UM Book

THE NATIONAL ACADEMIES PRESS
This PDF is available at http://nap.edu/1359 SHARE

Controlling Costs and Changing Patient Care?: The Role of

Utilization Management
DETAILS
328 pages | 6 x 9 | HARDBACK

ISBN 978-0-309-04048-8 | DOI 10.17226/1359
CONTRIBUTORS
GET THIS BOOK Committee on Utilization Management by Third Parties, Institute of Medicine
FIND RELATED TITLES

Visit the National Academies Press at NAP.edu and login or register to get:
Access to free PDF downloads of thousands of scientic reports

10% off the price of print titles
Email or social media notications of new titles related to your interests
Special offers and discounts

Distribution, posting, or copying of this PDF is strictly prohibited without written permission of the National Academies Press.
(Request Permission) Unless otherwise indicated, all materials in this PDF are copyrighted by the National Academy of Sciences.
Copyright National Academy of Sciences. All rights reserved.

Controlling Costs and Changing Patient Care?: The Role of Utilization Management
Controlling Costs and

Changing Patient Care?
The Role of Utilization Management
Committee on Utilization Management

by Third Parties
Division of Health Care Services
INSTITUTE OF MEDICINE
Bradford H. Gray and Marilyn J. Field, editors
NATIONAL ACADEMY PRESS

Washington, D.C. 1989
ii
National Academy Press 2101 Constitution Avenue, N.W. Washington, D.C. 20418
NOTICE: The project that is the subject of this report was approved by the Governing Board of the
National Research Council, whose members are drawn from the councils of the National Academy
of Sciences, the National Academy of Engineering, and the Institute of Medicine. The members of
the committee responsible for the report were chosen for their special competences and with regard
for appropriate balance.
This report has been reviewed by a group other than the authors according to procedures
approved by a Report Review Committee consisting of members of the National Academy of Sci-
ences, the National Academy of Engineering, and the Institute of Medicine.
The Institute of Medicine was chartered in 1970 by the National Academy of Sciences to enlist
distinguished members of the appropriate professions in the examination of policy matters pertain-
ing to the health of the public. In this, the Institute acts under both the Academy's 1863 congres-
sional charter responsibility to be an adviser to the federal government and its own initiative in
identifying issues of medical care, research, and education.
This project was supported by the John A. Hartford Foundation of New York City, Grant No.
87343-H, and the Pension and Welfare Benefits Administration, U.S. Department of Labor, Con-
tract No. J-9-P-8-0067. The Honeywell Corporation provided funds for a roundtable meeting with
corporate benefits managers. The views presented are those of the Institute of Medicine Committee
on Utilization Management by Third Parties and are not necessarily those of the funding organiza-
tions. The Andrew W. Mellon Foundation contribution to independent Institute of Medicine funds
was also used to support the project.
Library of Congress Cataloging-in-Publication Data
Institute of Medicine (U.S.). Committee on Utilization Management by Third Parties.
Controlling costs and changing patient care? : the role of utilization management / Insti-
tute of Medicine, Division of Health Care Services, Committee on Utilization Manage-
ment by Third Parties ; Marilyn J. Field and Bradford H. Gray, editors.
p. cm.
Includes bibliographical references.
ISBN 0-309-04048-5.ISBN 0-309-04045-0 (pbk.)
1. Medical careUnited StatesCost control. I. Field, Marilyn Jane. II. Gray, Bradford
H., 1942- . III. Title.
RA410.53.I58 1989
338.4`33621/0973dc20 89-39638
CIP
No part of this book may be reproduced by any mechanical, photographic, or electronic process, or
in the form of a phonographic recording, nor may it be stored in a retrieval system, transmitted, or
otherwise copied for public or private use, without written permission from the publisher, except for
the purpose of official use by the U.S. government.
Printed in the United States of America

iii
COMMITTEE ON UTILIZATION MANAGEMENT BY

THIRD PARTIES
JEROME H. GROSSMAN,* Chairman, President, New England Medical

Center Hospitals, Boston, Massachusetts
HOWARD L. BAILIT,* Vice President for Health Research and Policy, Aetna
Life and Casualty, Hartford, Connecticut
ROBERT A. BERENSON, Washington, D.C.
JOHN M. BURNS, Vice President, Health Management, Honeywell, Inc.,
Minneapolis, Minnesota
RICHARD H. EGDAHL,* Director, Boston University Medical Center, Boston,
Massachusetts
JOHN M. EISENBERG,* Chief, General Internal Medicine, University of
Pennsylvania, Philadelphia, Pennsylvania
DEBORAH ANNE FREUND, Professor and Chair of the Health Care Faculty,
and Director, The Center for Health Services Research, School of Public
and Environmental Affairs, Indiana University, Indianapolis, Indiana
PAUL M. GERTMAN, President, ClinMan, Inc., Waltham, Massachusetts
ALICE G. GOSFIELD, Attorney, Alice G. Gosfield & Associates, P.C.,
Philadelphia, Pennsylvania
MICHAEL E. HERBERT, President, Physicians Health Services, Trumbull,
Connecticut
NATHAN HERSHEY,* Professor of Health Law, Graduate School of Public
Health, University of Pittsburgh, Pittsburgh, Pennsylvania
NEIL HOLLANDER, Vice President, Corporate Health Strategies, Blue Cross
of Western Pennsylvania, Pittsburgh, Pennsylvania
KAREN IGNANI, Associate Director, Department of Occupational Safety,
Health, and Social Security, AFL-CIO, Washington, D.C.
CAROL ANN LOCKHART, Executive Director, Greater Phoenix Affordable
Health Care, Phoenix, Arizona
ARNOLD MILSTEIN, President, National Medical Audit, San Francisco,
California
ALAN R. NELSON,* Associate, Memorial Medical Center, Salt Lake City, Utah
ROBERT PATRICELLI, President & Chief Executive Officer, Value Health,
Inc., Avon, Connecticut
CYNTHIA L. POLICH, President, InterStudy, Excelsior, Minnesota
* Institute of Medicine member

iv
DONALD M. STEINWACHS, Director, Health Services Research and

Development Center, School of Hygiene and Public Health, The Johns
Hopkins Medical Institutions, Baltimore, Maryland
BRUCE S. WOLFF, Partner, Proskauer Rose Goetz & Mendensohn, New
York, New York
STUDY STAFF
KARL D. YORDY, Director, Division of Health Care Services
BRADFORD H. GRAY, Study Director (through December 1988)
MARILYN J. FIELD, Study Director (through January 1989)
SUSAN E. SHERMAN, Research Associate
MARGARET WALKOVER, Research Associate
DON TILLER, Administrative Assistant
WALLACE K. WATERFALL, Editor, Institute of Medicine
EILEEN CONNOR, Consultant
SHARON ROSEN, Consultant

CONTENTS v
Contents
Preface vii
Executive Summary 1
1 Utilization Management: Introduction and Definitions 13
2 Origins of Utilization Management 28
3 The Utilization Management Industry: Structure and Process 58
4 Impact of Prior Review Programs 91
5 High-Cost Case Management 119
6 Conclusions and Recommendations 143
Acknowledgments 163
Appendixes
A Legal Implications of Utilization Review 169
William A. Helvestine
B Utilization Management and Quality Assurance in Health 205
Maintenance Organizations: An Operational Assessment
Joan B. Trauner and Sibyl Tilson
C Utilization Management in Peer Review Organizations 246

CONTENTS vi
D Summary of Public Hearings 250

E Summaries of Committee Site Visits to Utilization Manage- 253
ment Organizations
F Analysis of Agreements Between Utilization Management 282
Organizations and Their Clients
G Glossary and Acronyms 288
H Biographies of Committee Members 293
Index 301

PREFACE vii
Preface
The past decade has seen great growth in efforts by purchasers of health
care to understand and influence the treatment of patients. In large measure,
these efforts reflect purchasers' concerns that their increasing expenditures are
not matched by increasing value and even that a significant amount of care is
inappropriate and wasteful. Clinicians and researchers, too, are acutely
concerned about unexplained variations in practice patterns and lack of
evidence of treatment effectiveness. To respond to these concerns, we must
focus on how the health care system works as well as how individual patients
are served.
Utilization management brings patient-level and system-level concerns
together and represents a new nexus of relations among payers, practitioners,
hospitals, and patients. Because it is new, at least as broadly applied, and
because it is changing rapidly, utilization management needs to be watched.
This report is a preliminary effort in that direction, and the committee
hopes that it will inform private and public policymakers alike. Whether the
current organizational forms of utilization management remain or subside, the
function of managing utilization will remain a central challenge. Therefore, this
committee's conclusions and recommendations bear both on generic issues of
knowledge and values and on issues specific to current organizations and
procedures for influencing patient care decisions.
The first step in this study of utilization management by third parties was a
workshop held in the summer of 1987 to identify policy issues and research
questions, consider roles for the Institute of Medicine, and

PREFACE viii
discuss the organization and methods of utilization management. This workshop

provided initial direction and encouragement for the committee and staff, who
held seven meetings between December 1987 and February 1989. In addition to
soliciting presentations from knowledgeable persons at committee meetings, the
project sought views, data, and ideas through
12 site visits to organizations that provide utilization management services
(including freestanding firms, insurer subsidiaries, health maintenance
organizations, and peer review organizations);
a June 1988 hearing in which written and oral testimony was presented by
over 30 consumer, provider, industry, employer, and other organizations;
four papers commissioned by the committee from experts in different
aspects of utilization management (two of these papers appear in
Appendixes A and B of this report); and
review of published and, in some cases, unpublished literature on the goals,
processes, effects, and operating contexts of utilization management.
Additional information was obtained from ongoing discussions with
people involved in utilization management as purchasers of utilization
management services, suppliers of these services, and subjects of utilization
management review, that is, physicians, hospitals, and patients. Reflecting the
competitive nature of the utilization management industry, all of the site visits
and some other discussions required commitments that the committee would
keep confidential any information that would identify the source.
The committee on utilization management plans further work to evaluate
the continued course of utilization management and related means of improving
the appropriateness and cost-effectiveness of medical services. Other
committees and councils of the Institute of Medicine are studying strategies for
quality review and assurance, methods for monitoring and improving access for
the uninsured, priorities and processes for technology assessment, and problems
facing employer-sponsored health benefit plans. This and related work grows
out of a continuing commitment by the Institute of Medicine to an agenda of
evaluation that encompasses the cost, quality, and availability of health services.
JEROME H. GROSSMAN
CHAIRMAN, COMMITTEE ON UTILIZATION MANAGEMENT BY
THIRD PARTIES

EXECUTIVE SUMMARY 1
Executive Summary
With great speed and relatively little public awareness, a significant

change has occurred in the way some decisions are made about a patient's
medical care. Decisions that were once the exclusive province of the doctor and
patient now may be examined in advance by an external reviewersomeone
accountable to an employer, insurer, health maintenance organization (HMO),
or other entity responsible for all or most of the cost of the care. Depending
upon the circumstances, this outside party may be involved in discussions about
where care will occur, how treatment will be provided, and even whether some
treatments are appropriate at all.
Such "utilization management" is part of a complex balancing act created
by society's struggles with two important questions. First, how do we ensure
that people get needed medical care without spending so much that we
compromise other important social objectives? Second, how do we discourage
unnecessary and inappropriate medical services without jeopardizing necessary
high-quality care?
Experience indicates that these questions have no fixed answers. Rather,
we find a series of working hypotheses and partial solutions that are continually
revised, discarded, and even reinvented as changes occur in medical
technology, social values, economic conditions, and other circumstances. In this
preliminary report, the Committee on Utilization Management by Third Parties
of the Institute of Medicine examines one current working hypothesisthat
external review of the appropriateness of proposed medical services for
individual patients can improve the way care is provided and, as one
consequence, help constrain health benefit costs.

EXECUTIVE SUMMARY 2
The validity of this hypothesis is of interest to everyone involved with

health carepatients and potential patients, practitioners and others who
provide medical services, employers, unions, insurers, and makers of public
policy.
Utilization management has become a growth industry, spurred by
purchasers' search for control over rapidly escalating expenditures for health
care. One recent survey reported average cost increases from 1987 to 1988 of
14 percent for employers with insured health benefit plans and 25 percent for
employers with self-insured plans. In the private insurance sector, many
commercial insurers, Blue Cross and Blue Shield plans, and HMOs have seen
substantial losses, and some commercial insurers are withdrawing from the
group health insurance market.
To the dismay over rising health care costs has been added a growing
perception that a significant amount of medical care is unnecessary and
sometimes harmful. The studies that have contributed to this perception have
also produced some optimism that external review of physician practice
decisions could detect unnecessary care, influence physician behavior, and
reduce costs without jeopardizing access to needed services. Such review has
also appeared to offer an alternative to retrospective denials of claims for
benefits and across-the-board cutbacks in health plan coverage.
In this preliminary report, the Committee on Utilization Management by
Third Parties examines several questions.
How effective is utilization management in limiting utilization and
containing costs?
Are there unintended positive and negative consequences of bringing an
outside party into the process of making decisions on patient care?
Are utilization management organizations and purchasers sufficiently
accountable for their actions or are new forms of oversight, perhaps
government regulation, needed?
What are the responsibilities of health care providers and patients for the
appropriate use of health services?
The focus is on the private sector, in which two-thirds of the nonelderly
population are covered directly or as dependents under employer-sponsored
health plans. An estimated one-half to three-quarters of the individuals in these
plans are subject to utilization management.
CURRENT STATUS OF UTILIZATION MANAGEMENT

Early in its discussions the committee realized that the term utilization
management has no single, well-accepted definition. As with the labels cost
containment and managed care, different people may mean different things by
the term. In this report, the committee considers utilization management as a set
of techniques used by or on behalf of purchasers of health benefits

EXECUTIVE SUMMARY 3
to manage health care costs by influencing patient care decision-making

through case-by-case assessments of the appropriateness of care prior to its
provision.
The dominant utilization management strategy is prior review of proposed
medical or surgical services, which includes several related techniques such as
preadmission review, continued-stay review, and second surgical opinions.
Prior review provides advance evaluation of whether medical services planned
for a specific patient conform to provisions of health plans that limit coverage
to medically necessary care. Typically, all elective hospital admissions are
subject to such review before the patient enters the hospital, all emergency
admissions must be reviewed within a short period following admission, and the
need for continued hospital care is assessed periodically.
High-cost case management is a more focused strategy that concentrates
on the relatively few people in any group who have generated or are likely to
generate very high expenditures. The management process involves an
assessment of an individual's health care needs and personal circumstances to
determine whether extra assistance in planning, arranging, and coordinating a
specialized treatment plan will permit appropriate but less costly care. If the
individual's health plan does not cover some elements of the treatment plan,
individual exceptions to these coverage limitations may be approved.
Empirical evidence on the effects of utilization management is fairly
limited and suffers from a number of methodological weaknesses. Despite these
limitations, the committee believes that available evidence, taken together,
indicates that utilization management has had some impact on health care use
and costs. Specifically:
Utilization management has helped to reduce inpatient hospital use and to
limit inpatient costs for some purchasers beyond what could be expected
from other factors such as growth in outpatient resources, changes in
benefit plan design, and shifts in methods for paying hospitals. Employee
groups with higher initial levels of hospital use tend to show more change
than groups with lower initial hospital utilization.
The impact of utilization management on net benefit costs is less clear.
Savings on inpatient care have been partially offset by increased spending
for outpatient care and program administration. Some of this offset is an
expected and acceptable result of utilization management (and other
factors), and some is an unwanted consequence of moving care to
outpatient settings, where fewer controls on use and price now operate.
Although it probably has reduced the level of expenditures for some
purchasers, utilization managementlike most other cost containment
strategiesdoes not appear to have altered the long-term rate of

EXECUTIVE SUMMARY 4
increase in health care costs. Employers who saw a short-term moderation

in benefit expenditures are seeing a return to previous trends.
Systematic evidence about the impact of utilization management methods
on the quality of care and on patient and provider costs is virtually nonexistent.
Purchasers have not demanded such evidence, and researchers have found the
measurement of these effects even more costly, time-consuming, and uncertain
than the measurement of effects on purchasers' costs. During the course of this
study, the committee did not locate documented anecdotes or other information
to suggest that prior review programs in the private sector are jeopardizing
patient safety. However, the processes of prior review and associated changes in
courses of treatment may cause anxiety and inconvenience to some patients.
And utilization management does add to the administrative burdens on
practitioners and institutional providers and contributes to resentment about
reduced professional autonomy and satisfaction. More positively, the committee
has some confidence that high-cost case management is easing some financial
and emotional burdens on catastrophically ill patients and those who care for
them.
Several features of utilization management are important to keep in mind.
First, utilization management as it currently operates in the private sector is
highly variable, which makes generalizations difficult. Second, until recently
utilization management has focused on the site, duration, and timing of medical
care. The unnecessary use of the hospital, rather than the actual need for a
particular service, has been the main target. The primary strategy has been
discussion and negotiation about appropriate care. Refusals to certify benefits
appear uncommon, perhaps 1 to 2 percent of cases.
Third, utilization management in the private sector operates under few
explicit legal restrictions. There is, however, considerable awareness among
review organizations and major purchasers of the legal risks inherent in efforts
to influence patient care decisions and operationalize the terms of health benefit
plans. And there appears to be growing recognition of the conventionalbut
not infallibleprotections offered against liability by good management, good
judgment, good faith, and good documentation.
The lack of good research on the effectiveness and impact of utilization
management is a frequent theme in this report, which likewise notes that
research on the effectiveness of many medical procedures is also limited. As
utilization management expands its review of the actual need for specific
procedures, the clinical foundation for such assessments becomes more
important. Good research is a critical base for good utilization management.
Moreover, the research on feedback and education strategies to influence
physician decisions suggests that utilization management criteria will be

EXECUTIVE SUMMARY 5
more likely to win acceptance and change behavior if they are based on clear
clinical evidence from respected academic and professional sources.
HOW UTILIZATION MANAGEMENT IS EVOLVING

The continuing evolution of utilization management is most evident in its
scope and its operational efficiency. The reasons for these developments are
several. First, the initial savings from shifting the site and timing of care have
largely been realized, and the survival of review organizations may depend on
their continuing ability to affect benefit costs. Second, review organizations are
being influenced by researchers' beliefs that much care is still inappropriate and
unnecessary. Third, the administrative and other costs of review programs,
including physician dissatisfaction and employee confusion, make
simplification and efficiency important objectives.
Thus, based on survival instincts and evidence of continuing utilization
problems, the emphasis of utilization management is beginning to expand from
the site and duration of care to include the actual need for specific types of
inpatient and outpatient services. Again, the availability of sound clinical
criteria for assessing medical necessity is one constraint on this movement.
Legal concerns are another factor.
At this time, the committee does not see utilization management moving
toward intentional rationing of clinically necessary medical services. A decision
not to approve payment for an unnecessary service is not rationing per se.
However, the committee recognizes that there may be instances when review
nurses or physicians may apply implicit cost-effectiveness judgments. In high-
cost case management, such judgments may be explicit, but the intention is to
determine whether services normally excluded from a benefit plan should be
covered to permit less costly but still appropriate care for a particular patient.
With respect to administrative costs, frequently mentioned priorities
include greater computerization, expanded use of treatment protocols in high-
cost case management, and greater targeting of reviews to high-payoff
categories of problems and services. In some cases, gains in operational
efficiency should reduce administrative burdens on patients, physicians, and
institutional providers of care.
ISSUES FOR THE FUTURE

The committee has identified some shortcomings in utilization
management or gaps in the knowledge of it that raise concerns about patient
protection, particularly given the growing focus on the appropriateness of

EXECUTIVE SUMMARY 6
specific proposed procedures. If the positive potential of utilization

management for improving the cost-effective use of health resources is to be
encouraged, then the committee believes that several issues need attention.
Objective methods and resources for evaluating the impact of utilization
management on health care costs, use, and quality are limited and must be
improved. There needs to be more confidence about what works and what
does not and under what circumstances.
The review criteria used by those engaged in utilization management
(including hospitals and HMOs) should be available for outside scrutiny.
Physicians, purchasers, and patients should know the basis for judgments
about the site, timing, and need for care.
Systematic investigation of the effects of utilization management not only
on purchaser costs but also on patient and provider costs and attitudes
should be a higher priority. A more complete picture of costs and benefits
is needed.
The opportunities for patients and physicians to appeal review decisions
should be clearly described and free from unreasonable complexity, delay,
or other barriers. This is an essential protection for patients.
The explicit links between utilization management and quality assessment
and assurance mechanisms need to be clarified and strengthened. Review
organizations should have standard operating procedures for responding to
the quality problems they uncover.
RECOMMENDATIONS FOR THE NEAR TERM

The committee believes that utilization management has sufficient promise
that a number of short-term and long-term efforts should be made to promote its
positive potential and guard against its shortcomings. A prudent course in the
near term is for the parties involved in utilization managementpurchasers,
review organizations, physicians, and patients to accept greater responsibility
for the reasonable and fair conduct of utilization management and the
appropriate use of medical care.
Responsibilities of Employers and Purchasers

As financers of both utilization management and health services,
employers are in the best position to exert influence on the conduct of
utilization management. Although such an effort may be beyond the resources
of small employers, larger purchasers should investigate the operating
procedures and capabilities of the organization or organizations from which
they purchase review services. This review should include organizations such as
HMOs that provide prior review and high-cost case management as part of a
broader package of services. Purchasers can visit review organizations,

EXECUTIVE SUMMARY 7
request detailed information and references, and seek advice from business
coalitions, consultants, and other similar resources. Human resources staff
should be trained to respond to employee questions, assist with problems, and
handle grievances.
Employers should also examine other aspects of their health benefit plans
for impediments to the appropriate use of medical services or the rational
payment for these services. Moreover, workers must be clearly informed of
their responsibilities and rights. Also, although employers have the right and
responsibility to take vigorous actions to manage the costs of employee health
benefits, they should respect both the confidentiality of medical information
about employees and the primary obligation that physicians have to serve their
patients.
Responsibilities of Utilization Management Organizations

Any supplier of services has responsibilities to purchasers that are intrinsic
to the concept of a buyer-seller relationship. It is in the business interest of
review organizations to anticipate and respond to purchaser demands for
information about the organization, its services, and its results. Further, it is in
the legal interest of these organizations to manage their activities rationally, to
act in good faith, and to maintain careful records.
Although good business and legal judgment should dictate prudent
behavior, those who provide utilization management services also have a moral
obligation not to harm the patients whose medical care they review and
influence. Harm includes discouraging appropriate care and mishandling
confidential information. When organizations perform prior review and high-
cost case management for individually purchased insurance plans (with no
employer sponsorship), they have a particular responsibility to provide good
educational materials and appeals processes for beneficiaries who have no
employer or other sponsor to act as their agent and aid. They should also
develop guidelines for what to do when they discover quality of care problems.
With respect to practitioners and individual providers of care, good business
sense should dictate that review organizations encourage provider acceptance
and cooperation by
using sound clinical criteria that are open to examination,
involving the medical community in criteria development,
minimizing the administrative burdens placed on hospitals and physicians,
and
clarifying and simplifying processes for appealing negative judgments.
The committee is aware that further steps, in particular, making clinical
criteria available, raise difficulties given the competitive environment of

EXECUTIVE SUMMARY 8
benefit plan administration. Organizations that have invested their own

resources in developing criteria will be reluctant, on the one hand, to make them
available to competitors with less initiative and, on the other hand, to reveal
some details to practitioners and institutions for fear of their ''gaming the
criteria'' by providing misleading information. Though these are reasonable
concerns, on balance, they are outweighed by the need to move toward open
criteria and standards.
Responsibilities of Practitioners and Institutions

The committee found the responsibilities of physicians and other health
care providers in utilization management the most troublesome to analyze and
define, a situation typical of many current ethical and policy issues in health
care today. On five basic issues, the committee agreed that health care
practitioners and institutions are responsible for
cooperating with the reasonable efforts of payers, including utilization
management, to ensure that payments are for appropriate care within the
terms of a patient's benefit plan;
constructively challenging unreasonable utilization management programs
and specific decisions that threaten patient safety or damage patient privacy;
informing patients about treatment options, risks, and benefits and then
considering their preferences;
seeking to ensure that patients get needed services, which may mean
locating an alternative source of care if the patient cannot pay and the
provider cannot give free treatment; and
staying current with scientific literature on the necessity and effectiveness
of medical services in their areas of practice.
Although some difficult situations with insurers and review organizations
may be more conveniently and quickly dealt with in the short term by
misrepresenting patient symptoms, diagnoses, or treatments, the committee
believes that it is in the patient's, physician's, and society's interest over the long
term for physicians to deal honestly with reviewers and claims administrators
and to challenge questionable criteria, procedures, and decisions directly.
Manipulation and evasion can have serious risks. Specifically, incorrect
information may enter the patient's medical records or insurance history with
later negative consequences. Moreover, perceptions by purchasers that
physicians are gaming the system undermine professional credibility and
stimulate the sorts of auditing, second-guessing, and external oversight to which
practitioners object.

EXECUTIVE SUMMARY 9
Responsibilities of Patients
In many respects, patients and potential patients are the weakest strand in
the web of responsibilities for the appropriate use of medical services. When ill,
individuals may not be able to act in an informed and prudent way. And
whether well or ill, individuals may find both their benefit plans and their
medical care difficult to understand and evaluate. Nonetheless, health plan
members should try to understand their responsibilities under the plan. The
challenge for those involved in health care delivery and financing is to help all
kinds of patients make informed decisions about getting or not getting care.
LONGER-TERM RECOMMENDATIONS AND QUESTIONS

As noted earlier, the committee views utilization management as,
essentially, a working hypothesisone of several partial and overlapping
strategies for balancing health care expenditures, access, and quality. When
knowledge advances, economic conditions change, and social values shift, these
partial strategies are revised, integrated, and sometimes discarded. However,
even if some of the techniques now employed by utilization management
organizations are abandoned or the organizations themselves change,
improvements in the criteria for judging appropriate care and for monitoring the
provision of care will continue to be relevant. The longer-term
recommendations of the committee focus on the foundations for improving
effective and safe decision-making about patient care: knowledge development,
knowledge utilization, and value clarification. The recommendations cover
three areas:
research on the effectiveness of medical services and the effectiveness of
utilization management and related techniques,
formulation and dissemination of guidelines for medical practice and
criteria for utilization review, and
possible oversight of utilization management.
Research On Effectiveness
Utilization management can be no better than the clinical evidence and
expertise on which it is based. Although review organizations today may not be
effectively using all available research, they are still constrained by the large
areas of undocumented impact and clinical uncertainty involving many major
medical procedures.
Policymakers are increasingly recognizing that the free market system is
unlikely to invest sufficiently in outcomes research and data collection because
those making the investment cannot capture all the benefit but

EXECUTIVE SUMMARY 10
must share it with those who have not invested. Since the public gains from
investments in such research, public financing and priority setting are
appropriate, although they should add to rather than replace initiatives being
undertaken by private researchers, health care organizations, and others.
The creation of an agenda to strengthen knowledge of what is effective in
medical care is well under way, and the Institute of Medicine is actively
involved with many other private and public organizations in developing and
implementing this agenda. It is, however, important to have realistic
expectations.
In the first place, there are practical and ethical limits on clinical
effectiveness researchtoo few researchers, long time horizons, and numerous
procedures where clinicians would balk at research protocols that require
withholding treatments generally thought to be useful or providing treatments
generally thought to be inappropriate. Second, much care does not really focus
on the effectiveness of care in an average setting or population, nor does it
evaluate the impact of care on quality of life and many other outcomes that
society now considers important. Third, effectiveness research that relies on
existing claims and other records, although less expensive and time-consuming
than most clinical trials, is not quick or suitable for many questions. Fourth,
research cannot resolve some questions, for example, whether use of a specific
procedure for a specific problem is prudent given other uses to which limited
financial resources could (and would) be put. Fifth, sound clinical research does
not automatically affect behavior.
The research on what works in medical care should be complemented by
research on how to ensure that such knowledge is used effectively and
efficiently. Such programmatic research is, for the most part, a low priority
today. It is expensive, methodologically troublesome, and slow to pay off. As
part of the overall strategy for containing total health care costs and improving
the appropriateness of health care for all citizens, the committee urges federal
and private consideration of carefully targeted research projects to test prior
review and case management strategies and build methodologies for
documenting the effects of ongoing programs.
Practice Guidelines and Review Criteria

Translating effectiveness research into valid, reliable, and usable
guidelines for medical practice and utilization management is a complicated
undertaking. The committee has identified a number of questions about the
process of developing, disseminating, and updating practice guidelines that
need attention. They include the following:
Should there be some kind of oversight of guidelines developed by different
sourcesa sort of quality review mechanism that assesses both

the method and the substance of specific guidelines? What should happen
when different sources develop conflicting guidelines?
To what extent should patient preferences or cost-effectiveness analyses be
considered in the development of practice guidelines? How should these
issues of value be dealt with in the application of guidelines or in other
strategies?
Should adherence to guidelines provide physicians with protection against
malpractice charges? Over the long term, should a role for community or
local standards continue?
What considerations should apply in the translation of guidelines into
criteria used in prospective or retrospective review programs?
These questions are relevant to much of the Institute of Medicine's work.
Further exploration of these issues is under way and will draw on the expertise
of the Committee on Utilization Management by Third Parties, the Committee
to Design a Strategy for Quality Review and Assurance in Medicare, the
Council on Health Care Technology, and other parties inside and outside the
Institute.
Oversight of Utilization Management

The protections offered by caveat emptor, self-regulation, and tort liability,
although important, do not respond to all concerns about the impact of
utilization management on patients, providers, and overall health care costs.
The committee believes that incompetent review organizations need to be
weeded out and that some form of oversight seems advisable. However,
premature or misguided regulation to accomplish this could stifle worthwhile
innovations, lock in ineffective methods, or so paralyze utilization management
that purchasers abandon it for more onerous methods of controlling their costs.
The experience of the federal government in overseeing peer review
organizations (PROs) shows how difficult the oversight function is.
At this time, the committee feels that neither it nor other parties are in a
good position to make sound specific recommendations about oversight for
utilization management. The committee has posed several questions that it
intends to explore further.
How do we decide whether oversight is necessary and feasible?
If an oversight mechanism is necessary and feasible, should it be public or
private?
What should be its focus?
Should utilization management conducted by different kinds of
organizations, for example, HMOs, be subject to different kinds of
oversight?

Can anything be learned from government oversight of PROs or the

accreditation process managed by the Joint Commission on the
Accreditation of Health Care Organizations?
These are not simple questions. Answering them demands more
information and more thoughtful debate over how to judge the strengths and
weaknesses of utilization management versus those of other strategies to control
costs and influence patient care decisions. This, in turn, will depend on better
evidence about the impacts of different cost containment strategies. Finally,
recommendations about oversight will require more deliberation about the
legitimate but sometimes conflicting needs, interests, and values of the parties
involved in utilization management.
In a broader sense, the limits of utilization management and any other
single strategy, even any combination of strategies, need to be recognized. The
issue is not whether utilization management does everything that needs to be
done but whether it produces desirable results in reasonable ways at an
acceptable cost. Is it, on balance, better to use it than to discard it? How can it
be improved, and what other strategies are needed to supplement it? This report
provides a progress report and some preliminary views on these issues.
The Institute of Medicine will continue its efforts to better define what role
utilization management might play in helping society find an acceptable balance
of efficiency, access, and appropriateness in health care. This clearly must be a
shared venture. Fortunately, the quest to know what is useful and how to apply
it universally are now central issues in medical research and public policy.

UTILIZATION MANAGEMENT: INTRODUCTION AND DEFINITIONS 13
1
Utilization Management: Introduction and
Definitions
Prior to having the cholecystectomy recommended by her physician, Greta

Harrison calls an 800 telephone number to notify the organization that does
utilization management for her employer. That organization gets in touch with
the surgeon's office to discuss various aspects of the care that is proposed for
her. Is hospitalization necessary or can the surgery be done as an ambulatory
procedure? How long will the patient need to be in the hospital? In this case,
the reviewer agrees that inpatient care is clearly appropriate but questions the
plan to admit the patient two days prior to surgery. Since the patient lives in
the same town as the hospital and can easily have preoperative tests performed
on an outpatient basis, the surgeon agrees to admit her on the day of the surgery.
After Michael Travers is admitted to the hospital following a myocardial
infarction, the hospitalaware of his benefit plan's requirementsnotifies the
appropriate utilization management organization. The length of stay is
discussed, but no explicit target date for discharge is set. However, the hospital
is then called every third day by the organization, which evaluates information
about the patient's need for further hospitalization. The calls continue until Mr.
Travers, who has a difficult recovery, is improved enough to be discharged to
his home. The physician has not had to adjust the treatment plan but feels
irritated at the "red tape" involved. And Mr. Travers has worried on some
occasions that payment for part of his hospital stay might be denied.
With their daughter depending on a ventilator to breathe and receiving other
hospital care for muscular dystrophy, the parents of Patty Simon are contacted
by a case manager for the insurance company that covers the family. The
question is whether they and their physician would like to explore
arrangements for home care, which is possible in this case but considerably
more complex than usual. With the parents' and physician's cooperation, the
case manager works out a plan for transfer that includes assessment of the
home's wiring (which is adequate for the equipment), provision for two shifts
of home nursing care every day, and purchase of appropriate medical
equipment and supplies. This requires some expenditures not normally covered
by the benefit plan, but the employer agrees with the insurer to make an
exception in this case because the arrangements will not only be less costly
than hospital care but will also improve the quality of life for the family.

