 418
shootings, 548 Forensic psychiatry, defined, 409
Federal Rules of Evidence, 435, 436 Frustration, 109
Fetal alcohol syndrome, 219 in elderly persons, 382
Fetal substance exposure, 213 impulsive aggression and, 213
Fetishism, 441 low tolerance for, in brain-injured
“Fight, flight, or freeze” responses, in patients, 187
posttraumatic stress disorder, 124 Frye v. United States, 435, 437
Fighting. See also Aggression; Assaults
amphetamine-induced, 9 GABA (γ-aminobutyric acid), 215
antisocial personality disorder and, Gabapentin, for aggression, 82, 253
11 in elderly patients, 403
Gait abnormalities, 6 Gun safety management, 6, 372, 373,

Galantamine, for Alzheimer’s disease, 557, 562, 563
399 Gun-related violence, 6
Gang activity, 359, 365, 375 among adolescents, 79, 362, 365,
alcohol and, 146 372, 373
youth homicide and, 362 controlling patient access to
Gas exposure, 197 weapons, 6, 372, 373, 557, 562,
Gasoline inhalation, 9 563
Gender and violence risk, 11, 23, 253 management of armed patient, 476
childhood sexual abuse, 442 mass murder, 424–425
in children and adolescents, 359, murder-suicide, 422
360–361 posttraumatic stress disorder and,
among inpatients, 260 128, 129, 133
intellectual disability and, 220 school shootings, 79, 359, 365, 371,
intermittent explosive disorder, 216 541–549, 551
murder-suicide, 422 Columbine High School, 543–544
pyromania, 218 FBI report on myths about, 548
rape, 415 Jonesboro, Arkansas, 544–545
sexual sadism, 446 managing threats or predictions
stalking, 411 of, 548–549
suicide, 89, 223 National Research Council
Genetics report on, 549
depression, aggression, and notable cases of, 541–543
serotonin transporter operant styles of perpetrators of,
polymorphisms, 93 547
twin and adoption studies of profiles of perpetrators of, 547–
violence, 364 548
Gilles de la Tourette syndrome, 195– social causation model of, 541
196 Virginia Tech, 546–547
attention-deficit/hyperactivity suicide, 85
disorder and, 195
self-injurious behaviors in, 225 Hallucinations, 284
Glue inhalation, 9 during alcohol withdrawal, 9
Golden, Andrew, 544–545, 547 command, 108–109, 267, 267–268,
Gooden v. Tips, 532 295
Gordis, Enoch, 155 violence and, 108–109
Grandiosity and violence in schizophrenia, 10
in bipolar disorder, 83 Hallucinogens, 9
cocaine-induced, 9 Haloperidol, 12
violence and, 108 anti-hostility effects of, 313
Graves’ disease, 387, 389 for chronic aggression, 202
Grigson, James, 435 intramuscular, 13, 294, 302, 304, 305
Group therapy combined with lorazepam, 302,
for paraphilias, 452–453 305, 388
prison-based, 325 during seclusion and restraint, 351
Guided imagery, for posttraumatic for violent inpatients, 114, 254
stress disorder, 132 Harassment, 60
Index ❘ 587
Hare Psychopathy Checklist (PCL), 62– FBI profiling of organized and

63 disorganized crime scenes,
cross-cultural validity of, 63 417–418, 418
in perpetrators of sexual violence, intermittent explosive disorder and,
445, 448 217
recidivism risk, 451 mass murder, 11, 37–38
screening version of, 62 murder-suicide, 421–425
Harris, Eric, 543–544 patients’ disavowal of tendencies
HCR-20, 21–23, 28, 30, 39, 46 toward, 327
case example of use in outpatient against physicians, 462
setting, 240–241 during pregnancy, 484
combining risk factors for, 22 random but purposeful, 177
generating final risk estimate with, in schools, 79, 359, 365, 371, 539–549
22–23 (See also School violence)
predictive validity of, 21 serial killing, 38
Psychopathy Checklist–Revised sexually motivated, 447–450
and, 62 brain abnormalities in
selecting and measuring risk factors perpetrators of, 447
for, 22 child victims of, 448
use in clinical practice, 28, 29 impetus for, 448–449
Head banging, 225 measuring sexual arousal in
Head injury. See Traumatic brain injury perpetrators of, 449–450
Health Insurance Portability and serial killings, 448–449
Accountability Act, 513 sexual sadism and, 442, 445–450
Helping Patients Who Drink Too Much: XYY syndrome in perpetrators
A Clinician’s Guide, 151, 152 of, 447–448
Helplessness, 213, 331 by strangulation, 446, 447
of elderly persons, 382 studying autopsy reports of victims
learned, of victim of intimate of, 327
partner violence, 495 suicide and, 91–92, 328–329
Hepatolenticular degeneration, 195 targeted, impulsive, 173–174
Herpes simplex virus (HSV) in workplace, 501, 505, 506, 509
encephalitis, 192–193 youth, 37, 359, 362, 375, 539
Hispanics, 35 Honesty in responding to patients, 468
culturally appropriate assessment Hopelessness, 284
of interpersonal violence in, 52 suicidality and, 87, 90, 525, 526
myths about youth violence among, Hormonal therapy, 335
37 Hospitalization. See also Inpatients
risk for violence or homicide, 37 assaults by psychotic patients
Histrionic personality disorder, 252 during, 109–118
Homicide, 7, 12, 39 decision for admission to, 3, 13
of abuser by victim of intimate of high-risk youth, 373
partner violence, 494–495 involuntary, 430
alcohol intoxication and, 142 for acute psychosis, 523–524
cocaine and, 147 expert testimony at hearing for,
by elderly persons, 384–385 438–439
ethnicity and, 37–38
Hospitalization (continued) Impulsive aggression, 68, 211–228

involuntary (continued) in autism, 221–222
for imminent risk of violence, in borderline personality disorder,
244–245 11
predicting future violence and, drinking and, 144
438–439 etiology of, 213–216
management of violent psychotic childhood trauma, 214
patients during, 113–118 neurochemistry, 215–216
patient transfer from outpatient prenatal development, 213–214
therapist to, 336 traumatic brain injury and brain
risk management for discharge dysfunction, 214–215
from, 13, 14, 559–560, 561 by inpatients, 262
for risk of imminent violence, 244 in intermittent explosive disorder,
violence among intellectually 216–217
impaired patients during, 189– neuropsychological testing of
190 persons with, 213
Hostility, 212 in personality disorders
antipsychotic effects on, 313 nontargeted, impulsive violence
distinction from anger, 68 incidental to emotional
instruments for assessment of, 68 escape, 174–176
HSV (herpes simplex virus) targeted, impulsive violence
encephalitis, 192–193 and, 172–174
Huntington’s disease, 195, 204–205 in persons with intellectual
Hydrocarbon inhalants, 9 disabilities, 219–221
5-Hydroxyindoleacetic acid, 92 in posttraumatic stress disorder, 124
Hyperarousal, 109 in pyromania, 217–219
Hyperthyroidism, 387 reactive assaults due to, 110
Hypervigilance, 335 in schizophrenia, 10
Hypoglycemia, 198 self-injurious behaviors due to, 224–
Hypotension, antipsychotic-induced, 226
402 suicide, 222–224
aripiprazole, 310 treatment of, 253
olanzapine, 308 Impulsivity, 252
Hypothalamic-pituitary-adrenal axis instruments for assessment of, 68–
effects of childhood abuse and 69, 72
trauma exposure on, 78 past history of, 7
mood symptom, violence and, 92 phencyclidine-induced, 9
Hypoxia, 196 Incarcerated persons, 326
elderly, 382–383
Ibn-Tamas v. United States, 494 group therapy for, 325
Ictal aggression, 191 posttraumatic stress disorder in, 126
IED. See Intermittent explosive substance abuse and mental illness
disorder among delinquent youth, 364
Iich’aa, 45 Incest perpetrators, 450, 451–452
Imminence of risk, assessment of, 244– Incoherence, amphetamine-induced, 9
245 Incompetent stalkers, 413
Impatience, 6 Infanticide, 423, 448
Index ❘ 589
Informed consent least restrictive environment for,

and confidentiality for interview of 113
adolescent, 367–368 pharmacological, 113, 114
for pharmacological treatment of physical space, 114–115
elderly patients, 401 predicting violence, 113, 114
for treatment during 72–hour hold, principles of, 113–114
288 seclusion and restraint, 113–114,
Inhalant abuse, 9 259
Inpatients. See also Hospitalization treatment team for, 113
acute management of high-risk managing threats against clinicians
youth, 373 by, 473–474
aftercare for, 14 measures of control for, 264, 266
assaults by, 109–118, 259–272 protecting personal property of, 115
criminal prosecution for, 269– violence among intellectually
270, 271 impaired patients, 189–190
environmental risk factors for, Insanity defense, 410, 524
114–115, 260–261 Institute of Medicine, 538, 549
gender and, 260 Instrumental violence, 109–110
impulsive vs. premeditated, 262 among adolescents, 370–372
long-term risk factors for, 261 Intellectual disability, 189–190, 219–
motivations for, 109, 113, 261–270 221, 227
on nursing staff, 259, 263–265 and aggression in institutional vs.
patient characteristics associated community settings, 189–190,
with, 109, 259–260 220–221
prevention of, 114 case example of, 219
reactive vs. instrumental, 109– definition of, 219
110, 262–263, 264 in DSM, 219
research on classification of, 110– gender and aggression in, 220
111 negative effects of aggression in
short-term risk factors for, 260 persons with, 220
staffing ratios, ward atmosphere prevalence of aggression in persons
and, 261 with, 220
threatening behaviors before, self-injurious behaviors and, 225
112 skill deficits and probability of
videotape recording of, 111–112, aggression in, 221
262 types of aggressive behavior
cold threats by, 268–269 associated with, 220
discharge of, 13, 14, 559–560, 561 Intensive Case Management, 284
Classification of Violence Risk Intent to harm, 6
for, 25–26 Interictal aggression, 191–192
ensuring safety of, 13, 14, 113, 115 Intermittent explosive disorder (IED),
management of violent behavior 11, 13, 216–217, 227, 317
among, 113–118 age distribution of, 216
case examples of, 116–118 aggressive episodes in, 217
criteria for, 113 frequency and duration of, 217
interpersonal interactions, 115– physical and affective symptoms
116 of, 217
Intermittent explosive disorder assessment for victims of, 489–490