With great rapidity and relatively little public awareness, a significant

change has taken place in the way some decisions are made about a patient's
medical care. Many decisions like those just described, once the exclusive
province of the doctor and patient, now have to be examined in advance by an
external reviewer, someone who is accountable to an employer, insurer, health
maintenance organization (HMO), preferred provider organization (PPO), or
other entity responsible for paying all or most of the cost of the care. Depending
upon the circumstances, this outside party may be involved in discussions about
whether a service is needed, how treatment will be provided, and where care
will occur.
This preliminary Institute of Medicine (IOM) report describes the nature of
this change in medical decision-making and assesses its impact on patients,
providers, and purchasers of medical services. It focuses on the utilization
management efforts of the private sector, which provides health benefits for
most Americans under age 65.1
Prior review of proposed medical care is not entirely new in the 1980s.
Review organizations for Medicare were performing some preadmission review
in the 1970s, and some private payers made limited use of the technique even
earlier. However, widespread application of this approach to managing health
care utilization is a phenomenon of the 1980s.
A survey conducted in 1983 reported that only 14 percent of corporate
benefit plans required prior approval of nonemergency admissions to hospitals
(Equitable Life Assurance Society of the United States, 1983). By 1988, another
survey found 95 of 100 large firms had such programs (Corporate Health
Strategies, 1988). Perhaps half to three-quarters of employees nationwide are
now covered by such programs, up from only 5 percent in 1984 (Foster
Higgins, 1987; Gabel et al., 1988).
What accounts for this rapid spread of utilization management through
external assessments of the need for proposed medical services? The most
obvious factor is rapidly rising health care costs. Purchasers' search for effective
ways to limit their financial liability for health benefits stems directly from their
belief that costs are out of control.
The trends responsible for this view are painfully familiar to everyone
concerned with health care financing. In 1987, the latest year for which
statistics are available, total spending on health care reached an estimated $500
billion, up from $234 billion just 5 years earlier (Levit and Freeland, 1988).
This spending has been increasing at a rate considerably above the rate of
general inflation (Table 1-1), and the share of the gross national product
attributed to health services went from 5.9 percent in 1965 to 11.1
1 Public programs have been the subject of several reports in recent years (for
example, General Accounting Office, 1983, 1988a, 1988b; Health Care Financing
Administration, 1979; Physician Payment Review Commission, 1988, 1989, and Project
HOPE, 1987).
TABLE 1-1 Consumer Price Index in the United States (Annual Average, 1967 = 100.0)
Item 1960 1967 1975 1980 1985 1986
All 88.7 100.0 161.2 246.8 322.2 328.4
Medical care 79.1 100.0 168.6 265.9 403.1 433.5
NOTES: Some data are revised. Data for 1960 are based on urban wage and clerical workers.
Data for other years are for all urban consumers.
SOURCE: Health Insurance Association of America, Sourcebook of Health Insurance Data (1987,
1988). Based on Bureau of Labor Statistics, U.S. Department of Labor, CPI Detailed Report,
various issues.
percent in 1987. Spending for health care by business as a percentage of

the gross private domestic product grew from 1.1 percent in 1965 to 3.4 percent
in 1987 (Levit et al., 1989).
High health care costs for employers have been cited as one factor
impairing American competitiveness in world markets and a reason why many
small firms do not provide health benefits for workers. In 1987, spending for
health care by business equaled about 6 percent of total labor compensation
compared with about 2 percent in 1965 (Figure 1-1) (Levit et al., 1989). A
recent survey of nearly 800 employers of all sizes reported average premium
increases from 1987 to 1988 of 11 percent for conventional insurance plans and
between 8 and 10 percent for HMOs (Gabel et al., 1989). Another survey cited
average increases from 1987 to 1988 of 14 percent for employers with insured
programs and 25 percent for employers with self-insured programs (Foster
Higgins, 1989). Companies that self-insure assume all or most of the financial
risk of paying for covered medical services used by employees and their
dependents instead of paying an outside insurance to accept that risk. In the
private insurance sector, many commercial insurers, Blue Cross and Blue Shield
plans, and HMOs have seen significant underwriting losses$3.6 billion for
commercial carriers and $1.1 billion for Blue Cross and Blue Shield plans in
1988 (Donahue, 1989). Some commercial insurers, for example, Kemper,
Provident Mutual, Allstate (for large groups only), and Transamerica
Occidental, are withdrawing from the group health insurance market (Meyer
and Page, 1988).
To the dismay over rising health care costs has been added a growing
perception that much medical care is unnecessary and sometimes harmful. The
studies that have contributed to this perception have also produced some
optimism that external review of physician practice decisions could detect
unnecessary care, influence physician behavior, and reduce costs without
jeopardizing access to needed services (Eisenberg, 1986; Schwartz,

Figure 1-1 Expenditures by private industry for health services and supplies as
a percent of total labor compensation, 1965-1987.
Source: Levit et al. (1989, p. 9).
1984; Wennberg, 1984; Wennberg et al., 1977). In addition, experience
has suggested that review of some care prospectivelyprior to its provision
would be more palatable and effective than retrospective review has been. This
set of perceptions and expectations is, in essence, the hypothesis of utilization
management, a hypothesis of interest to patients, practitioners, purchasers, and
policymakers.
The IOM Committee on Utilization Management by Third Parties has
examined the utilization management hypothesis by asking several questions.
How effective is utilization management in limiting utilization and
containing costs?
Are there unintended positive and negative consequences of bringing an
outside party into the process of making decisions about patient care?
Are utilization management organizations and purchasers sufficiently
accountable for their actions, or are new forms of oversight, perhaps
government regulation, needed?

What are the responsibilities of health care providers and patients for the
appropriate use of health services?
The committee's investigatory approach has been described in the preface.
Chapters 2 through 5 discuss the committee's findings about why utilization
management has become so widespread, how utilization management actually
operates and appears to be evolving, and what is known about its effects. In
Chapter 6, the committee assesses the current status of utilization management,
including its strengths and shortcomings, and recommends near-term and
longer-range actions that could help utilization management realize its
objectives of controlling costs and reducing inappropriate services without
undermining patient access to needed care.
WHAT IS UTILIZATION MANAGEMENT?

In its study of utilization management, the committee found that the term
has no single, well-accepted definition. As with the labels cost containment and
managed care, different people may mean different things by the same term. In
this report, the committee considers utilization management as a set of
techniques used by or on behalf of purchasers of health care benefits to manage
health care costs by influencing patient care decision-making through case-by-
case assessments of the appropriateness of care prior to its provision.
Three points about the committee's focus are worth underscoring. First, the
committee examines methods that rely on case-by-case assessments of care.
Second, the focus is on review prior to the provision of services. Third, this
report stresses actions taken to reduce costs for third-party purchasers of care.
The first characteristic distinguishes utilization management from methods that
analyze aggregate utilization patterns to identify potential problems or that rely
on across-the-board limits on health care benefits that take no account of
individual patient characteristics. The second characteristic differentiates
utilization management from the retrospective review of claims or medical
records submitted after care has been provided. The third characteristic directs
attention to purchaser-sponsoredrather than provider-sponsoredutilization
management efforts, except when providers explicitly share the financial risk
with purchasers of care, as they do in HMOs.
The dominant utilization management strategy is prior review of proposed
medical services, which includes several related techniques. A second, more
focused, strategy is high-cost case management (see Table 1-2).
Prior Review
Prior review provides advance evaluation of whether medical services
proposed for a specific person conform to provisions of health plans that

TABLE 1-2 Utilization Management Methods

Prior Review Techniques
Preadmission Review
Admission Review
Continued-Stay Review
Discharge Planning
Second Opinion
High-Cost Case Management
limit coverage to medically necessary care.2 Most prior review programs

include an integrated set of review steps, not all of which will apply to any
single patient. The focus may be on the site of care, the timing or duration of
care, or the need for a specific procedure or other service.
The first point of assessment, often called preadmission review, may occur
before an elective hospital admission. This is what Greta Harrison and her
physician experienced in one of the vignettes that opened this chapter. In this
case, the review did not challenge the need for the procedure itself or the need
for hospital care, but it did challenge the proposed admission 2 days before
surgery. The terms preservice review and preprocedure review are sometimes
used to indicate that the focus of review is the need for a procedure, regardless
of whether it is to be performed on an inpatient or an outpatient basis.
For emergency or urgent admissions to the hospital when prior review is
not reasonable or feasible, admission review may be required within 24 to 72
hours after hospitalization to check the appropriateness of the admission as
early as possible. The vignette describing Mr. Travers involved this technique
as well as continued-stay review or concurrent review, which assesses the
length of stay for both urgent and nonurgent admissions. Reviewers may press
for timely discharge planning by hospital staff and, in some instances, assist in
identifying and arranging appropriate alternatives to inpatient care.
In addition, a patient may be required to get a second opinion on the need
for certain proposed treatments from a practitioner other than the patient's
physician. Increasingly, preadmission review or preservice review is used to
screen patients so that referrals for second opinions are focused on patients for
whom the clinical indications for a service are dubious.
To encourage patients covered by a health plan to cooperate in the
2 Medical necessity is another term that is used differently by different people in
different contexts. Some use it generally to cover assessments of the site and duration of
care as well as the clinical need for a particular procedure, whereas others use it only in
the latter sense. Those who use the term more restrictively tend to apply the term
appropriateness to the former assessments. For a discussion of legal interpretations of
medical necessity, see the paper by William A. Helvestine in Appendix A of this report.
prior review process, a financial penalty, such as higher cost-sharing, may apply
when individuals fail to obtain necessary certifications. Chapter 3 provides
more details about the mechanisms of prior review.
Although terms like prior review, predetermination, precertification, and
prior authorization of benefits are often used interchangeably, the approval of
benefits in advance of service provision may be contingent rather than final. For
example, if a retrospective claims review suggests that the information on
which the predetermination was based was seriously flawed, payment of a claim
may be denied upon further investigation. Or if a utilization management firm
does not have access to the details of the benefit plan for a group, it might
authorize services not covered by the contract. A review of claims prior to
payment might then result in denial of benefits. Since this latter practice usually
makes patients unhappy, many utilization management firms try to consider
restrictions in a client's health plan in their determinations. Retrospective
denials of claims following prior certification appear to be rare, as are refusals
to preauthorize services.

High-cost case managementalso called large case management, medical
case management, catastrophic case management, or individual benefits
managementfocuses on the relatively few beneficiaries in any group who
have generated or are likely to generate very high expenditures. This small
percentage of individualsperhaps 1 to 7 percent of a groupmay account for
30 to 60 percent of the group's total costs. For the United States as a whole in
1980, 1 percent of the population accounted for 29 percent of total health care
spending (Berk et al., 1988).
Case management for individuals with high-cost illnesses is similar to
other forms of social and health case management, in that it involves assessing a
person's needs and personal circumstances and then planning, arranging, and
coordinating the recommended services. It differs in its targets, those very
expensive cases for which specialized attention may encourage appropriate but
less costly alternative forms of treatment.
In contrast to prior review programs, high-cost case management programs
are usually voluntary, with no penalties for patient failure to become involved
in the process or comply with its recommendations. (In the third vignette, Patty
Simon's parents could have refused the alternative course of care suggested for
her.) In addition, more effort is generally devoted to reviewing the patient's
particular condition and circumstances and exploring, even arranging,
alternative modes of treatment. Finally, exceptions to limitations in benefit
contracts may be authorized in advance if this will permit appropriate but less
expensive care. For instance, additional home nursing benefits may be arranged
so that an individual can avoid further

hospitalization. In unusual cases, benefits may be provided for other than health
care services, such as construction of a wheelchair ramp or rewiring a patient's
home, if these expenditures will allow home care or self-care to be substituted
for institutional services at a lower total cost. (The assessment of the wiring in
Patty Simon's home would have been covered in this fashion.)
Retrospective Utilization Review

Utilization management techniques, particularly prior review methods,
attempt to overcome the disadvantages and unhappiness associated with
retrospective review and denial of claims after services have already been
provided. Retrospective claims and medical record reviews can, however,
support and reinforce utilization management by
monitoring the accuracy of information provided during prior review and
identifying problem areas,
examining claims that are unsuitable for predetermination (generally those
with high volume and low unit costs), and
analyzing patterns of practitioner or institutional care for use in provider
education programs and selective contracting arrangements.
Retrospective utilization review methods have a longer history of general
application than do prospective methods (Blum et al., 1977; Congressional
Budget Office, 1979, 1981; Institute of Medicine, 1976; Law, 1974). Its
strengths and weaknesses have been scrutinized in a number of studies before
this one and are not explicitly considered in this report. However, constraints on
retrospective review have been a key stimulus for the development of prior
review methods. Many of the concerns raised by the committee about the
clinical soundness of review criteria, the fairness of procedures, and other
matters described apply to both prospective and retrospective reviews.
Other Cost-Containment Methods

The techniques of prior review and high-cost case management are but a
subset of the cost-containment methods that can influence decisions about
patient care. Other methods, some of which are discussed in Chapter 2 and
Appendix B, include the following:
benefit design (including patient cost-sharing and coverage exclusions),
consumer education, and other approaches that shape patient demand for
care;
financial incentives (for example, capitation or bonuses) that are designed
to reward physicians or institutions for providing less costly care;

contracts with health care practitioners and institutions that establish limits
on payment for care provided to health plan enrollees;
use of gatekeeping, triaging, and other devices to manage patient flow to
specialists and expensive services; and
physician education and feedback on standards of care and patterns of
practice.
Utilization management shares with the last four strategies a recognition of
the physician's central role as the player-manager of the health care team who is
responsible for organizing and directing the production process and providing
some of the productive input (Eisenberg, 1986). The different strategies for
influencing decisions about patient care, however, vary in their emphasis or
reliance on different models of control (such as professional self-regulation,
informed consumerism, or prudent purchasing), their techniques of influence
(such as education, financial incentives, peer pressure, or external oversight),
and the parties involved (that is, patients, primary care practitioners, or
specialists).
As will be described in Chapter 2, different strategies for cost containment
have been tried, abandoned, and revived as third-party financing of health care
has expanded. This history reflects both the difficulties of the task and an
appreciation that there is no single solution to problems of health care costs,
quality, or access. Many strategies have a place, each of which has different
strengths and weaknesses and each of which needs monitoring and adjustment
as circumstances change and people adapt to various attempts to shape their
behavior.
TWO NOTES OF CAUTION
Obstacles To Evaluation
This report laments the limited evidence on utilization management and
calls repeatedly for more and better assessments. Nonetheless, the committee is
well aware that sound evaluation of utilization management programs faces
several obstacles. Some are intrinsic to the research problem, some reflect
common organizational behaviors, and some involve particular pressures faced
by market-driven organizations. Rigorous evaluation also tends to be quite
expensive. In Appendix B of this report, the commissioned paper by Joan B.
Trauner notes that evidence about the impact of physician financial incentives
on patient care decisions and quality of care is also quite limited.
Intrinsic Conceptual and Methodological Problems

A number of problems in evaluating utilization management and other cost-
containment programs are predictable difficulties faced, to one degree

or another, in much social and evaluation research (Eddy and Billings, 1988;
Wennberg, 1987). One such problem is that there are no uniformly accepted and
applied rules for measuring health care utilization or adjusting data for
differences in the characteristics of groups being compared. Other
methodological difficulties involve (1) data quality and availability; (2)
definitions and measurements of program characteristics, group characteristics,
outcomes, and other variables; (3) projections of what would have happened
without the interventions; and (4) generalizations to other programs and settings.
Common Behavioral Biases Against Evaluation

Under this heading come obstacles to systematic evaluation that are typical
of organizations whether they be public or private, for-profit or not-for-profit,
big or small (Eddy and Billings, 1988; Hatry et al., 1973; March and Simon,
1958; Suchman, 1967). They include preferences for
action over evaluation, for example, developing, selling, and running a
program rather than seeing if it works;
quick payoff rather than long-term products or results;
easy rather than difficult actions (for example, using data on inputs and
procedures that are simpler to collect rather than data on outputs or
outcomes);
compelling anecdotes, consensus, or tradition over careful and complex
analyses; and
positive rather than negative results.
In addition, faced with limited resources, managers are frequently reluctant
to allocate funds for evaluation instead of wages and benefits, shareholder
dividends, or other activities. The committee has no information about what
utilization management firms spend on evaluation (for internal use or for
clients) or how much different employers invest in systematically assessing the
impact of prior review or other cost-containment strategies.3
Competition and Evaluation

The normal individual and organizational biases against systematic
evaluation may be both mitigated and intensified in competitive environments.
3 The private sector is not alone in providing meager resources for program
evaluation. The utilization and quality review components of Medicare's peer review
organization (PRO) program have not been very rigorously examined (General
Accounting Office, 1988a; Physician Payment Review Commission, 1988). The Health
Care Financing Administration does have performance standards for PROs, but they tend
to emphasize process rather than outcome and tend to involve measures of impact that
are more appropriate for ongoing monitoring rather than systematic evaluation of the
review techniques.
Certainly, competition can be a powerful stimulus for internal evaluation of

how well a product is working and what makes it work better. Also, clients of
utilization management organizations have a strong interest in obtaining reports
on results and in shifting their business to other firms if they cannot get such
reports.
Balanced against these forces are several threats posed by evaluation. Most
obviously, an evaluation may be negative and thereby reduce a firm's chances
for retaining clients or winning new clients.4 Moreover, when an evaluation is
publicly available, a firm's competitors gain information that could help them
build a case to inform potential clients that the competitor could provide better
results or, at least, better reports. Further, evaluations of utilization management
programs may provide competitors with statistical norms or even provider-
specific information that would not be readily available to them otherwise.
Likewise, if firms that invest in relatively sophisticated research and
development reveal their work, they may give a free ride for competitors to
copy or build on the resulting review criteria, analytic methodologies, or other
products. In a new and rapidly evolving industry, this can seem a significant
issue for more experienced organizations.
Forces Behind Rising Health Care Costs

The Committee on Utilization Management by Third Parties also
recognizes that the forces behind rising health care costs are exceptionally
strong and difficult to constrain through moderate means. Many believe that, for
the foreseeable future, health care costs will continue to increase faster than
costs in the rest of the economy.
Clinical judgments about the value of treatment for various categories of
patients are changing as new treatments or new evidence of treatment
impact emerges. For example, women who underwent mastectomy for
breast cancer and had no evidence that the cancer had spread were until
recently not expected to benefit from chemotherapy, but some new
analyses suggest such treatment does increase survival rates. It also
increases initial treatment costs (Early Breast Cancer Trialists'
Collaborative Group, 1988). Recent guidelines for the use of
mammography screening could greatly expand the amount of such
screening but some professional sources question whether the guidelines
are clinically warranted (McIlrath, 1989).
4 Even when the reported results were positive, the committee encountered
considerable reluctance by review organizations to have their analyses published.

New tests may reduce diagnostic uncertainty but not add any information
that aids in treatment decision-making (Kassirer, 1989). Advances in
screening techniques may catch individuals much earlier in the course of
disease and reduce the numbers who will receive later expensive
treatments. The question is, will the costs of screening and early treatment
offset the savings? Will real survival rates increase? Researchers involved
with cancer point to methods under development to screen for very early
traces of dozens of different kinds of cancer, not all of which are more
successfully treated if they are detected earlier.
The work force and the general population are aging, and the use of both
acute-care and long-term-care services is higher for people in the older age
groups.
Between 1980 and 2000, the number of physicians has been projected to
increase from 171 to 260 per 100,000 population (Graduate Medical
Education National Advisory Committee, 1981; U.S. Department of Health
and Human Services, 1985). Whether this will bring a surplus of physicians
is a matter for debate (Ginsburg, 1989; Schwartz et al., 1989). Nonetheless,
one estimate, now many years out of date, is that every additional physician
results in $400,000 in additional yearly expenditures for medical services.
The concern about the millions of Americans who have no routine health
insurance coverage is generating various proposals to protect these
individuals through, for example, state-sponsored insurance pools,
mandated employer-based insurance, expansions of Medicaid, and
universal federal health insurance (Congressional Research Service, 1988).
What are the short-term costs (and for whom) of increasing access? What
long-term costs and benefits can be expected?
Reducing increases in health care costs such that they are much closer to
the level of general inflation would appear to demand radical changes in
American health policy, either major restructuring of the financing and delivery
systems or major cutbacks through large shifts in costs to patients, severe
limitations on patients' choices of hospitals and physicians, and explicit
rationing of some technologies for all or some individuals. Society may not be
willing to make such changes, particularly in the short run (Curran, 1987). It
may continue the search, described in the next chapter, for more moderate
strategies to control health care expenditures. Utilization management is one
such strategy.
It is an unfortunate reality, however, that most cost-containment strategies
eventually disappoint their supporters and evaluators to some degree. Even
when these strategies seem to reduce costs initially, trend projections do not
appear to show an appreciably lower increase in total costs over the longer term
(Prospective Payment Assessment Commission, 1989). Given

the effort and optimism it generally takes to commit a corporation or a

government to a new program, it is not surprising that excessively high
expectations often give way eventually to disillusionment. Unwarranted or
excessive negativism can, in turn, be counterproductive and lead to premature
abandonment of modest but still helpful strategies.
Cognizant of these hazards, the Committee on Utilization Management by
Third Parties has tried to approach its initial evaluation of utilization
management with reasonable expectations. To this end, the committee has
reviewed the development of third-party financing of health care in the United
States and the ways in which various strategies to manage costs have evolved.
The next chapter summarizes this review.
REFERENCES
Berk, Marc L, Monheit, Alan C., and Hagan, Michael M., ''How the U.S. Spent Its Health Care
Dollar, 1929-1980,'' Health Affairs, Fall 1988, pp. 46-60.
Blum, John D., Gertman, Paul M., and Rabinow, Jean, PSROs and the Law, Germantown, MD:
Aspen Systems Corporation, 1977.
Congressional Budget Office, The Effects of PSROs on Health Care Costs: Current Findings and
Future Evaluations, Washington, DC, 1981.
Congressional Budget Office, The Impact of PSROs on Health Care Costs: Update of CBO's 1979
Evaluation, Washington, DC, 1981.
Congressional Research Service, Health Insurance and the Uninsured. Background Data and
Analysis, Washington, DC, June 9, 1988.
Corporate Health Strategies, The Health Poll, Fall 1988, p. 1.
Curran, William J., The Health Care Cost-Containment Movement: A Reconsideration, Report of a
conference sponsored by Medicine in the Public Interest, Baltimore, MD, March 16 and
17, 1987.
Donahue, Richard, "Health Premiums Soared Past the $100 Billion Mark in 1988," National
Underwriter, June 5, 1989, p. 1.
Early Breast Cancer Trialists' Collaborative Group, "Effects of Adjuvant Tamoxifen and of
Cytotoxic Therapy on Mortality in Early Breast Cancer," New England Journal of
Medicine, December 29, 1988, pp. 1681-1692.
Eddy, David M., and Billings, John, "The Quality of Medical Evidence and Medical Practice,"
Paper prepared for the National Leadership Conference on Health Care, Washington, DC,
1988.
Eisenberg, John M., Doctors' Decisions and the Cost of Medical Care, Ann Arbor, MI: Health
Administration Press, 1986.
Equitable Life Assurance Society of the United States, The Equitable Healthcare Survey: Options
for Controlling Costs, conducted by Lou Harris and Associates, Inc., New York, August
1983.
Foster Higgins, Health Care Benefits Survey, New York, 1987.
Gabel, Jon, DiCarlo, Steven, Fink, Steven, and de Lissovoy, Gregory, "Employer-Sponsored Health
Insurance in America," Research Bulletin of the Health Insurance Association of America,
Washington, DC, January 1989.
Gabel, Jon, Jajich-Toth, Cindy, de Lissovoy, Oregory, and Cohen, Howard, "The Changing World
of Group Health Insurance," Health Affairs, Summer 1988, pp. 48-65.
General Accounting Office, Improving Medicare and Medicaid Systems to Control Payment for
Unnecessary Physicians' Services, GAO/HRD-83-16, Washington, DC, February 8, 1983.

General Accounting Office, Medicare: Improving Quality of Care Assessment and Assurance, GAO/
PEMD-88-10, Washington, DC, May 1988a.
General Accounting Office, Medicare PROs: Extreme Variation in Organizational Structure and
Activities, GAO/PEMD-89-7FS, Washington, DC, November 1988b.
Ginzberg, Eli, "Physician Supply in the Year 2000," Health Affairs, Summer 1989, pp. 84-90.
Graduate Medical Education National Advisory Committee, Report to the Secretary, Vol. 1, DHEW
Publication No. HRA 81-6510, Washington, DC, 1981.
Hatry, Harry P., Winnier, Richard E., and Risk, Donald M., Practical Program Evaluation for State
and Local Governments, Washington, DC, The Urban Institute, 1973.
Health Care Financing Administration, Professional Standards Review Organizations: 1978
Program Evaluation, Washington, DC, U.S. Department of Health, Education, and
Welfare, 1979, pp. 34-37.
Health Insurance Association of America, 1986-1987 Source Book of Health Insurance Data,
Washington, DC, 1987.
Health Insurance Association of America, 1988 Update: Source Book of Health Insurance Data,
Institute of Medicine, Assessing Quality in Health Care, Washington, DC, National Academy of
Sciences, 1976.
Kassirer, Jerome P., "Our Stubborn Quest for Diagnostic Certainty," New England Journal of
Medicine, June 1, 1989, pp. 1489-1491.
Law, Sylvia, Blue Cross: What Went Wrong?, New Haven, CT, Yale University Press, 1974.
Letsch, Suzanne W., Levit, Katherine R., and Waldo, Daniel R., "National Health Expenditures,
1987," Health Care Financing Review, Winter 1988, pp. 109-122.
Levit, Katherine R., and Freeland, Mark S., "National Medical Care Spending," Health Affairs,
Winter 1988, pp. 124-136.
Levit, Katherine R., Freeland, Mark S., and Waldo, Daniel, "Health Spending and Ability to Pay,"
Health Care Financing Review, Spring 1989, pp. 1-12.
March, James G., and Simon, Herbert A., Organizations, New York, John Wiley & Sons, Inc., 1958.
McIlrath, Sharon, "11 Groups Endorse Mammogram Guidelines," American Medical News, July 14,
1989, pp. 3, 35.
Meyer, Harris, and Page, Leigh, "New Era in Utilization Review," American Medical News,
December 9, 1988, pp. 1, 45.
Physician Payment Review Commission, Annual Report to Congress, Washington, DC, March 1988.
Physician Payment Review Commission, Annual Report to Congress, Washington, DC, April 1989.
Project HOPE, A Study of the Preadmission Review Process, Report prepared for the Prospective
Payment Assessment Commission, Chevy Chase, MD, November 1987.
Prospective Payment Assessment Commission, Medicare Prospective Payment and the American
Health Care System: Report to Congress, Washington, DC, June 1989.
Schwartz, J. Sanford, "The Role of Professional Medical Societies in Reducing Variations," Health
Affairs, Summer 1984, pp. 90-101.
Schwartz, William B., Sloan, Frank A., and Mendelson, Daniel N., "Debating the Supply of
Physicians: The Authors Respond," Health Affairs , Summer 1989, pp. 91-95.
Suchman, Edward A., Evaluative Research, New York, Russell Sage Foundation, 1967.
U.S. Department of Health and Human Services, Projections of Physician Supply in the U.S.,
ODAM Report No. 3-85, Washington, DC, Bureau of Health Professions, 1985.
Waldo, Daniel R., Levit, Katherine R., and Lazenby, Helen, "National Health Expenditures," Health
Care Financing Review, Fall 1986, pp. 1-21.
Wennberg, John E., Blowers, Lewis, Parker, Robert and Gittelshon, Alan M., "Changes in
Tonsillectomy Rates Associated with Feedback and Review," Pediatrics, June 1977, pp.
821-826.
Wennberg, John E., "Dealing with Medical Practice Variations: A Proposal for Action," Health
Wennberg, John E., "Use of Claims Data Systems to Evaluate Health Care Outcomes," Journal of
the American Medical Association, February 20, 1987, pp. 933-936.

ORIGINS OF UTILIZATION MANAGEMENT 27
2
Origins of Utilization Management
In introducing this report, the committee has emphasized the seriousness of

the health care cost problem and the challenges involved in constraining
expenditures. The next step is to put the problems and strategies in a historical
context. Without understanding how or why utilization management and other
cost-containment strategies have developed, policymakers may oversimplify
both the difficulties and opportunities they face in managing health care
benefits. This chapter
reviews the development of third-party financing of health services in the
United States;
summarizes early efforts to control costs for health benefits;
describes the growth of government and private employer involvement in
cost containment; and
identifies some factors behind the change in societal attitudes about medical
care and the acceptance of external assessments of the appropriateness of
health care use.
THE GROWTH OF THIRD-PARTY FINANCING OF HEALTH

CARE
Extensive third-party financing of individual health services is relatively
new, emerging in this country largely in the past 60 years. Its development
reflects more than a century of scientific progress in the diagnosis and treatment
of illness. This progress has stimulated substantial growth and change in health
care institutions and professions, raised public expectations about

the benefits of medical care, and increased the costs of that care. These
developments, in turn, have led to broad-based demands that individuals and
families be protected against medical expenses by means other than charity.
Although third-party payment was and is among the causes of escalating health
care costs, it is also a crucially important social invention to deal with that
problem. Some key dates in the move to third-party payment are summarized in
Table 2-1. Table 2-2 shows the shift in funding sources for selected medical
care expenses over the last 60 years.
Before the 1930s, few Americans had anything resembling modern health
benefit plans (Anderson, 1968, 1972; Somers and Somers, 1961). Concerns
about medical costs were defined largely in personal rather than governmental
or corporate terms. For example, the Committee on the Costs of Medical Care
(CCMC) reported in the early 1930s that less than 15 percent of American
families bore the burden of more than half of all annual family expenditures for
illness (Anderson, 1968). For individuals, medical care expenses were highly
unpredictable, ranging at that time "from five dollars to one thousand dollars or
more" for a single illness (Rorem, 1982, p. 62). When patients and families
(first parties) could not pay these costs, health care providers (second parties)
absorbed them with varying amounts of assistance from other organizations
(third parties) such as local governments, religious groups, and private charities.1
Among actuaries and others involved with the commercial life and
casualty insurance industry that had developed in the nineteenth century, there
was considerable doubt that medical care was an insurable risk (Donabedian,
1976). A classic text on insurance (McIntyre, 1962) describes traditional
conditions for insuring a hazard or risk. The insured event (1) must be
susceptible to unambiguous description, (2) must not be something the insured
person wants or can control, and (3) must be a relatively uncommon occurrence
for individuals but have a predictable incidence for a group. Medical care
presents problems in all three areas.
Nevertheless, in the early 1930s, a number of individuals, influenced by
the analyses of the CCMC, began a virtual social movement to organize and
promote new kinds of "third-party" financing for health carealthough they did
not use the term explicitly (Anderson, 1968, 1975; Rorem, 1982; Somers and
Somers, 1961). They believed that medical expensesat least, hospitalization
expensesfor a group of people could be projected with some accuracy so that
a group could do what the individual could not:
1 In 1961, Somers and Somers wrote: "The term 'third party,' usage of which varies
from a technical concept of contract law to a popular epithet, means... an organization or

institution involved in the financing or provision of medical care, other than the two
primary parties: the vendor doctor or hospitaland the patient. For most practical
purposes the third party is a private insurance carrier, prepayment plan, or government
agency." Today, employers should be added to that list.
TABLE 2-1 Key Dates in the Move from First-Party to Third-Party Payment
1798 U.S. Maritime Hospital Service established; deductions from seamen's pay
cover part of cost
1847 First company to issue health insurance organized in Boston
1847 American Medical Association organized
1863 Travelers Insurance Company offers accident insurance
1870s Employee benefit associations formed; offer small death and disability
payments
1883 Germany passes broad social insurance laws
1898 American Hospital Association organized
1910s Physician service and industrial health plans established in Northwest and
remote areas
1912 First model law developed for regulating health insurance
1915 Referenda to establish compulsory state health insurance defeated
1917 American College of Surgeons sets standards for hospitals
1927 Committee on the Costs of Medical Care established
1929 Stock market crash followed by Depression
1929 Ross-Loos group founded (first prepaid group practice)
1929 Baylor hospitalization plan founded (first Blue Cross plan)
1934 Roosevelt puts low priority on public health insurance during planning of
Social Security legislation
1937 Blue Cross Commission established
1939 Federal Security Agency (predecessor of U.S. Department of Health and
Human Services) established
1940 Predecessor of Group Health Association of America founded
1940s Federal wage freeze increases union interest in fringe benefits
1945 Kaiser Foundation Health Plan opens to non-Kaiser groups
1946 Hill-Burton hospital construction program established
1946 Blue Shield Commission and Health Insurance Council organized
1949 Supreme Court decisions allow employee benefits to be part of collective
bargaining
1951 Joint Commission on the Accreditation of Hospitals founded
1954 San Joaquin County Foundation for Medical Care established
1965 Medicare and Medicaid legislation
1971-1974 Economic Stabilization Program (wage and price controls)
1972 Professional standards review organization legislation
1972 John Deere & Co. begins to self-fund health benefits
1973 Wennberg article on small area variations published in Science
1973 HMO legislation passed
1974 Employee Retirement Income Security Act passed
1974 Washington Business Group on Health organized (predecessor)
1978 Labor-Management Group paper on health care costs
1978 General Motors cost-containment reports initiated
1983 Medicare Prospective Hospital Payment legislation passed
SOURCES: Health Insurance Association of America (1987); Somers and Somers (1961); Starr
(1982); U.S. Department of Health, Education, and Welfare (1976); and Wilson and Neuhauser
(1974).

TABLE 2-2 Expenditures for Personal Health Care, Hospital Care, and Physician
Services, by Major Sources of Funds, in Percentages, 1929-1987
Year Type of Care Direct Patient Private Insurance Government
1929 All personal health care 88.4 NA 9.0
(PHC)
1950 All PHC 65.5 9.1 22.4
Hospital care 29.9 17.7 48.9
Physician services 83.2 11.4 5.2
1960 All PHC 54.9 21.1 21.8
1970 All PHC 40.0 24.1 34.4
1980 All PHC 28.7 30.7 39.4
1987 All PHC 27.8 31.4 39.6
SOURCE: Adapted from Gibson (1980) and Letsch (1988). Government expenditures include both
direct services and public insurance. Figures do not add to 100%.
budget the costs of medical care and share the risk of expense among the
well and the unwell.
Real growth in health insurance began during the crisis of the Depression.
The federal government rejected health insurance as a priority in developing
Social Security legislation, but in the private sector communitywide hospital
benefit plans began to be organized for employed groups. These plans collected
monthly per-employee payments (premiums) that were independent of
individual episodes of ill health. Early premiums ran about 50 to 75 cents per
month per member.
Once private health insurance had a chance to prove itself, it quickly
became regarded as a necessity and enrollments grew rapidly (Figure 2-1). By
the end of World War II, more than 30 million people had private hospital
insurance, and employment-based insurance was becoming the norm in major
companies.2 Also, as early as 1940, the movement for prepaid group practices
(PGPs) had helped organize enough PGPs, such as Group Health Association of
Washington, D.C., to warrant establishment of
2 Although some predecessors of Blue Shield plans, the physician service bureaus in
the Northwest, predate the Depression, insurance for physicians services grew more
slowly than did coverage for hospital services. In part, this reflected the lesser expense of
such services compared with that of institutional services. It also reflected resistance
from physicians, many of whom agreed with one leader's 1939 quote from George
Washington: "He who would surrender liberty for security is likely to lose both"
(Fishbein cited in Rorem, 1982, p. 94).
Figure 2-1 Private health insurance enrollment. Percentage of civilian

population with hospital expense, surgical expense, "regular medical," and
"major medical" coverage, 1940-1959. Data are based on end-of-year
population and enrollment figures. Sources: Somers and Somers (1961). Based
on data from the Health Insurance Council and the Social Security
Administration of the U.S. Department of Health, Education, and Welfare.
the trade association that eventually became the Group Health Association
of America (Somers and Somers, 1961). Insurance coverage continued to
expand during the next quarter century, reaching a peak of over 188 million
people in 1982 (Health Insurance Association of America, 1987, 1988).
Recently, after 50 years of growth, the reach of private health insurance
has begun to decline. AS shown in Table 2-3, the percentage of the nonelderly
population covered by employment-based plans dropped from 67.4 percent in
1979 to 64.8 percent in 1986 (Congressional Research Service, 1988). In the
same period, the number of individuals with neither public nor private insurance
expanded to an estimated 37 million in 1986

TABLE 2-3 Sources of Health Insurance Coverage for Nonelderly Population, 1979,
1983, and 1986
Percentage of Nonelderly Population
Coverage Source 1979 1983 1986
Employment-based plans
Covered on own job 33.1 32.5 33.4
Covered through someone else 34.3 32.1 31.4
Total employment-based 67.4 64.6 64.8
Other planspa 17.9 18.5 17.7
Uninsured 14.6 16.9 17.5
Total 100.0 100.0 100.0
a Excludes persons covered by employment-based plans.
SOURCE: Congressional Research Service (1988, p. 181).
(Chollet, 1988; Congressional Research Service, 1988). Those people

represent 17.5 percent of the civilian population, up from 14.6 percent in 1982.
(A recent revised estimate puts the number of uninsured at 31.1 million in 1987
[Rich, 1989].) Some individuals move in and out of coverage as they take
seasonal or other periodic employment, but many of the uninsured are full-time
workers and their dependents. In 1986, 15 percent of workers reported no health
insurance from any source (Monheit and Schur, 1988).
Although a variety of economic and social conditions lie behind the
decline in insurance coverage, high costs for health care certainly contribute to
it. In fact, the continuing escalation of health care costs is once again prompting
questions about the insurability of medical care and the viability of private
health benefit plans (Abramowitz, 1989; Freudenheim, 1989; National Health
Policy Forum, 1988).
EARLY COST-MANAGEMENT EFFORTS BY THIRD PARTIES

The transfer of many health care costs from individuals to third parties has
been accompanied both by a shift and a recasting of the cost problem. Insurance
has added new complexities to the problem, along with new resources for
managing it. Table 2-4, which shows changes in the general consumer price
index and selected medical care items from 1929 to 1960, reveals a sharp
postwar increase in hospital rates compared to physician fees.
Between 1920 and 1965, many of the basic elements of today's strategies
for managing health benefit costs were identified, even if they were not
persuasively articulated or successfully applied. These include

TABLE 2-4 Consumer Price Indexes for All Items, All Services, and Medical Care Items,
Selected Years, 1929-1960 (Average Annual Index, 1947-1949 = 100.0)
Item or Service 1929 1935 1940 1950 1955 June 1960
All items 73.3 58.7 59.9 102.8 114.5 126.5
Medical care 73.5 71.4 72.7 106.0 128.0 156.1
General practitioners' fees 73.9 74.7 104.0 124.3 147.5
Surgeons' fees - 73.8 74.0 104.5 116.4 129.3
Hospital rates 47.1 50.4 114.6 164.4 222.7
Hospitalization insurance a 115.5 174.3
NOTE: Values are annual averages, except for those in June 1960.
a Index for December 1952 is 100.
SOURCE: Adapted from Somers and Somers (1961, p. 545). Data from consumer price index,
Bureau of Labor Statistics.
management of the risk pool,

design of the benefit plan,
controls on payments to health care providers,
constraints on the supply of health care resources, and
review of the appropriateness of utilization.
Management of the Risk Pool

The founders of health insurance plans in the 1920s and 1930s had to abide
by a fundamental principle of all insurance: The composition of the risk pool is
critical to the cost and survival of a plan. If the people who buy health insurance
are disproportionately those who expect high expenses for health care, then
insurance will be, at best, a form of group budgeting for the ill minus the critical
feature of risk-sharing with healthy individuals.
The choice of the employment group as the foundation for private health
insurance was a key element in managing the risk pool. The employee group
was attractive because it existed for reasons other than the purchase of insurance.
Many of the early nonprofit health insurance plans were also committed to
"community rating." That is, they charged the same amount per individual
based on the projected expenditures for all those covered in the community.
However, "experience rating" soon emerged as a way for insurers to compete by
segmenting risk. Insurers could attract business by offering lower premiums to
groups with lower health cost experience or lower projected costs. This left
smaller employers, groups with older and less healthy workers, and self-
insuring individuals to pay higher premiums.

To make it feasible to sell insurance directly to individuals, most insurers

have devised various controls. These include medical underwriting (that is,
excluding those with conditions such as cancer or charging them higher rates)
and waiting periods for coverage of preexisting medical problems. Such
controls are designed to limit or compensate for the disproportionate selection
of an insurance product by less healthy individuals.3 Even so, individually
purchased insurance was and is more expensive than comparable group
insurance.
Recently, with many employers offering several health plans to their
employees, the management of risk within and across these plans has become a
major issue. The long-term impact of this new fragmentation of the risk pool on
benefit costs and on the stability of health plans is the subject of spirited
controversy (Scheffler and Rossiter, 1985). Efforts to create sound risk pools
for small groups, the self-employed, the fragmented employer group, and others
remain an elusive goal for those trying to extend health benefits to the nonpoor
uninsured.
Design of the Benefit Plan

Like management of the risk pool, the centrality of benefit design was also
quickly appreciated as a vehicle to control health plan costs. One way to limit
expenses is to require patients to bear some of the cost of care themselves
through deductibles, copayments (a flat payment per service), coinsurance (a
percentage of a charge), and dollar maximums on benefits for all services or
specific categories of service. Cost-sharing has two objectivesfirst, the simple
transfer of some liability for costs to the patient and, second, the discouraging
of patient demand for care.
Plan administrators also concluded that premiums could be held in check
by excluding coverage for experimental and ineffective treatments, for
treatments whose use was highly discretionary or difficult to monitor, for
extended or custodial care for chronic conditions, and for relatively low-cost
services that could be scheduled and budgeted. Early benefit plans often limited
coverage to selected hospital services. These coverage limitations applied
uniformly to the covered group and did not require patient-by-patient judgments
of medical appropriateness. Later, as insurers
3 The difference between individual and group coverage risk pools is illustrated by
statistics from a recent report issued by Independence Blue Cross and Pennsylvania Blue
Shield (1988). The Blue Cross plan offers individual coverage without medical
underwriting on an open enrollment basis throughout the calendar year for residents of
its five-county service area. In 1987, individual subscribers were 6 years older on
average (44 versus 38 years), had a hospital admission rate that was 46 percent higher
than that for all group subscribers, and incurred costs that were 60 percent higher. Fifty-
five percent of the individual subscribers were age 50 or over compared with 35 percent
of the group subscribers.
and employers became more comfortable with sponsoring health plans,

coverage began to include nonhospital services.
Controls On Payments To Providers

During the financially uncertain years of the 1930s, contracting and risk-
sharing with providers were important economic elements of prepaid group
practice arrangements and some health insurance plans. One of the earliest
prepaid group practices, the Ross-Loos Medical Group, was started in 1929 and
later followed by the Group Health Association of Washington, D.C., the Kaiser
Foundation Health Plan (as a community-based organization), and the Health
Insurance Plan of Greater New York. Also in 1929, what is considered the first
Blue Cross plan was established (for a public teachers group) and
''underwritten'' by Baylor University Hospital. Later Blue Cross plans also had
provisions for some sharing of risk by their contracting hospitals or physicians,
and negotiated limits on fees were elements in many plans (Donabedian, 1976;
Hellinger, 1976).
Strong contractual relationships that included some risk-sharing or limits
on payments to providers, however, were hard to establish and maintain
(Werlin, 1973). The expansionary postwar decades stimulated hospital
restiveness with the contractual relationship that guaranteed service to Blue
Cross enrollees at a negotiated price (Anderson, 1975). Physicians continued to
fight prepaid group practice plans and other forms of contracting and risk-
sharing. However, some medical societies took a less negative approach,
establishing Foundations for Medical Care (FMCs), beginning with the San
Joaquin County (California) Foundation in 1954. Those FMCs that were based
on contracts and physician acceptance of financial risk are the predecessors of
today's independent practice associations (IPAs) (Egdahl, 1973).4
Constraints On Supply
Another approach to cost containment was developed under the rubric of
health planning. Health planning had received much of its initial nationwide
impetus as a tool for guiding the expansion in community hospital resources
under the Hill-Burton program established after World War II. Beginning in the
late 1950s, however, the growing supply of hospital resources came to be
viewed as a source of rising health care costs (Roemet
4 Egdahl (1973) distinguishes two types of FMCs. The comprehensive FMCs
sponsored prepaid health insurance and performed peer review of health care quality.
The other type provided peer review services to insurers and other organizations and
became the model for government-mandated professional standards review organizations
(PSROs) and peer review organizations established in the 1970s and 1980s.
and Shain, 1959), and health planning was seen as a way to limit excessive
capital investment (Somers and Somers, 1961). By the 1970s, a full-blown
federal, state, and local health planning program was attempting to regulate
resource development (IOM, 1981; Yordy, 1972). Many insurance officials and
employers were active supporters of these efforts (Ennes, 1974; Herod, 1974).
Using a tactic pioneered by Blue Cross of Northeast Ohio as early as 1950 (U.S.
Department of Health, Education, and Welfare, 1976), some insurers warned
that unless hospitals cooperated with public or voluntary health planning, "we
will not pay full reimbursement or continue our contract with a hospital"
(Walter McNerney, quoted in Somers, 1969, p. 138). In the 1980s, faith waned
that health planning could be effectively implemented as a cost-control tool
(Schwartz, W., 1981), and much of the federal and state legal framework for
health planning was dismantled.
Utilization Review
Historically, payers have concentrated their cost-containment energies on
the unit price of medical services and have directed less attention to the volume
of those services provided by institutions and practitioners. Although some
hospitals used committees to monitor utilization in an effort to cope with the
short supply of hospital beds during World War II, the first explicit use of
retrospective utilization review to control fee-for-service payments for
unnecessary and inappropriate hospital services seems to have begun in the
1950s (Payne, 1987, p. 724).5 In 1954, Fred Carter, a physician, wrote in The
Modem Hospital, "Why not appoint a standing hospital staff committee
designated as the 'hospital utilization committee' to do in the field of hospital
and medical economics what the tissue committee does . . . in the field of
surgery. Abuses in the use of hospital services and facilities coming to the
attention of this hospital utilization committee could be disciplined to the point
of near deletion" (cited in London, 1965,
5 To a considerable degree, this utilization review was built on earlier efforts to
standardize medical care delivery and establish mechanisms for overseeing the quality of
care (Payne, 1976; Starr, 1982). Before the turn of the century, Florence Nightingale had
pioneered techniques for assessing and improving the quality of medical services and
spurred advances in the training and use of nurses. In the early 1900s, the Flexner report
(1910) stimulated reforms in medical education. Ernest Codman developed techniques to
audit medical care and identify corrective strategies for treatment deficiencies. Studies
by the American College of Surgeons (ACS) identifying problems in the quality of
hospital care led in 1917 to the Hospital Standardization Program of the ACS, which was
refashioned after World War II into the Joint Commission on the Accreditation of
Hospitals (Roberts et al., 1987). The ACS also initiated the professional activities studies
to develop abstracting techniques to help audit committees. Over time other professional
societies, such as the American College of Physicians and the American Society of
Anesthesiologists, have also worked to develop standards of care and guidelines for
medical practice.
p. 77). Apparently, high optimism about the impact of utilization review was
born with the idea itself.
The 1950s also appear to have seen the first attempt to establish provisions
in health benefit plans to encourage or require second surgical opinions
(Rutgow and Sieverts, 1989). The United Mine Workers Union tried to institute
second opinion requirements but failed because of resistance from organized
medicine. It was not until the 1970s that such provisions were successfully
introduced by the Store Workers Health and Welfare Fund working with
Eugene McCarthy and his colleagues at the Cornell University Medical College
in New York City (McCarthy and Widmer, 1974).
The San Joaquin County Foundation for Medical Care (FMC), founded in
1954, not only served as a model for many IPAs but also helped inspire several
medical societies to organize peer review of health care utilization and quality.
FMCs pioneered many utilization review tools, including model treatment
profiles to assess physician performance, protocols for reviewing ambulatory
care, and computerized screening of claims. By 1973, there were 61 FMCs in
27 states (Egdahl, 1973).
Utilization review also spread in other settings (Werlin, 1973). In the early
1960s, more than 60 Blue Cross plans reported programs to review hospital
claims for the appropriateness of admissions, and more than 50 looked at the
length of stay.6 Some required physicians to certify at admission that hospital
care was necessary for cases such as diagnostic and dental admissions, and
more than two dozen required physicians to certify the need for continued
hospital care after a specified length of stay (Fitzpatrick, 1965; Young, 1965).
In an observation echoed many times over the next 25 years, one speaker
at a 1964 conference on cost-containment programs observed that insurer staffs
had varying opinions on the effectiveness of utilization review, lamenting that
"specific data are lacking" (Fitzpatrick, 1965, p. 24). In a prescient comment,
Odin Anderson noted in 1968 that as payers showed increasing interest in
medical practice patterns, "the central concern of the medical profession today
and in the years ahead might well be 'bureaucracy'" (Anderson, 1968, p. 161).
To summarize, as third-party payment for medical care services expanded
from the 1930s into the 1960s, payersprimarily insurers and
6 State insurance commissioners, who often have considerable power over Blue Cross
plans by virtue of their power to approve rate increases, spurred some of the utilization
review and other cost-containment initiatives of these plans. This was true, for example,
in Michigan, Pennsylvania, and New York (Anderson, 1968; Law, 1974). However,
these initiativesat least their outward formsseem to have spread rather quickly
elsewhere without intervention by state officials (Fitzpatrick, 1965). In the late 1950s,
many Blue Cross plans were in precarious financial condition (Anderson, 1975).
health maintenance organizations (HMOs)tried various tools to control these

costs. These tools may have had some impact, but they often were neither
rigorously applied nor rigorously evaluated. In general, concerns about
controlling costs were still overshadowed by society's desire to expand access
and improve health outcomes through the development and implementation of
advances in medical care.
GOVERNMENT AND EMPLOYER INVOLVEMENT

Two major developments differentiate third-party involvement in
containing health care costs during the periods before and after 1965. The first
is the entry of the federal government as a powerful force for both cost
escalation and cost control. Second, and somewhat later, private employers
began serious efforts to control their expenditures for employee health benefits.
Federal Government Initiatives

The federal government's interest in health care cost containment was
prompted primarily by its creation in the 1960s of two major health benefit
programs: the federal Medicare program for the elderly and the federal and state
Medicaid program for some of the poor. (The latter program also gave states an
increased stake in containing health care costs.) Both programs responded to
inadequacies in private group insurance as a means for covering high-risk or
low-income individuals. Government's stake in health costs consequently
increased. State and federal spending for personal health care services rose from
22 percent of total spending on these services in 1960 to 40 percent by 1980
(Table 2-2).
When Congress established the Medicare and Medicaid programs in 1965
to increase access to health services, it also recognized the need for constraints
against overuse of services. The initial strategy was to expand and strengthen
provider-based utilization review. Hospitals and extended-care facilities were
required, as a condition of participation in Medicare, to have operational
utilization review committees to assure the medical necessity and quality of
care "without involving government in day-to-day hospital operations" (Mills,
1967). This was consistent with the preamble to the Medicare legislation, with
its famous prohibition against federal "supervision or control over the practice
of medicine or the manner in which medical services are provided" (P.L. 89-97).
The costs of the Medicare program soon climbed at rates far above what
was predicted when the program began. As one result, policymakers came to
view the strategy of delegating utilization review as ineffective, "more form
than substance" (U.S. Congress, Senate, 1970). In addition, the

medical profession and some consumer groups were complaining about the
claims reviews and payment denials instituted by the private contractors that
processed Medicare claims. The U.S. Department of Health, Education, and
Welfare had required these activitieswhich some contractors already used in
their private businesswhen they saw that expenditures were quickly
exceeding projections (Blum, et al., 1977; Law, 1974).
In the Social Security Amendments of 1972 (P.L. 92-603), Congress
provided for the establishment of PSROs to control costs and improve the
quality of care through independent peer review. Building on many of the
concepts developed by the Foundations for Medical Care and the Experimental
Medical Care Review Organizations (federal demonstration projects beginning
in 1971), PSROs were to be physician-controlled community organizations that
would develop and apply professional standards to the review of institutional
health services (Blum et al., 1977; Ermann, 1988; Gosfield, 1975; Nelson,
1984). The law required PSROs to perform concurrent review, but preadmission
and retrospective review were optional. In their decade or so of existence, the
PSROs designed and refined many of the data collection and analysis
techniques used by successor organizations to serve both public and private
purchasers.
Congress also got interested in the mid-1970s in second surgical opinion
programs as a means of reducing unnecessary surgery. A congressional
subcommittee stated that there were more than 2 million unnecessary operations
each year, a figure it reached by extrapolating from rates of nonconfirming
second opinions found in early research (American College of Surgeons, 1982).
Although researchers disavowed this kind of projection, the figures got a lot of
attention and helped persuade Congress to authorize demonstration projects to
test second-opinion programs for Medicare and Medicaid beneficiaries.
Despite some positive accomplishments, the PSROslike hospital-based
utilization reviewdisappointed policymakers (Berman and Gertman, 1981;
Chassin, 1978; Congressional Budget Office, HCFA, 1979, 1981; Institute of
Medicine, 1976; Nelson, 1984). Most PSROs delegated review to hospitals
(Ermann, 1988), and the PSROs had little power to penalize physicians for
inappropriate practices. Also, general problems in the economy of the United
Statesthe combination of high inflation and low economic growth
complicated efforts to contain costs.
In 1982, Congress replaced PSROs with statewide "utilization and quality
control peer review organizations," PROs for short (P.L 97-248) (Gosfield,
1989). Three years later, Congress added a requirement for PROs to manage a
focused second-opinion program, a provision not actually implemented until
1989. Appendix C of this report describes some ways in which PROs tend to
differ from private review organizations. Many PROs also have review
contracts with private employers.

Utilization review by PROs was intended to supplement and monitor

another reform that most viewed as much more importantthe introduction of
the prospective payment system (PPS) for hospital services provided to
Medicare beneficiaries. PPS itself was another response by government
policymakers to the frustration with earlier cost-containment efforts, for
example, the Economic Stabilization Program with its wage and price controls
and the health planning program, which tried to limit major capital investments.
The end result of the former, in one evaluator's assessment, was considerable
effectiveness in "reducing rates of increase in hospital employees' wages but
[little impact] . . . on hospital costs" (Ginsburg, 1978).
Private Purchasers Become an Force

In the private sector, purchasers more gradually came to understand that
they too had to be aggressive and prudent buyers of health benefits for their
employees. Indicative of employers' low profile in the 1960s was a National
Conference on Medical Costs, convened at the request of President Johnson to
consider the general problem of rising costs. This 1967 conference had no
corporate members (outside the health industry) on its advisory committee,
listed no business associations among the groups that were consulted, and
included only one corporate representative on its agenda (U.S. Department of
Health, Education, and Welfare, 1968). As one employee benefits manager put
it later:
In the past, many of us in the business community routinely paid bills for
medical care with no questions asked. Unknowingly, we contributed to the
continuation of many abuses in the medical care system by simply signing the
cheek and looking the other way (Chinsky, 1986).
The oil embargo in 1973 and subsequent jumps in oil prices, rising interest
rates, stiffer foreign competition, and other economic shocks combined with
even sharper increases in health benefit costs finally overcame the employer
passivity. By 1976 the Council on Wage and Price Stability was holding
hearings, later published, on private sector concerns and initiatives to control
costs (Council on Wage and Price Stability, 1976). The share of national
spending for health services and supplies accounted for by business grew from
17 percent in 1965 to around 30 percent in 1987 (Levit et al., 1989).
One event that helped influence the thinking of the business community
was the publication in 1978 of a 37-page report entitled, simply, "Labor-
Management Group Position Paper on Health Care Costs" (Labor-Management
Group, 1978). This labor-management group included former Secretary of
Labor John Dunlop, the chairmen of eight major corporations, and presidents of
seven influential labor unions. Their report, issued after 9

months of meetings and staff work, supported such utilization management

precursors as retroactive utilization review (supported by explicit provisions in
collectively bargained benefit plans) and coverage of alternatives to inpatient
hospital treatment. It also promoted a variety of other initiatives, including
prospective reimbursement, HMOs, health planning, technology assessment,
medical malpractice reform, and health education.
In 1987, a new task force of the Labor-Management Group issued a
statement that emphasized preadmission review, (high-cost) case management,
and review of nonhospital care as key strategies for cost containment (Labor-
Management Group, 1987). A related group, the so-called Dunlop Group of
Six, has met on an ongoing basis to investigate and develop cost-containment
proposals. The group consists of the American Hospital Association, American
Medical Association, Health Insurance Association of America, Blue Cross and
Blue Shield Association, Business Roundtable, and the AFL-CIO (Cronin, 1988).
Another early manifestation of greater business interest in health benefit
costs was the move to self-insurance by employers large enough to bear the
risk. In 1975, perhaps 3 percent of companies were self-insured (Gabel et al.,
1986). Among companies surveyed in 1986, the proportion of those that were
self-funded had jumped from 19 to 59 percent between 1980 and 1986 (The
Wyatt Company, 1986). Many of the advantages of self-insurance lie not in
reduced costs for health care but in reduced underwriting, tax, and other
expenses associated with the insurance mechanism. Moreover, the Employment
Retirement and Income Security Act allows self-insured employers to avoid
paying benefits under state-mandated coverage for a variety of services and
providers. These mandates raise both health care and administrative costs,
particularly for multistate employers. Despite the apparent advantages of self-
insurance, a recent study of Bureau of Labor Statistics data found that monthly
costs for family coverage were higher for self-insured groups than for those
with Blue Cross and Blue Shield or commercial insurance (Jensen and Gabel,
1988). It also reported that groups converting from insured to self-insured status
showed higher cost increases (19 percent over 4 years) than did those remaining
insured (10 percent) or those already self-insured (16 percent).
Once larger employers began to assume more active roles in managing
their health benefit plans, their initiatives multiplied (Sullivan, 1984). These
initiatives have included both individual and collective efforts and have often
involved participation by other groups, such as labor representatives. A few
examples can illustrate the wide variety of these activities.
Beginning in 1979, General Motors started joint labor-management efforts
to monitor the cost-containment activities of participating Blue Cross plans,
implement pilot cost-containment projects, and undertake special

studies. In 1984, management and labor agreed to a target 10 percent

reduction in health benefit costs (in constant dollars) over 3 years (Labor-
Management Group, 1987).
The state of Pennsylvania and Council 14 of the American Federation of
State, County, and Municipal Employees agreed in 1984 to institute a
statewide plan for prior review of the medical necessity (not just site of
care) of certain elective procedures. Other activities involve employee
education, HMOs, preferred provider organizations, and program
evaluation (Labor-Management Group, 1987).
The Massachusetts Business Roundtable helped forge revisions in the
state's regulation of hospital rates (Bergthold, 1988).
The Greater Phoenix Affordable Health Care Foundation developed a
software package for local physician offices (with monthly updates) that
lists features of hundreds of employer-based benefit plans (Lockhart, 1988).
Business coalitions, such as the Washington Business Group on Health and
the Midwest Group on Health, have been formed to pursue a broad range of
educational, data collection, and other activities.
More and more corporate benefits managers now talk about health benefits
as a complicated instance of a familiar problem: the procurement of the supplies
and services needed to produce the companies' products. A vivid illustration of
this shift was described in 1986 by Patricia M. Nazemetz, Xerox's benefits
development manager,
We realize that the purchase of poorly defined products and services at
undefined prices is not consistent with good business practice. It does not meet
the standards of the quality improvement process that is applied to every other
aspect of our business. While we have taken steps toward cost containment
through plan redesign and expansion of alternative delivery options, we have
continued to purchase off-the-shelf [health care] products that do not
necessarily meet our customer requirements. A logical next step will be to
define our customer requirements [and then] request proposals from providers
and select those who can conform best to our specifications ....
We realize the importance of developing measurement systems... that allow all
the parties involved to determine if... service and quality standards are being
met and if benefits are being delivered at the right level for the negotiated
price. Conformance to service standards will be determined through
employees' input .... Measuring quality will be much more difficult. The key
will be to identify and agree upon those things that are appropriate quality
indicators. For example, we may need to consider time lost from work,
hospital readmissions, relapses, infections, etc. (Nazemetz, 1986).
Although many larger businesses can take advantage of these approaches,
it should be noted that small businesses often lack the financial and managerial
resources and the economic leverage necessary to implement many cost-
management strategies. Many small employers see cutting or eliminating health
benefits as their major option.
The ability of large purchasers to secure better terms for their purchase

of medical care is attributable, in part, to shifts in the relative supply of

providers versus users of care. The perceived shortage of health care resources
following World War II enhanced the position of hospitals and physicians.
More recently, the greatly increased ratios of hospital beds and physicians to
population have turned the resource shortage into a surplus in most areas. What
is in relatively shorter supply today is patients. Occupancy levels in community
hospitals, for example, have dropped from a national average of 76 percent in
1980 to 65 percent in 1987 (American Hospital Association, 1988). Since
ability to pay now depends largely on public or private insurance, one result is
enhanced power for government and employers, at least large employers.
Many approaches now being used by payers to bring about modifications
in patterns of care have been initiated by Medicare, the largest single payer. On
the private side, the first use of utilization review requirements also appears to
have occurred where certain purchasers were in a particularly powerful position
vis--vis health care providers, for example, among manufacturers with large
facilities in small or mid-sized cities (such as Caterpillar, Inc.) or among
insurance plans with large local market shares (such as the Blue Cross and Blue
Shield plans in Pennsylvania) (Richards, 1984). Although many individual
employers in large cities lack leverage on their own, business coalitions in a
number of communities are attempting to combine employer purchasing power.
An estimated 200 to 250 business coalitions now operate; at least 72 have paid
staff (Cronin, 1988).
OTHER FACTORS GIVING RISE TO UTILIZATION

MANAGEMENT
The intensified involvement of government and private purchasers in
efforts to contain health care costs has been a key element in the rise of more
direct and assertive programs to influence patient care decisions. Purchasers
have acted as both a stimulus and a lever for the development and application of
new approaches to modifying medical practice patterns and limiting the
unnecessary use of services (Webber and Goldbeck, 1984).
In addition, at least three other factors have played a role in the transition
from the provider-based utilization review of the 1960s and 1970s to utilization
management by third parties in the 1980s. These factors include
a growing body of research suggesting that many medical services are
unnecessary or inappropriate and could safely be eliminated;
an increasing emphasis by purchasers on linking cost containment anti
quality assurance; anti
a proliferation of information resources, assessment tools, and
organizations that make case-by-case review of proposed services feasible
on a large scale.