(continued) case examples of, 41–42, 47, 82–83,
alcohol abuse and, 217 485–489
bipolar disorder and, 85 couples or family therapy for, 492
case example of, 216 crisis management for, 492
diagnostic criteria for, 216 culturally appropriate assessment
concerns about validity of, 217 of, 47–53
driving and, 522 A–E model for, 48, 49
DSM classification of, 216 in African Americans, 52–53
gender and, 216 in Asian Americans, 49–50, 51
helping patients recognize origins in Cambodians, 51–52
of rage in, 329 case example of, 47
pharmacotherapy for, 317 in Latinos, 52
prevalence of, 216, 317 in Muslims, 47
psychiatric comorbidity with, 217 in Somalis, 48–49
Internet in Vietnamese, 50–51
child sexual abuse and, 443–444 in cultures that consider women to
case example of, 444–445 be inferior, 48
pornography on, 444 cycle of violence in, 483–484
stalking and, 412 definition of, 41
Interpersonal violence, 37–38, 41–42. depression in victims of, 80–81, 485,
See also Aggression; Assaults; 488, 489
Fighting emotional abuse, 483–484
definition of, 41 ethnicity and, 80
ethnicity and risk for, 37–38 exposure of children to, 78, 80
among inpatients, 115–116 factors affecting patient’s decision
case studies of, 116–118 to leave violent relationship,
intimate partner violence, 41–42, 489, 492
47–53, 80–81 forms of physical abuse, 483
as means of control, 48 in gay couples, 41
posttraumatic stress disorder and, health consequences of, 41
124 helping patients recognize origins
among combat veterans, 125 of, 329
InterSept Scale for Suicidal Thinking, 295 legal issues related to, 492–495
Interviews arrests, 493
to assess short-term potential for battered women’s syndrome as
violence, 4, 13 legal defense, 494
with caregivers, 384 protection and reporting, 492–494
in outpatient settings, 238–239 psychiatrist as expert, 494–495
of significant others, 4, 13 restraining orders, 493–494
telephone, 465 perpetrated by veterans with
Intimacy-seeking stalkers, 413 posttraumatic stress disorder
Intimate partner violence (IPV), 41–42, or depression, 127
80–81, 327, 483–495 posttraumatic stress disorder in
alcohol-related, 82–83, 145–146 victims of, 80, 485
assessing risk for commission of, during pregnancy, 41, 484
490–491 prevalence of, 41, 80, 484
Index ❘ 591
rape, 484, 488 in intermittent explosive disorder,

risk factors for, 484, 491 217
screening for, 41, 490 in mood disorders, 83
settings for, 489 in posttraumatic stress disorder, 132
severe mental illness and, 485–489 sleep deprivation and, 198
spousal/consortial murder-suicide,
422–423 Johnson, Mitchell, 544–545, 547
stalking and, 411 Joint Commission for the Accreditation
suicidality and, 80–81, 82, 485, 489 of Healthcare Organizations
theories of causation of, 486–487 (JCAHO), regulations for use of
treating assaultive partner in, 492 seclusion and restraint, 339–340,
treating victims of, 82, 489, 491–492 341, 344, 346, 351, 353
victims’ reluctance to discuss, 490 Jonesboro, Arkansas, school shootings,
women as perpetrators of, 491, 492 544–545, 547
Intimate stalkers, 412, 414 Judgment impairment during
Intoxication intoxication, 9
alcohol, 9 Judicial system, 5, 430. See also Legal
impaired judgment due to, 9 issues
inhalant, 9 dependence on accurate assessment
murder-suicide and, 422, 424 of dangerousness, 430, 437–438
phencyclidine, 9 for juveniles, 359, 367–368
self-injurious behaviors and, 225 waiving juveniles to adult court,
signs of, 6 375
vehicular crashes due to, 522
violence, personality disorders and, Kaczynski, Ted, 546
167, 171, 173, 178 Kirk v. Michael Reese Hospital and
Investment theory of intimate partner Medical Center, 532
violence, 487 Klebold, Dylan, 543–544
Involuntary hospitalization, 430 Klüver-Bucy syndrome, 193
for acute psychosis, 523–524
expert testimony at hearing for,
LA Suicide Prevention Center, 418
438–439
Lack of remorse, 11, 40, 61
for imminent risk of violence, 244–
Lamotrigine, 253, 303, 315
245
Language impairments, in elderly
predicting future violence and, 438–
persons, 390
439
Latinos. See Hispanics
Involuntary outpatient commitment,
Lead exposure, 197
547
Least restrictive environment, 113
IPV. See Intimate partner violence
Legal issues, 326, 429–440
Irritability
civil commitment, 430, 438–439
in adolescents, 374
consent and confidentiality for
anabolic steroid use and, 9
interview of adolescent, 367–
antipsychotic-induced, 200
368
brain tumors and, 194
criminal prosecution for inpatient
in brain-injured patients, 187, 188
assaults, 269–270, 271
hypoglycemia and, 198
duties of clinician to prevent
inpatient assaults and, 260
vehicular crashes, 531–534
Legal issues (continued) Lo, Wayne, 546

duty of clinicians to warn and Long-term care settings, 186, 382, 392,
protect potential victims, 246, 396–397
247, 269, 333–334, 373, 430–434, managing identified precipitants of
524, 557–559, 560 violence in, 396–398
expert testimony of clinicians, 430, Long-term violence risk
434–439 actuarial risk assessment of, 18
forensic evaluations, 409–426 psychological testing for
insanity defense, 410, 524 assessment of, 61
juvenile justice system, 359, 367–368 Lorazepam
legal responsibility and violence abuse of, 268
associated with personality for alcohol withdrawal, 305
disorder, 165 for chronic aggression, 203
legal risks associated with intramuscular, 13, 302, 304, 305, 386
workplace violence, 507–508 advantages and disadvantages
legal standard for competency to of, 305
stand trial, 410 combined with aripiprazole, 310
related to intimate partner violence, combined with haloperidol, 302,
492–495 305, 388
related to potential violence, 246–247 withdrawal from, 316
restraining orders against stalkers, Loud patients, 5, 106
415 Love obsessional stalkers, 412
seclusion and restraint, 353 Lu, Gan, 546
structured risk assessment for Luteinizing hormone–releasing
violence after discharge from hormone (LHRH) agonists, for
treatment facility, 4 paraphilias, 453, 454
waiving juveniles to adult court, 375
Lesch-Nyhan syndrome, 345 MacArthur Violence Risk Assessment
Leuprolide acetate, for paraphilias, Study, 23, 28, 38, 67–68, 69
453, 454 Mal de pelea, 45
Lewis, Kurt, 239 Malingering, false allegations of
Lewy body dementia, 186, 198 stalking due to, 412
LHRH (luteinizing hormone–releasing Manganese exposure, 197
hormone) agonists, for Mania, 13, 25. See also Bipolar disorder
paraphilias, 453, 454 aggression and, 84–86
Licensed independent practitioner deliberate self-harm and, 88
(LIP), to initiate and monitor in elderly persons, 386, 393
seclusion and restraint, 344 legal problems associated with, 84
Limbic system, in alcohol effects, 143 multiple sclerosis and, 196
Limit setting by clinician, 469–470 poststroke, 189
Lithium, 314, 374 psychosis in, 10
for akathisia, 315 subtypes of, 84
antisuicide effect in bipolar Marijuana. See Cannabis
disorder, 91, 92 Marital power theory of intimate
combination therapy with partner violence, 486
antipsychotics, 315 Marital status and suicide risk, 89
for elderly patients, 386 Marital/couples therapy, 557
Index ❘ 593
for intimate partner violence, 492 Memantine, for Alzheimer’s disease,