More generally, the 1960s and 1970s saw the growth of doubts that greater
expenditures on medical care were improving health status. As one sociologist
has put it, this idea "hit with the force of a thunderclap" and added to pressure to
constrain medical care expenditures (Starr, 1982).
Inappropriate Utilization as an Cost-Containment Target

The move by third-party payers to become more aggressive about
detecting and eliminating the inappropriate use of medical services has been
strongly influenced by research suggesting that some costly patient care is
unnecessary or inappropriate and that less costly and more appropriate ways of
providing care are possible. Support for these views comes from two directions:
studies that document substantial variations in the use of medical services
across different settings and studies that assess medical services by using
explicit criteria of appropriateness.
Variations in Utilization
Some of the most influential research on medical care utilization has
examined how the variations in the volume of medical services relate to
variations in ways of organizing and financing medical care. For example,
several studies suggest that the volume of physician services tends to increase
when fee-for-service payment rates are frozen but levels off or decreases when
prices are allowed to rise (Gabel and Rice, 1985; Mitchell, 1988; Rice, 1983;
Rossiter and Wilensky, 1983; Schwartz, M., et al., 1981). Most widely known
are the array of studies reporting that hospital use tends to be 5 to 35 percent
lower in HMOs than in fee-for-service systems. Lower use rates in HMOs have
been linked to (1) the financial incentives of capitated or salaried payment for
services; (2) the advantages of group practice in providing peer pressure,
feedback, and practice protocols; (3) the combination of organizational
selection and self-selection that leads practitioners with different styles of
practice into different practice settings; and (4) the tendency of healthier
individuals to be more accepting of health benefit plans that limit their choice of
physician (Epstein et al., 1986; Luft, 1987; Luft and Miller, 1988; Manning et
al., 1984; Norbrega, 1982; Siu et al., 1986; Ware et al., 1986). The latter
explanation is the only one in which differences in the clinical status of patients
might play a significant role in explaining differences in utilization.
The view that medical practice patterns are not a precise and unvarying
function of medical science is also supported by studies showing substantial and
largely unexpected variations in patient care patterns across geographic areas.
The studies date back to the late 1960s and early 1970s (Bunker, 1970; Lewis,
1969; Vayda, 1973). John Wennberg's 1973 article in Science on variations in
small geographic areas attracted much attention, both

for its findings and for its methodological innovations (Wennberg, 1973). Small-
area studies frequently show two-to fourfold differences in the use of various
procedures (Eisenberg, 1986; Lohr et al., 1985; Sandrick, 1984; Wennberg,
1984). They have led purchasers to ask increasingly pointed questions about the
rationale for such variations.7
Small-area studies tend not to label any particular level of utilization as
appropriate or inappropriate (Wennberg, 1986). Certainly, some variation in use
is thought to be due to different population needs for care (Blumberg, 1987;
Wennberg, 1987). For example, a small-area study sponsored by Blue Cross
and Blue Shield of Minnesota found one rural county with unusually high rates
of hospitalization for acute myocardial infarction (80 percent higher than the
statewide average) and several other heart conditions or symptoms that are not
thought to involve much discretion in hospital admission decisions (Thomas,
1989). The insurer asked academic epidemiologists to investigate, and they
came up with evidence to suggest disproportionate rates of coronary heart
disease in the county. (In response, the insurer joined with others to fund a
$300,000 wellness program for the county.)
Inappropriate Utilization
Studies of variations in utilization levels have been complemented by
research that attempts to distinguish clinically between appropriate and
inappropriate care and to compare actual with correct use of services. These
studies have examined hospital admissions, days of inpatient care, ancillary
services, and surgical and other medical procedures.
This research has used a number of methods for identifying inappropriate
use, including review of pathology specimens after surgery, second opinions on
the need for recommended surgery, and audits of medical records. Some
studies, for example, those employing second surgical opinions and early audits
of medical records, have relied on clinical judgment without explicit written
criteria. For other methods, such as certain kinds of medical record audits,
explicit written criteria are central to the assessment process (Payne, 1987).
Taken together, research on appropriateness suggests that perhaps one-
quarter to one-third of medical services may be of little or no benefit to patients
(Brook and Lohr, 1986). For example, several studies have suggested that
between 20 and 40 percent of ancillary services in hospitals
7 To help PROs understand area utilization patterns and target potential problems, the
federal government is funding a 12-state demonstration project to make small-area

analyses available for use by PROs (Tokarski, 1989). In addition, small-area studies have
been initiated by public and private organizations in such states as Maine, Minnesota,
North Carolina, Iowa, California, and New York (Caper, 1987, Tokarski, 1989,
Wennberg, 1984).
were unnecessary or inappropriate (Eisenberg, 1977; Hughes et al., 1984;

Schroeder et al., 1984). Other studies have found similar rates of inappropriate
hospital admissions and days of care (Restuccia, 1984). Recent work by the
Rand Corporation suggests that one-third or more of coronary angiographies
and coronary artery bypass surgeries may be performed for patients whose risk
from the procedure equals or exceeds the expected benefit (Winslow et al.,
1988b). Carotid endarterectomies (which remove clots in the arteries leading to
the brain) were judged appropriate in only 35 percent of cases (Chassin et al.,
1988; Winslow et al., 1988a).
Some inappropriate use of medical services is probably an inescapable
consequence of the inexactitude of medical science, and geographic variations
are not necessarily correlated with variations in appropriate use (Chassin et al.,
1987). Nevertheless, the great variability in the use of medical services and the
studies suggesting that many services are unnecessary have led purchasers to
target such services for cost containment.
Linking Cost and Quality

Purchasers have also recognized that a single-minded concern with the
costs of inappropriate utilization can be counterproductive. It can create
suspicion, antagonism, and resistance rather than cooperation among payers,
practitioners, and patients. By tying utilization management to quality of care as
well as cost, purchasers do two things:
they underscore the responsibilities of health care institutions and
physicians for providing patients with safe, effective care, and
they emphasize to policymakers and health plan beneficiaries that payment
for poor-quality care and unneeded services is a bad use of limited resources.
Researchers have noted that where there is strong professional consensus
on the appropriate use of particular services, such as surgery for cancer of the
bowel or hospitalization for hip fracture, there tends to be less variation in
levels of utilization. Where consensus is low, for example, on the need for
hysterectomy and prostatectomy, utilization is more variable (Caper, 1984).
Although differences among populations in the use of particular services can
reflect differences in the incidence of the relevant medical problems, many
variations appear to be due to differences in what Wennberg (1984) calls
physician ''practice styles.'' To strengthen consensus and reduce inappropriate
care, the federal government, many medical societies, and other organizations
are encouraging the development of practice guidelines or protocols for the
appropriate and cost-effective use of specific medical services.

The value of linking cost and quality concerns is suggested by several

developments over the past two decades. For example, the growing support for
research on the effectiveness of medical servicessupport that cuts across
purchaser, provider, and consumer interestsmight not have materialized based
on only cost or quality considerations. In addition, the legislative history of
PROs and HMOs suggests that congressional support for these programs
depends on their perception that quality of care as well as cost-containment
goals are being served. When Congress banned hospitals from making incentive
payments to physicians as an inducement to reduce services to Medicare
beneficiaries, it also set restrictions on some financial incentives used by HMOs
that were to take effect in 1989 (P.L. 99-509). The effective date has been
shifted to 1990, and other changes may result from further study of these
incentives, but the basic concern about the cost containment/quality link
remains (Ready, 1989). The paper on utilization management in HMOs in
Appendix B of this report discusses these concerns at greater length.
Improving the Tools and Structures for Utilization

Management
Changing perceptions about the extent of unnecessary medical care and the
impact of such care on both cost and quality have been intertwined with
changes in information resources, analytical tools, and organizational structures.
Although there is substantial need for improvement in each area, these technical
and organizational developments have made it more feasible to move from
problem identification to problem solving.
Information Resources
The past two decades have seen many advances in the quality and
availability of data on the cost and use of health services. These advances include
greater, though still incomplete, standardization of cost and utilization data
from different provider and payer sources;
more rationality and uniformity in the definition and coding of medical
diagnoses and services;
creation of more large utilization and cost data bases covering hospital,
physician, and other services; and
development of computer software to manage and analyze information.
Each of these developments has increased the ability of purchasers,
researchers, and others to describe utilization and cost patterns, track changes
across time, and identify anomalies. This kind of descriptive

information has been a powerful stimulus for purchaser efforts to reduce

inappropriate utilization.
The committee recognizes that, despite considerable progress, many
problems with the availability and quality of data remain. Studies continue to
document imprecise or inaccurate diagnosis and procedure coding, lack of
diagnostic codes on most claims forms, only scattered documentation about
entire episodes of treatment or illness, errors and ambiguities in preparation and
processing of claims data, and limited information on patient and population
characteristics (Larks, 1989; PPRC, 1988). However, the momentum for
solving these problems is substantial and stems from multiple public and private
interests and needs. These include the demands of purchasers for better
documentation on claims for payment, the specter of malpractice, the interest of
policymakers in data to develop and monitor changes in methods of paying
providers, and technological advances in computer hardware and software.
Assessment and Education Strategies

Assessment and analytic tools that have helped the move toward utilization
management include (1) explicit criteria for assessing the appropriateness of
medical services, (2) statistical methods for setting utilization norms, and (3)
methods for assessing the health status of different patients and population
groups. On this analytic base are built feedback techniques for informing
physicians about how their practice patterns compare with statistical norms and
criteria for appropriate care (Eisenberg, 1986; Wennberg, 1984; Wennberg et
al., 1977).
Explicit criteria for judging the appropriateness of medical services include
methods that are diagnosis-specific and those that are not (Payne, 1987). The
studies of carotid endarterectomies and cardiac pacemakers cited earlier rely on
the former, whereas studies of the need for hospital admission or clays of
hospital care rely on the latter.
Statistical norms have been used most frequently to assess lengths of
hospital stay and frequencies of diagnostic or treatment procedures. Average
lengths of stay for patients in the western United States, which are lower than
those for patients in other regions, are frequently used as a standard in
continued-stay review.
Since one reason that health care use and costs may differ across groups is
because the health status of group members differs, much effort has gone into
methods to assess how severely ill individual patients are, whether they need to
be hospitalized, and how institutions and practitioners vary in the overall
severity of their mix of cases. These methods use various approaches, including
patient resource consumption, organ system involvement, disease stage, and
diagnosis (Aronow, 1988; Hornbrook,

1982). Some rely on claims data; others require abstracted information from
medical records.
Two cautions should be offered about the current interest in developing
medical practice guidelines to influence physician behavior. First, the quality of
the relevant medical evidence and the quality of the guideline development
process can have major impacts on the soundness and utility of the guidelines.
A number of researchers have written persuasively about the limits of clinical
research, the lack of fit between some guidelines and relevant clinical evidence,
and the problems with methods for generating guidelines (Bunker, 1988;
Chassin, 1989; Eddy and Billings, 1988; Schwartz, 1984; Wennberg, 1984).
Moreover, empirical research itself can clarify but not fully resolve some
questions, for example, whether it is an appropriate use of limited resources to
perform cataract surgery to correct small visual deficits.
A second caution about the influence of guidelines is that the success of
education programs in changing physician behavior is mixed (Donabedian,
1982; HCFA, 1979; Myers and Gleicher, 1988; Restuccia, 1982; Schroeder, et
al., 1984; Wennberg et al., 1977; White et al., 1975). One thorough review of
research in this area concluded that just presenting facts is not likely to
accomplish much in the absence of other facilitating conditions (Eisenberg,
1986). These conditions include a supportive professional environment, a
perceived need for the information, an active involvement by respected peers
and professional leaders, and careful targeting of quality, utilization, and cost
problems.
The growth of new contracting and payment techniques that put physicians
at financial risk is increasing the perceived need for information feedback to
practitioners. Analyses and comparisons of practice patterns may have a greater
impact on practitioners when reenforced by the specter of future claims denials,
exclusion from an HMO or PPO, or loss of a year-end bonus. Moreover, the
skilled use of feedback techniques may avoid the tensions created when the
more negative or punitive strategies are actually used.
New Organizations
The demand on the part of large public and private purchasers of care for
ways to manage, rather than only review, health care utilization has been
matched by the emergence of many "suppliers" wanting to take on this role (de
Lissovoy et al., 1987; Freund, 1987; Gardner and Scheffler, 1988; InterStudy,
1989; Mayo Clinic, 1988). These suppliers fall into two broad categories: first,
organizations that integrate utilization and cost control with service delivery
and, second, organizations that offer specialized utilization management
services to both health care providers

and purchasers. One 1987 survey of over 700 employers found that about one-
third of those covered by the employers' health plans were enrolled in "managed
fee-for-service" programs and over a quarter more were enrolled in either
HMOs or PPOs (Gabel et al., 1988).
The 1980s have seen substantial growth in the number and types of
organizations that integrate, to some degree, cost-management and service
delivery functions. The number of HMOs has grown from 175 in 1976 to over
600 in 1988, and membership has risen from 6 million to over 32 million
(Group Health Association of America, 1988; InterStudy, 1989). A recent
survey placed the number of PPOs at over 600 (American Medical Care and
Review Association, 1988). Declining hospital occupancy rates and competition
for the business of large purchasers has stimulated some of the country's largest
multihospital systems and insurers to develop integrated service/insurance
products. By 1986, all six of the largest multihospital systems, nine of the
largest ten group insurance carriers plus the Blue Cross and Blue Shield system,
and six of the eight largest HMO systems had developed some kind of
integrated product or products (Patricelli, 1986).
These organizations differ in the weight they place on financial incentives,
favorable unit prices, utilization management, and other means to control costs.
Some, including many PPOs, develop and manage contractual arrangements
with health care providers but do not underwrite the cost of services. Others,
such as many HMOs, accept risk for the cost of services, although they
typically share that risk with their participating hospitals and physicians.
The financial incentives used by many of these organizations, such as
capitation and bonuses, target both elements of utilization risk, that is, the
number of services per episode of care and the number of episodes. Although
these arrangements may reassure payers that overuse of services is not being
encouraged, that reassurance may be countered by concerns that some needed
care will be discouraged (AFL-CIO, 1986; Halowell, 1989).8 These worries are
reflected in congressional requirements for PRO monitoring of the quality of
care provided to Medicare beneficiaries in capitated HMOs and PPOs (P.L.
99-272) and in proposals that people be informed about financial arrangements
for participating providers before they enroll in such health plans (Physician
Payment Review Commission, 1989).
8 One early critic of bonuses and withholds linked them pejoratively to such historic
arrangements as "split fees, kickbacks, rebates, bribes, and so forth." However, these
latter incentives have generally been regarded as unethical inducements to over-rather
than underutilization (Geist, 1974, p. 1306).
Disappointing earnings, and perhaps increasing concern about regulatory

restrictions, have led many sponsors and investors to drop their new products or
move out of certain geographic markets. Still, the array of organizations
offering integrated service delivery and cost management is larger and more
diverse than it was a decade ago and gives employers more leverage than they
had in the past.
Some of the vertically integrated organizations that remain are promoting
and underwriting an integrated health benefit package, the so-called triple
option. Such programs, which are not yet widespread, typically allow
employees to enroll in an HMO, a PPO, or a traditional benefit plan based on
their own weighing of coverage, premium, and freedom of choice of provider.
Ideally, since all options are purchased from only one supplier, the employer
gets a less complex program to manage plus an integrated premium that helps
adjust for any tendency of sicker individuals to choose more generous and
expensive plans while healthy individuals choose more restrictive and less
expensive alternatives. A variant on this concept is the open-ended HMO that
provides limited coverage for nonemergency use of providers outside the HMO
panel.
A second market response to purchaser demands for cost containment has
involved lower entry costs and lower risks than the product integration and
triple-option strategies (McGraw-Hill, 1987). Many organizations have been
established to offer freestanding utilization management services to providers,
insurers, and employers. Although some purchasers, such as prepaid group
practices, are well situated to undertake utilization management internally,
others, such as many independent practice associations, have found it easier to
obtain the services from an external agent. Also, some purchasers prefer not to
buy an integrated product.
The utilization management industry, although growing fast and coveting
an increasing number of benefit plans and beneficiaries, has received little
systematic study. This report is designed to help correct that deficiency.
CONCLUSION
Much of this century has been devoted, first, to improving the
effectiveness of medical treatments, second, to increasing individual protection
against the financial burdens of medical services, and, third, to trying to control
the economic costs of expanding access to more sophisticated medical care. In
each endeavor, various strategies or working hypotheses rise and wane in
popularity as initial promising ideas or results are subjected to empirical
evaluation, trials of administrative feasibility, or tests of political acceptability.

Most current approaches to controlling costs and improving the

appropriateness of care have been tried and revised, with varying degrees of
enthusiasm and sophistication, over several decades. Utilization management
builds on the gradual accumulation of experience and data that suggest that
externally applied assessments of the appropriateness of proposed medical
services can constructively influence how care is provided and, as one
consequence, help constrain health care costs. The next three chapters describe
what we know about the varied ways in which utilization management works.
REFERENCES
Abramowitz, Michael, "Governments Pinched by Health Care," The Washington Post, May 28,
1989, p. C1.
American Federation of Labor and Congress of Industrial Organizations (AFL-CIO), Department of
Occupational Safety, Health and Social Security, Health Care Cost Containment Crisis:
Bargaining for Improved Health Care Benefits, Washington, DC, 1986.
American College of Surgeons, "Second Surgical Opinion Programs: A Review and Progress
Report," Chicago, 1982.
American Hospital Association, Hospital Statistics, Chicago, 1988.
American Medical Cam and Review Association, Directory of Preferred Provider Organizations,
Bethesda, MD, 1988.
Anderson, Odin, Blue Cross Since 1929: Accountability and the Public Trust, Cambridge, MA:
Ballinger, 1975.
Anderson, Odin, Health Care: Can There Be Equity? New York: Wiley, 1972.
Anderson, Odin, The Uneasy Equilibrium, New Haven, CT: College and University Press, 1968.
Aronow, David B., "Severity of Illness Measurement," Medical Care Review, Fall 1988, pp. 339-366.
Bergthold, Linda A., "Purchasing Power: Business and Health Policy Change in Massachusetts,"
Journal of Health Politics, Policy and Law, Fall 1988, pp. 425-451.
Berman, David E., and Gertman, Paul M., "Cost Containment and Quality Assurance," in Federal
Health Programs, ed. by Altman, Stuart, and Sapolsky, Harvey, Lexington, MA:
Lexington Books, 1981.
Blum, John D., Gertman, Paul M., and Rabinow, Jean, PSROs and the Law, Germantown, MD:
Aspen Systems Corporation, 1977.
Blumberg, Mark S., "Inter-area Variations in Age-Adjusted Health Status," Medical Care, April
1987, pp. 340-353.
Brook, Robert H., and Lohr, Kathleen N., "Will We Need to Ration Effective Medical Care?" Issues
in Science and Technology, Fall 1986, pp. 68-77.
Bunker, John P., "Is Efficiency the Gold Standard for Quality Assessment?" Inquiry, Spring 1988,
pp. 51-58.
Bunker, John P., "Surgical Manpower: A Comparison of Operations and Surgeons in the United
States and in England and Wales," New England Journal of Medicine, January 15, 1970,
pp. 135-144.
Caper, Philip, "Variations in Medical Practice: Implications for Health Policy," Health Policy,
Summer 1984, pp. 110-119.
Caper, Philip, Presentation to the Physician Payment Review Commission, Washington, DC,
September 1987.
Chassin, Mark, "The Containment of Health Care Costs: A Strategic Assessment," Medical Care,
October 1978 (Supplement), pp. 1-55.

Chassin, Mark, "Developing Good Medical Standards We All Can Live With," The Internist, May-
June 1989, pp. 6-9.
Chassin, Mark R., Kosecoff, Jacqueline B., and Park, R. E., "Does Inappropriate Use Explain
Geographic Variation in the Use of Health Services?" Journal of the American Medical
Association, November 13, 1987, pp. 2533-2537.
Chassin, Mark, Kosecoff, Jacqueline, Brook, Robert H., and Solomon, David H., "How Coronary
Angiography Is Used," Journal of the American Medical Association , November 13,
1988, pp. 2543-2547.
Chinsky, David, Employee Insurance Division, Ford Motor Company, Speech on February 19, 1986.
Chollet, Deborah, Uninsured in the United States: The Nonelderly Population Without Health
Insurance, 1986. Washington, DC: Employee Benefit Research Institute, October 5, 1988.
Congressional Budget Office, The Effects of PSROs on Health Care Costs: Current Findings and
Future Evaluations, Washington DC, 1979.
Congressional Budget Office, The Impact of PSROs on Health Care Costs: Update of CBO's 1979
Evaluation, Washington, DC, 1981.
Congressional Research Service, Health Insurance and the Uninsured, Report to Congress, Vol. 1,
Washington, DC, June 9, 1988.
Cronin, Carol, "Business Wields Its Purchase Power," Business and Health, November 1988, pp. 4-8.
Council on Wage and Price Stability, Executive Office of the President, The Complex Puzzle of
Rising Health Care Costs: Can the Private Sector Fit It Together? Washington, DC,
December 1976.
de Lissovoy, Gregory, Rice, Thomas, Gabel, Jon, and Gelzer, Heidi J., "Preferred Provider
Organizations One Year Later," Inquiry, Summer 1987, pp. 127-135.
Donabedian, Avedis, Explorations in Quality Assessment and Monitoring, Ann Arbor, MI: Health
Donabedian, Avedis, Benefits in Medical Care Administration, Cambridge, MA: Harvard University
Press, 1976.
Eddy, David M., and Billings, John, "The Quality of Medical Evidence and Medical Practice,"
Paper prepared for the National Leadership Conference on Health Care, Washington, DC,
1988.
Egdahl, Richard, "Foundations for Medical Care," New England Journal of Medicine, March 8,
1973, pp. 491-498.
Eisenberg, John M., Doctors' Decisions and the Cost of Medical Care, Ann Arbor, MI: Health
Eisenberg, John M., et al., "computer-Based Audit to Detect and Correct Overutilization of
Laboratory Tests," Medical Care, 1977, pp. 915-921.
Ennes, Howard, Testimony in hearings on national health planning and resources development held
by the Subcommittee on Public Health and Environment, Committee on Interstate and
Foreign Commerce, U.S. House of Representatives, May 8, 1977, in Hearing Record,
Serial No. 93-91, pp. 709-718.
Epstein, Arnold M., Begg, Colin B., and McNeil, Barbara J., "The Use of Ambulatory Testing in
Prepaid and Fee-for-Service Group Practices," New England Journal of Medicine, April
24, 1986, pp. 1089-1094.
Ermann, Danny, "Hospital Utilization Review: Past Experience, Future Directions," Journal of
Health Politics, Policy and Law, Winter 1988, pp. 683-704.
Fitzpatrick, Thomas B., "Utilization Review and Control Mechanisms: From the Blue Cross
Perspective," Inquiry, September 1965, pp. 16-29.
Flexner, Abraham, Medical Education in the United States and Canada, New York: Carnegie
Foundation for the Advancement of Teaching, 1910.
Freudenheim, Milt, "A Health Care Taboo Is Broken," The New York Times, May 8, 1989, p. D1.
Freund, Deborah A., "Competitive Health Plans and Alternative Payment Arrangements for
Physicians in the United States: Public Sector Examples," In Search of an Optimal
Incentive Structure in Health Care, special issue of Health Policy, 1987, pp. 163-173.

Gabel, Jon, Jajich-Toth, Cindy, and de Lissovoy, Gregory, "The Changing World of Group Health
Insurance," Health Affairs, Summer 1988, pp. 48-65.
Gabel, Jon, et al., "The Emergence and Future of PPOs," Journal of Health Politics, Policy and Law
Summer 1986, pp. 305-321.
Gabel, Jon R., and Rice, Thomas H., "Reducing Public Expenditures for Physician Services: The
Price of Paying Less." Journal of Health Politics, Policy and Law, Winter 1985, pp.
505-609.
Gardner, Laura B., and Scheffler, Richard M., "Privatization in Health Care: Shifting the Risk,"
Medical Care Review, Fall 1988, pp. 215-254.
Geist, Robert W., "Incentive Bonuses in Prepayment Plans," New England Journal of Medicine,
December 12, 1974, pp. 1306-1308.
Gibson, Robert, "National Health Expenditures, 1979," Health Care Financing Review, Summer
1980, pp. 1-36.
Ginsburg, Paul, "Impact of the Economic Stabilization Program on Hospitals," in Hospital Cost
Containment, ed. by Zubkoff, Michael, Raskin, Ira E., and Hanft, Ruth S., New York:
PRODIST for the Milbank Memorial Fund, 1978, pp. 293-323.
Gosfield, Alice, PSROs: The Law and the Health Consumer, Cambridge, MA: Ballinger Publishing
Co., 1975.
Gosfield, Alice, "PROs: A Case Study in Utilization Management and Quality Assurance," 1989
Health Law Handbook, New York: Clark Boardman Co., Ltd., 1989.
Group Health Association of America, HMO Industry Profile, Washington, DC, December 1988.
Hallowell, Elizabeth, "In Del., Challenging HMO Financial Incentives," The Philadelphia Inquirer,
March 28, 1989, p. B1.
Health Care Financing Administration, Professional Standards Review Organizations: 1978
Program Evaluation, Washington, DC, U.S. Department of Health, Education, and
Welfare, 1979, pp. 34-37.
Health Insurance Association of America, 1986-1987 Source Book of Health Insurance Data,
Health Insurance Association of America, 1988 Update, Source Book of Health Insurance Data,
Hellinger, Fred J. "An Empirical Analysis of Several Prospective Reimbursement Systems," in
Hospital Cost Containment, ed. by Zubkoff, Michael, Raskin, Ira E., and Hanft, Ruth S.,
New York: PRODIST for the Milbank Memorial Fund, 1978, pp. 370-400.
Herod, James C., Testimony in hearings on national health planning and resources development
held by the Subcommittee on Public Health and Environment, Committee on Interstate and
Foreign Commerce, U.S. House of Representatives, May 8, 1974, in Hearings Record,
Serial no. 93-91, p. 381.
Hornbrook, Mark C, "Hospital Case Mix," Medical Care Review, Spring 1982, pp. 1-43, and
Summer 1982, pp. 73-123.
Hughes, Robert A., et al., "The Ancillary Services Review Program in Massachusetts: Experience of
the 1982 Pilot Project," New England Journal of Medicine, October 5, 1984, pp. 1727-1732.
Independence Blue Cross and Pennsylvania Blue Shield, Independence and Leadership in Health
Care: Community Health Care Report 1988, Philadelphia, 1988.
Institute of Medicine, Assessing Quality in Health Care, Washington, DC: National Academy of
Sciences, November 1976.
Institute of Medicine, Health Planning in the United States, Washington, DC: National Academy
Press, January 1981, two volumes.
InterStudy, InterStudy Edge, Excelsior, MN, 1989.
Jensen, Gail, and Gabel, Jon, "The Erosion of Purchased Health Insurance," Inquiry, Fall 1988, pp.
328-343.
Labor-Management Group, Position Paper on Health Care Costs, 1978.
Labor-Management Group, Policy Issues in Health Care Costs and Six Case Studies, Washington,
DC, 1987.
Larks, Marlen, Fostering Uniformity for Health Care Assessment Data Gathering, Washington, DC,
National Association of Health Data Organizations, May 4, 1989.

Law, Sylvia, Blue Cross: What Went Wrong? New Haven, CT: Yale University Press, 1974.
Letsch, Suzanne W., Levit, Katherine R., and Waldo, Daniel R., "National Health Expenditures,
1987," Health Care Financing Review, Winter 1988, pp. 109-129.
Levit, Katherine R., Freeland, Mark S., and Waldo, Daniel R., "Health Spending and Ability to
Pay," Health Care Financing Review, Spring 1989, pp. 1-12.
Lewis, Charles, "Variations in the Incidence of Surgery," New England Journal of Medicine,
October 16, 1969, pp. 880-884.
Lockhart, Carol Ann, "Physicians Operation Procedures: A Project for Enhancing Patient Flow
Through a Competitive Health Care System," Phoenix, AZ: Greater Phoenix Affordable
Health Care Foundation, 1988.
Lohr, Kathleen N., Lohr, William R., and Brook, Robert H., Geographic Variations in the Use of
Medical Services and Surgical Procedures: A Chartbook, Washington, DC, George
Washington University National Health Policy Forum, 1985.
London, Morris, "Medical Staff Utilization Committees," Inquiry, September 1965, pp. 77-95.
Luft, Harold S., Health Maintenance Organizations: Dimensions of Performance, New Brunswick,
NJ: Transaction Books, 1987.
Luft, Harold S, and Miller, Robert H., "Patient Selection in a Comprehensive Health System,"
Health Affairs, Summer 1988, pp. 97-119.
Manning, Willard G., et al., "A Controlled Trial of the Effect of a Prepaid Group Practice on Use of
Services," New England Journal of Medicine, 1984, pp. 1505-1510.
Mayo Clinic, "The 'Cost' of Effective Utilization Review Programs," Statement prepared for the
Institute of Medicine Committee on Utilization Management by Third Parties, May 20,
1988.
McCarthy, E.G., and Widmer, Got, "Effects of Screening by Consultants on Recommended Elective
Surgical Procedures," New England Journal of Medicine, 1974, pp. 1331-1335.
McGraw-Hill, Inc., Review Resources: Sourcebook of Private Independent UR Companies,
McIntyre, Duncan, Voluntary Health Insurance and Rate Making, Ithaca, NY: Cornell University
Press, 1962.
Mills, Wilbur D., "Rising Health Costs: A View from Congress," in Report of the National
Conference on Medical Costs, June 27-28, 1967, Washington, DC: U.S. Department of
Health, Education, and Welfare, 1967.
Mitchell, Janet, Wedig, Gerard, and Cromwell, Jerry, "The Medicare Physician Fee Freeze," Health
Affairs, Spring 1989, pp. 21-33.
Monheit, Alan C., and Schur, Claudia L., "The Dynamics of Health Insurance Loss," Inquiry, Fall
1988, pp. 315-327.
Myers, Stephen A., and Gleicher, Norberet, "A Successful Program to Lower Cesarian-Section
Rates," New England Journal of Medicine, December 8, 1988, pp. 1511-1516.
National Health Policy Forum, Health Insurance Institute, Washington, DC, November 30-
December 1, 1988.
Nazemetz, Patricia M., "Health Care Management Directions at Xerox," Speech at Government
Research Corporation Health Leadership Conference, Washington, DC, April 1986.
Nelson, Alan, "Lessons from the Past," The Internist, January 1984, pp. 7-8.
Norbrega, F. T., "Hospital Use in a Fee-for-Service System," Journal of the American Medical
Association, February 12, 1982, pp. 806-810.
Patricelli, Robert, "Musings of a Blind ManReflections on the Health Care Industry," Health
Affairs, Summer 1986, 128-134.
Payne, B. C., et al., The Quality of Medical Care: Evaluation and Improvement, Chicago: Hospital
Research and Educational Trust, 1976.
Payne, Susan M. C., "Identifying and Managing Inappropriate Hospital Utilization," Health
Services Research, December 1987, pp. 709-769.
Physician Payment Review Commission, Annual Report to Congress, Washington, DC, 1988.
Ready, Tinker, "House Panel Modifies Physician-Incentives Law," Managed HealthCare, July 13,
1989, p. 5.

Restuccia, Joseph D., et al., ''A Comparative Analysis of Appropriateness of Hospital Use,'' Health
Affairs, Summer 1984, p. 130-138.
Restuccia, Joseph D., "The Effect of Concurrent Feedback in Reducing Inappropriate Hospital
Utilization," Medical Care, January 1982, p. 46.
Richards, Glenn, "Business Spurs UR Growth," Hospitals, March 1, 1984, p. 96.
Rice, Thomas H., "The Impact of Changing Medicare Reimbursement Rates on Physician-Induced
Demand," Medical Care, August 1983, pp. 803-815.
Rich, Spencer, "Health Plans May Cover More Than Was Estimated," The Washington Post, July 1,
1989, p. A-5.
Roberts, James S., Coale, Jack G., and Redman, Robert A, "A History of the Joint Commission on
the Accreditation of Hospitals," Journal of the American Medical Association, August 21,
1987, pp. 936-940.
Rorem, C. Rufus, A Quest for Certainty: Essays on Health Economics, 1930-1970, Ann Arbor, MI:
Health Administration Press, 1982.
Roemer, Milton I., and Shain, Max, Hospital Utilization Under Insurance, American Hospital
Association Monograph Series No. 6, Chicago, 1959.
Rossiter, Louis F., and Wilensky, Gail R., "A Reexamination of the Use of Physician Services: The
Role of Physician-Initiated Demand," Inquiry, Summer 1983, pp. 162-172.
Rutgow, Ira M., and Sieverts, Steven, "Surgical Second Opinion Programs," in Socioeconomics of
Surgery, ed. by Rutgow, Ira M., St. Louis, MO: The C.V. Mosby Co., 1989.
Sandrick, Karen M., "Blue Cross and Blue Shield of Michigan's Efforts to Change Practice
Patterns," Quality Review Bulletin, November 1984, pp. 349-352.
Scheffler, Richard M., and Rossiter, Louis F., Biased Selection in Health Care Markets, Advances
in Health Economics and Health Services Research, Vol. 6, Greenwich, CT: JAI Press,
Inc., 1985.
Schroeder, Steven A, Myers, L P., and McPhee, S. J., "The Failure of Physician Education as a Cost
Containment Strategy: Report of a Prospective Controlled Trial at a University Hospital,"
Journal of the American Medical Association, 1984, pp. 225-230.
Schwartz, J. Sanford, "The Role of Professional Medical Societies in Reducing Practice Variations,"
Health Affairs, Summer 1984, pp. 90-101.
Schwartz, Michael, et al., "The Effect of a Thirty Percent Reduction in Physician Fees on Medicaid
Surgery Rates in Massachusetts," American Journal of Public Health, 1981, pp. 370-375.
Schwartz, William B., "The Regulation Strategy for Controlling Hospital Costs," New England
Journal of Medicine, November 19, 1981, pp. 1249-1255.
Siu, Albert L., et al., "Inappropriate Use of Hospitals in a Randomized Trial of Health Insurance
Plans," New England Journal of Medicine, November 13, 1986, pp. 1259-1266.
Somers, Anne R., Hospital Regulation: The Dilemma of Public Policy, Princeton, NJ: Industrial
Relations Section, Princeton University, 1969.
Somers, Herman M., and Somers, Anne R., Doctors, Patients, and Health Insurance, Washington,
DC: The Brookings Institution, 1961.
Starr, Paul, The Social Transformation of American Medicine, New York: Basic Books, Inc., 1982.
Sullivan, Sean, Managing Health Care Costs: Private Sector Initiatives, Washington, DC,
American Enterprise Institute, 1984.
Thomas, M. Carroll, "Where Overutilization Does Not Mean Bad Doctoring," Medical Economics,
May 1, 1989, pp. 65-68.
Tokarski, Cathy, "HCFA Conducting Market-by-Market Study of Physician Practice Patterns,"
Modern Healthcare, February 17, 1989, p. 30.
U.S. Congress, Senate, Medicare and Medicaid: Problems, Issues and Alternatives , Report of the
Staff to the Committee on Finance, U.S. Senate, Washington, DC, February 9, 1970.
U.S. Department of Health, Education, and Welfare, Trends Affecting U.S. Health Care System,
prepared by the Cambridge Research Institute, DHEW Pub. No. HRA 76-14503,
Washington, DC, January 1976.