for pedophilia, 445 399, 402
for posttraumatic stress disorder, Memory impairment in elderly
134, 137 persons, 381, 387
Marquis de Sade, 448 assessment of, 390–392
Mass murder, 11, 37–38 with dementia, 390
categories of, 424, 425 Mental handicap. See Intellectual
characteristics of perpetrators of, disability
424–425 Mental health professionals.
instrumental, 425 See Clinicians
massacres, 425 Mental Health Triage Scale, 283
school shootings, 79, 359, 365, 371, Mental retardation. See Intellectual
541–549, 551 disability
victim-specific, 425 Mental status examination of elderly
MCIT (mobile crisis intervention persons, 389–394
team), 290–291 attentional problems, 389–390
MCMI-III (Millon Clinical Multiaxial cognitive and memory impairment,
Inventory–3), 63, 64, 65, 163 390–392
Medical disorders, 106, 196–205. See disorders of mood and affective
also Neurological disorders regulation, 392–394
characteristics of aggression in, 186 thought processing, 392
delirium, 197, 197 Mercury exposure, 197
in elderly persons, 386–388, 395–396 Metabolic syndrome, atypical
hypoglycemia, 198 antipsychotic–induced, 403
medication side effects, 197–198 Metal detectors, 249
misdiagnosis in elderly persons, Methadone maintenance treatment, 154
386–388 Methamphetamine, 147
psychosis and, 10 Michigan Alcoholism Screening Test,
rheumatic diseases, 198 445
sleep disorders, 198 Millon Clinical Multiaxial Inventory–3
toxin exposure, 197 (MCMI-III), 63, 64, 65, 163
treatment of, 199–205 Minimal brain dysfunction, 196
assessment and quantification of Mini-Mental State Examination
aggressive episodes, 199 (MMSE), 285, 390, 392, 402
behavioral strategies, 201, 204 Minimum Data Set, 397
pharmacotherapy, 199–201 Minnesota Multiphasic Personality
for acute aggression and Inventory–2 (MMPI-2), 28, 63–65,
agitation, 199–200 67, 72, 163, 416, 491
for chronic aggression, 200– Clinical and Supplementary Scales
201, 202–203 of, 64
Medical education in management of for evaluation of test-taking style, 64
violent patients, 326 Overcontrolled Hostility Scale of,
Medical history of elderly persons, 64–65
386–388 profile most commonly associated
Medication side effects, 197–198 with violence, 64
Medroxyprogesterone acetate, for Psychopathic Deviance Scale of, 65
paraphilias, 453, 454 Minor violence, defined, 108
Miscarriage, 41 driving safety and, 522

MMSE (Mini-Mental State for outpatients with impulsive
Examination), 285, 390, 392, 402 aggression, 317
Mobile crisis intervention team for violent youth, 374
(MCIT), 290–291 Moral reasoning training, for students
Modafinil, 150 at risk for school violence, 550
Mode of death, 419 Motivational enhancement therapy, for
Monoamine oxidase-A, 364 alcoholism, 155
Mood disorders, 77–94. See also Bipolar Motivations for inpatient assault, 261–
disorder; Depression; Mania 270
in adolescents, 374 Movement disorders, 194–196
biology of aggression in, 92–93 Huntington’s disease, 195, 204–205
in genesis of violence, 83–86 Parkinson’s disease, 195
among elderly persons, 392–394, Tourette’s syndrome, 195–196
399–400 Wilson’s disease, 195
intermittent explosive disorder and, Multiple sclerosis, 196
217 Multisystem atrophy, 198
substance abuse and, 84, 85 Multisystemic therapy, for delinquent
violence in genesis of, 77–83 adolescents, 374
community violence, war, and Murder-suicide, 421–425
terrorism, 81 definition of, 421
direct abuse or neglect, 79–81 extrafamilial, 424–425
adult assault, 80 categories of mass killing, 424,
intimate partner violence, 80– 425
81 characteristics of perpetrators of,
physical abuse, 79 424–425
sexual abuse, 79–80 school shootings, 543, 546
verbal and emotional abuse, 79 gender and, 422
passive exposure, 77–79 motivators for, 422
parental substance abuse, 78– of other family members, 423–424
79 as altruistic, 423, 424
witness to domestic violence, familicide, 424
77–78 filicide, 423
prevention of, 82 prevalence of, 421–422
treatment approaches for, 81–83 spousal/consortial, 422–423
and violence toward self, 86–92 time frame for, 421
deliberate self-harm, 86–89 typology for, 422
bipolar disorder and, 88–89 by vehicular crash, 526
depression and, 86–88 Muslims, 35
suicide, 89–92 culturally appropriate assessment of
bipolar disorder and, 90–91 interpersonal violence in, 47, 49
depression and, 89–90
homicide and, 91–92 Nadolol, 316
Mood stabilizers, 84–85, 203, 334. See Naidu v. Laird, 524
also Anticonvulsants Nail biting, 225
combination therapy with Naltrexone, for alcoholism, 152–153
antipsychotics, 314–315 in elderly persons, 399
Index ❘ 595
Narcissism, 252 in mood disorders, 92–93

Narcissistic personality disorder, 11 neurochemistry, 92–93, 215–216
purposeful, instrumental violence of alcohol, 143
and, 168–169 Neuroimaging
targeted, impulsive violence and, of aggression in dementia, 187, 400,
172, 174 402
violence related to chronic paranoia of brain regions associated with
or related misconception and, impulsivity and aggression,
179 214–215
Narcolepsy and driving safety, 522, 531 Neuroleptics. See Antipsychotics
NASH acronym for mode of death, 419 Neurological disorders, 185–196. See
National Comorbidity Replication also Medical disorders
Study, 89 central nervous system tumors, 194
National Comorbidity Survey, 222 characteristics of aggression in, 186
National Crime Victimization Survey, congenital brain disorders and
415, 461 developmental disorders, 189–
National Highway Traffic Safety 190, 219–221 (See also
Administration, 523 Intellectual disability)
National Institute of Mental Health dementia, 185–187
Clinical Antipsychotic Trials of driving safety and, 522, 531
Intervention Effectiveness, encephalitis, 192–194
107–108 epilepsy, 190–192
Epidemiologic Catchment Area among inpatients, 261
study, 106, 148 movement disorders, 194–196
National Institute of Occupational Huntington’s disease, 195, 204–
Safety and Health, 504 205
National Institute on Alcohol Abuse Parkinson’s disease, 195
and Alcoholism, 151, 152, 155 Tourette’s syndrome, 195–196
National Institute on Drug Abuse, 151 Wilson’s disease, 195
National Research Council, 538, 549 psychosis and, 10
National Triage Scale, 283 self-injurious behaviors, 225
National Vietnam Veterans stroke, 188–189
Readjustment Study (NVVRS), traumatic brain injury, 187–188, 204
125–126 treatment of, 199–205
National Violent Death Reporting assessment and quantification of
System, 421 aggressive episodes, 199
Native Americans, 35. See also Alaska behavioral strategies, 201, 204
Natives; American Indians pharmacotherapy, 199–201
substance abuse and suicidality for acute aggression and
among, 144 agitation, 199–200
Natural History of Alcoholism, The, 155 for chronic aggression, 200–
Neglect of children. See Childhood 201, 202–203
abuse or neglect Neurological soft signs, 196
Neonaticide, 423 Neuropsychological testing
Neurobiology of persons with impulsive
of addiction, 149–150 aggression, 213
of aggression of violent offenders, 214
Neurotransmitters, 215–216 ODD (oppositional defiant disorder),

definition of, 215 363, 369
norepinephrine, 215–216 Office safety, 248–249, 332, 463–464,
serotonin, 92–93, 215, 328 465–466, 476–477
Nicotine patch, for elderly patients, Olanzapine, 12, 311
399, 400, 403 for aggressive elderly patients, 401
Nightmares, in posttraumatic stress anti-hostility effects of, 313
disorder, 132, 133–136 for chronic aggression, 202
Nitrous oxide exposure, 197 intramuscular, 303, 304, 305, 306,
Noncompliance with medications, 11, 309–310
12, 114, 302–303, 335 adverse effects of, 309
Norepinephrine, aggression, dosage of, 304, 309
impulsivity and, 215–216 effectiveness of, 311
Novaco Anger Scale, 67, 72 indications for, 309
Nurses transition to oral administration,
“cold” threats against, 268–269 311
de-escalation techniques used by, long-term treatment with, 312–314
265, 266 valproate and, 314–315
inpatient assaults on, 259, 263–265 for violent psychotic inpatients, 114,
debriefing after, 265 254
psychological support for, 265 Olfactory aversion, for paraphilias, 453
prevalence of violence against, 461 Ondansetron, for alcoholism, 148
in psychiatric emergency room, 282 Opiate analgesics, 198
Nursing home placement, 186, 392 Opioid dependence, 147–148
NVVRS (National Vietnam Veterans maintenance treatment of, 154
Readjustment Study), 125–126 withdrawal from, 147–148
Opportunistic rapists, 415
Observation of patient Oppositional defiant disorder (ODD),
agitated elderly persons, 394 363, 369
extended psychiatric observation Organized and disorganized crime
services, 285–289 scenes, FBI profiling of, 417–418,
during seclusion and restraint, 292, 418
342, 344, 348–349 OSHA (Occupational Safety and
Obsessional stalkers, 411–412 Health Act), 507
Obsessive-compulsive disorder, 195 Osteopenia/osteoporosis, anti-
Obsessive-compulsive personality androgen–induced, 454–455
disorder Outline for Cultural Formulation
nontargeted, impulsive violence (OCF), 43–44, 44, 46, 54
incidental to emotional escape Outpatient settings, violence risk
and, 176 assessment in, 237–255, 465–466
targeted, impulsive violence and, actuarial assessment, 237–238
172, 174 assessing imminence of risk, 244–245
Occupational Safety and Health Act Axis I diagnoses and, 250–252
(OSHA), 507 clinical assessment, 237–239
OCF (Outline for Cultural conceptual framework and case
Formulation), 43–44, 44, 46, 54 example of, 239–243
O’Connor v. Donaldson, 438 environmental variables and, 253
Index ❘ 597
gender and, 253 targeted, impulsive violence and,