U.S. Department of Health, Education, and Welfare, Report of the National Conference on Medical
Costs, June 27-28, 1968, Washington, DC.
Vayda, E., "A Comparison of Surgical Rates in Canada and in England and Wales," New England
Journal of Medicine, 1973, pp. 1224-1229.
Ware, John E., Jr., et al., "Comparison of Health Outcomes at a Health Maintenance Organization
with Those of Fee-for-Service Care," Lancet, May 31, 1986, pp. 1017-1022.
Webber, Andrew, and Goldbeck, Willis B., "Utilization Review," in Health Care Cost
Management: Private Sector Initiatives, ed. by Fox, Peter B., Goldbeck, Willis B., and
Spies, J. J., Ann Arbor, MI: Health Administration Press, 1984, pp. 69-81.
Wennberg, John E., Blowers, Lewis, Parker, Robert, and Gittelshon, Alan M., "Changes in
Tonsillectomy Rates Associated with Feedback and Review," Pediatrics, June 1977, pp.
821-826.
Wennberg, John E., "Population Illness Rates Do Not Explain Population Hospitalization Rates,"
Medical Care, April 1987, pp. 354-359.
Wennberg, John E., and Gittelsohn, Alan M., "Variations in Medical Care Among Small Areas,"
Science, December 14, 1973, pp. 1102-1108.
Wennberg, John E., "Which Rate Is Right?" New England Journal of Medicine, January 30, 1986,
pp. 310-311.
Wennberg, John E., "Dealing with Medical Practice Variations: A Proposal for Action," Health
Affairs, Summer, 1984, pp. 6-32.
Werlin, Stanley, "Cost Control Methods in Health Care Delivery" in PSRO: Organization for
Regional Peer Review, ed. by Decker, Barry, and Bonner, Paul, Cambridge, MA: Ballinger
Publishing Co., 1973.
White, J. J., Satillana, M., and Haller, J.A., Jr., "Intensive In-Hospital Observation: A Safe Way to
Decrease Unnecessary Appendectomy," American Surgeon, December 1975, p. 794.
Wilensky, Gail R., and Rossiter, Louis F., "The Relative Importance of Physician-Induced Demand
for Medical Care," Milbank Memorial Fund Quarterly, Spring 1983, pp. 252-277.
Wilson, Florence A, and Neuhauser, Duncan, Health Services in the United States, Cambridge, MA:
Ballinger, 1974.
Winslow, Constance M., Solomon, David H., Chassin, Mark R., Kosecoff, Jacqueline, and Brook,
Robert H., "The Appropriateness of Carotid Endarterectomy," New England Journal of
Medicine, March 24, 1988a, pp. 721-727.
Winslow, Constance M., Kosecoff, Jacqueline, Chassin, Mark R., Kanouse, David E., and Brook,
Robert H., "The Appropriateness of Performing Coronary Angiography and Coronary
Artery Bypass Surgery," Journal of the American Medical Association, July 22/29, 1988b,
pp. 505-509.
The Wyatt Company, 1986 Group Benefits Survey, New York, 1988, p. 18.
Yordy, Karl D., "National Planning for Health: An Emerging Reality," Bulletin of the New York
Academy of Medicine, January 1972, pp. 32-38.
Young, Lee A., "Utilization Review and Control Mechanisms: From the Blue Shield Perspective,"
Inquiry, September 1965, pp. 5-15.

THE UTILIZATION MANAGEMENT INDUSTRY: STRUCTURE AND PROCESS 58
3
The Utilization Management Industry:
Structure and Process
Analyzing the changes brought about by utilization management requires a

basic understanding of how utilization management actually works. What
decisions do utilization management organizations make and on what basis? To
whom are they accountable and for what? How do they try to be effective in
controlling costs without harming patients? How standardized are their methods
and criteria?
To answer these questions, the committee set out to learn more about the
workings of utilization management. What it found about prior review
procedures is described in this chapter. Chapter 5 covers high-cost case
management. Judgments and recommendations are reserved primarily for
Chapter 6.
From the standpoint of the employers or governmental programs that
purchase medical care, utilization management can be brought into play in one
of three ways. The purchaser can
engage in utilization management directly, as some Medicaid programs do;
contract with another organization for utilization management services, as
Medicare and many employers do; or
shift some of the financial risk to another party, such as an insurer, an
HMO, or a PPO, which is then faced with the same three choices of doing
utilization management directly, contracting for the services, or shifting
risk to still another party (for example, physicians or hospitals).

The committee did not focus on utilization management activities

undertaken directly by employers. Rather, it looked at the broad and varied
array of organizations that either contract directly with employers to provide
utilization management services or, like some HMOs and insurers, provide
utilization management as part of a larger package of services. The products,
target markets, philosophies, and technologies of these organizations vary
widely, but the industry is evolving so rapidly that summary is difficult. What
follows is a snapshot of the field of utilization management.
INDUSTRY OVERVIEW
In 1980, there was no utilization management industry to speak of,
although some of the building blocks existed in HMOs, professional standards
review organizations (PSROs), insurance plans, and hospital utilization review
programs. Now hundreds of organizations offer utilization management
services to thousands of clients who employ perhaps half to two-thirds of all
American workers (Foster Higgins, 1987; Gabel et al., 1988). A precise count
of utilization management organizations is virtually impossible because the
industry is changing constantly, and no single trade association or industry
information source exists. A 1987 publication listed 158 private independent
utilization review companies (McGraw-Hill, 1987). However, this list did not
include the utilization management departments of those commercial insurers
and Blue Cross and Blue Shield plans that do not have separate utilization
management subsidiaries. Also not included were the internal utilization
management departments of HMOs, independent practice associations (IPAs),
and PPOs. A 1989 survey by Business Insurance reported 125 review
organizations (Business Insurance, 1989). Again, utilization management
departments of many insurers were not listed. The companies listed in this
second survey covered from 10,000 to over 11 million individuals, and review
services accounted for 2 to 100 percent of company revenues. The ten largest
firms are listed in Table 3-1.
Figure 3-1 charts the industry's growth, a growth vividly illustrated by the
experience of the Mayo Clinic (Mayo Clinic, 1988). In 1984, the Mayo Clinic
was dealing with only one utilization management programprecertification
for Medicare beneficiaries. Four years later it was working with approximately
1,000 utilization review plans. This does not equate to 1,000 review
organizations, however, because many review companies alter some details of
their programs to fit particular client preferences. (The Mayo Clinic, for
example, reported dealing with over 200 Blue Cross and Blue Shield review
programs, but there are fewer than 75 Blue Cross and Blue Shield organizations
in existence.) The American Hospital Association (1989) reports that hospitals
may deal with from 50 to 250 organizations doing prior and retrospective review.

TABLE 3-1 The 10 Largest General Service Utilization Review Firms

No. of Full-Time Staff
Company No. of Lives Total Physicians Registered Physicians
(Ownership) Serviced Nurses on Retainer
Intracorp 11,500,000 600 NA NA NA
(CIGNA Corp.)
Corporate 6,200,000 243 6 228 7
Health
Strategies
(Metropolitan
Life Insurance
Co.)
HealthCare 5,000,000 433 27 300 NA
COMPARE
Corp.
(independent)
Peer Review 4,000,000 45 7 16 65
Analysis, Inc.
(privately held)
Corporate 3,750,000 236 9 93 23
Health Care
Management
(EQUICOR,
Inc.)
Cost Care, Inc. 2,900,000 268 19 136 6
(independent)
Value Health, 2,800,000 119 5 9 7
Inc. (privately
held)
The 2,683,000 293 2 89 0
Sunderbrunch
Corp.
(privately held)
Western 2,000,000 40 1 22 50
Medical
Review
(nonprofit)
August 1,979,750 230 3 48 45
International
Corp.
(privately held)
NOTE: NA, not available.

SOURCE: Business Insurance (1989).
Some utilization management organizations are spin-offs and descendants

of the PSROs developed for Medicare after 1972. Many PSROs had private
clients, as do many of the successor PROs. Over three-quarters of the PROs
were listed in a 1987 compilation of private independent utilization review
companies (McGraw-Hill, 1987), although only a handful appeared in a 1989
directory compiled by Business Insurance (Business Insurance, 1989).
Insurance companies, which are coping with the loss of business that
resulted from employers' decisions to self-insure, offer utilization management
services both to control costs for underwritten accounts and to provide new
products for both insured and self-insured groups. From 1982 to 1986, the
percentage of Blue Cross plans reporting prior review programs

Figure 3-1 Founding years for utilization management organizations. All

organizations founded before 1970 were founded to perform other functions
such as insurance or claims administration. Some organizations founded later
are subsidiaries of older insurance companies or other organizations.
Regardless of founding data, most initiated utilization management services
after 1982. Source: Business Insurance (1988).
jumped from 28 to 95 percent (Scheffler et al., 1988). Third-party
administrators, which began by specializing in processing claims for self-
insured employers, have also diversified into prior review. In addition, a few
community coalitions are administering utilization management programs for
employers.
The strategies, lines of business, market niches, and sophistication of
organizations in the utilization management field are quite varied. Some
organizations specialize in a particular service, such as high-cost case
management or review of mental health services, whereas others offer

a broad range of services, including underwriting, claims administration, and

provider contracting. Certain organizations concentrate on developing software
packages for analyzing claims data and assessing the appropriateness of
services but actually administer programs for relatively few clients.
The relative youth of utilization management organizations, their limited
track records, and the inexperience of purchasers have created a small offshoot
industry: firms that review the reviewers. These firms provide consulting
services to employers and other purchasers. They also conduct independent
audits of hospital medical records and compare their results with those of
different review systems or organizations (Milstein et al., 1987).
The description of utilization management that follows is drawn largely
from a series of site visits during the summer of 1988 to 12 organizations
engaged in utilization management (see Appendix E for summaries of these
visits). The organizations visited were selected for variability, not because they
were believed to be exemplary. They included both industry leaders and
organizations that are not as well known. Site visit teams were made up of two
staff members and one or two members of the committee. The visits generally
consumed a full day and focused heavily on the details of the operation of
preadmission review and high-cost case management programs. The committee
asked the utilization management organizations that it visited for examples of
contracts with clients. Appendix F in this report presents a review of the
agreements that were submitted.
The organizations visited included three independent utilization
management companies, three insurance company subsidiaries, two third-party
administrators, two PROs engaged in private review, and two HMOs (one staff
model and one IPA). This mixture of organizations conveys something of the
variety in the utilization management field but not the extent of cross-
fertilization that is built into these organizations. The backgrounds of the
organizations and their founders includes HMOs, PSROs, the insurance
industry, claims administration, employee benefit management, disability
management, and academic research. These differences in origins and
backgrounds translate in complex and subtle ways into differences in corporate
cultures, data systems, computerization, attitudes toward physicians and
institutional providers, expectations about what utilization levels can be
achieved, and sensitivity to the employee relations issues that utilization
management can create for employers.
Although any simple categorizations inevitably oversimplify the situation,
it is helpful to define several basic ways in which the utilization management
function can relate to the functions of health care delivery, insurance, and
benefit plan administration. The following four categories of utilization
management are described below: (1) freestanding service,

(2) insurer-based service without provider contracts, (3) insurer-or broker-based

service with provider contracts, and (4) provider-based service.
In the first category, utilization management is carried out by an
organization that (1) does not have contracts or other formal relationships with
providers and (2) is not at risk for the costs of medical care services. The party
at risk may be an employer, an insurer, or even an HMO or PPO that contracts
with the organization for utilization management services. Even though the
review function may be freestanding, the organization itself may be a subsidiary
or an affiliate of an insurance company or a subdivision of a third-party
administrator rather than a completely independent organization. What
distinguishes these programs from those that are integrated into the insurance or
claims administration function is that they may be marketed as a freestanding
product to purchasers that use other companies for insurance or claims
administration or that perform those functions themselves.
The organizations that engage in utilization management as a freestanding
service have gained so much attention that they are often thought of as
constituting the utilization management industry. And, in fact, except for two
HMOs, all of the organizations visited in this studyincluding the PROsfit
into the freestanding category.
In the category of an insurer-based service without provider contracts,
utilization management is done by a party that is at risk for the costs of
additional medical services but has no contractual relationship with the
hospitals or physicians who are subject to its utilization management efforts.
Many insurers, acting in their insurance capacity,1 engage in utilization
management as a way of controlling claims costs, with the administrative
expense for the function built into their premiums. Alternatively, they may offer
prior review as a separate option to their customers, but they do not market their
service on a freestanding basis to clients whom they do not insure.
The third category of utilization management covers organizations that
establish and manage explicit networks or panels of providers. These
organizations include some HMOs, some PPOs, and some Blue Cross and Blue
Shield plans. The specific arrangements vary from highly integrated group
practice HMOs to PPOs and other models with weaker ties between the
providers and the broker or other entity that organized the network. The entity
that organizes and manages the network may act simply as a broker that does
not accept economic risk for the cost of services delivered by network
providers. Alternatively, the organization may accept risk, typically
1 The implied distinction arises bemuse insurance companies and Blue Cross and Blue
Shield plans may either act as an insurer (accepting actuarial risk) or as an entity that
provides only administrative services for self-insuring clients.
sharing it with the providers in the network. Generally, the provider agrees to
initiate the prior review process and to not bill the patient for any claim denied
for lack of medical necessity.
When an organization has the kinds of contractual relationships with
physicians described above, it has at least three types of potential influence over
physicians. The first lies in its control of continuing access to the network's
patients, which can be effective with doctors who need the patients furnished by
the network more than the network needs access to the services provided by that
doctor. Second, because the organization generally has the provider's agreement
not to bill patients when claims are denied for lack of medical necessity, the
threat of a payment denial can be significant. The organization does not have to
face unhappy patients whose bills are not being covered or unhappy clients
whose employees are complaining about denied claims. Third, the organization
can establish economizing or risk-sharing incentives, for example, capitation or
bonuses, as part of its compensation arrangements. (The commissioned paper
by Joan B. Trauner and Sybil Tilson in Appendix B describes utilization
management in HMOSs in more details.)
Provider-based utilization management, the fourth category, is undertaken
out of self-interest by hospitals or other health care organizations that have
formal relationships with physicians, for example, employment or admitting
privileges. One incentive for health care organizations to undertake utilization
management is to reduce the potential for retrospective denial of payment for
services judged medically unnecessary. In addition, hospitals paid on a
prospective per-case basis have an incentive to minimize costs once a patient is
admitted. Institutions may also undertake utilization management as an
alternative to having the function performed by an external organization, as may
happen with some HMOs or PPOs.
For provider-based utilization management, effectiveness may rest less on
the threat of payment denial than on administrative pressure, a perceived
mutuality of interest, and the risk that the institution might sever its relationship
with the physician, for example, by not renewing an employment contract or by
withdrawing admitting privileges. Although several payers (most importantly,
Medicare) provide institutions with economic incentives to minimize
unnecessary days of care, little documentation is available on the extent to
which health care institutions actually use the methods of utilization
management (Burda, 1989; Project HOPE, 1987).
The committee notes that those who consider the full research and policy
questions raised by utilization management must recognize the existence of
provider-based utilization management. The committee, however, has not
focused on the utilization management activities of institutional providers of
health care for several reasons. Most hospitals have been responsible for
utilization review and discharge planning for many years, so hospital-based
prior review is a less dramatic change from past practices

than are programs undertaken by external organizations. Institutional providers

also face comparatively small changes in their legal responsibilities for patient
care, which exists independently of any prior review programs.
In theory, the differences among the various organizational contexts of
utilization management described above could have important implications for
an organization's efficiency and effectiveness, its sensitivity to the complexities
and costs created for patients and physicians, its ability to assess the impact of
its activities, and the extent to which it is exposed to legal risk. Unfortunately,
as described in Chapter 4, virtually no research exists on the effects of these
organizational differences. As Appendix B points out, this also holds true for
many organizational differences among HMOs.
Even though the organizations visited do not represent the full range of
organizations that carry out utilization management activities, the committee
found many noteworthy variations in the ways that utilization management is
designed and implemented. The description that follows covers several areas of
variation in prior review processes. The short history and rapid evolution of the
industry and the flux in employer-sponsored health benefits make it impossible
to predict how long this description will hold.
THE REALITIES OF PRIOR REVIEW: HOW IS IT

ACTUALLY DONE?
"We have the same things that everyone else has ... a book, a WATS line, and
nurses."2
Despite certain basic similarities implied by this quote and by the
definitions offered in Chapter 1, each utilization management organization's
programs are a mixture of invention, imitation, borrowing, and adaptation by an
organization that has its own history, market niche, and strengths and
weaknesses. Moreover, the industry is subject to little explicit regulation and
has not developed any extensive form of self-regulation during its relatively
brief existence. For these reasons, it is not surprising to find variations in how
such "standard" methods as preadmission and continued-stay review are carried
out by different organizations. Even the PROs that contract with Medicare,
which are subject to very detailed oversight and performance specifications
from the federal government, vary a great deal in their organization and
activities (General Accounting Office, 1988).
2 Robert E. Becker, M.D., founder of HealthCare COMPARE Corp., at the June 6,
1988, hearing of the Committee on Utilization Management by Third Parties. Dr. Becker
followed this characterization with his views on why his company is, in fact, different.
Likewise, the means used by HMOs to influence patient care decisions

vary widely, as described in Appendix B of this report.
To date, little or no documentation exists regarding the nature, extent, or
consequences of such variations. This report attempts to correct some of that
descriptive void, but the committee's charter did not and could not include
empirical research on the impact of organizational differences.
Certain elements appear to be common to all or most programs in which
prior authorization or certification must be obtained from an external
organization regarding the necessity or appropriateness (different programs use
different terms) of hospitalization or other services. First, all programs offer
some or all of the prior review and high-cost case management activities
described in Chapter 1.
Second, all rely on telephone rather than face-to-face contact with the
patient or physician. The major exception is that some programs for continued-
stay review programs use review nurses who are employed by the utilization
management organization but who work on-site in the health care facilities that
are subject to review. This arrangement was typical of PSROs and continues to
be common in PROs. It is also used by HMOs, particularly those that have
contractual relationships with a limited number of hospitals. In addition, for
admissions planned weeks in advance, written forms may be used, with the
telephone used for follow-up as needed.
Third, programs use a two-stage assessment process. In the first stage, the
review organization obtains information about the patient and the proposed
services; services that pass the screening criteria are certified as necessary or
appropriate. This first stage is handled by nurses in most programs, although
experienced medical secretaries perform this function in one program that the
committee visited. Services that do not pass the screening criteria pass to the
second stage, which involves physician review. Nurses may authorize services,
but only physicians may deny authorization.
Fourth, the current focus of prior review is on the appropriateness of the
site of care, the timing of an admission, or the duration of care. A few
organizations examine the need for specific procedures, but most currently do
not. During preadmission review, the cost of alternative care is not evaluated on
a case-by-case basis. (As described in Chapter 5, high-cost case management
includes such an assessment.)
The typical steps in the prior review process are described in Figure 3-2.
Beyond these elements, organizations vary widely, as described below.
How Prior Review Is Integrated with Other Administrative

Functions
Because utilization management is done by organizations with different
purchaser and provider relationships, there are significant differences in the
extent to which the prior review function is integrated with other

Figure 3-2 Typical basic steps in the prior review process.

aspects of benefit plan administration, such as determination of whether a

proposed or alternative service falls in a category covered by a patient's benefit
plan. Some organizations have no information about benefit plan provisions and
thus can make judgments only about the appropriateness of proposed services,
not whether a general category of care is covered for a particular patient. Such
organizations typically tell the patient or the provider that they must find out
from the employer, insurer, or claims administrator whether or not the benefit
plan covers the particular services. Even organizations that have information
about benefit plan provisions do not always have on-line information about
whether a particular patient is still eligible for benefits or has left the employer,
been dropped as a dependent, or joined another health plan.
Moreover, some review organizations have no connection with the claims
payment function and possess little or no information about the actual use of
services. Thus, they can report only on their activities (for example, the
numbers of admissions requested and authorized), not on services actually used
or paid for. They also cannot readily determine whether adherence to utilization
management requirements is being checked during the claims payment process.
At the other extreme are organizations that do both utilization management
and claims payment and that have fully integrated data systems. Such
integration provides greater certainty that unauthorized services are not paid for
inadvertently, a problem that may occur more often when one organization does
utilization management and another pays claims. It also makes possible the
generation of reports based on actual claims experience. Ready access to
particular patients' claims histories, which some organizations possess, may
permit more sophisticated judgments about the appropriateness of proposed
treatment.
For example, a third-party administrator that the committee visited had a
system in which the patient's claims history appeared automatically on the
computer screen when the nurse reviewer began to enter data during a telephone
request for prior authorization. The example that was demonstrated involved a
child for whom a tonsillectomy was proposeda procedure that was subject to
a ''focused second opinion'' program under the insured's health benefits. The
child's claims history showed that a series of office visits for tonsillitis had
occurred over the previous year. Since this suggested a recurrent problem that
was not responding to less intense treatment, the second-opinion requirement
was waived because it was likely to concur. This saved the family some
inconvenience and the purchaser the cost of the second opinion.
Systems that integrate utilization management with claims information
also facilitate identification of potential cases for high-cost case management,
for example, by flagging repeat hospitalizations. Integrated systems

also make possible, at least in theory, a kind of outcome measurement, because

future services will show up in additional claims. Although none of the
organizations that were visited reported doing outcome studies, such
assessments are not far beyond the capacity of several of them.
Another form of data integration takes place at some utilization
management organizations whose activities are highly concentrated in a
particular geographic area. PROs, Medicaid programs, Blue Cross and Blue
Shield plans, HMOs, and PPOs operate largely in particular states or
metropolitan areas, whereas the larger commercial insurance companies, third-
party administrators, and independent review organizations often do business
nationally and deal with a much more dispersed group of providers. The
geographic concentration of a relatively high volume of business provides an
opportunity, if data systems are properly organized, to accumulate considerable
experience with particular hospitals and physicians and to better target review
activities on problem services or providers.
None of the review organizations visited in this study confined their prior
review or case management programs to providers who had been identified as
problems. However, focusing on problem providers has been an important
feature of the retrospective review activities of the PSRO, PRO, and carrier
programs conducted for Medicare. One PRO visited by the committee reported
that they had tried unsuccessfully to interest their private customers in a focused
precertification program.
How Basic Logistical Matters Are Handled

For utilization management organizations, like most service companies,
the ways in which basic operations are handled can have a major effect on
relations with clients and suppliers. Purchasers often insist on visiting a review
organization in an effort to assess how well the program is run, and many ask
for periodic reports on service levels, for example, the percentage of telephone
calls answered within five rings. Although the impact of logistical differences
on benefit costs cannot now be documented empirically, the committee found a
number of plan characteristics that are widely regarded as important.
Initiating Review
For organizations that do not have contracts with providers (and for some
that do), the enrollee is usually responsible for initiating the review process.
Otherwise, the enrollee may incur a financial penalty. One of the biggest
challenges for purchasers and suppliers of utilization management services is to
educate employees about this responsibility. Mere discussion in a benefits
handbook is generally regarded as insufficient, and an array of additional
information efforts are often attempted. These include

highlighted notices on benefit plan identification cards, discussions during new

employee orientation, and information letters that must be returned to the
personnel office with a signed acknowledgment of receipt from the employee.3
As a practical matter, physician and hospital offices often make the initial
calls to the utilization management organization since they generally end up
discussing the case with the reviewer anyway and also want to reduce the risk
of a denied claim or bad debt. For many years most hospitals have had standard
routines to check a patient's insurance coverage before admission, and contact
with review organizations can be incorporated into this process. Some
utilization management firms say that they try to obtain all the information they
need from the patient or the patient's family, but many make it a practice to talk
to the physician's office to confirm the information, obtain additional
information, and describe directly their review policies or guidelines.
Handling Telephone Calls

In some utilization management organizations, the calls from patients or
providers are answered centrally and passed along to the first available review
nurse. In other organizations, review nurses answer calls directly. Before the
actual admission takes place, as many as three or four telephone conversations
may take place, since review organizations may be notified of a pending
admission by more than one party and may have to talk to the physician's office
more than once, particularly if the proposed services do not pass the
organization's screens for certification.
How quickly the caller gets through to the correct party will depend upon
the organization's staffing levels, phone system, and other factors. Some firms
with clients nationwide provide a single toll-free 800 number, but calls are
automatically routed to different regional offices, depending on where the calls
originate. Many but not all review organizations have computerized telephone
systems that allow supervisors to monitor telephone waiting time and track how
often callers encounter busy signals or hang up because the phone was not
answered after several rings. Attention to these matters varies enormously from
organization to organization. For example, some organizations lack the
equipment for systematic telephone
3 For example, the identification card for the indemnity health plan offered by the
National Academy of Sciences includes one warning of special plan review requirements
on the front of the card (in red capital letters) and two warnings (in boldface capital
letters) at the top and the bottom of the card's reverse side. The reverse side also includes
brief summaries of the plan requirements and relevant phone numbers.
monitoring, whereas others not only have the equipment but also have specific
performance standards and incentives. In general, sophisticated phone systems
are a high priority for major utilization management organizations.
Computerization
The degree of computerization also varies greatly across utilization
management organizations. At one extreme there are organizations that use
computers off-line only for recordkeeping purposes. Some organizations have
gone to this approach after finding that their software was too slow to be
efficient for data entry while the patient or provider was on the telephone. Since
flexible and fast software can be very expensive to develop or license, some
organizations have been slow to upgrade their systems. At the other extreme are
organizations that use networked computers into which are built (1) questions
for patients or providers, (2) decision protocols and criteria to assess the
information and determine whether to certify care or refer it for further review,
(3) details of client benefit plan limitations, (4) individual claims histories, (5)
automatic diagnosis and procedure coding, (6) diagnosis or procedure-specific
length-of-stay norms, and (7) addresses for practitioners and institutional
providers. These systems may also allow the automatic preparation of letters to
patients, physicians, and hospitals to confirm the results of the review. These
aspects of computerization affect not only the comparative efficiencies of
different utilization management programs but also the kinds of analyses and
reports that can be generated.
Because multiple calls involving multiple parties may take place, logistics
can be important for the patient or provider. In a highly computerized
organization, any nurse reviewer or physician adviser can take up a case exactly
where it was left on a previous call. Organizations that use pencil and paper or
inflexible telephone and computer systems may require the person who took the
initial call to take follow-up calls. Since nurse reviewers and physician advisers
spend much of their time on the telephone, this means that the hospitals,
doctors, or patients who deal with the latter type of organization may have to
make many calls before they reach the right individual. Because this can be
time-consuming, physicians often turn the function over to employees. Thus,
the prior review process often becomes a discussion between a nurse reviewer
at the utilization management organization and a secretary or nurse in the
physician's office.
How Nurse Reviewers Work

Nurses play an integral role in most prior authorization programs. This role
usually includes obtaining all requisite information about the patient and
proposed treatment, entering the information into the computer,

assigning diagnosis and procedure codes, reviewing the proposed treatment

against the organization's criteria of appropriateness, and conveying approval
for treatment that conforms to those criteria. Two organizations departed from
this pattern. One uses experienced medical secretaries in this role. The other
gives staff physiciansassisted by computerized screening criteriathe
responsibility for reviewing treatment against approval criteria and conveying
approval.
Several organizations report that nurse reviewers handle 40-50 calls per
day, doing anywhere from 10 to 20 certifications (certification may require
several calls to complete). In most organizations, nurse reviewers do other
functions, such as high-cost case management and answering inquiries from
beneficiaries. Some organizations assign those functions to a separate group or
groups, and others rotate nurses through different functions.
Even in the organizations that use nurses for apparently similar functions,
the nature of their authority varies substantially. No organization visited gives
nurses the authority to refuse prior certification of care proposed by attending
physicians, but most give nurses the authority to try to negotiate conformity
with review criteria. However, two organizations instruct nurses not to negotiate
with attending physicians or their employees; anything that does not fit the
organization's criteria of appropriateness must be referred to a physician
consultant. Even in these organizations, the nurse reviewer's questions may
prompt the physician to modify a proposed plan of care.
Organizations also vary in whether they direct nurse reviewers not to
coach attending physicians on how to gain approval of proposed care. In one
organization that the committee visited, the attitude is that attending physicians
do not always know the right language to be used to gain approval of an
appropriate admission. If nurse reviewers' believe that a patient probably needs
to be admitted, they can give the attending physician's office the cues needed to
gain approval. Other organizations explicitly instruct nurse reviewers not to do
this.
Another area in which nurse reviewers' authority varies from organization
to organization is whether they are permitted to make exceptions to review
criteria on the basis of their clinical judgment. At least two organizations visited
permit review nurses to certify an admission that does not meet review criteria
if, in their judgment, it is nevertheless appropriate. Many utilization
management organizations do not permit this and require referral of all such
cases for review by physicians.
Most organizations prefer to hire nurse reviewers who are registered nurses
with at least 5 years of experience; however, one required only one year of
experience. Most organizations also prefer reviewers with hospital experience.
One mentioned experience in utilization review in hospitals

as being important; another prefered nurse reviewers to have emergency room

or intensive care unit experience.
The training of new nurse reviewers can be quite variable. Some are
trained in one-to-one sessions with a supervisor; others go through formal
training periods that last anywhere from 2 to 6 weeks. Most organizations report
that after some initial loss of nurses who simply do not like the work, there is
very little turnover. Recently, one national association implemented a training
and monitoring program to instruct nurse reviewers how to analyze and listen to
patient or provider requests and calls, explain insurance plan requirements, pose
questions effectively, and manage difficult calls ("New Blues Program to Train
Nurse Reviewers," 1989).
How nurse reviewers are monitored also varies. Supervisors in most
organizations are responsible for anywhere from 6 to 12 nurse reviewers, but in
a few organizations supervision is much less close. One PRO that was visited
had 76 nurses and one supervisor. Monitoring ranges from informal oversight
done by walking around and overhearing the nurse's end of conversations to
very rigorous and systematic monitoring. Several organizations monitor
telephone activity by using a computer (which can compute such things as the
average length per call and number of calls per day) and by silently listening in
on a sample of calls. Several organizations also systematically review a sample
of cases handled by nurse reviewers by looking at the appropriateness and
consistency of decisions made and the accuracy of coding. In a few
organizations, the physicians who receive cases on referral from nurse
reviewers play a role in monitoring their performance.
Several of the organizations visited provide incentives for accuracy and
productivity. When asked, none reported attaching any rewards to such
measures as days or dollars saved or utilization levels achieved.
What Role Do Physician Advisers Play?

All utilization management organizations that were visited refer cases that
fail screens to physician advisers. These physician advisers are expected to use
their clinical judgment and experience to determine whether the services should
be certified as medically appropriate. One organization that was visited noted
that under its state's law, this function could be construed to be the practice of
medicine and thus must be carried out by a physician who is licensed to practice
in the state. All organizations visited said that their physician advisers are not
bound by the organization's criteria or rules about what services are appropriate.
This is quite consistent with the logic of the entire processotherwise,
decisions not to certify services could be made by computers. Thus, the
physician adviser's role is crucial.

However, responsibilities and work arrangements vary in some important

ways across organizations.
One important source of variation involves the nature and sources of the
information on which the physician advisers base their decisions. In most of the
organizations visited, the physician advisers call the attending physician to
discuss cases that the nurse reviewers cannot certify. Companies say that
physician advisers have excellent success in establishing direct contact with
attending physicians. One reason for this is that, in at least some companies, the
physician adviser calls the patient's physician's office directly, announces
himself as Dr. X without disclosing the purpose of the call, and asks to speak
with the attending physician. (The committee's site visitors were unable to
ascertain physicians' reactions to receiving a call from another physician and
finding out when they come on the line that the calling physician was from a
review company.) Companies also say that by approaching attending physicians
as colleagues, physician advisers, are often able to obtain important additional
information that was not provided to the nurse reviewer, and they can work out
a satisfactory agreement with the attending physicians about the services that
will be certified as appropriate.
Some utilization management organizations, however, use their physician
advisers in quite a different way. In a pattern that may be particularly
characteristic of PROs, some physician advisers have no contact with attending
physicians. Ordinarily, they make their decisions based only on the information
obtained by the nurse reviewers and then convey their results to these nurse
reviewers, who carry on from there.
The relationship between physician advisers and utilization management
organizations varies. Four arrangements were seen during the site visits:
The medical director and physician advisers are full-time employees of the
utilization management organization.
The medical director and physician advisers are part-time employees and
part-time practitioners.
The medical director is employed by the utilization management
organization, and physician advisers are outside consultants.
The medical director and physician advisers are outside consultants.
In the first arrangement, the medical director and physician advisers are
full-time employees who work from the utilization management organization's
offices. This requires a certain volume of business to be practical, but it
provides advantages in standardizing operations such as recordkeeping,
monitoring performance, and developing a corporate culture. A potential danger
is that the corporate culture will include a significant degree of cynicism about
the abilities and motivations of practicing physicians. Also,

this type of arrangement is inconsistent with a strongly held, though

nonvalidated, belief of practicing physicians (and some people in the utilization
management business) that physicians who leave practice to work as reviewers
lose touch with important realities of medical practice. The opposing argument,
also unverified, is that physicians who work for a large review firm deal with
thousands of clinical situations each year and therefore not only keep current
but have a broader and deeper understanding of what goes on in medical
practice (such as variations among practitioners).
The second arrangement, the use of reviewers who are part-time
employees and part-time practitioners, appears to combine the advantages of the
first arrangement while avoiding the criticism about the loss of touch with the
realities of medical practice. Of course, the use of part-time physician advisers
is a less stable situation, since their practices may grow or they may assume full-
time responsibilities elsewhere. However, with the growing physician supply
nationally, review organizations in many areas of the country are able to find
practitioners who can commit a certain number of hours per week to review. By
law, physician advisers for PROs must be in active practice.
The third arrangement involves the use of an in-house medical director
who is responsible for the organization's medical decision-making and consults,
as needed, with a panel of practicing physicians in particular specialties, most of
whom work from their own offices. The medical director may seek advice from
these advisers or may refer to them an occasional case that they need to deal
with fully, including discussion with the attending physician. Because most
decisions and most feedback from practicing physicians pass through a single
individual, this arrangement may provide for a comparatively high degree of
consistency in the organization's decisions. However, the arrangement also has
dangers if the medical director's judgment is idiosyncratic and not balanced by
other on-site physicians.
The final arrangement, which was observed in several organizations and
may be the most common of all, particularly in small organizations, involves
the use of physician advisers (and even medical directors) who do their work
for the utilization management firm from their own offices. Although this is
probably the least expensive way to organize the function, it has several
potential weaknesses. It can be difficult to monitor reviewers' performance,
maintain consistency, and develop a corporate culture. Some organizations say
that it can be difficult to reach a physician adviser who has a busy practice, to
get him or her to attend to cases in a timely fashion, and to produce the
necessary documentation. Some organizations have found that physicians who
agree to be on their panel are reluctant to call the attending physicians or to
raise the necessary questions with them. This reluctance may be a source of the
delays in the review process complained of by some attending physicians.