general principles for, 237–239 172, 174
interview questions for, 238–239 violence related to chronic paranoia
legal aspects of potential violence, or related misconception and,
246–247 179–181
in patients with past history of Paranoid stalkers, 171, 179
violence, 243–244 Paraphilias, 327, 382, 441, 442
personality traits and, 252 coercive, 446
preventive action to reduce risk, cognitive distortions and, 453
245–246 treatment of, 452–455
risk of violence to clinician, 248– algorithm for, 454
249, 254, 259, 332 cognitive-behavioral therapy,
screening questions for, 238 452–453
threats to patients, 247–248 pharmacological, 453–455
treatment variables and, 253–254 relapse prevention, 453
use of HCR-20 for, 240–241 Parasomnias, 198
Overdose of drug Parasuicide, 222
cocaine, 147 Parent management training, 374
by suicidal patient, 295 Parents
Overt Aggression Scale, 7, 8, 188, 199, 215 adolescent depression related to
Oxcarbazepine, for aggression, 85, 253 maternal depression, 78
children learning how to manage
PADs (Protective Aggression Devices), aggression from, 363
352 effects on children of passive expo-
PAI (Personality Assessment sure to violence between, 78
Inventory), 63, 64, 65, 163 effects on children of substance
Correctional Report of, 65 abuse by, 78–79
Pain, 41 murder of children by, 448
Paint/paint thinner inhalation, 9 murder-suicide, 423–424
Paliperidone, 312 poor parenting by, 363
PANSS (Positive and Negative Parkinsonism, drug-induced, 402
Syndrome Scale), 313, 314 Parkinson’s disease, 195, 198
Paraneoplastic encephalitis, 193 Paroxetine, for posttraumatic stress
Paranoia, 168 disorder, 132
assaultiveness and, 107 Past history of violence or impulsive
in bipolar disorder, 83 behaviors, 5, 7
stimulant-induced, 9, 147 in adolescents, 368–369
workplace violence and, 511 assessing potential future violence
Paranoid defenders, 171 in patients with, 243–244
Paranoid delusions, 10, 12, 64, 118, “dissecting” details of, 7
265–266, 267, 284 information to obtain about, 7
Paranoid personality disorder, 11 as risk factor for future violence, 7,
nontargeted, impulsive violence 22, 23, 259–260
incidental to emotional escape severity of, 7
and, 175 Patriarchy theory of intimate partner
purposeful, targeted, defensive violence, 486
violence and, 171–172 Paulhus Deception Scales (PDS), 69–70
PCL. See Psychopathy Checklist definition of, 161, 165

PCL-R. See Psychopathy Checklist– in delinquent youth, 369
Revised kinds of violence associated with,
PCL-YV (Psychopathy Checklist– 167–181
Youth Version), 62, 369, 370 nontargeted, impulsive violence
PCP (phencyclidine), 9, 148 incidental to emotional
PDS (Paulhus Deception Scales), 69–70 escape, 174–176
Pedicide, 423 case examples of, 175
Pedophile Index, 445, 449 purposeful, instrumental
Pedophilia, 329, 441, 442–445. See also violence, 167–169
Sexual abuse in childhood case examples of, 168
assessment of, 444–445 purposeful, noninstrumental
child homicide and, 448–449 violence, 169–170
measuring sexual arousal in case examples of, 170
perpetrators of, 449–450 purposeful, targeted, defensive
hormonal treatment of, 335 violence, 170–172
treatment of, 445, 452–455 case example of, 171
Peer effects and youth violence, 363, random but purposeful
364, 365, 375 violence, 176–177
Penile plethysmography (PPG), 70–71, case example of, 177
416, 445 targeted, impulsive violence,
Persecutory delusions, 10, 108, 266, 267 172–174
Personality assessment, 63–65, 72 case examples of, 172–174
Millon Clinical Multiaxial violence related to chronic
Inventory–3, 63, 64, 65, 163 paranoia or related
Minnesota Multiphasic Personality misconception, 179–181
Inventory–2, 28, 63–65, 67, 72, case example of, 179–180
163 violence related to perceived/
Personality Assessment Inventory, feared loss or
63, 64, 65, 163 abandonment, 177–179
of specific characteristics, 67–69 case example of, 178
aggression, 68 prognosis for, 161
anger, 67–68 self-injurious behaviors and, 225
hostility, 68 treatment and management of, 166–
impulsivity, 68–69 167
of test-taking style, 63–64 pharmacotherapy, 317
Personality Assessment Inventory variations in diagnosis-related
(PAI), 63, 64, 65, 67, 163 behavior in, 164–165
Correctional Report of, 65 violence and, 163–164
Personality changes intoxication and, 167, 171, 173,
in posttraumatic stress disorder, 124 178
in Wilson’s disease, 195 legal responsibility and, 165
Personality disorders, 11, 13, 22, 27, Personality traits and violence, 252
161–182, 301. See also specific Pervasive developmental disorder, 369.
personality disorders See also Autism spectrum disorder
among adolescent offenders, 364 Pervasively angry rapists, 416
anger attacks and, 252 Phallometry, 449
Index ❘ 599
Pharmacotherapy Phenytoin, for impulsive aggression, 253

driving safety and, 522–523, 532–533 Physical abuse
for paraphilias, 453–455 during childhood, mood disorders
for posttraumatic stress disorder, related to, 79
131–132, 133–136 by intimate partner, 483–495 (See
for sleep disturbances, 134 also Intimate partner violence)
for students at risk for school psychiatric disorders in victims of,
violence, 550 80
for substance addiction, 152–154, suicide risk and, 89
399 Physical distance from violent patient,
Pharmacotherapy for agitation/ 468
aggression, 84–85, 253–254, 301– Physical space interventions, in
318. See also specific drugs and inpatient settings, 114–115
classes Physicians. See also Clinicians
during alcohol withdrawal, 302 violence against, 461–463
for chemical restraint, 351 Physician’s Guide to Assessing and
court-ordered, 114 Counseling Older Drivers, 528–530
depot medications for, 11 Pibloktoq, 44, 45
in elderly persons, 400–404 Poisoning of self, 225
acute/emergent management, Police
394–395 interviewing of, 4
cholinesterase inhibitors for racial stereotyping by, 46
Alzheimer’s disease, 302, to subdue violent patient in
303, 399, 402 emergency department, 13
among inpatients, 113, 114 Police reports, 5
long-term treatment, 312–316 Pornography on Internet, 444
in medical or neurological illness, Positive and Negative Syndrome Scale
199–201 (PANSS), 313, 314
acute aggression and agitation, Positron emission tomography, 187, 402
199–200 Postictal aggression, 191
chronic aggression, 200–201, Posttraumatic stress disorder (PTSD),
202–203 123–137
noncompliance with, 11, 12, 114, in combat veterans, 81, 123, 124–125
302–303, 335 diagnostic criteria for, 123
in psychiatric emergencies, 302, “fight, flight, or freeze” responses
303–311, 306–307 in, 124
intramuscular medications for, history taking in, 133
303–311, 304 increased expectation of danger in,
with seclusion and restraint, 340, 124
343, 351–352 intimate partner violence and, 41,
psychotherapy and, 334–335 80, 82, 485
refusal of, 302–303 management of violent patients
therapeutic alliance for, 114 with, 81–83, 130–137
use on 72-hour extended case examples of, 133–137
observation unit, 287 clinician safety and, 130–131
in violent youth, 374–375 cognitive-behavioral therapy,
Phencyclidine (PCP), 9, 148 132–133, 134
Posttraumatic stress disorder child sexual abuse, 443

(continued) toward clinicians, 461
management of violent patients intimate partner violence, 41, 80, 484
with (continued) long-term pharmacotherapy for, 312
in emergency department, 130– murder-suicide, 421–422
131 in persons with intellectual
marital/couples therapy, 134, 137 disability, 220
pharmacological, 131–132, 133– in psychiatric illness, 148–149, 250
134 bipolar disorder, 107, 250
psychiatric comorbidity with, 124, depression, 106
128, 129 posttraumatic stress disorder, 126
screening patients with trauma schizophrenia or psychosis, 106–
history for, 129 107, 250–252
suicide and, 128–129 rape, 143
symptoms of, 123–124, 133–136 in substance-abusing patients, 107
terrorism and, 81 symptoms of mental illness and,
violence and, 124–137 107–109
assessment of, 129–130, 130, 131 in workplace, 505–507
family violence, 127 Prevention
firearm-related, 128, 129, 133 Blueprints for Violence Prevention,
incarcerated men, 126–127 374
risk factors for, 125–127, 133 de-escalation techniques for
in violent youth, 369 agitated inpatients, 265, 266
Poverty, 11 duty of clinicians to warn and
Power-assertive rapists, 416 protect potential victims, 246,
Power-reassurance rapists, 416 247, 269, 333–334, 373, 430–434,
PPG (penile plethysmography), 70–71, 524, 557–559, 560
416, 445 legal duties of clinician to prevent
Prazosin, for posttraumatic stress vehicular crashes, 531–534
disorder, 132, 133–134, 135 of mood disorders in violence-
Predatory stalkers, 413 exposed persons, 82
Predatory violence among adolescents, of psychotic assaults by inpatients,
370–372 267
Prednisone, 198 to reduce violence risk in outpatient
Pregnancy settings, 245–246
fetal substance exposure during, 213 of school violence, 549–550
homicide during, 484 of short-term risk of violence, 3, 14
intimate partner violence during, Private stranger stalkers, 412, 414
41, 484 Prodrome of violence, 467
Premature labor, 41 Profiling
Premeditated aggression, 212–213, 312 by FBI of disorganized and
by inpatients, 262 organized crime scenes, 417–
psychopathy and, 213 418, 418
Prenatal risk factors for impulsivity of perpetrators of workplace
and aggression, 213–214 violence, 509–510
Prevalence of violence of school shooters, 547–548
among adolescents, 359, 360–362 Project Combine, 153
Index ❘ 601
Project MATCH, 155 in community, 289–291