Variations among utilization management organizations in the use of local

versus nonlocal medical advisers also have important implications. Again,
several arrangements exist. First, there are organizations that operate in a
relatively confined geographic area, like PROs, and that use local physicians as
reviewers. These reviewers may be more reluctant to challenge their peers
because of social or professional contacts.
A second variation is for an organization to do prior review on a national
basis but draw physician advisers from the areas where the firm's offices are
located. (An unusual version of this model involves a large insurance company
that uses its own nurse reviewers but contracts with organizations located
elsewhere for the services of physician advisers.) During the site visits, such
organizations did not complain of having difficulty in getting their physician
advisers to make the necessary phone calls. On the other hand, this arrangement
can lead to problems when practice standards that are commonplace where the
reviewers are located differ from the standards where the attending physician is
located. For instance, several organizations reported substantial regional
differences in whether surgeons will admit patients on the morning of rather
than the day before surgery or whether they will perform tonsillectomies on an
outpatient basis.
Probably the least common situation is for a company to do business over
a large geographic area with little or no use of local physician advisers. In part,
this approach may reflect the logistical difficulties that a review organization
can face in developing and monitoring physician advisers scattered around a
larger region. In one state that the committee visited, an entrepreneurial group
of physicians put together a statewide network of physicians who were wining
to do reviews on a contract basis. It can ensure that the physician reviewer who
is assigned to a case is from a different part of the state than the attending
physician.
There are also major differences among utilization management
organizations in how much physicians are involved in prior review. On the low
side, the proportion of cases that go to physician advisers for review runs about
1 or 2 percent in two organizations visited and about 10 to 15 percent in five
organizations visited. In contrast, 40 percent of cases go to physician advisers in
one of the oldest and best established firms, and another firm advertises 100
percent physician review.
Given these differences, it is not surprising that caseloads for physician
advisers vary enormously from organization to organization. The physician
advisers who work on-site at one organization handle about 10 cases per full-
time reviewer per day, whereas those at another organization do 200-250
certifications per day. Note that for a hypothetical 6-hour "review day" the latter
rate appears to allow a physician, on average, less than 2 minutes per review.
This suggests that many cases may be just scanned and signed in such
organizations.

Except for the organizations in which the physicians' determinations are

based purely upon a written or computerized record, all organizations report
that the process of physician review is collegial. Rarely do the attending
physician and the reviewing physician fail to reach agreement about a case.
Most organizations report that denials of certification are issued in fewer than 1
or 2 percent of all cases, but a much larger proportion of cases involve some
modification of proposed services at some stage in the review process. By way
of comparison, based on both prospective and retrospective review, PROs deny
payment for about 2 percent of all Medicare admissions (PPRC, 1989).
In hiring physician advisers, most organizations that the committee visited
state that the main credentials they seek are board certification and experience
in practice. Some organizations mention additional factors, such as professional
reputation or experience in utilization review. Investigations of qualifications do
not appear to be extensive. One organization recruits some reviewers from
among the physicians with whom they have contact as subjects of review.
The training of physician advisers is minimalfrom 1 day to several days
and performance monitoring varies greatly. In some organizations, up to 10
percent of each reviewer's cases are reviewed either by the medical director or a
quality assurance committee. Alternatively, a sample of cases may be
exchanged among reviewers for mutual review. One organization uses an
outside panel to review a sample of its reviewers' decisions. In contrast, some
organizations do not engage in such systematic oversight. They may only
review cases in which an attending physician appeals the physician adviser's
decision. One organization states that its main review of its physician advisers'
performance was by the nurse reviewers to whom the decisions are conveyed.
Physician advisers are paid in several different ways, reflecting the
diversity of their arrangements with the review organizations. The basis of
payment seen in the site visits included payment per call made, per case, per
hour, per month, and by an annual salary. Although explicit or implicit
productivity incentives are common, none of the organizations that were visited
indicated that they tie payments to physician advisers to such measures as
denial rates or the amount of money that they save the purchaser. Virtually all
of the organizations visited report little turnover among their physician advisers.
To summarize, the utilization management physicians in whose hands rest
decisions about the necessity or appropriateness of services
may be full-time or part-time employees of either the utilization
management organization or a subcontractor;
may or may not have an active medical practice;

may work from their own offices or from the review organization's office;
may review only local cases, only cases from other regions of the country,
or both;
may deal directly with attending physicians or may rely completely upon
information assembled by nurse reviewers; and
may review as few as 10 or as many as 200 cases per day.
These differences may well affect such matters as the reviewers' toughness
in trying to eliminate inappropriate services, their sympathy with the position of
the attending physician, the efficiency of the program, and the consistency of
the organization's decisions.
What Style Is Used with Attending Physicians?

Utilization management organizations generally state a preference for
negotiating agreement rather than attempting to impose decisions on attending
physicians. This preference stems, in part, from reluctance among review
organizations to risk legal liability for patient care decisions, but it also may
reflect client wishes. In addition, employers who purchase utilization
management services vary in their desire to maximize the economizing effects
of these activities and in their willingness to risk employee relations problems
by backing up negative review decisions. Numerous factors may contribute to
these variations, including corporate culture, the composition of the labor force,
and competitive pressures. Companies that have an expanding market and a
highly educated labor force may put quite different demands on a review
organization than may a company whose profits are threatened by health care
costs and whose labor force is poorly educated and decentralized.
One organization, most of whose clients had unionized labor forces,
particularly stresses its practice of trying hard to convince attending physicians
to accept its determinations of appropriateness or agree to a negotiated
alternative. They say they want to avoid putting patients in the middle between
the provider and the payer. Several organizations stated that some clients want
to be contacted before denial letters are issued, presumably so they can make
case-by-case decisions about whether to risk employee relations problems.
However, other organizations take pride in backing their reviewers'
determinations and believe that their clients expect them to refuse payment
authorization for inappropriate services. The impact of such a policy has not
been empirically compared with more lenient policies.

What Criteria Are Used to Assess Care?

The results of prior review of medical services by utilization management
organizations are shaped by criteriaan imprecise term employed here to refer
to the written policies, decision rules, or guides used to evaluate what services
are appropriate or necessary for patients in different clinical circumstances
(Donabedian, 1981). Organizations refer to several different types of criteria,
although the boundaries between categories are not necessarily well-defined.
Lists of Procedures
Most utilization management organizations have lists of procedures that
are used to target services for assessment. The most common lists involve (1)
surgical or diagnostic. procedures for which payment is ordinarily authorized
only if the procedure is done on an outpatient basis without an overnight stay in
the hospital and (2) surgical procedures or other services for which a second
opinion may be required. These lists may be included in the benefit plan
information provided to patients to alert them to the general policies of the
utilization management program. The utilization management organization then
uses the lists in evaluating proposed inpatient services, a step that relieves the
patient of keeping track of a long procedure list. (Depending on whether the
benefit plan includes preprocedure review, the patient may still need to be
aware of outpatient procedures for which second opinions are required.) To
consider whether an admission might be warranted for a target procedure, the
reviewer relics on exceptions criteria.
Exceptions Criteria
Exceptions criteria specify the circumstances under which certain
utilization management policies or requirements might be waived. For
procedures that are ordinarily certified only as outpatient procedures, exceptions
might include patients with a history of bleeding disorders or heart disease.
Also common are defined exceptions to the general rule against authorizing a
patient's admission to the hospital prior to the day of surgery. Both of these
types of exceptions criteria tend to be generic, and they are applied without
reference to the particular diagnosis or procedure for which an admission is
proposed.
Hospitalization Criteria
Utilization management organizations use a variety of criteria to determine
whether an individual needs inpatient care. Most are based on criteria that were
originally developed for retrospective review in the PSRO

program or for research on unnecessary use of the hospital (Payne, 1987). These
criteria have been refined and are marketed in hard copy or computer disk
versions by a handful of major vendors for use in both prospective and
retrospective reviews.
Most of the review organizations the committee visited base their
hospitalization criteria either on the appropriateness evaluation protocol (AEP)
or on the ISD-A criteria set (intensity of services, severity of illness, discharge,
and appropriateness screens). These tools, which were originally developed for
retrospective research on the extent of inappropriate hospital use, have been
adapted for prospective use to discourage inappropriate hospitalization
(Restuccia, 1986). A 1987 study of PROs indicated that the AEP and ISD-A
criteria are also widely used by PROs (Project HOPE, 1987).
The AEP criteria for adult medical/surgical services and for pediatric and
elective surgery, which were initially prepared with federal financial support in
the late 1970s, are in the public domain and have been the subject of several
studies that have indicated that they are reliable and valid (Gertman and
Restuccia, 1981; Strumwasser et al., 1987). More recent work on the AEP has
been proprietary, although it is readily available and relatively inexpensive. The
ISD-A criteria are proprietary (InterQual, Inc., 1987), and no specific reports on
their reliability appear to have been published.
The AEP, which has been revised and expanded since it was developed in
the mid-1970s, is comprised of lists of symptoms (such as sudden acute loss of
vision) or services (such as the need for vital signs monitoring every 2 hours or
more often). The lists are specific to major clinical services (for example,
surgery and pediatrics) and are designed so that any one symptom or service
suffices to justify a day in the hospital. Table 3-2 includes an excerpt from a
recent training manual describing criteria for determining when inpatient
surgery is warranted based on respiratory status. The ISD-A criteria operate by
a logic similar to that of the AEP but are specific to different body systems.
Most organizations purport to use a modified version of whichever criteria
set they have adopted. The committee cannot judge whether these modifications
are designed to make the criteria tougher, more appropriate locally, or easier for
reviewers to use.
To explore variations in two specific areas, site visitors inquired about the
circumstances that would warrant certification of admission to the hospital for
combined tonsillectomy and adenoidectomy and for nonsurgical back pain.
Some organizations consider their criteria and guidelines to be proprietary, but
all responded to these specific questions. As summarized in Table 3-3, the
responses varied considerably in content and specificity. It should be noted,
however, that tonsillectomy and adenoidectomy and lower

TABLE 3-2 Excerpt from the Surgery Appropriateness Evaluation Protocol Reviewer's
Manual
V. CRITERIA OF ADMISSIONLOCATION OF SURGERY/RISK FACTOR
ASSESSMENT
This section includes those factors felt to identify patients and procedures requiring that
surgery be performed on an inpatient basis rather than on an ambulatory or day of surgery
admission basis. These criteria for determining location fall into three categories: (1)
comorbidity (the presence of concurrent medical problems which place the patient at
special risk, no matter what kind of surgery is to be done), (2) potential for complications
post-operatively, and (3) the need for intensive post-operative care. If any one of these
criteria is met, inpatient admission for the surgical procedure is deemed appropriate.
Conversely, if none of the criteria are satisfied, inpatient admission is not justified. The
reviewer, however, does have the option to override the criteria in either direction (see
pages 9-10).
Comorbidity
The diagnoses or measurements in this section must be documented as being present and
the data (whether test results or clinical observations) should be from within the past four
months prior to admission. Do not use ''suspected'' diagnoses.
A. Respiratory Status
1. Significantly abnormal pulmonary function measurements
Independent of what particular respiratory disease is present any one
of the following four measures of pulmonary function will suffice:
a. Functional Vital Capacity (FVC) of < 1.0 liters
b. Forced Expiratory Volume in the First Second
(FEV1) of < 50% FVC
c. Arterial pCO2 > mmHg
This test must have been done while the patient
was breathing room air, not an atmosphere higher
in oxygen concentration.
d. Arterial pO2 < 50 mmHg
This test must have been done
while the patient was breathing
room air, not an atmosphere
higher in oxygen concentration.
2. Sleep apnea.
Documented as present regardless of how severe or frequent
SOURCE: Reprinted with permission from Restuccia (1986). Copyright 1986 by Restuccia.
back pain were picked deliberately because the committee knew they were
particularly likely to show variation.
Length-Of-Stay Norms
To assess proposed hospital lengths of stays, most utilization management
organizations use statistical norms based either on data published by the
Commission on Professional and Hospital Activities (CPHA) or on data
compiled by the review organization itself. These statistics are generally
disaggregated by geographic region and diagnosis. Most of the organizations
visited base their continued-stay review on the CPHA's average length of stay
for the CPHA Western region, where lengths of stay are shortest. Many review
organizations do continued-stay review either on the

TABLE 3-3 Circumstances Warranting Hospital Admission as Identified by Various

Organizations Visited by the Committee
Site Tonsillectomy and Adenoidectomy
1 Distance from home to facility; complications
2 Physician insistence on admission
3 Complications
4 Age; comorbidities
5 Physician insistence; high risk; 2 hours or more of travel time
6 Bleeding disorders; distance
7 Physician request for admission
8 Patient's history
9 None specified
10 One-night stay permitted
12 No special conditions required
Site Lower Back Pain
1 Patient requires intramuscular injections for pain control; inability to move
2 Need for intramuscular injections for pain control; motor defects
3 Pain; complications
4 Severe pain; need for traction
5 Three days of traction allowed for pain
6 ISD-A criteria
7 Failure of outpatient treatment
8 Neurological deficit
9 If physician requests, having tried outpatient treatment first
10 Physician insistence on admission
12 Rare, only for intensity of service
day by which 50 percent of patients with a given diagnosis are discharged

in the Western CPHA region or on the day before that. However, one HMO
visited initiates continued-stay review on the day when, on average, the first 10
percent of patients are or have been discharged. Most of the organizations
visited do not use the length-of-stay norms to certify a specific length of stay as
appropriate but, instead, use them as guides for scheduling continued-stay
reviews. They then assess whether a patient is ready for discharge based on
AEP, ISD-A, or similar criteria. Some organizations do not employ statistical
norms for scheduling continued-stay review but rather check back on a regular
basis, for example, every third day after admission.
Thus, although some organizations still tell providers and patients how
many days of stay they can certify at the outset, others simply state when they
will check back to see how the patient is doing. Two reasons have been cited for
this practice. First, when a particular length of stay is certified, physicians and
patients may interpret it as a fixed limit, not a guide. Patients who do not
understand the process may get anxious when they know that 3

days of stay have been authorized and then their physician tells them on the
third day that they are not ready for discharge. Second, some physicians may
regard a statistical norm as a minimum length of stay or may focus on the
length-of-stay numbers rather than on the justification for continued
hospitalization.
Necessity of a Procedure
The least commonly used criteria at this time are those designed to assess
whether a specific procedure (usually surgical, but sometimes diagnostic) is
appropriate for a particular patient. Although considerable research evidence
has accumulated to suggest that certain procedures insertion of pacemakers,
carotid endarterectomies, and othersare used much more commonly than is
medically justifiable, utilization management organizations have moved
cautiously to require prior review of procedures based on medical necessity
grounds. PROs are now required to review the need for care on a prospective
basis for several procedures (see Appendix c). Fear of legal liability
undoubtedly contributes to the reluctance of other organizations to accept the
responsibility that such determinations would entail. Nevertheless, a growing
number of programs are expanding or planning to expand in that direction.
The most commonly mentioned criteria for this kind of review involve the
medical review system developed by former Rand Corporation researchers
(Value Health Sciences, Inc., 1989). In this system, the review protocols for
procedures like tympanotomy tube insertion or knee arthroscopy involve a
series of questions on such matters as the chronicity of the patient's problem,
prior treatment by drugs or specified nonsurgical options, and physical findings.
The answers are evaluated through scaling and other techniques to determine
whether the proposed care is indicated (Value Health Sciences, Inc., 1989).
At present, it appears that most organizations that make medical necessity
judgments combine prior review with provisions for a nonbinding second-
opinion in cases in which the clinical indications for a procedure are dubious. In
certain cases in which clinical indications for a specific procedure are
nonexistent (based on information provided to the reviewer), some review
organizations say they now refuse to certify the necessity of the services.
How Criteria Are Adopted and Modified

Many approaches are used to adopt or modify criteria, but virtually all
involve the organization's medical director and physician advisers (or a
committee thereof), who are supplemented in some cases by an external
advisory group. Some organizations describe a fairly formal and extensive

internal process for developing criteria that includes using systematic review of
the clinical literature on a topic and consensus panels. However, criteria are
often borrowed, licensed, or adapted from outside sources or are developed less
formally through discussions among physician advisers.
Utilization management organizations generally report that their criteria
are in a continuous process of modification, although some also schedule
periodic reviews for particular sets of criteria. The process of modification can
be rather informal in smaller organizations. In large organizations in which
many different nurses and physicians use the criteria, the process is necessarily
more formal.
The decision to revise a criterion may be prompted in several ways.
Members of the medical staff may raise an issue because of differences among
themselves over how certain situations are being handled, because they are
encountering a great deal of resistance from attending physicians, or because
new information has come to light in medical literature or conferences. In
addition, pressure from customers to reduce medical care utilization may
prompt a new examination of criteria to search for areas in which additional
savings are possible.
Several organizations mentioned that they revise criteria based on their
own utilization data. They may find, for example, that although they are still
authorizing hospitalization for a particular procedure, the majority of the
patients covered by their program undergo the procedure on an outpatient basis.
Length-of-stay criteria may be revised when data show that many patients are
being discharged before the scheduled concurrent review. Examples like these
suggest that utilization management organizations tend not to be the pioneers in
the move to more parsimonious forms of care but are, instead, consolidators of
movement in that direction.
How Organizations Use Criteria

Utilization management organizations most commonly use written or
explicit criteria to differentiate patient conditions that warrant certification in a
first level of review (usually by a nurse reviewer) from those that require more
detailed review, usually by a physician based on implicit professional standards.
The written criteria are, as a result, often described as screens, although that
term is sometimes reserved for computerized edits of claims that flag certain
types of claims for more detailed examination by nurse reviewers.
The main users of criteria are the nurse reviewers who are empowered by
most (but not all) programs to certify services that conform to the criteria or to
negotiate changes in proposed services so that they conform to the criteria.
Criteria for outpatient surgery, presurgical days, or appropriate hospital days
may be applied more or less strictly, depending on

the utilization management organization's philosophy and the purchaser's

requirements. In most instances, physician advisers make little direct use of
such formal screening criteria. They are expected to make decisions based upon
their own clinical judgment. However, physician advisers are expected to be
familiar with the formal criteria and not contradict them without sound clinical
justification. It does not ordinarily suffice for a physician adviser to certify
services on the grounds that he or she does not agree with the criteria.
In some cases, medical directors and physician advisers cannot agree on
how particular circumstances should be handled, and it is left to the discretion
of the physician adviser who receives the case.4 In some organizations, the
nurse reviewers have the power to select the physician adviser to whom a case
will be referred. With the passage of time, the nurses can learn a great deal
about different physician advisers' tendencies and can make referrals based on
their own assessment of whet decision should be made and their knowledge of
which physician adviser is most likely to make that decision.
Appeals Processes
The appeals process used by organizations that engage in utilization
management are quite diverse, although it appears that all organizations have
some mechanism for appeals. Among the organizations visited, variations in
these mechanisms appear particularly significant along two dimensions. One
dimension is the elaborateness of the process. The other is whether parties
external to the organization are involved.
PROs have a very formal and extensive appeals process. (For a brief
overview of some general differences between the PRO program and private
programs, see Appendix c.) This is a function of the requirements that PROs
accept as part of their contracts with the federal government to review care for
Medicare beneficiaries.
In most of the organizations the committee visited, an appeal involves a
review of the case by a new physician adviser, the medical director, or a
4 A particularly interesting situation was described in one review organization visited
in which the two specialists to whom a particular type of case was referred had different
preferences regarding the two treatment options that were available. Depending upon
which physician adviser received the case, either treatment A or treatment B would be
authorized. For the organization to send all such cases to either of the physicians would
have involved a policy decision that the organization was not prepared to make. The
organization decided to alternate cases, without regard for the fit between the therapeutic
preferences of the attending physician and the physician adviser. So the luck of the draw
determined whether an attending physician received authorization with no difficulty or
might not be able to obtain authorization at all.
committee. Additional discussion with the attending physician may or may not
be a part of the process.
Several organizations indicated that their clients, usually an employer, may
get drawn into the appeals process when the issue is whether the proposed
service will be covered by the patient's benefit plan. In at least one organization,
appeals are shifted directly to the clients.
Reporting and Feedback Mechanisms

In inquiring about the impact of the programs operated by the
organizations the committee visited, it became apparent that interpretation of
results and comparisons across organizations are extremely difficult. What can
be made of information from one organization that the number of hospital days
per 1,000 covered individuals ranges from 300 days for one client to 600 days
for another and from a second organization that reports a range from 200 to
1,300 days per 1,000? For a third organization, the range is 487 to 1,583 days,
but the review organization at least knows that the second number applies to a
group of retirees. Despite the well-recognized differences in medical care cost
and use experience for different age, sex, and other groups and the considerable
variation across employee groups, there are no standard adjustments or
reporting conventions used by utilization management organizations to account
for these differences.
Organizations that provide review services under contract generally
provide their clients with periodic reports on their activities. These reports vary
considerably in usefulness. At one extreme are organizations that calculate
savings based on the difference between services requested and services
authorized. At the other extreme are organizations that can provide detailed
analyses of inpatient and outpatient utilization for the covered group. Some
firms have clients that demand relatively sophisticated, or at least extensive,
data, and they risk losing these clients if they cannot respond to those demands.
(Problems with different measures of program results are discussed in Chapter 4.)
Discussions of reporting requirements during the site visits conveyed a
clear sense that clients evaluate utilization management organizations based on
their apparent success in reducing hospital utilization or holding hospital
utilization to comparatively low levels. Client reaction to reports of cost and use
experience are, thus, an important feedback mechanism to review organizations.
Review organizations also receive several other kinds of feedback. One
source is appeals and objections from physicians and patients regarding the
organization's decisions. Criteria that produce too much negative reaction and
consume too much of the physician advisers' time may be revised.
None of the organizations that provide utilization management services

under contract report any routine mechanism to evaluate the impact that their
decisions have on the quality of patient care. For example, no systematic
mechanisms are in place to track the experience of patients whose care is
shifted from an inpatient to an outpatient basis. Similarly, no organization has
any systematic process to assess the burden that their procedures place on
providers of care. Some organizations, however, report that they survey patients
who have gone through prior review. Also, clients may have (or may be
planning) programs to monitor quality of care based, for example, on
retrospective reviews of certain types of claims such as those showing
rehospitalization within a specified number of days after discharge.
SECOND-OPINION REQUIREMENTS
Second-opinion programs were not the subject of much systematic inquiry
during the site visits because few of the organizations regard second opinions as
a major element in their package of services. Second-opinion programs, which
are generally confined to a list of 15-30 surgical procedures, have two major
objectives. First, they are intended to save money by reducing the likelihood
that unnecessary procedures will be recommended and performed. Second, they
are an educational device designed to increase the amount of information
available to the patient who is contemplating elective surgery.
There are several types of second-opinion programs. Some employers and
insurers have voluntary programs that pay for a second opinion if the patient
wants it. Other programs are mandatory, that is, they make obtaining a second
opinion a requirement for full payment of benefits, but they do not condition
payment on the receipt of a confirming opinion. Some benefit plans require a
third opinion if the second opinion does not confirm the first. Most programs,
like those for preadmission review, require that the patient initiate the process
with a call to the utilization management organization.
Several permutations were encountered in the site visits. For example,
organizations vary in whether they refer patients to particular physicians for
second opinions. Some have panels of physicians who have agreed to provide
second opinions, and these organizations generally offer patients the names of
three participating physicians whom they can call for an appointment. Other
organizations leave the selection to the patient. With regard to procedures about
which there may be differences of opinion across medical specialties (for
example, between cardiac surgeons and a cardiologists), some programs require
that the second opinion be obtained from the nonsurgical specialist. One
organization has a second-opinion program that consists of record reviews and
telephone consultations by

their regular panel of physician advisers. No face-to-face consultation is required.

Although older second-opinion programs required (with few exceptions) a
second opinion for any targeted procedure, several organizations now waive the
requirement when preadmission or preservice review determines that the
clinical indications for the procedure are clear. A typical example would be a
patient with uterine cancer recommended for a hysterectomy. The screening of
cases during preadmission or preservice review is intended to reduce the cost
and inconvenience of the program. However, by retaining the provision for
second opinions for cases in which indications are in doubt, the review
organization has a means of dispute resolution that can supplement the usual
appeals process.
LEGAL ISSUES
In Appendix A of this report is a commissioned paper by William A.
Helvestine that discusses the legal implications of utilization review, primarily
prior review, and continued-stay review techniques. It describes the general lack
of explicit legislative, administrative, or judicial rules involving prior review. It
also describes standards of behavior that have been applied by the courts to
other aspects of health plan administration that are, in many respects, similar to
prior review.
In one well-known case, Wckline v. California, a state court has held that a
review organization (in this case the state Medicaid program) could be held
liable for "defects in the design and implementation of cost containment
mechanisms" that result in the denial of medically necessary services. In this
particular instance, the organization was not held liable because the harm to the
patient was attributed to the negligence of the attending physician. A later case,
Sarchett v. Blue Shield of California, explicitly upheld the right of an insurer to
challenge an attending physician's decisions about medically necessary care.
A few states have passed laws regulating private review organizations
(American Hospital Association, 1989). Maryland and Arkansas have
established (but have not implemented at the time of this writing) a certification
process that requires submission by organizations of information about review
standards, appeals provisions, personnel, confidentiality policies, information
for patients and providers, and accessibility to patients and providers (for
example, business hours). The laws do not require detailed disclosure of review
criteria. Other states have considered, but not passed, legislation that would
require that review physicians be licensed within the state, that no penalties (for
example, reductions in benefit payments) be imposed on patients or providers
for failing to follow review procedures, and that all reviews (including those
now performed by nurses) be done by licensed physicians. The efforts to
impose state regulation have been attributed, in part, to concern about
controlling a few "bad apples"

and getting accountability, especially from the smaller organizations. However,

physician resistance to outside review and unwillingness to change challenged
practice patterns is also cited (Page, 1989). Some proposed regulations could
make operations much more difficult for organizations operating on a
nationwide basis, and some might make reviews prohibitively expensive.
Despite the lack of much explicit regulation, the committee found that
review organizations are aware of the legal risks inherent in efforts to influence
patient care decisions and to operationalize the terms of health benefit contracts.
There also appears to be a general appreciation of the conventionalbut not
infallibleprotections offered against liability by good management, good
judgment, good faith, and good documentation. The contract provisions
analyzed in Appendix F reflect the sensitivity of utilization management firms
to these legal issues.
CONCLUSION
As might be expected for a new and evolving activity, prior review
programs are not standardized. Aside from a few common elementsthe
absence of face-to-face contact with patients, the focus on site, timing, and
duration of care, and the requirement that negative determinations be made by a
physicianall aspects of the process vary markedly from organization to
organization. This variation includes the roles and responsibilities of the nurses
and physicians who are involved in the review process, the logistics of the
process, the nature and availability of the criteria used in prior review, the types
of decisions that are made, the appeals process, and reports of impact.
Although review organizations vary in their inventiveness and willingness
to tackle new issues of appropriate use, they generally follow the lead of
researchers or medical groups in selecting targets for attention. The evolving
focus on assessing the clinical indications for selected procedures is a case in
point.
Some variation among review organizations is prompted by clients, some
by the origins or roots of the organization. Variations in state regulation, which
affect many aspects of insurance company operations, do not account for
differences among review organizations since little regulation of these programs
exists. One exception involves PROs, which face many requirements for their
federal business that also govern their private contracts. However, PROs still
vary considerably. It might be expected that the outside consulting firms that
assess review programs for employers would have a certain standardizing
influence over time, but the committee has seen no obvious evidence of this to
date.

Through its site visits, the committee developed a clear sense that some
organizations have better designed programs or more effectively implemented
programs than others. However, as the next chapter reports, the limited
evidence on the impact of prior review programs does not include any
assessment of the link between differences in program characteristics and
differences in program results.
REFERENCES
American Hospital Association, Private Utilization Review, State Issues Forum Monograph Series,
August 1989.
Burda, David, "Changing Physician Practice Patterns," Modern Healthcare, February 17, 1989, pp.
18-26.
Business Insurance Survey, "Directory of Utilization Review Companies," Business Insurance,
February 20, 1989, pp. 33-68.
Donabedian, Avedis, "Criteria, Norms and Standards of Quality: What Do They Mean?" American
Journal of Public Health, April 1981, pp. 409-412.
Gabel, Jon, Jajich-Toth, Cindy, de Lissovoy, Gregory, and Cohen, Howard, "The Changing World
of Group Health Insurance," Health Affairs, Summer 1988, pp. 48-65.
Activities, GAO/PEMD-89-7FS, Washington, DC, November 1988.
Gertman, Paul M., and Restuccia, Joseph D., "The Appropriateness Evaluation Protocol," Medical
Care, August 1981, pp. 855-871.
InterQual, Inc., The ISD-A Review System, Chicago, 1987.
Mayo Clinic, "The Cost of Effective Utilization Review Programs," Statement for the Institute of
Medicine Committee on Utilization Management by Third Parties, Washington, DC, May
20, 1988.
McGraw-Hill, Inc., Review Resources: Sourcebook of Private Independent UR Companies,
Meyer, Harris, "Two States Lead Move to Regulate Utilization Review," American Medical News,
April 21, 1989, pp. 1, 45.
Milstein, Arnold, Oehm, Marvis, and Alpert, Geraldine, "Gauging the Performance of Utilization
Review," Business and Health, February 1987, pp. 10-12.
"New Blues Program to Rain Nurse Reviewers," American Medical News, April 14, 1989, p. 49.
Page, Leigh, "AMA, Insurers Agree on Guidelines for Hospital Admission Review," American
Medical News, July 21, 1989, pp. 3, 34-35.
Payne, Susan M. C., "Identifying and Managing Inappropriate Hospital Use," Health Services
Research, December 1987, pp. 706-769.
Project HOPE, A Study of the Preadmission Review Process, Prepared for the Prospective Payment
Assessment Commission, Washington, DC, November 1987.
Restuccia, Joseph, Appropriateness Evaluation Protocol Reviewer's Manual, Boston: Boston
University, 1986.
Scheffler, Richard M., Gibbs, James O., and Gurnick, Deborah, The Impact of Medicare's
Prospective Payment System and Private Sector Initiatives: Blue Cross Experience
1980-1986, HCFA Grant No. 15-C-98757-50-1, Research Program in Health Economics,
University of California, Berkeley, July 1988.
Strumwasser, Ira S., et al., Estimates of Non-Acute Hospitalization, Final Report, HCFA Grant No.
18-C98582/5-01 and 2, Washington, DC: Health Care Financing Administration, 1987.
Value Health Sciences, Inc., Medical Review System, Santa Monica, CA, 1989.

IMPACT OF PRIOR REVIEW PROGRAMS 91
4
Impact of Prior Review Programs
The explosive growth of utilization management activities, organizations,

and expenditures raises an obvious question: Does utilization management
work? More specifically, does it do what it is intended to do at a reasonable cost
without unacceptable effects on quality or access to care? And under what
circumstances do programs have better results? How have patients, physicians,
and other involved parties fared?
Unfortunately, in its search of the literature, the committee found that
relatively little rigorous empirical evidence is publicly available on the effects
of these programs. What is available only covers those elements of utilization
management that have been in place long enough for assessment data to have
been accumulated. Prior review of the actual medical need for specific
procedures is mostly too new to have been evaluated.
This chapter summarizes the evidence available to the committee,
describes weaknesses in the evidence, and considers the positive and negative
aspects of prior service for patients, providers, and purchasers. Most of the
evaluations cited here focus on preadmission review and continued-stay review.
A review of evidence specific to the effects of high-cost case management is
presented in Chapter 5. Appendix B of this report discusses the limited
information about the effects of different methods used by HMOs to influence
patient care decisions.
The committee has not attempted a global assessment of the broad societal
effects of utilization management. At this point, the base is simply too weak to
estimate how utilization management has specifically affected

total health care expenditures, overall use of health services, and attitudes
toward medical care.
DIRECTION OF AVAILABLE EVIDENCE: IMPACT ON

UTILIZATION AND COST
Although reports of the impact of prior review programs on utilization and
cost suffer individually from an array of methodological weaknesses, they
present a reasonably consistent pattern of positive results. Taken together, the
studies show:
a substantial initial drop in inpatient hospital utilization following
introduction of prior review, with use rates tending to decline at a lesser
rate or to level off thereafter;
an increase in the use of outpatient facilities and physician office services
following the introduction of prior review;
a greater decline in inpatient utilization for reviewed groups than for
nonreviewed groups during a period of generally declining hospital use;
among groups covered by prior review, a more sizable drop in inpatient
utilization for groups that started with higher than average initial utilization
rates compared with those with lower than average initial utilization rates;
and
a lower rate of increase in the short-term in per-employee medical care
costs for groups covered by prior review compared with those that were
not, but no long-term reduction in the rate of growth in total medical care
spending.
Evidence discussed in this chapter about the effects of prior review
programs was discovered through site visits by the committee, computerized
literature searches, presentations to the committee, and less formal efforts. The
majority of reports take the form of marketing materials, press releases, and
client reports prepared by review organizations. Much scarcer are more
sophisticated assessments prepared by review organizations or academic
researchers.
The earliest studies typically were simple one-group before-and-after
studies with no comparison groups. Several recent studies are more
methodologically sophisticated, although they too are imperfect. Limitations of
available research are described at the end of this section and in the appendix to
this chapter.
Before-and-After Studies
Most of the early and influential attempts to demonstrate the effects of
prior review programs were based on simple before-and-after comparisons.

These comparisons are typical of the promotional materials, conference

presentations, and materials that most utilization management companies
release to the public to show changes in utilization for client groups.
The trend data presented in summary form by one utilization management
company that the committee visited are illustrative. For a large food and
chemical company, the number of hospital days per 1,000 covered individuals
dropped from 579 days before initiation of prior review management to 486 and
then 389 days in the subsequent 2 years. For an aerospace company, the
comparable figures were 627 for the preprogram year and 514 and 416 for the
next 2 years. And for a large consumer products company, the number of
inpatient days per 1,000 covered individuals went from 1,065 days before to
889 to 703 days after the program. Many other companies report similar drops
in inpatient use after the introduction of prior review.
In 1984, the journal Hospitals included some typical press release
information in a story on preadmission review programs (''Preadmission Review
Cuts Hospital Use,'' 1984). The story noted that Blue Cross and Blue Shield of
North Carolina reported that a pilot preadmission review program helped cut
hospital days by 37 percent and that Blue Cross of Northeast Ohio reported a 23
percent decrease in hospital days and savings of $30 million during the first 5
months of a preadmission review program.
More fully described is the experience of Deere & Company of Moline,
Illinois, as presented in a case study prepared by the Health Systems
Management Center at Case Western Reserve University (Kauer, 1983). Deere
initiated a utilization management program in 1977 by using PSROs to perform
preadmission and concurrent review. Over 36 months, inpatient days dropped
21 percent, and the company estimated savings of $11 for every program dollar
spent. The case study noted that one PSRO performed its own reviews, whereas
another PSRO delegated the review responsibility to each area hospital. The
nondelegated program showed greater initial results, with the delegated
program catching up somewhat in later months.
The Deere study mentions that company staff accompanied PSRO officials
to discuss utilization problems with individual hospitals, and the company
added HMO options to its benefit offerings in 1980. These activities point to a
major problem with before-and-after studies. While prior review programs were
being introduced, other changes were also occurring that affected utilization
patterns within particular companies and across society more generally.
Comparative Studies
Over the past decade, several factors, including the growing safety and
acceptance of procedures done on an outpatient basis, the increasing supply of
outpatient resources, and restrictions on inpatient payments, have

TABLE 4-1 Some Factors That May Affect the Impact or Assessment of Prior Review
Benefit Plan Characteristics
Coverage of noninpatient benefits
Levels and types of beneficiary cost-sharing
Choice among alternative benefit plans
Other Initiatives
Quality of care monitoring
Selective contracting
Hospital payment method and financial incentives
Physician payment method and financial incentives
Retrospective utilization review
Provider audit, feedback, and education
Consumer education and health promotion
Government Regulation
Medicare prospective payment system, PRO programs
State rate setting and other hospital regulation
Mandated benefits and other insurance regulation
Medicaid program features
Health planning
Individual, Group, and Community Characteristics
Age, income, education, race, and sex
Union membership
Marital and family status
Other health insurance coverage
Health status
Occupation and industry
Geographic region
Urban or rural location
Health Care Delivery System,
Supply of hospital and other institutional resources
Supply and distribution of physicians and other practitioners
Medicare and Medicaid market shares
HMO and PPO market shares
Blue Cross and Blue Shield market shares
Proportion of self-paying and uninsured patients
changed the balance of incentives for using inpatient versus outpatient care
and have contributed to substantial general reductions in hospital use
(Table 4-1). Data from the American Hospital Association, the Hospital
Discharge Survey, the Health Interview Survey, and the Blue Cross and Blue
Shield Association indicate that hospital days per 1,000 people under age 65
were beginning to level off and then drop in the latter part of the 1970s (Lerner
et al., 1983b). The drop accelerated in the 1980s, before it appeared to level off
again recently. Simple before-and-after studies are not capable of distinguishing
the impact of utilization management from the impact of other factors and, thus,
may wrongly credit those programs with changes that would have occurred
anyway.

Some studies have attempted to control for the impact of systemwide

influences by comparing the utilization and cost experience of groups covered
by prior review programs with groups that were not. The RCA organization
tried this approach within its own organization. In 1985, it instituted an array of
utilization management programs (called Plan for Health) for about half of its
employees, early retirees, and dependents while continuing the old plan for the
other half. Although inpatient hospital utilization dropped for both groups, it
dropped more for the utilization management group. Inpatient days for medical
care patients showed the same general pattern. However, surgical days
unexpectedly increased for both groups, although more for the nonmanaged
group. Nevertheless, net benefit costs for employees under utilization
management dropped by 4 percent, whereas they rose by 6 percent for the
nonmanaged group. The company estimated a 4:1 return on its investment in
the utilization management program (O'Donnell, 1987).
In 1984, Blue Cross and Blue Shield of Massachusetts added a
comprehensive utilization management option for its fee-for-service business
and then compared what happened to hospital utilization for its subscribers who
were covered by the option, those who were not covered by the option, and
those who were covered under an HMO. In 1983, the number of hospital days
per 1,000 covered individuals was 680 for the traditional plan and 409 for the
organization's HMOs. By 1985, traditional plan days had dropped to 600 (a 12
percent drop), "managed" fee-for-service days were 520 (down 24 percent), and
HMO days dropped to 374 (a 9 percent drop) (Getson, 1987).
An analysis prepared for the Service Employees International Union and
the State of Michigan took a somewhat different approach. The evaluation of
the group's precertification program attempted to measure "the difference
between what experience was expected to occur if no intervention took place,
and what actual experience was given implementation of the program" (SEIU,
1988). The decline in hospital use for the program group was higher than that
for a statewide comparison group for the first year of the program but not for
the second year. However, in both years the program had lower hospital
utilization than predicted on the basis of preprogram trends for the union group.
The analysis estimated a gross savings/cost ratio for the program of 2.27 to 1.
The savings were not reduced by the cost of alternative services but did exclude
the estimated fixed costs of avoided admissions (Service Employees
International Union, 1988).
Yet another approach was used by Blue Cross of Greater Philadelphia,
which established a utilization management program as part of its 1985
contractual agreement with area hospitals. Because this agreement covered all
enrollees unless a client explicitly rejected application of the utilization
management features (which was rare), the company did not

have an adequate base to compare enrollees with and without prior review.
Instead, the company compared inpatient days per 1,000 covered individuals for
contracting hospitals, which dropped 23 percent from 1984 to 1986, with
inpatient clays for noncontracting hospitals (mostly specialized mental health,
rehabilitation, and similar facilities), which saw inpatient clays increase by 28
percent. One consequence was that the share of inpatient clays accounted for by
noncontracting hospitals went from 17 to 27 percent of the total inpatient days.
Total inpatient days per 1,000 covered individuals for all hospitals decreased by
20 percent from 1984 to 1987. Combined inpatient and outpatient surgical rates
(excluding physician office surgery), which had increased from 48.2 procedures
per 1,000 in 1980 to 53.1 in 1983, declined to 47.9 in 1986 (Blue Cross of
Greater Philadelphia and Pennsylvania Blue Shield, 1986, 1987).
Multivariate Studies
A potential contaminant in simple comparative studies is the fact that the
introduction of utilization management may have been accompanied by other
changes that could affect utilization and costs. For example, other cost-
containment methods such as increased cost-sharing by beneficiaries have often
been implemented along with prior review, and sales and acquisitions can
change the work force composition, which in turn can affect health care use.
Also, the purchasers that self-select utilization management could differ in
various ways from those that do not. Although not all such factors can be
controlled, some multivariate studies have attempted to rule out alternative
explanations for changes in cost and use after the introduction of prior review.
One utilization management company structures its comparisons by trying
to match review and nonreview groups by industry, geographic region, and
other characteristics. In an analysis for one insurance company client, the firm
reported that hospital days per 1,000 covered individuals dropped 14 percent for
review groups, whereas for nonreview groups they dropped 7 percent. The
percentage increase in expenditures for groups with prior review was lower than
the rate of medical inflation, whereas the opposite was true for comparison
groups (Health Data Institute, 1988).
An analysis of early state programs to contain Medicaid costs reported that
prior authorization for elective surgery and specific services appeared to reduce
growth in real hospital expenditures by controlling growth in the number of
beneficiaries receiving hospital care. For the period 1977-1984, states with prior
authorization showed a 1.4 percent average annual increase in real total
inpatient expenditures, whereas states without this policy experienced a 6.5
percent increase. The number of beneficiaries receiving inpatient care dropped,
on average, by 1.2 percent a year for

prior authorization states but rose 1.9 percent a year for the other states. A
multivariate statistical analysis linked prior authorization with a 2.7 percent
annual reduction in beneficiaries who received inpatient care after several other
changes in Medicaid policy and state environments were controlled
(Zuckerman, 1987).
Feldstein et al. (1988) and Wickizer et al. (1989) analyzed claims data for
one insurer to compare utilization and costs for clients with and without
utilization management. Using a combination of cross-sectional and
longitudinal analyses that included controls for some differences in case mix,
employee characteristics, market factors, and benefit plan features, the authors
concluded that for the period 1984-1986 prior review reduced admissions by 13
percent, inpatient days by 11 percent, and total medical expenditures by 6
percent. (Because some groups included in this part of the analysis had a
utilization management program in place before the first year for which data
were analyzed, these results do not reflect straightforward preprogram-
postprogram utilization.) From their statistical analysis, the authors concluded
that the review programs produced a "onetime" reduction in use and costs but
had little impact on growth in use and costs over time. In another analysis, the
researchers compared groups with high utilization prior to adoption of
utilization review to groups with low prior utilization. The former were found to
have significant decreases in use and costs, but the latter did not (Feldstein et
al., 1988).
After introduction of a utilization management product called
HEALTHLINE, Aetna Life and Casualty compared postimplementation
utilization and cost experience for employee groups with (122,299 employees)
and without (296,519 employees) the program. The time period for the study
was a six-quarter period ending December 31, 1987. The study controlled for
some claimant, group, and benefit plan differences across the two samples (for
example, claimant age, total covered employees, and coinsurance rates). The
multivariate analysis also included a time trend variable. The researchers found
that hospital admission rates dropped nearly 8 percent for the program sample
but only I percent for the nonprogram sample; overall hospital days per 1,000
covered individuals dropped about 4 percent for the former and about 2 percent
for the latter. Surgical outpatient costs per employee increased at about the
same rate for both samples (15.5 and 16.1 percent, respectively), but inpatient
medical and surgical costs rose about 5 percent for the sample with utilization
management and 9 percent for the sample without.1 Combined inpatient costs
and outpatient surgical costs rose 6 and 10 percent, respectively. The study
estimated gross savings from utilization management to be around 12 percent
1 Inpatient costs included room and board, ancillary services, and physician services.
Surgical outpatient costs included physician but not facility charges.

during the six-quarter period. The researchers emphasize that the results apply
only to the study period and should not be used to predict program performance
in other periods or for other employer groups. Further analyses will include
preprogram data, data on nonsurgical outpatient use and costs, and other control
variables (Allen and Khandker, 1988; Harris Allen, Aetna Casualty and Life,
personal communication, November 3, 1988).
Scheffler et al. (1988) studied the impact of the Medicare prospective
payment system (PPS) and Blue Cross benefits cost-management programs
using quarterly claims data (1980-1986) on Blue Cross inpatient admissions and
lengths of stay, hospital outpatient visits, total inpatient and total outpatient
benefit payments, and total payments per member (adjusted for inflation). When
controlled for the effects of the Medicare PPS, several other Blue Cross cost-
management programs, and an array of environmental variables (for example,
state regulation, state average income, and state age composition), the study
found that (1) preadmission review was associated with lower hospital
admission and outpatient visit rates but did not affect payments per 1,000
members for inpatient services or outpatient visits; (2) concurrent review was
linked to lower inpatient and higher outpatient payments per 1,000 members
and had a negative but not statistically significant impact on total payments per
member; and (3) retrospective denial of payment to hospitals for inappropriate
utilization was correlated with lower values for all utilization and payment
measures.2 In addition, the study attributed significant declines in utilization
rates and rates of payment increase to the spillover effects of Medicare PPS and
Blue Cross hospital payment policies (Scheffler et al., 1988).
Impact of Second-Opinion Programs

The committee did not closely examine the impact of second-opinion
programs. In part, this choice reflects the fact that second-opinion programs
have a longer history of application and assessment than do the other utilization
management techniques investigated by the committee. (For an up-to-date
review of the evidence on the impact of second-opinion programs, see Rutgow
and Sieverts [1989]).
2 This statistical analysis is complicated by characteristics of some of the independent
variables. In all but 2 years between 1980 and 1986, over 95 percent of the Blue Cross
plans reported that they had (and had used) policies to deny reimbursement for
inappropriate use. In all years, over 95 percent had retrospective utilization review. The
percentage of plans with concurrent review went from 52 percent in 1980 to 90 percent
in 1986. No information was available on preadmission review in 1980 or 1981, but
between 1982 and 1986 the percentage of plans with this program went from 28 to 95
percent. The analysis was not able to consider variations in the scope of programs nor
what percentage of enrollees were covered by review requirements. Thus, a plan with 1
percent of its enrollees covered would rank the same as one with 90 percent.
The research on private sector second-opinion programs is even less likely

than the research on other prior review techniques to include comparisons with
groups not subject to second-opinion requirements. Moreover, refinements in
the use of second opinions, particularly prior screening of cases for medical
necessity before referral, are too new for much data to have accumulated on
their effects; and it is not clear from discussions with those involved in such
programs that the second opinions themselves are expected to have a major cost-
containing effect. The major expectation is that referred patients will have more
information on which to base their decisions about whether or not to have the
recommended surgery.
During its site visits and other discussions of utilization management, the
committee found relatively little interest in second-opinion programs and very
mixed opinions about their cost-effectiveness. To some degree, this may reflect
the inclination of purchasers and vendors to pay more attention to newer rather
than older programs. Discussions with purchasers, however, suggest that
programs that include no screening prior to mandatory referral for a second
opinion are increasingly seen as wasteful.
In sum, the committee felt that a thorough, independent investigation of
traditional second-opinion programs was not a high priority. The following
reflects its assessment of such programs:
Both voluntary and mandatory second-opinion programs may enhance
consumer knowledge and affect some patient decisions about whether to
have surgery. The effect may be to encourage surgery in certain cases (for
the reluctant patient who gets a confirming opinion) and discourage it in
others (for the patient who gets a strong nonconfirming opinion).
Compared with voluntary programs, mandatory programs tend to show
much higher rates of confirmation; that is, the second opinion confirms the
first opinion or, more specifically, does not reject some form of surgical
intervention.
Voluntary second-opinion programs generally have much lower rates of
participation by patients than do mandatory programs that include penalties
if a patient does not seek a second opinion.
Evidence on the net impact of second-opinion programs on utilization and
costs is less supportive of the conclusion that it contains costs than is the
(also imperfect) evidence about other utilization management techniques.
WEAKNESSES IN THE EVIDENCE ON EFFECTS OF PRIOR

REVIEW
Because prior review programs have been developed and implemented in
an operational rather than a research context, rigorous evaluation has not been a
high priority for most organizations, and studies by outside researchers have
been limited. One consequence is that much of the

evidence about the impact of programs is of marginal value in answering the

major questions that are asked by clients, policymakers, and other interested
parties. All available studies suffer one or more of the following deficiencies in
research design and measurement:
absence of comparisons between groups with and without prior review that
use such methodologies as random selection of study participants and
random assignment of participants to program and nonprogram groups;
use of short time series (for example, 1 year of preprogram data and 1 or 2
years of postimplementation data) that do not allow confident identification
and assessment of preexisting trends, cyclical patterns characteristic of the
health insurance industry, and long-term effects of utilization management;
heavy reliance on inpatient utilization data with little or no evidence about
(1) potentially offsetting changes in inpatient prices, (2) shifts in
noninpatient prices, utilization, and overall costs, (3) changes in net benefit
costs, and (4) savings relative to program costs;
failure to control statistically or otherwise for nonprogram variables (for
example, other cost-containment activities, scope of benefits, shifts in
group composition, and market area characteristics) that may affect
utilization and costs;
knowledge by evaluators of whether data are from groups with or without
utilization management;
absence of comparisons of the relative impact of different program
elements (for example, preadmission review versus continued-stay review)
or alternative program designs (for example, greater emphasis on physician
rather than nurse judgment); and
failure to specify conditions associated with better or worse program results
(for example, type of client and the supply of local health care resources).
Most importantly, all of the studies reviewed above confine their focus to
utilization patterns and costs to the purchaser. The committee found no
empirical research on the possible effects of private sector prior review
programs on quality of care, patient out-of-pocket costs, patient convenience,
patient-physician relationships, and attitudes and administrative costs of health
care practitioners. (Four reports on the effects of Medicare payment and review
programs are cited later in this chapter.) The appendix at the end of this chapter
discusses methodological issues in more detail.
The committee did not study actuarial estimates of the impact of prior
review that insurers and others employ in setting premiums or projecting benefit
expenditures. Informal conversations suggest that actuaries may initially have
estimated a 5-10 percent lower premium for companies that

adopted prior review. Rate reductions of 1-5 percent are said to be more
common for groups that have only recently adopted prior review. One
explanation for the changes in estimates is that actuaries initially
underestimated how much of the reduction on inpatient use would be offset by
higher inpatient prices and higher outpatient use and prices and thus set
premiums too low. This is consistent with the recent report of the Prospective
Payment Assessment Commission (1989) which concluded that Medicare
prospective payment and other public and private cost-containment efforts have
reallocated health expenditures but not slowed their growth.
Another explanation for changes in actuarial estimates of prior review
savings is that late-adopting groups have already gained some benefit from
changes in physician behavior produced by the prior review (and other)
programs implemented earlier by other purchasers. Also proposed is an analog
of "herd immunity," in which coverage of one-half to three-quarters of the
insured population by prior review may produce "spillover" protection for the
rest. There has been no systematic test of this proposition. However, one study
of an insurer's program to promote inhospital ambulatory surgery in the late
1970s and early 1980s suggested spillover effects for other payers whose
inpatient surgery rates also began to drop. Nonetheless, the decrease for the
sponsoring insurer started earlier and the gap in rates persisted over several
years (Lerner et al., 1983a).
The committee notes that many reports on the impact of prior review come
from those with an interest in positive results. This is most obvious for the
organizations that provide prior review services. However, even union and
corporate benefits managers may incline evaluations toward favorable results
unions may do this because utilization management can be seen as an
alternative to benefit cutbacks and benefits managers may do this because good
results can make them look effective to higher levels of management. Some
purchasers of prior review services, recognizing that utilization management
firms may be biased or may lack expertise in evaluation, ask that data be turned
over for analysis by outside consultants. The study committee did not have
access to the reports of these outside consultants.
EFFECTS OF PRIOR REVIEW ON SPECIFIC PARTIES

Although empirical data are very limited, it is important to consider how
private sector prior review programs may positively or negatively affect
different partiesenrollees of benefit plans, health care practitioners and
institutions, and purchasers. The following discussion is based on committee
members' experiences, testimony at the committee's June 1988 hearings, site
visits of the committee members, a roundtable discussion

among benefits managers of several major companies with prior review

programs, and discussions among committee members over the course of
several meetings.
The committee is aware that efforts to tease out the impact of prior review
are complicated by the multiple cost-containment programs to which
individuals and organizations are exposed. It may be impossible to isolate the
effects of utilization management from the effects of prospective payment,
benefit redesign, or provider and consumer education. And private review
programs may be confused with those involving Medicare and Medicaid
patients. Also, changes in clinical and management technologies would
inevitably have spurred some shifts in clinical and administrative practices in
the absence of utilization management.
Effects on Enrollees and Patients
Quality of Care
The foremost question about the effects of prior review on enrollees and
patients is whether and how it affects the quality of care. Does prior review,
which is explicitly aimed at unnecessary care, discourage necessary care as well?
The evidence on these points, which is very limited and pertains primarily
to Medicare beneficiaries, does not suggest that prior review, as implemented
thus far, has threatened the quality of care. The Prospective Payment
Assessment Commission, prompted by concerns similar to those of this
committee, contracted for a qualitative assessment of the impact of
preadmission review on Medicare beneficiaries that found no evidence of harm
(Project HOPE, 1987). Another study followed patients whose proposed
hospitalizations had failed to receive certification by a PRO and also found no
evidence of appreciable harm (Imperiale et al., 1988). In contrast, a more
general study of what happened in one city to Medicare patients with hip
fractures following the introduction of prospective payment indicated that
patients were being discharged with significantly less functional capacity after
the initiation of PPS and were more likely to remain in nursing homes 1 year
later (Fitzgerald et al., 1988). Another study of PPS (not prior review)
documented a shift in the location of death for elderly patients from the hospital
to nursing homes but included no assessment of medical appropriateness (Sager
et al., 1989).
On the other hand, a recent American Medical Association survey found
that almost one-third of the responding physicians said they believed their
patients had suffered an aggravation of illness or injury resulting from delayed
or denied prior authorization for coverage (American Medical Association,
1989b). The response rate to this survey was only 30 percent, and

it seems likely that physicians who had clinical, practical, or philosophical

problems with prior review were most likely to have responded.
Prior review could be improving the quality of care, although this also is
not documented. Unnecessary care can expose patients to needless risks from
anesthesia, blood transfusions, hospital errors and infections, adverse drug
reactions, and other hazards. It may also cause people to miss excess time from
work and other customary activities. Still, unnecessary is not always harmful,
and it may sometimes add to a patient's psychological or physical comfort (for
example, the extra hospital day for the overburdened mother).
The absence of documented cases of quality problems linked to prior
review might be seen as indirect evidence that these programs are, in fact, not
harming patients. This is obviously a tenuous inference, and better monitoring
and study of prior review's relationship to quality of care is needed.
Navigating Health Care and Health Benefits

Most employers and utilization management firms make some effort to
educate employees about prior review requirements, although the level of effort
appears to vary considerably. Many have added notices to insurance
identification cards and taken other steps to help remind patients and providers
about prior review requirements. Nevertheless, health plan requirements tend to
be more complicated today than they were before, and some lapses in patient
compliance seem inevitable. Even when a review organization or employer
waives a penalty for employee failure to comply with requirements, the extra
procedural responsibilities can cause anxiety or resentment.
Anxiety may also arise on the relatively uncommon occasions when a
physician recommends a preoperative day of care that the review organization
will not certify or when the patient receives a second opinion that disagrees
with the attending physician's recommendation. To some extent, these anxieties
are analogous to the stresses of trying to be an informed consumer in a complex
medical care system. Utilization management requirements, like some patient
education materials, may bring to patients or prospective patients an unwelcome
awareness of the subjectivity and variability of many medical decisions.
Although prior review is one more bureaucratic hurdle in an increasingly
complex health care system, it may also help patients deal with this complexity.
Many employers and review companies report that benefit plan members often
use nurse reviewers as a source of descriptive information about medical
conditions, treatment options, alternative sites of care, and other issues related
to a planned course of care. Some review companies

have recognized this as a public relations opportunity and a job-enrichment

vehicle for reviewers, and they have added training for reviewers in handling
questions and providing descriptive information. (The companies emphasize
that reviewers do not attempt to diagnose problems or recommend treatment.)
Patient Costs, Comfort, and Convenience

If prior review does help control benefit costs and deflect more onerous
cost-containment strategies, then the aggregate economic impact on employees
should be positive. That is, employees collectively may avoid higher
deductibles, coinsurance, or other forms of reduced benefits.
In an individual case, depending on the structure of an individual's
coverage, out-of-pocket costs may be lower or higher if utilization management
changes the course of treatment. For example, out-of-pocket costs could be
higher for patients in health plans that have higher coinsurance rates for
outpatient care and do not cover outpatient prescription drugs. Patients' costs
may be lower when health plans have equivalent coverage for inpatient and
outpatient care or impose steep per-admission deductibles on hospital care. In
some programs, patient failure to comply with review requirements may prompt
a financial penalty.
When inpatient care is avoided, outpatient care may be either more or less
convenient and comfortable for patients and patients' families. This depends on
home conditions, income, distance from sites of outpatient treatment, and the
complexity of alternative care needed. Some review organizations appear to be
more willing than others to consider these sorts of circumstances in making
decisions, as indicated in the presentation in Chapter 3 of circumstances
considered to warrant hospitalization for lower back pain and tonsillectomy and
adenoidectomy (see Table 3-3).
Some review organizations survey a sample of those who have been
subject to utilization management to determine their views of the process. This
information may be used in reporting to clients and refining programs. It has
not, to the committee's knowledge, been published in any systematic way.
To summarize, prior review has both potential positive and negative
implications for patients, depending on the specifics of their coverage and their
understanding of it, the nature of the alternatives available to them, and the
design and implementation of the review program affecting them. Physician and
hospital reactions to utilization management may also affect patients.

Effects on Health Care Practitioners and Institutions

Most clinicians and hospital administrators now see prior review as a fact
of life (Weinstein, 1988). In a recent American Medical Association (1989b)
survey, for example, physician respondents reported that over one-third of their
cases required prior review as a condition of payment. The growth of the
utilization management industry, which depends on cooperation from hospitals
and physicians, is itself an indicator of acceptance. Most physicians understand
that complete prohibitions against payer ''supervision or control ... over the
manner in which medical services are provided'' (as stated in P.L. 89-97, the
original Medicare legislation) are no longer realistic.
This recognition, however, has not come without some resistance. At a
minimum, complaints about prior review requirements continue, and instances
of noncooperation are still reported by review organizations. This physician's
response in a focus group discussion of Medicare is illustrative of the resistor's
stance: "I don't think that at this point in my life I have changed what I would
do based on the fact that someone is pushing me. It just adds time to my day
because I have to take telephone calls from so-and-so. And I resist the urge to
be nasty on the phone saying, 'what do you think I'm going to treat this case
with? I'm going to treat them with whatever I see fit, thank you'" (cited in
Walker, 1989).
Physician-Patient Relationship
For those cases subject to prior review, third parties have a role in
decisions on patient care that alters the context of the physician-patient
relationship. However, the potential ramifications are easier to enumerate than
to document systematically and almost certainly affect some specialties, such as
surgery, more than others. The testimony provided by medical organizations at
a hearing sponsored by this committee (summarized in Appendix D of this
report) reflects both positive and negative views of utilization management.
Some physicians feel that prior review compromises their professional
autonomy and increases their vulnerability to malpractice suits. Some are
resentful when they must soothe patients worried about prior review
requirements or act as advocates for their patients when review organizations
promote a course of care that they see as less safe, comfortable, or convenient.
This resentment could spill over into their relations with their patients. Some
observers worry that certain physicians may acquiesce to a reviewer's
judgments because they would rather avoid hassles than challenge a
questionable review determination. On the other hand, some physicians note
that utilization management may be contributing to improved patient relations
that they are explaining proposed services more carefully, jointly

exploring more options, and documenting treatment rationales and details more
carefully (Carroll, 1988; London and Anderson, 1988).
Physician-Hospital Relations
Utilization management, in combination with prospective hospital payment
and other purchaser strategies to contain health care costs, has also affected
relationships between physicians and hospitals. With hospital occupancy at or
near record lows in most areas, physicians are more important than ever as the
source of patients. However, physicians who are profligate orderers of resources
once patients are admitted may make hospitals vulnerable to retrospective
denials of payment or losses under a per-case payment system.
Hospitals clearly have increased their monitoring of physician. behavior.
Many have initiated their own preadmission review processes and have
strengthened efforts to ensure that physicians adequately document their actions
and decisions (Anderson, 1988). The objectives are to see that payer
requirements have been met and to protect the hospital against after-the-fact
payment denials for care judged to be medically unnecessary. More generally,
utilization management reinforces other trends that are increasing the severity
of hospital case mix and encouraging hospitals to diversify, sometimes
successfully and sometimes not, into areas such as home health, health
promotion, health insurance, and administrative services for physicians and
other institutions.
Overall, utilization management appears to be one of many factors
contributing to tighter integration of hospital medical staffs and more physician
involvement in hospital management. Most physicians recognize their stake in
the fiscal soundness of the hospitals in which they practice, although hospital-
based review and related actions may nonetheless provoke irritation. For
example, this reaction to hospital pressure came in a focus group discussion:
"You see, what's happening is you're putting the doctor in a vice between what
he feels he wants to do for the patient and what the hospital administrator wants
to tell the doctor about how he should take care of the patient. There are a lot of
MBAs between us and the patient telling us what care we can administer, for
how many days, what they will pay for and what they won't pay for" (cited in
Walker, 1989).
Provider-Purchaser Relations
Physician and hospital relationships with employers, insurers, and other
third-party payers have also changed dramatically. More time and money are
being invested by hospitals and physicians to keep up with a broad array of
payer requirements, including utilization management. And it is not only the
number of requirements but also the number of organizations

involved that makes demands on institutions and practitioners. Much of the

organized response to prior review, as reflected in the committee's June 1988
hearing, shows provider interest in greater standardization of utilization
management processes across organizations, as well as better communication of
requirements, criteria, and results. Timely handling of review requests and
appeals is another major issue.
The American Hospital Association says that hospitals may deal with 50 to
250 organizations doing prior and retrospective review (American Hospital
Association, 1989). The Mayo Clinic, which employs over 800 physicians and
scientists, estimates that it deals with hundreds of review programs and would
spend at least $0.5 million in 1988 to meet the requirements of utilization
management programs (Mayo Clinic, 1988). The American Medical
Association survey referred to earlier (American Medical Association, 1989b)
showed that while "complex administrative services" related to patient health
benefit plans took 1 hour of office time (physician and staff) for about one-
quarter of responding physicians and 2 hours for another quarter, one-fifth of
the respondents reported spending more than 6 hours a day on such activities.
These results were not categorized by type or size of physician practice.
Payer questions about the appropriateness of physician decisions and
physicians' concerns about the appropriateness of payer judgments of
appropriatenesshave prompted physician organizations to become more
involved in the development of guidelines and standards of care (Meyer and
Page, 1988; Physician Payment Review Commission, 1988, 1989). The
American College of Physicians is notable for its work in this area for more
than a decade (American College of Physicians, 1986; Schwartz, 1984).
Recently, the American Medical Association has discussed work with the Rand
Corporation to develop practice guidelines, and there is a complementary effort
with the Blue Cross and Blue Shield Association and the Health Insurance
Association of America to incorporate the guidelines into insurers' utilization
management programs (McIlrath, 1988; Meyer and Page, 1988).3 And, to name
one other initiative, physician organizations have been cooperating with the
efforts of the Physician Payment Review Commission to improve the definition
and coding of medical services and to suggest methodological standards for
developing practice guidelines (Physician Payment Review Commission, 1988).
On the administrative side, the American Hospital Association is working
with the American Medical Association, the Blue Cross and Blue Shield
Association, and the Health Insurance Association of America to develop
3 In addition, some former Rand Corporation researchers are now working
independently with a for-profit organization, Value Health Sciences, to develop better

criteria and supporting software for use in both prospective and retrospective reviews
(Findlay, 1989; Michaelson, 1988).
voluntary guidelines for private review (American Hospital Association, 1989).