Property destruction, 7, 27 definition of psychiatric emergency,
autism and, 222 277, 278
inpatient assaults and, 260 description of, 277–278, 279
intellectual disability and, 220 extended psychiatric observation
intermittent explosive disorder and, services, 285–289
217 case example of, 287
Propranolol, 203, 204, 316 referral to on-call mental health
Protective Aggression Devices (PADs), specialist, 278
352 safety in, 13, 14, 281, 282, 291, 462–
Psychiatric disorders and violence, 22, 463, 470
162–163, 310. See also specific service delivery models for, 278,
disorders 280–285
actuarial risk assessment of, 4 psychiatric emergency room,
among adolescent offenders, 364 280–284
association with symptoms of design of, 281
mental illness, 107–109 seclusion area of, 280, 350
Axis I diagnoses, 250–252 staff of, 282–283
in elderly persons, 398–400 weapons search in, 46, 249, 462
mood disorders, 392–394 Psychiatrists. See Clinicians
ethnicity and diagnosis of, 35 Psychological autopsy, 418–420
among inpatients, 261 indications for, 419
intermittent explosive disorder, 11, 13 of murder-suicide, 421–425
intimate partner violence, 41, 485 of suicide by cop, 420–421, 421
mania, 10 techniques for, 419–420
mood disorders, 77–94, 149 Psychological testing, 59–73, 317
personality disorders, 11, 161–182 actuarial assessment and, 59–60
posttraumatic stress disorder, 123– case example of, 60
137 in civil vs. criminal settings, 61
prevalence of, 148–149, 250 of a criminal defendant, 60
psychotic disorders, 10, 105–119, 149 instruments for violence risk
public fear of, 148 assessment, 61–73
substance abuse and, 148–149 defensiveness, 69–70
suicide and, 89 general personality assessment,
vehicular crashes, 521–523 63–65
in workplace, 511 psychopathy, 61–63
Psychiatric emergency services (PESs), Rorschach test, 66–67
277–296. See also Emergency sexual offenders, 60, 70–72
settings specific personality
assessment issues in emergency characteristics, 67–69
setting, 291–295 substance abuse, 69
restraints, 292–293, 294–295 of special populations, 60
safety, 291 time frame of concern for, 61
seclusion, 291–292, 293–294 Psychologists. See Clinicians
suicide risk, 294–295 Psychopathy
case examples of, 279–280, 284–285, assessment of, 62–63
462–463 definition of, 61
Psychopathy (continued) court-mandated, 327

instrumental assaults and, 110, 111 for depression and posttraumatic
premeditated-predatory aggression stress disorder resulting from
and, 213 violence exposure, 82
Psychopathy Checklist (PCL), 62–63, 445 deranged transferences in, 331–332
cross-cultural validity of, 63 determining feasibility of, 330–331
in perpetrators of sexual violence, for elderly patients, 399
445, 448, 451 fears and liabilities in, 327
screening version of, 62 goals and strategies of, 329–331
Psychopathy Checklist–Revised (PCL- for paraphilias, 452–453
R), 27, 62–63, 68, 213 for personality disorders, 166, 169
Psychopathy Checklist–Youth Version pharmacotherapy and, 334–335
(PCL-YV), 62, 369, 370 for posttraumatic stress disorder,
Psychosis, 10, 13, 105–119. See also 132–133
Delusions; Hallucinations; problem of agency in, 327
Schizophrenia setting for, 332
brain tumors and, 194 for substance addiction, 154–155
driving safety and, 522, 523–524, 533 supervision of, 335
in elderly persons, 399, 402 transference threats during, 472
ethnicity and diagnosis of, 35 unstable treatment situations and,
exacerbation by seclusion, 349 335–336
inpatient assaults due to, 109, 110– vectors of violence and, 328–329
111, 262–263, 264, 266–267 with victims, 333–334
command hallucinations and, with violent youth, 374
108–109, 267, 267–268 Psychotic identification with victim, 10
indicators of impending assault, PTSD. See Posttraumatic stress
267 disorder
strategies for prevention of, 267 Public-figure stalkers, 412, 414
motivations for assault in patients Pupillary dilation, 6
with, 109, 113 Pyromania, 217–219, 227
poststroke, 189 among arsonists, 218
prevalence of violence in, 106–107 case example of, 217–218
psychiatric disorders associated definition of, 218
with, 10 DSM classification of, 218
in rheumatological diseases, 198 gender and, 218
steroid-induced, 198 helping patients recognize origins
violence associated with symptoms of, 329
of, 107–109 prevalence of, 218
workplace violence and, 511 psychiatric comorbidity with, 219
Psychosocial history of elderly risk factors for, 218
persons, 389
Psychotherapy, 325–337 QT interval prolongation, ziprasidone-
for aggression, 317 induced, 308
clinical inexperience of clinicians Quetiapine
with violent patients, 326–327 for aggressive elderly patients, 382,
countertransference issues in, 332– 400, 401, 403
333 for chronic aggression, 202
Index ❘ 603
in emergency setting, 303 Risk factors for suicide, 89

long-term treatment with, 312–314 Risk factors for violence
Quiet patient, 6 in adolescents, 364–365, 366, 370
Quinolone antibiotics, 198 available means to inflict harm, 6–7
behavioral, 28
Race. See Ethnicity/race on Classification of Violence Risk, 23
Racing thoughts, in intermittent clinician survey of relevance of, 28
explosive disorder, 217 against clinicians, 465–466, 466
Ramelteon, for sleep problems, 134 combining of, 21
Rape. See Sexual assault/rape cultural, 35
Rapid eye movement behavior demographic factors, 11–12
disorder, 198 generating final risk estimate based
Rapid Risk Assessment for Sexual on, 21
Offender Recidivism (RRASOR), on HCR-20, 22
29, 437 by intimate partner, 484
Rapists, 415–417. See also Sexual intimate partner violence, 491
offenders minor violence, 108
assessment for future past history of violence, 7, 22, 23,
dangerousness, 416 259–260
diagnostic classification of, 446 personality disorders, 11
factors associated with reoffense, 417 in posttraumatic stress disorder,
recidivism risk for, 417, 450 125–126, 133
typology of, 415–416, 416 protective factors and, 37
Reactive assaults, 109–110 psychosis, 10
Reasoning abilities, assessing in vs. risk markers, 37, 38
elderly persons, 391 in schools, 551
RECON stalker typology, 412, 414 selection and measurement of, 20
Rejected stalkers, 413 serious violence, 108
Relationship instability, 11, 22 sexual violence, 70
Religion. See Spirituality and religious treatment noncompliance, 11
beliefs on Violence Risk Appraisal Guide,
Reporting, of intimate partner 26–27
violence, 492–494 in workplace, 508–512
Resentful stalkers, 413 Risperidone, 387
Resident physicians, violence against, for aggressive elderly patients, 391–
461–463 392, 393, 401
Resource theory of intimate partner anti-hostility effects of, 313
violence, 486 electroconvulsive therapy and, 316
Restlessness liquid, 311
antipsychotic-induced, 200 long-term treatment with, 312–314
after traumatic brain injury, 187, 188 valproate and, 314–315
Restraining orders for violent inpatients, 254
in cases of intimate partner for violent youth, 374
violence, 493–494 Rivastigmine, for Alzheimer’s disease,
against stalkers, 415 399
Restraint. See Seclusion and restraint Robert Wood Johnson Foundation, 151
Rheumatic diseases, 198 Rorschach Inkblot test, 29, 66–67, 72
RRASOR (Rapid Risk Assessment for Schizophrenia, 4, 13, 27, 105–119, 301
Sexual Offender Recidivism), 29, antisocial personality disorder and,
437 250–251
availability of potential victim to
Sadism, 330. See also Sexual sadism patient with, 7
Safety causes of violence in, 10
of aggressive elderly persons and clinical violence risk assessment in, 12
their caregivers, 395 command hallucinations in, 108–
of clinician, 248–249, 254, 332, 461– 109, 267, 267–268
478 conduct disorder and, 250–251
of clinician’s office, 248–249, 332, delusions in, 10, 12, 265–266
463–464, 465–466, 476–477 diagnosed in emergency setting, 279
vs. confidentiality, 241, 247 driving safety and, 523–524
driving, 521–535 escalating pattern of violence in, 7
duty of clinicians to warn and inpatient management of violent
protect potential victims, 246, patients, 113–118
247, 269, 333–334, 373, 430–434, case examples of, 116–118
524, 557–559, 560 intimate partner violence and, 485
in emergency settings, 13, 14, 281, motivations for assault in patients
282, 291, 462–463, 470 with, 109, 113
gun safety management, 6, 372, 373, pharmacotherapy for aggression/
557, 562, 563 agitation in, 113, 114, 302–303
management of armed patient, 476 prevalence of violence in, 250–252
of patient in restraints, 347 psychosis in, 10
protecting victims and reporting of pyromania and, 219
intimate partner violence, 492– reversible risk factors for violence
494 in, 106
seclusion room design for, 350 self-injurious behaviors in, 225
weapons checks for, 46, 249, 332, 462 substance abuse and, 250
of workplace, 507 violence and positive symptoms of,
Sarcoidosis, 198 252
SASSI (Substance Abuse Subtle violence assessment in, 105–113
Screening Inventory), 69, 72 association with symptoms of
Satiation (masturbatory), for mental illness, 107–109
paraphilias, 453 classification of assaults, 109–111
SAVD (School-Associated Violent research on classification of
Death) study, 539 assaults, 110–111
SAVRY (Structured Assessment of videotape recording of inpatient
Violence Risk in Youth), 370 assaults, 111–112
Scheduling appointment for violence Schizotypal personality disorder, 174,
risk assessment, 466 176, 179–180
Schizoaffective disorder, 12, 251, 302 Schneidman, E.S., 418–419
Schizoid personality disorder School threat assessments, 371–372
purposeful, targeted, defensive School violence, 79, 359, 365, 371, 537–
violence and, 171 552
targeted, impulsive violence and, background of, 538
174 bullying, 540–541
Index ❘ 605
criteria for, 538 unique restraints, 352

current extent of, 538–540 disallowance of standing orders for,
homicides, 539–540 345
interventions for students at risk documentation of, 343, 345
for, 550, 551 of elderly persons, 395
legal restrictions on administrators’ emergency medication during, 340,
response to, 541 343, 351–352
prevention of, 549–550 emotional impact of, 349–350
risk factors for, 551 explaining purpose to patient, 347
school size and, 539 fluids for patient in, 349
shootings, 79, 359, 365, 371, 541– forensic aspects of, 353
549, 551 frequency of use, 340
at Columbine High School, 543– inclusion in hospital policy manual,
544 345
FBI report on myths about, 548 indications for, 342–343
at Jonesboro, Arkansas, 544–545 informing families about
managing threats or predictions institutional policies and
of, 548–549 practices regarding, 346
National Research Council initiation of, 343–345
report on, 549 by licensed independent
notable cases of, 541–543 practitioner, 344
operant styles of perpetrators of, instructors for, 346
547 JCAHO regulations for, 339–340,
profiles of perpetrators of, 547– 341, 344, 346, 351, 353
548 long-term use of, 345
social causation model of, 541 meals for patient in, 349
at Virginia Tech, 546–547 of minority patients, 38
suicides, 539 new governmental guidelines for
unreliability of data on, 537 use of, 341–342
School-Associated Violent Death observation and monitoring of
(SAVD) study, 539 patient during, 292, 342, 344,
Schuster v. Altenberg, 532 348–349
Seclusion and restraint, 339–354 as part of patient’s treatment plan,
of adolescents, 292 346
care of patient in, 349–350 patient positioning in restraint, 347–
case example of, 293–294 348
chemical restraint, 351 reasons for, 341
CMS requirements for, 341, 344– removal from, 350–351
345, 351 seclusion room design, 280, 350
committee to oversee use of, 353 self-injurious behaviors during, 350,
contraindications to, 291, 352 352
controversy about, 339 staff certification for, 345–346
danger and injury associated with, staff implementation of, 291, 292,
352–353 346–347
debriefing after use of, 293, 348 standard of care for, 339
definition of, 292 studies of, 340–341
devices for, 346 techniques for, 345–348
Seclusion and restraint (continued) self-hitting, 225

time parameters for, 292, 342, 344– self-poisoning, 225
345, 348 skin burning, 225, 330
toileting of patient during, 349 skin cutting, 224–225, 226
use in emergency settings, 13, 280, skin picking or scratching, 225
291–295, 340–341, 462–463 stereotypic, 225
of violent inpatients, 113–114, 259 substance abuse and, 226
Seizures while in seclusion, 350, 352
during benzodiazepine Serial killing, 38
withdrawal, 305 Serious violence, defined, 108
driving safety and, 522, 531, 533 Serotonin, 92–93, 215, 328
epilepsy and violence, 190–192 Sertraline, 384
poststroke, 189 for chronic aggression, 202
Selective serotonin reuptake inhibitors for paraphilias, 454
(SSRIs) for pedophilia, 445
for anger attacks in depressed for posttraumatic stress disorder,
outpatients, 254 131–132, 134
for chronic aggression, 202, 316 Seventy-two–hour observation bed,
for paraphilias, 453–454 psychiatric, 288–289
for pedophilia, 445 admission criteria for, 288
for posttraumatic stress disorder, patient consent to treatment during,
131–132, 134 288
Self-esteem patient rights and, 288
childhood verbal and emotional personnel who can place patient in,
abuse and, 79, 83 288
intimate partner violence and, 491, Severity of violence, 7
492 Sexual abuse in childhood, 87, 214, 545
self-injurious behaviors and, 226 among adult perpetrators of sexual
Self-injurious behaviors (SIBs), 224– violence, 442
226, 228. See also Deliberate self- convictions for, 443
harm; Suicide deliberate self-harm and, 87
among adolescents, 225–226 impulsive aggression and, 214
autism and, 222 incest perpetrators, 450, 451–452
case example of, 224–225 Internet as tool for, 443–444
childhood trauma exposure and, 214 case example of, 444–445
compulsive, 225 mood disorders related to, 79–80
definition of, 225 prevalence of, 442–443
diversity of, 225 programs for prevention of, 443
impulsivity and, 223, 225, 226 verbal and emotional abuse and
intellectual disability and, 220 risk for sexual victimization as
in Klüver-Bucy syndrome, 193 adult, 79, 82–83
major, 225 Sexual acting out, 7
in posttraumatic stress disorder, 124 Sexual aggression
prevalence of, 225–226 against children, 333, 441
psychiatric disorders associated definition of, 441
with, 226 by elderly persons, 382
psychological functions of, 226 against women, 441
Index ❘ 607
Sexual arousal/interest assessment Sexual sadism, 176, 441, 445–450

Abel Assessment of Sexual Interest, brain abnormalities and, 446
71–72 characteristics of, 446
penile plethysmography, 70–71, gender and, 446
416, 445 reliability of diagnosis of, 445–446
in perpetrators of sexual sadism sexually motivated homicide and,
and sexually motivated 442, 445–450
homicide, 449–450 child victims of, 448
phallometry, 449 measuring sexual arousal in
Sexual assault/rape, 237 perpetrators of, 449–450
alcohol and, 143 XYY syndrome in perpetrators
childhood sexual abuse and, 80, 82 of, 447–448
as crime of power and control, 448– Sexual violence, 441–456. See also
449 Sexual abuse in childhood; Sexual
date rape, 143 assault/rape
gender and, 415 definition of, 441–442
hormonal therapy for, 335 history of childhood sexual abuse
by intimate partner, 484, 488 among perpetrators of, 442
legal elements of rape, 415 intellectual disability and, 220
number of victims of, 415 Internet as tool for, 443–445
prevalence of, 143 recidivism risk for, 29, 416–417,
robbery preceding, 363 450–452
of stalking victims, 411 sexual sadism and sexually
typology of rapists, 415–416, 416 motivated homicide, 442, 445–
Sexual deviations. See Paraphilias 450
Sexual Fantasy Checklist, 445 suicide risk and, 89
Sexual offenders, 164, 442 treatment for perpetrators of, 452–
bipolar disorder among, 84 455
case example of, 327 women as perpetrators of, 80
on Internet, 444–445 Sexual violence risk assessment, 60,
recidivism risk for, 29, 416–417, 70–72
450–452 Abel Assessment of Sexual Interest,
factors associated with, 417, 451 71–72
Rapid Risk Assessment for of Internet child pornographers,
Sexual Offender 444–445
Recidivism, 29, 437 of known rapists, 415–417, 416
Static-99, 452 Pedophile Index, 445, 449
structured violence risk penile plethysmography, 70–71,
assessment of, 29 416, 445
victim age and, 451–452 in perpetrators of sexual sadism
treatment of, 452–455 and sexually motivated
cognitive-behavioral therapy, homicide, 449–450
452–453 Rapid Risk Assessment for Sexual
pharmacological, 453–455 Offender Recidivism, 29
relapse prevention, 453 for recidivism, 29, 417, 450–452
typology of rapists, 415–416, 416 structured, 29
Sexual rapists, 416 Sexually transmitted infections, 41
Sexually Violent Predator (SVP) Spousal/consortial murder-suicide,

commitment statutes, 70 422–423
Short-term violence risk amorous-jealous subtype of, 423
clinical assessment for, 3–14 declining-health subtype of, 423
psychological testing for psychiatric disorders and, 422
assessment of, 61 SSRIs. See Selective serotonin reuptake
time frame for, 4, 18 inhibitors
SIBs. See Self-injurious behaviors Stalking, 60, 247, 411–415
Single-photon emission computed assaults related to, 411
tomography, 400 circumstances of, 412
Situational threats, 472 clinical management of, 475
Sjögren’s syndrome, 198 of clinician, 474–475
Skilled nursing facilities, 396–397 danger posed by stalker, 413
Sleep apnea, driving safety and, 531 definition of, 474–475
“Sleep attacks,” driving safety and, 531 duration of, 411
Sleep deprivation, 198 elements of, 411
Sleep disturbances, 198 false allegations of, 412
driving safety and, 531 gender and, 411
pharmacotherapy for, 134 interventions to reduce impact of,
in posttraumatic stress disorder, 413–415
133–136 paranoid stalkers, 171, 179
Smoking prevalence of, 411, 475
cessation strategies for elderly psychological impact of, 475
persons, 399, 400, 403 restraining orders against stalkers,
fetal tobacco exposure, 213, 214 415
Social learning theory of intimate stalker typologies, 411–412, 413
partner violence, 486 according to relationship with
Social skills training, for students at victim and public-figure
risk for school violence, 550 context, 412, 414
Socioeconomic status, 11–12, 37, 38 threats and, 475
Sodium amytal, 303, 305 via electronic communication, 412–
Solvent exposure, 197 413
Somalis, culturally appropriate Standard Family Violence Index, 127
assessment of interpersonal State/Trait Anger Expression
violence in, 48–49 Inventory, 67
South African Truth and Reconciliation Static-99, 437, 445, 452
Commission, 82 Stimulant drugs. See also
Spirit possession, 45 Amphetamine; Cocaine
Spirituality and religious beliefs, 35 abuse of, 147
about subservience of wives, 35 agitation induced by, 9
as mitigating factors against driving safety and, 522
violence, 252 side effects of, 197–198
self-injurious behaviors and, 225 Strangulation, 446, 447
support of African American Stroke, 188–189, 196, 389
women through, 81 antipsychotics and, 403
Spousal abuse. See Intimate partner Structured Anchored Clinical
violence Judgment–Minimum, 437
Index ❘ 609
Structured Assessment of Violence psychotherapy, 154–155

Risk in Youth (SAVRY), 370 12-Step facilitation therapy,
Structured Guide for the Assessment 155
of Violence, 285 12-Step programs, 152, 155
Structured violence risk assessment, 3– alcohol, 7–9, 141–146
4, 17–31. See also Actuarial among assault victims, 80, 82
violence risk assessment cannabis, 85, 86, 107, 148
Classification of Violence Risk, 18, children of substance-abusing
23–26, 28, 28, 30 parents, 78–79
compared with clinical risk cocaine, 9, 13, 39, 147
assessment, 3–4, 17–21, 28 ethnicity and, 35
comparison of tools for, 28, 30 among female victims of intimate
components of, 20–21, 31 partner violence, 485, 488
combining risk factors, 21 methamphetamine, 147
generating final risk estimate, 21 opioids, 147–148
selecting and measuring risk phencyclidine, 9, 148
factors, 20 prevalence of violence and, 250
continuum of, 18 psychiatric comorbidity with, 148–
court and legislative openness to, 149, 301
29, 31 bipolar disorder, 84, 85, 251
HCR-20, 21–23, 28, 28–29, 30, 39, 46, intermittent explosive disorder,
240–241 217
predictive validity of, 30, 31 personality disorders, 167
psychological testing and, 59–60 posttraumatic stress disorder,
of sexual offenders, 29 129, 133
use in clinical practice, 27–29, 30–31, psychosis, 10
59 pyromania, 219
Violence Risk Appraisal Guide, 18, schizophrenia, 250
26–27, 28, 28–29, 30 self-injurious behaviors and, 226
Substance abuse disorders, 4, 7–9, 13, signs of intoxication, 6
22, 23, 39, 141–156. See also suicide and, 89, 92, 143–144, 294
Alcohol use/abuse; specific among college students, 143
substances of abuse risk factors for, 144
addiction, 149–155 workplace violence and, 510, 511
definition of, 149 youth violence and, 364, 369–370
as disease, 155 Substance Abuse Subtle Screening
identification of, 151–152 Inventory (SASSI), 69, 72
media dramatization of, 151 Suicidal gestures, 222
modafinil for patients with, 150 Suicide, 7, 12, 211, 222–224, 227. See also
neurobiology of, 149 Deliberate self-harm; Self-
treatment of, 152–155 injurious behaviors
in elderly persons, 399 antidepressants and risk for, 90, 92
evidence-based strategies for assessing potential for, 4, 6, 12
relapse prevention, 152 to prevent death of child, 424
patient–treatment matching assisted, 423
for, 155 borderline personality disorder
pharmacological, 152–154 and, 11
Suicide (continued) lethality of, 419

bullying and, 79 as predictor of eventual death by
childhood trauma and, 80, 214, 223 suicide, 89, 223
cocaine and, 147 prevalence of, 222–223
drinking age and, 143 psychiatric comorbidity with, 223
emergency room management of recurrence of, 223
suicidal patient, 294 vehicular crashes as, 522, 524–526
case example of, 294–295 “Suicide by cop,” 180, 420–421
gender and, 89, 223 behavioral clues to, 420–421
homicide and, 91–92, 328–329 characteristics of, 420, 421
clinician’s anxiety about therapy definition of, 420
with suicidal patients, 327– of mass murderer, 425
328 prevalence of, 420
hopelessness and, 87, 90 verbal clues to, 420
impulsive, 222–224 Supervision of therapy, 335
aggression as risk factor for, 224 Supportive therapy, 82
case example of, 222 Surgical history of elderly persons,
definition of, 223 386–388
methods of, 224 Suspiciousness
incidence of, 89 assaultiveness and, 107–108
intent for, 419 cocaine-induced, 9
intimate partner violence and, 80– in elderly persons, 381
81, 82, 485, 489 SVP (Sexually Violent Predator)
mercy, 423 commitment statutes, 70
mood disorders and, 85, 89–92, 223 Syncope
motive for, 419 driving safety and, 531
murder-suicide, 421–425 olanzapine-induced, 308
national rate of, 294 Systemic lupus erythematosus, 198
other deliberate self-harm and, 86,
92, 222 Tachycardia, olanzapine-induced, 308
past suicide attempts as predictor Tarasoff v. Regents of the University of
of, 89, 223 California, 246, 269, 328, 373, 430–
posttraumatic stress disorder and, 434, 493, 524, 532, 557–558
128–129 TBI. See Traumatic brain injury
psychological autopsy of, 419 Telephone interview for violence risk
in schools, 539 assessment, 465
serotonin level and, 92, 93, 215 Telephone threats, 473
sociodemographic risk factors for, Temper tantrums
89 in childhood, 216
substance abuse and, 89, 92, 143– intellectual disability and, 220
144, 294 Terrorism, 81, 501, 505
alcohol, 89, 146 Testosterone, 92
among college students, 143 anti-androgens for suppression of,
risk factors for, 144 453
Suicide attempts, 222–223 Test-taking style, assessment of, 63–64
definition of, 222 Thapar v. Zezulka, 433–434
impulsive, 223–224 Theft, in inpatient facilities, 115
Index ❘ 611
Therapeutic alliance Transference threats, 472

duty to warn and, 557 Traumatic bonding theory of intimate
inpatient assaults and, 260 partner violence, 487
Therapists. See Clinicians Traumatic brain injury (TBI), 187–188,
Thought insertion, 107 196
Threat assessment alcohol response and, 149
in schools, 371–372 case example of, 204
in workplace, 512–516 impulsivity, aggression and, 214–
crisis management teams for, 514 215
difference from clinical relation of postinjury behavior to
evaluations, 512–513 preexisting aggressive
disclosure of information from, tendencies, 215
513 Trazodone
goal of, 512–513 for aggressive elderly patients, 403
information used for, 513, 515 for sleep problems, 134
mental illness and, 513 Treatment interventions, 253–254
process of, 514–516 for elderly persons, 394–404
“Threat book,” 269 electroconvulsive therapy, 316
Threats for intimate partner violence, 491–
to clinicians, 335, 463–465, 471–475 492
(See also Violence toward involuntary hospitalization, 244–245
clinicians) involuntary outpatient treatment, 254
case example of, 464 lack of response to, 22
“cold,” 268–269 noncompliance with, 11
to inpatient staff, 268–269 for pedophilia, 445
to patients, 247–248 pharmacotherapy, 301–318
of school violence, 548–549 psychotherapy, 325–337
by stalkers, 475 racial differences in use of, 38–39
in workplace, 502–503 seclusion and restraint, 339–354
Time frame for validity of risk for violent inpatients, 113–118, 253–
assessment, 4, 18–19 254
Token economy, 201 for violent youth, 359–360, 372–374
Toluene exposure, 197 in schools, 550, 551
Topiramate, for aggression, 85 for workplace violence risk, 515–516
Tourette’s syndrome, 195–196 Tremors, 6
attention-deficit/hyperactivity Trichotillomania, 225
disorder and, 195 12-Step facilitation therapy, 155
self-injurious behaviors in, 225 12-Step programs for substance
Toxin exposure, 197 abusers, 152, 155
Transference Twenty-three–hour observation bed,
deranged, 331–332 psychiatric, 286
stalking due to, 475 Twin studies, 364
diluting of, 473
monitoring of, 472–473 Unabomber, 546
psychotic, violence to clinician due Unstructured violence risk assessment.
to, 464–465 See Clinical violence risk
racism and, 40 assessment
U.S. Secret Service profiling of school duty of clinicians to warn and

shooters, 547–548 protect potential victims, 246,
U.S. Surgeon General reports 247, 269, 333–334, 373, 430–434,
Mental Health: Culture, Race and 524, 557–559, 560
Ethnicity, 36 with intellectual disability, 220
Report on Youth Violence, 36 mental health professionals as, 248–
249, 332, 461–478
Vaillant, George E., 155 of patient with antisocial
Valproate, 12, 84–85, 203, 303, 312, 387 personality disorder, 11
for aggressive elderly patients, 403 patients as, 247–248
for outpatients with impulsive perpetrators’ lack of empathy for, 61
aggression, 317 psychiatric disorders of, 80
with risperidone or olanzapine, psychotherapy with, 333–334
314–315 psychotic identification with, 10
for violent youth, 374 Videotape recording of inpatient
van Gogh, Vincent, 88–89 assaults, 111–112, 262
Vasculitis, 198 Vietnamese, 35
Vehicular crashes, 521–535 culturally appropriate assessment of
dementia and, 527–531 interpersonal violence in, 50–51
depression and, 524–526 Vindictive rapists, 416
disturbances of consciousness and, Violence. See also Aggression; Agitation
531 Axis I diagnoses and, 250–252
intoxication and, 522 among children and adolescents,
legal duties of clinicians to prevent, 359–376
531–534 clinical inexperience with patients
medications and, 522–523, 532–533 exhibiting, 326–327
psychosis, schizophrenia and, 523– definition of, 237
524 dynamics of, 467
suicidal drivers, 522, 524–526 helping patients recognize origins
Venlafaxine, 284, 393 of, 329
Verbal abuse during childhood heterogeneity of persons exhibiting,
mood disorders related to, 79 325
and risk of adult sexual patients’ disavowal of tendencies
victimization, 79, 82–83 toward, 327
Veterans prodrome of, 467
depression in, 81 serious, 237
head injury and aggression among, vectors of, 328–329
215 Violence Risk Appraisal Guide
National Vietnam Veterans (VRAG), 18, 26–27, 28, 30
Readjustment Study, 125–126 combining risk factors for, 27
posttraumatic stress disorder in, 81, generating final risk estimate with, 27
123–137 (See also Posttraumatic predictive validity of, 26
stress disorder) Psychopathy Checklist–Revised
Victims and, 62
of bullying, 540–541 selecting and measuring risk factors
childhood abuse and, 79, 80 for, 26–27
of crime in schools, 539 use in clinical practice, 28, 29
Index ❘ 613
Violence risk assessment, 465–466, 466 preemptive practices, 555–556

actuarial, 3–4, 17–31, 46, 59 in schools, 550, 551
in adolescents, 365–372 in workplace, 515–516
assessing imminence of risk, 244– Violence toward clinicians, 461–478
245 additional management techniques
clinical, 3–14, 46 for, 468–469
components of, 20–21 affect management for, 463, 465,
continuum of structure for, 18 467–468
cultural competence in, 35–54 anticipation of, 466
exert testimony based on, 430, 434– assessing risk for, 465–466, 466
437 components of, 465
in inpatient settings, 259–272 scheduling appointment for, 466
in outpatient settings, 237–255, 465– setting for, 465–466
466 telephone interview, 465
in patient with past history of dynamics of, 467
violence, 243–244 in emergency settings, 462–463, 470
of patients with posttraumatic on inpatient unit, 470
stress disorder, 129–130, 130, limit setting and, 469–470
131 managing the armed patient, 476
psychological testing in, 59–73 office safety, 248–249, 332, 463–464,
reliability of assessments used for 465–466, 476–477
expert testimony, 437, 439 prevalence of, 461
for sexual violence, 60, 70–72 prodrome of, 467
structured, 3–4, 17–31 psychiatrist’s denial of risk for, 464,
in Texas criminal procedures, 438 465, 466
time frame for validity of, 4, 18–19 as result of psychotic transference
for violence against clinician, 465– in outpatient setting, 464–465
466, 466 threats of, 335, 463–465, 471–475
clinician denial of risk and, 466 case example of, 464
components of, 465 dynamics of, 471
scheduling appointment for, 466 forms of, 471
setting for, 465–466 management of, 473–474
telephone interview, 465 monitoring of, 472–473
in workplace, 512–516 settings for, 471
Violence risk management, 555–564 situational, 472
avoidant practices, 556 stalking, 474–475
case example of, 556–557 transference, 472
documentation, 562 Violent ideation, 6
duty to warn and protect Virginia Tech shootings, 546–547
potential victims, 246, 247, 269, Visuospatial abilities, assessing in
333–334, 373, 430–434, 524, elderly persons, 391
557–559, 560 Vitamin D, during anti-androgen
goal of, 555 therapy, 455
gun safety management, 6, 372, 373, von Economo’s disease, 192
557, 562, 563 Voyeurism, 441
patient discharge from inpatient VRAG. See Violence Risk Appraisal
facilities, 559–560, 561 Guide
War exposure homicide, 501, 505, 506, 509

mood disorders and, 81 legal risks associated with, 507–508
posttraumatic stress disorder in managing risk for, 515–516
combat veterans, 81, 123–137 methodological problems in studies
“Warn” assaults, 112 of, 508–509
Weapons checks, 46, 249, 332, 462 nonfatal assaults, 506
White matter disorders, 196 prevalence of, 505–507
Whitman, Charles, 546 profiles of perpetrators of, 509–510
WHO (World Health Organization), 144 psychiatric disorders and, 511
Wife abuse. See Intimate partner risk factors for, 508–512
violence individual, 510–511
Wilson’s disease, 195 organizational, 511–512
Women profiling approach to, 509–510
intimate partner violence static vs. dynamic, 512
perpetrated by, 491, 492 threat assessment in, 512–516
murder-suicide of children by, 423 threats of, 502–503
religious beliefs about subservience Types I to IV, 505–507
of wives, 35 types of, 501
sexual abuse perpetrated by, 80 workers’ compensation for injuries
sexual assault of arising from, 508
alcohol and, 143 wrongful accusations of, 504, 508
childhood sexual abuse and, 80 World Health Organization (WHO), 144
date rape, 143 Written threats, 473
prevalence of, 143
use of Psychopathy Checklist– XYY syndrome, 447–448
Revised in, 63
Youth violence. See Adolescents
as victims of intimate partner
violence, 41–42, 80–81, 483–495 Zar, 44, 45
alcohol-related, 82–83, 145–146 Ziprasidone
culturally appropriate for aggressive elderly patients, 401
assessment of, 47–53 intramuscular, 295, 303, 304, 306,
during pregnancy, 41, 484 308–309
treatment of, 82 contraindications to, 308
as victims of stalking, 411 dosage of, 304, 308
violence among, 253 effectiveness of, 311
Workplace violence, 501–516 indications for, 308
Americans with Disabilities Act QT interval prolongation
and, 508 induced by, 308
case example of, 502–504 transition to oral administration,
common-law negligence claims 311
related to, 508 use in renal disease, 308
definition of, 504–505 long-term treatment with, 312–314
disaster plan for, 507 Zolpidem, for sleep problems, 134

Violence Assessment and Management

Uploaded by

Copyright:

Available Formats

Violence Assessment and Management

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Violence Assessment and Management

Uploaded by

Copyright:

Available Formats

What are the terms of use for the book of the month?

What are the terms of use for the book of the month?

What information is included in the note section?

What information is included in the note section?

BOOK OF THE MONTH TERMS OF USE

PsychiatryOnline.com currently offers access in PDF format to

The Book of the Month is supplied as PDF; no other formats will be

Click here to continue on to the Book of the Month.

Library of Congress Cataloging-in-Publication Data

British Library Cataloguing in Publication Data

1 Clinical Risk Assessment of Violence. . . . . . . . . . . . . . . . . . . .3

2 Structured Risk Assessment of Violence . . . . . . . . . . . . . . 17

3 Cultural Competence in Violence Risk Assessment . . . . 35

4 Psychological Testing in Violence Risk Assessment . . . . 59

6 Schizophrenia and Delusional Disorder . . . . . . . . . . . . . . 105

7 Posttraumatic Stress Disorder . . . . . . . . . . . . . . . . . . . . . . .123

8 Substance Abuse Disorders. . . . . . . . . . . . . . . . . . . . . . . . . . 141

9 Personality Disorders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161

10 Neurological and Medical Disorders . . . . . . . . . . . . . . . . . 185

11 Impulsivity and Aggression . . . . . . . . . . . . . . . . . . . . . . . . . 211

12 Outpatient Settings. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237

13 Inpatient Settings. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .259

15 Psychopharmacology and Electroconvulsive

16 Psychotherapeutic Interventions . . . . . . . . . . . . . . . . . . . . 325

17 Seclusion and Restraint . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 339

18 Children and Adolescents . . . . . . . . . . . . . . . . . . . . . . . . . . . 359

19 The Elderly. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .381

20 Forensic Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 409

21 Legal Issues of Prediction, Protection,

23 Violence Toward Mental Health Professionals . . . . . . . . 461

24 Intimate Partner Violence and the Clinician. . . . . . . . . . 483

25 Workplace Violence and the Clinician. . . . . . . . . . . . . . . . 501

26 Vehicular Crashes and the Role of

27 School Violence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 537

28 Clinically-Based Risk Management of

Thomas Ala, M.D.

Karen E. Anderson, M.D.

Paul S. Appelbaum, M.D.

Peter Ash, M.D.

James C. Beck, M.D., Ph.D.

Carl C. Bell, M.D.

David M. Benedek, M.D., D.F.A.P.A.

John M.W. Bradford, M.B.Ch.B., D.P.M.

Rodney Burbach, M.D.

Leslie Citrome, M.D., M.P.H.

Martha L. Crowner, M.D.

Britt Darwin-Looney, J.D.

William R. Dubin, M.D.

Peggy L. El-Mallakh, Ph.D.

Rif S. El-Mallakh, M.D.

Paul Fedoroff, M.D.

Alan R. Felthous, M.D.

Philip Firestone, Ph.D.

Susan Hatters Friedman, M.D.

Thomas A. Grieger, M.D., D.F.A.P.A.

Anzalee Khan, M.S.

Gregory B. Leong, M.D.

Russell F. Lim, M.D.

Jean-Pierre Lindenmayer, M.D.

John R. Lion, M.D.

Name of patient Name of rater

Time incident began_ Duration ______