The last three organizations already have cooperated on guidelines for the
conduct of utilization management programs (American Medical Association,
1989a). There are eight guidelines specific to prior review.
The medical protocols as well as other relevant medical issues used in prior
authorization programs should be established with input from physician
advisers selected by the health benefit plan.
Prior authorization programs may be conducted on a targeted review basis
rather than attempting to prereview all services eligible for coverage.
All preadmission review programs should provide for immediate
hospitalization of any patient for whom the treating physician determines
the admission to be of an emergency nature, so long as medical necessity is
subsequently documented.
In the absence of any contractual agreement between physician and health
benefit plan, the responsibility for obtaining prior authorization required by
a claims administrator should be that of the enrollee.
The claims administrator and employee benefits manager should work
together to alert enrollees to the need to be aware of and to inform the
physician of any prior authorization requirements applying to their
insurance coverage.
In cases where a claims administrator requires prior authorization, the
claims administrator should respond promptly and efficiently to requests
for authorization. A physician or a patient should receive a response within
2 business days.
In any instance where authorization is questioned on the basis of medical
necessity, the attending physician should be able to review medical
necessity with the physician adviser representing the claims administrator.
To the extent that prior authorization programs are administered efficiently
with minimal disruption to the provision of medical care, additional
payment to physicians for complying with prior authorization requirements
should not be necessary.
Effects of Prior Review on Purchasers

Employers are the principal purchasers of private sector review programs
and are affected by its application in a variety of ways. Initial employer reaction
to prior review has been positive (Jennings, 1987; Vibbert, 1989). They feel it
has had some impact on benefit costs and has improved the value they receive
for their expenditures.
On the other hand, the fact that prior review programs seem not to have
shifted the basic slope of the cost curve intensifies employers' frustration with
rising expenditures. This frustration over basic trends

repeats employers' experiences with other cost-containment tools, including

increased employee cost-sharing and alternative delivery systems (such as
HMOs).
In general, the adoption of prior review programs by private purchasers has
reinforced their long-standing dissatisfaction with data about the health services
they are buying and has strengthened their efforts to obtain better information.
Most utilization management firms are under heavy pressure to report more
utilization and cost data and present more credible analyses of program impact.
Some employers are concerned that prior review exposes them to a new
risk of liability, particularly if they administer the programs directly. The case
of Wickline v. California held that third-party payers "can be held legally
accountable when medically inappropriate decisions result from defects in the
design or implementation of cost containment mechanisms . . ." (emphasis
added) (see the paper by William A. Helvestine in Appendix A of this report for
further discussion). Employers often seek protection from such liability in their
contracts with review organizations, but the latter are understandably reluctant
to promise to indemnify their clients against damages.
As purchasers move from retrospective to prospective review of care,
relationships with employees may change. Many employers recognize the need
to enlist the employee in their efforts to hold down costs. Some have developed
programs that teach employees about making informed health care decisions
and assist them with decisions about the course and place of treatment. At a
minimum, most prior review programs increase the administrative
responsibilities of employees
Purchasers of prior review services also may find themselves in a different
relationship with providers. Until recently, purchasers generally paid for
whatever treatment the doctor ordered. Now, in their new role as managers of
utilization and armed with data on the prices and practices of community
providers, they are using their purchasing power to influence community
practice standards.
The committee noted that purchaser decisions are shaped by many factors.
For example, although the committee is unaware of any systematic research on
this point, different industries and different firms within industries appear to
have different norms governing their decisions on employee benefits. For
example, in a roundtable discussion with benefits managers from several large
companies that the Institute of Medicine held in December 1988, some firms
were portrayed as slow to adopt prior review out of concern that they would
antagonize their white-collar employees. In areas where workers are in short
supply, less restrictive health benefits may help in recruiting new workers but
may be of little significance where unemployment is high. Employers that have
multiple locations and small

work sites may find it difficult to locate and coordinate effective utilization
management services.
Overall, employer reaction to prior review has been positive, but the long-
term picture is uncertain. This uncertainty has two aspects, neither empirically
documented. First, to the extent that prior review and other programs influence
provider practices throughout the community regardless of a particular benefit
plan's provisions, some employers may consider dropping programs in the hope
that the spillover benefit from other employers' programs will suffice. Others
may feel that prior review has had all the impact it is going to have, that
changes in practice patterns are now entrenched, and that they do not need to be
reinforced by continued application of these techniques.
Second, employer discontent with the reemergence of sharply rising
benefit expenditures may lead some to make drastic changes in their funding of
employee benefits. For example, if employers switch from a defined benefit to a
defined contribution program that limits their yearly increase for health benefits
to a level they can control, regardless of what happens to health care costs, then
their interest in various cost-control programs may diminish. Defined
contribution programs have become common for employee pension plans but
are not widespread for health benefit plans.
CONCLUSION
Although the evidence on prior review is generally not rigorous, it does
tend to be consistently positive about the short-term effects of prior review on
hospital use and expenditures. It focuses almost entirely on reviews of the site,
timing, and duration of care rather than on the medical necessity of specific
procedures, because the latter emphasis is too new to have produced adequate
data for evaluation.
The impact of prior review techniques on access and quality of care has
not been assessed systematically, but no serious suggestion of negative
consequences has come to the committee's attention. Qualitative assessments of
the impact of utilization management on patients, providers, and purchasers
suggest the potential for both positive and negative effects.
The committee recognizes that rigorous evaluations are expensive and
difficult. In the clinical arena itself, rigorous evaluations of the impact of
specific medical services are the exception, not the rule. Furthermore, the
committee recognizes that many health benefit programs are adopted,
maintained, or discontinued by private and public decision-makers on the basis
of evidence as weak as or weaker than that available for utilization
management. Nevertheless, the committee is concerned about the limited
commitment to systematic evaluation of utilization management. Its
recommendations on this point are contained in Chapter 6.

APPENDIX SOME METHODOLOGICAL ISSUES IN

ASSESSING THE EFFECTS OF UTILIZATION
MANAGEMENT PROGRAMS
An exhaustive review of methodological problems in evaluating the effects
of utilization management is beyond the scope of this report. However, a brief
overview of key issues illustrates the array of difficulties faced by purchasers,
program managers, policymakers, and others interested in the consequences of
utilization management. This overview includes a table describing typical
measures of program impact and their limitations (Table 4-2).
Claims Data
Health care cost and utilization data based on claims submitted by
providers or patients may suffer from a variety of defects (Wennberg, 1987).
The claim form itself may be improperly designed to capture the needed
information about the site of care, type of care, or diagnosis in an unambiguous
form. Information submitted on the claim form may be inaccurate because of
errors in medical records, transcription mistakes, imprecise diagnosis or
procedure codes, and deliberate provision of false information (for example,
recording a medical problem for an examination undertaken purely for
screening purposes). Claims information from different companies may be
difficult to merge for multigroup studies. Claims data alone are not sufficient to
establish the severity of illness for purposes of comparison over time or across
groups.
Even when data are accurate, they may not be available until months or
even years after care has been provided. This limits the efforts of program
managers to identify problems and adjust programs in a timely fashion. Also,
individual claims data may be accurate but limited in scope, often not reflecting
an entire episode of care for a patient, particularly if the patient uses some care
for which claims are not recorded (for example, care covered by a spouse's plan
or care from a noncovered provider). This means that actual utilization and
costs per episode of care may be underestimated for alternative modes of
treatment. Information that tracks multiple episodes of care for an individual is
even more limited, and the patient-level links between prior review decisions
and subsequent care are typically not examined.
Group Data
Important characteristics of employee groups and individuals covered by
utilization management may be unmeasured or measured inadequately.

TABLE 4-2 Measures of Prior Review Impact

ACTIVITY DATA
Number of admissions or days requested
Number of requests approved, negotiated, or denied
Number of requests referred to physician reviewers
Number of denials appealed and upheld or not upheld
Number of admissions or days averted (requested minus approved)
Number of second opinions obtained and confirmed or not confirmed
Number of nonconfirming opinions not followed by surgery
Source
Review organization
Comments
Do not measure health services use or costs
Are helpful in assessing workload and checking some administrative practices
Are relatively simple to collect
Can be manipulated by utilization management organizations to project
unrealistically favorable results
Can be manipulated by providers who request more days than are really wanted
Cannot tap "sentinel effect" (that is, admissions discouraged with no prior
authorization approval sought)
May not be matched to actual utilization (that is, may ignore days or admissions
approved but not used; days or admissions denied but approved upon appeal or after
an emergency admission; for second opinion. may ignore individuals encouraged by
second opinion to get surgery when they otherwise would not have)
INPATIENT AND OUTPATIENT UTILIZATION
Inpatient hospital days per 1,000 covered individuals
Inpatient admissions per 1,000 covered individuals
Average length of stay
Outpatient medical claims per 1,000 covered individuals
Inpatient and outpatient surgical claims per 1,000 covered individuals
Total outpatient claims per 1,000 covered individuals
Physician office visits per 1,000 covered individuals
Source
Claims data
Benefit plan enrollment statistics
Comments
Can correct some limitations of activity data
Are needed to help interpret changes in site of care, plan costs, and other variables
May involve questionable denominators for rates if the number of covered
dependents is unknown or is estimated by using outdated multipliers for family size
and if additional coverage for working spouses is not accounted for
May count only hospital outpatient facility services and not visits to physician
offices or other sites
Are generally not aggregated to show entire episodes of care

INPATIENT AND OUTPATIENT COSTS

Inpatient payments per admission or day of care
Inpatient payments per 1,000 covered individuals
Outpatient payments (medical, surgical, or total) per claim
Outpatient payments per 1,000 covered individuals
Payments for inpatient or outpatient physician services per 1,000 covered individuals
Source
Claims data
Hospital cost reports
Comments
Are useful for assessing changes in hospital payments, which have traditionally
accounted for the largest share of total outlays
May not adjust for differences in costs for days of care averted earlier versus later in a
hospital stay
May not adjust for fixed costs that are eventually absorbed by most retrospective cost-
or charge-based payment systems
Do not adjust for severity of remaining admissions and days of care
May have same problems with denominators of rates as utilization statistics
PROGRAM SAVINGS AND COSTS
Benefit payments per 1,000 covered individuals
Premium per covered individual
Administrative charge per contract (or other basis)
Ratio of review program costs to program savings
Benefit payments less program costs per 1.000 covered individuals
Number of hospital days averted and savings
Number of surgeries averted and savings
Source
Claims data
Plan contract, enrollment, and premium data
Review organization
Comments
Are needed to assess trends in overall benefit costs and to assess net savings
May not measure true costs to the supplier, depending on market strategy of utilization
management company (importance of this issue depends on evaluator's objectives)
May not reflect costs of alternatives to admissions or surgeries or costs from care only
temporarily averted
May exclude noncontractual costs to client for explaining program to employees, etc.
Do not include any costs or savings or other effects for providers or patients

If different employee groups or subgroups vary in their exposure to benefit

claims in unmeasured ways, then per-employee or per-covered-individual use
and cost comparisons may be misleading (see also Table 4-1). The scope of
health plan benefits for a group can have an important impact on the use of
services but is often unmeasured. Age, geographic location, and many other
factors can also have an impact.
A fundamental inadequacy of many data bases is imprecise specification of
the size of the groups covered by a health plan. Although the number of covered
employees (contracts) is generally known, the number of covered dependents
may only be estimated by "family" factors that ignore drops in average family
size and increases in alternate primary coverage for spouses. This imprecision
makes the denominators for utilization and expenditure rates problematic. Plans
such as HMOs that require explicit enrollment information are in a stronger
position to calculate use and payment rates. In addition, although plan managers
are making strenuous efforts to coordinate benefits (that is, see that care for
dependents with primary coverage from another source is not improperly
reimbursed under their employee's contract), the extent of such duplicate
coverage is often not known in the aggregate (Lerner et al., 1983b; Luft, 1981)).
Important information about the health status of group members is often
tapped in only the most rudimentary way, if at all. This has been a major
controversy in comparisons of HMOs and fee-for-service groups. Some claim
that HMOs are getting healthier enrollees, and thus price and use comparisons
are misleading (Luft, 1987; Scheffler and Rossiter, 1985).
Program Data
Information about the prior review program often is limited. Programs may
theoretically go into effect on a specified date, but actual implementation may
lag considerably. Differences in program quality, scope, techniques, and other
characteristics may not be described, much less assessed, in reports on prior
review program effects.
Savings Calculations
Many utilization management organizations estimate savings by taking the
number of hospital (lays requested by a physician or hospital, subtracting the
number of hospital days not authorized, and multiplying the result by the
average cost of a hospital day. This approach has many flawsthe days
requested may be overstated by providers who are trying to game the program,
the number of days approved may be exceeded without subsequent adjustment,
and the average cost of a hospital day may overstate the cost

of days avoided at the end of a hospital stay when care is normally less intense
than it is earlier in the stay.
Other Interventions
Prior review programs are often one of many efforts to deal with escalating
benefit costs. Their implementation frequently coincides or overlaps with
initiatives to collect more detailed and accurate information about use, costs,
and covered individuals. Although such initiatives typically improve subsequent
data, earlier data generally cannot be supplemented to provide a comparable
time series for analysis. In addition, retrieval of preprogram data is often time-
consuming and expensive, particularly when data are being aggregated for
several groups and must be obtained from a separate claims payer.
Utilization management also may be introduced simultaneously with a
major redesign of benefits (for example, increased cost sharing) and the
addition of a choice among multiple health plans. This complicates efforts to
distinguish any effects due independently to, or interactively with, utilization
management. The study by Scheffler et al. (1988) attempts to distinguish the
effects of Blue Cross cost-containment programs from each other and from the
Medicare prospective payment system. Any such effort is inevitably plagued
with problems in combining data from multiple sources, evaluating the
accuracy of these data, measuring the intensity and quality of programs, and
adjusting for statistical characteristics and quirks of the data.
Medical Care Prices

Costs are a function of both utilization and unit prices of services. If
inpatient prices are better controlled than outpatient prices, then savings from
shifting care may be illusory. Likewise, if hospitals increase their prices to
cover fixed costs in the face of reduced occupancy and if payers cannot limit
their exposure to these increases because they pay uncontrolled hospital charges
or costs, the reduced inpatient use may not bring a net savings for purchasers of
care. If some payers can limit their exposure to these increases, as Medicare and
some other payers do, then costs for payers without such protection may
increase even more.
Noneconomic effects
Sometimes claims data can be explored in an effort to assess possible
effects of prior review programs on quality of care. For example, emergency
admissions, readmissions, admissions following outpatient surgery, length

of home or nursing home care, and other events canwith varying degrees of
difficultybe identified and links with prior review decisions can be attempted.
PROs track some of these events in an effort to monitor the effects of
prospective hospital payment. Some utilization management organizations are
trying to build data systems that allow patient histories to be easily retrieved
and analyzed.
More direct information on health status and health outcomes is more
expensive and difficult to obtain. Medicare, other purchasers, and other
organizations, such as the Joint Commission on the Accreditation of Health
Care Organizations, are actively working on quality assurance programs to
identify potential quality problems at various sites of care. Much remains to be
done. The Institute of Medicine is engaged in a major project to help design a
quality assurance program for Medicare (Institute of Medicine, 1989).
Some employers and utilization management organizations survey
employees to assess their reactions to utilization management programs. This is
more the exception than the rule. Possibly, the absence of volunteered
complaints from employees is taken as evidence that employees are not
unhappy with the program. They are not adverse to complaining about other
aspects of benefit plan administration, for example, retrospective claims denials,
slow payment of claims, and confusing explanations of coverage.
Systematic assessments of provider problems and attitudes are even less
common than checks on beneficiary attitudes.
REFERENCES
Allen, Harris, and Khandker, Rexaul, ''Aetna's HEALTHLINE Program: Fourth Quarter, 1987
Update,'' Unpublished paper, September 30, 1988.
American College of Physicians, Clinical Efficacy Assessment Project, Philadelphia, PA, October
1986.
American Hospital Association, Private Utilization Review, State Issues Forum Monograph Series,
Washington, DC, August 1989.
American Medical Association, Guidelines for the Conduct of Prior Authorization Programs,
Chicago, 1989a.
American Medical Association, Summary Report: 1988 Payer Accountability Monitoring Survey,
Chicago, May 1989b.
Anderson, Suzanne, "Hospitals Can Improve Cash Flow by Managing Preauthorizations,"
Healthcare Financial Management, December 1988, pp. 56-59.
Blue Cross of Greater Philadelphia and Pennsylvania Blue Shield, Extending the Influence Beyond
the Source: Community Data Report 1987, Philadelphia, June 1987.
Blue Cross of Greater Philadelphia and Pennsylvania Blue Shield, State of the Art Health Care
Management: Community Data Report 1986, Philadelphia, July 1986.
Carroll, Robert P., "The Problem with PROs Is Hard Heads Like Me," Medical Economics,
December 19, 1988, pp. 103-115.

Feldstein, Paul J., Wickizer, Thomas M., and Wheeler, John R. C., "The Effects of Utilization
Review Programs on Health Care Use and Expenditures," New England Journal of
Medicine, May 19, 1988, pp. 1310-1314. (See also Zwarenstein, M. B., et al., letter to the
editor and E J. Feldstein response, New England Journal of Medicine, October 17, 1988, p.
1158.)
Findlay, Stevan, "Looking over the Doctor's Shoulder," U.S. News & World Report, January 30,
1989, pp. 70-73.
Fitzgerald, John F., Moore, Patricia S., and Dittus, Robert S., "The Care of Elderly Patients with Hip
Fracture," New England Journal of Medicine, November 24, 1988, pp. 1392-1397.
Getson, Jacob, "Reforming Health Care Delivery: The Massachusetts Blues' Role," Business and
Health, February 1987, pp. 30-35.
Health Data Institute, Inc., "OPTIMED Managed Care Program," Lexington, MA, April 1988.
Imperiale, Thomas, et al., "Preadmission Screening of Medicare Patients," Journal of the American
Medical Association, June 17, 1988, pp. 3418-3421.
Institute of Medicine, "Designing a Strategy for Quality Review and Assurance in Medicare:
Twelve Month Update," Washington, DC, February 1989.
Jennings, Susan, "Survey of PEW Corporate Fellows in Health Policy," Unpublished paper, Boston
University, December 1987.
Kauer, Robert, "Evaluating a Corporate Health Care Utilization Review Program: The Case of
Deere & Company," Working Paper No. 013, Cleveland, OH, Health Systems
Management Center, Case Western Reserve University, December 1983.
Lerner, Monroe, Salkever, David S., and Davis, Leonard, "Evaluation of Program to Move Care for
Certain Surgical Procedures to an Ambulatory Care Setting," Paper presented at the annual
meeting of the American Public Health Association , Dallas, November 14-17, 1983a.
Lerner, Monroe, Salkever, David S., and Newman, John E, "The Decline in Blue Cross Plan
Admission Rates: Four Explanations," Inquiry, Summer 1983b, pp. 103-113.
London, Alan E., and Anderson, Richard A., "Provider and Reviewer Speak Out on Utilization
Management," Federation of American Health Systems Review, July/August 1988, pp.
36-40.
Luft, Harold, "Divergent Trends in Hospitalization: Fact or Artifact?" Medical Care, October 1981,
pp. 979-994.
Luft, Harold, Health Maintenance Organizations, New Brunswick, NJ: Transaction Books, 1987.
Mayo Clinic, "The 'Cost' of Effective Utilization Review Programs," Statement submitted to the
Institute of Medicine Committee on Utilization Management, May 1988.
McIlrath, Sharon, "AMA, Rand Corp. Plan Joint Development of Practice Guidelines," American
Medical News, October 28, 1988, pp. 2, 27.
Meyer, Harris, and Page, Leigh, "New Era in Utilization Review," American Medical News,
December 9, 1988, pp. 1, 42-45.
Michaelson, Leslie, Medical Review System, Santa Monica, CA: Value Health Sciences, Inc., 1988.
O'Donnell, Peter S., "Controlling Costs Under a Fee-for-Service Plan," Business and Health, March
1987, pp. 38-41.
Physician Payment Review Commission, Annual Report to Congress, Washington, DC, March 1988.
"Preadmission Review Cuts Hospital Use," Hospitals, August 1, 1984, pp. 54-55.
Project HOPE, A Study of the Preadmission Review Process, Prepared for the Prospective Payment
Assessment Commission, Washington, DC, November 1987.
Prospective Payment Assessment Commission, Medicare Prospective Payment and the American
Health Care System: Report to Congress, Washington, DC, June 1989.
Rutgow, Ira M., and Sieverts, Steven, "Surgical Second Opinion Programs," in Socioeconomics of
Surgery, Ira Rutgow, ed., St. Louis: C. V. Mosby Company, 1989.

Sager, Mark A., et al., "Changes in the Location of Death After Passage of Medicare's Prospective
Payment System," New England Journal of Medicine, February 16, 1989, pp. 433-439.
Scheffler, Richard M., and Rossiter, Louis E, eds., Biased Selection in Health Care Markets,
Advances in Health Economics and Health Services Research, Vol. 6, Greenwich, CT: JAI
Press, Inc., 1985.
Scheffler, Richard M., Gibbs, James O., and Gurnick, Deborah, The Impact of Medicare's
Prospective Payment System and Private Sector Initiatives: Blue Cross Experience,
1980-1986, HCFA Grant No. 15-C-98757-5-01, Berkeley, University of California, July
1988.
Schwartz, J. S. "The Role of Professional Medical Societies in Reducing Variations," Health
Service Employees International Union, "Utilization Review and Case Management in Employee
Benefit Plans," Washington, DC, July 1988.
Vibbert, Spencer, "Is Utilization Review Paying Off?" Business and Health, February 1989, pp.
20-26.
Walker, Allison, "Findings of Physician Focus Groups," Unpublished paper prepared for the
Institute of Medicine Committee to Design a Strategy for Quality Review and Assurance
in Medicare, Washington, DC, January 1989.
Wennberg, John, "Use of Claims Data Systems to Evaluate Health Care Outcomes," Journal of the
American Medical Association, February 20, 1987, pp. 933-936.
Wickizer, Thomas M., Wheeler, John R. C., and Feldstein, Paul J., "Does Utilization Review
Reduce Unnecessary Hospital Care and Contain Costs?" Medical Care, June 1989, pp.
632-647.
Zuckerman, Stephen, "Medicaid Hospital Spending: Effects of Reimbursement and Utilization
Control Policies," Health Care Financing Review, Winter 1987, pp. 65-77.

HIGH-COST CASE MANAGEMENT 119
5
High-cost case management is a set of techniques to promote more cost-

effective and appropriate modes of care for patients with expensive illnesses.
Even more than prior review, these techniques have gained acceptance in a very
short period. As recently as 1983, a survey of employer cost-containment
programs conducted for Equitable included no questions about high-cost case
management (Equitable Life Assurance Society of the United States, 1983). At
that time, formal programs to manage expenditures for very expensive medical
cases were uncommon, except for the programs of private firms and state
agencies working with recipients of worker's compensation benefits.
The situation is different now. A 1988 survey of benefits managers for
large companies found that 83 percent said they used high-cost case
management (called "medical case management" in the survey) (Corporate
Health Strategies, 1988). Although the surveyed group is not a representative
sample, clearly this approach to utilization management has become common.
Virtually all major commercial insurers and Blue Cross and Blue Shield plans
offer the service either directly or through subsidiaries (Henderson and
Wallack, 1987), as do most of the larger utilization management firms and third-
party administrators. Some review firms are actually outgrowths of
organizations originally established to manage worker's compensation cases.
For health care institutions and other organizations trying to cope with the
multiple and complex needs of patients with acquired immune deficiency
syndrome (AIDS), high-cost case management

is one prominent strategy (Health Insurance Association of America, 1988;

Taravella, 1987).
Because the case manager is not managing care as a physician does but
rather is trying to see that health plan benefits are used wisely, some
organizations prefer to call their programs "individual benefits management."
Other organizations, reluctant to highlight cost objectives, use the term "medical
case management."
FOCUS OF HIGH-COST CASE MANAGEMENT

What distinguishes high-cost case management from other utilization
management or managed care programs is its intensive and specialized focus on
high-cost medical cases. Preadmission and continued stay review, in contrast,
cover a broader array of cases in a much less intense fashion.
The generic label "case management" seems to have arisen in the 1970s
with government projects to encourage the integration of services for clients of
social welfare programs (Monroe County Long Term Care Program, Inc.,
1986). These include the Services Interaction Targets for Opportunities program
in the 1970s and the channeling and Social/Health Maintenance Organization
demonstration projects for the elderly in the 1980s (Merrill, 1985). High-cost
case management shares with other case management approaches an emphasis
on assessing individual needs and circumstances and then planning, arranging,
and monitoring needed services. However, because it focuses on a relatively
small number of expensive cases, high-cost case management can be
distinguished from so-called primary-care case management approaches often
used in HMOs and Medicaid programs. In these programs, all individuals are
assigned to a case manager who ordinarily must authorize or provide most
services. And because health care cost containment is a central goal, high-cost
case management can be distinguished from social welfare case management
that is aimed at helping clients get the services they need from the complex
labyrinth of welfare programs and agencies.
In focusing on very expensive patients, high-cost case management
responds to an important characteristic of medical care: a small group of
individuals with very costly illnesses or injuries account for a large share of the
total expenditure. The experience of one company is typical: 2 percent of plan
members with yearly expenses of $10,000 or more accounted for 38 percent of
plan medical expenses (Rosenbloom and Gertman, 1984). In another company,
6 percent of those making benefit claims (and not all employees submit claims)
accounted for more than 55 percent of health plan expenditures (Alexandre,
1988). National data suggest that 1 percent of the U.S. population (all age
groups) accounts for 29 percent of total

health expenditures (Berk et al., 1988). High-cost case management focuses on

these expensive patients.
Two basic types of high-cost patients can be identified (Rosenbloom and
Gertman, 1984). The first are those who suffer a sudden, catastrophic event,
such as a heart attack, stroke, accident, or complicated birth, that may require a
long period of hospitalization. Recovery to normal or near normal functioning is
often possible, and even those with a less complete recovery may need little
special medical care. The second category of patients is those with serious,
often fatal, chronic conditions, such as muscular dystrophy, certain cancers,
AIDS, and amyotrophic lateral sclerosis (ALS), who may be hospitalized
repeatedly and become more severely ill over time.
Not all high-cost cases are appropriate for case management. Sometimes
no less costly modes or sites of care are suitable or the patient's plan of
treatment already includes these alternatives.
The opportunities for savings through high-cost case management by third
parties arise from the complexity and fragmentation of the medical care system
and the limits built into health benefit plans. It is not easy to chart the most cost-
effective path of service for patients with catastrophic medical problems.
Certainly, most patients and families with high-cost medical problems start out
with relatively little knowledge of what lies ahead of them, and they may not
always learn from their physician about the subsequent care alternatives that are
possible or available. Specialized high-cost case managers can help patients and
families to understand what these options are and what they mean in terms of
cost, quality, convenience, and comfort.
Although physicians and hospital staff typically are more informed than
patients and families, they too may lack detailed knowledge about the range of
options that might benefit specific patients, particularly during the recovery or
long-term-treatment stages. The care of catastrophically ill patients often
involves several specialists, each tending to focus on specific problems, not the
whole situation. An individual with severe burns, for example, may receive
services from an internist, nephrologist, pulmonary specialist, burn surgeon,
infectious disease specialist, orthopedist, neurologist, gastroenterologist, and
psychiatrist (Mazoway, 1987). Hospital discharge planners often aid in
arranging posthospital services, but identifying the most cost-effective option is
generally not their charge.1 A case
1 The prospective per case method of paying hospitals used by Medicare encourages
hospitals to discharge Medicare patients as soon as it is medically appropriate. PROs,

intermediaries, or carriers are not authorized to undertake case management. The Health
Care Financing Administration (HCFA), however, is planning demonstration projects to
assess the role it might play, given Medicare's special characteristics. For payers whose
method of paying hospitals does not, in itself, stimulate hospital discharge planning,
continued-stay review may be used to prompt its
manager who is accountable to a purchaser of services, on the other hand, does

have that responsibility.
Case managers can also deal with limitations of benefit plans. Some
coverage limits, such as restrictions on frequency of home nursing care, may
keep down overall expenditures in a benefit plan but nevertheless increase
outlays for specific patients. Many employers who shy away from across-the-
board expansions in coverage have welcomed high-cost case management
because it looks at cost-effectiveness on a patient-specific basis.
ROLE OF THE PURCHASER

As is true of other types of utilization management, purchasers can bring
high-cost case management into play in at least three different ways. First, the
purchaser may actually employ case managers, as, for example, Ciba-Geigy and
Zenith have (Henderson and Collard, 1988). Second, and more typical,
purchasers may contract with an outside organizationan insurer, a utilization
management firm, a PRO, or other entityfor high cost case management
services. Third, they may rely on an HMO, insurer, or other entity that bears the
financial risk for benefit costs to undertake the function as part of their regular
cost-containment activities. For purchasers who offer multiple benefit plans to
employees, both the second and third strategies may be in place for different
employees, depending on which kind of health plan they select.
For those employers who contract for or agree to case management
services, some become closely involved with the function. They look for cases
to refer, demand detailed reports, analyze data and estimate savings
independently, and insist on controlling all exceptions to coverage limitations.
Other employers want only general reports and refuse any role in decisions to
waive coverage limitations.2
Employers also vary in their objectives of high-cost case management.
Some are concerned almost exclusively with avoiding unnecessary costs. Others
have multiple objectives: lower costs, higher quality of care, more humane care,
and assistance to patients and families in negotiating the complexities of the
health care system during a stressful and difficult period.
initiation. For uncomplicated cases, discharge planning may be quite modest,

involving no need to identify or help in arranging posthospital services.
2 The federal Office of Personal Management (OPM), for example, has been relatively
uninvolved with case management. Although OPM has agreed to allow some plans
serving federal employees to make coverage exceptions and to provide other case
management services, they originally did not allow the program to be mentioned in
brochures distributed to employees. In addition, OPM does not ask for reports on costs or
savings achieved through high-cost case management (Eileen Thomas, Blue Cross and
Blue Shield Federal Employees Program, personal communication, November 22, 1988).
The latter group of employers may accept the possibility that high-cost
case management might increase the cost of care for some patients. For
example, a patient near death from cancer might be allowed, through case
management, to receive short-term but expensive support through 24-hour-a-
day nursing services so as to be able to die at homeeven though palliative
care in a hospital might be less expensive.
HOW HIGH-COST CASE MANAGEMENT WORKS

Permutations are numerous, but the basic approach of high-cost case
management is fairly straightforward: to bring a knowledgeable party who is
concerned with cost containment into the process by which patient care
decisions are made for patients with expensive medical problems. The most
common elements of high-cost case management programs are
efforts to find less costly alternatives to the care (usually hospital care) that
would be provided for specific patients in the absence of case management;
willingness to approve payment for services not covered by the patient's
benefit plan, if doing so will help reduce overall costs; and
concurrence from all relevant partiespatient, family, and physicianin
implementing cost-saving alternatives for meeting the patient's needs.
Although the first two of the above elements appear to be virtually
universal in high cost case management programs, the third element is not, even
though it is often described as part of a standardized model.
Figure 5-1 charts in more detail the steps that are often involved in high-
cost case management. The process starts with case identification, proceeds
through assessment of the patient's suitability for high-cost case management
and the availability of alternative courses of treatment, and continues with the
development of a specific plan that is agreed to by all involved parties and then
implemented and monitored. The final step is closure of the case. This may
occur when the patient's condition changes or when coverage under a benefit
plan is exhausted or otherwise lost. Table 5-1 provides two hypothetical case
summaries.
The four categories of utilization management described in Chapter 3
generally apply to high-cost case management. The distinction between those
organizations with and without provider contracts is probably less important for
high-cost case management because cooperation rather than contractual
compliance tends to be stressed in either situation. However, because case
identification, as described in the next section, is such an important step in high-
cost case management, freestanding programs may face more difficulties than
case management undertaken by the insurer

Figure 5-1 Examples of steps and variations in case management.

or claims administrator. There is no systematic evidence on this point,

however.
Operational Variations
The committee learned during its site visits that high-cost case
management programs are highly variable in ways that are not always easy to
describe. The narrowest programs are, in essence, hospital discharge planning
programs that operate by means of telephone and confine themselves
TABLE 5-1 Typical Case Management Summary and Analysis
CONFIDENTIAL CASE REPORT 1
Patient: YYYY Client: XXXX
Referred by: VVVV
Prepared by: WWWW
Diagnosis: Lymphoma Date Opened: 2/14/88 Date closed: 3/9/88
INITIAL CASE SUMMARY
Patient is a 56-year-old woman with terminal breast cancer. Chemotherapy has been
ineffective, and death is expected within a month of this referral. Patient requires pain
management and skilled nursing and personal care. Family can provide some care but not
24 hours a day. Physician will continue hospitalization unless adequate home support can
be arranged.
OUTCOME
Patient was transferred home with support from registered nurse and home health aide.
Death occurred 14 days later.
COST ANALYSIS ASSUMPTIONS
Patient would have required continued hospital care.
Case management permitted discharge and maintenance at home.
Hospital days averted: 14 days
Dollars per hospital day averted: $600 (average billed this admission to date)
Actual home health expenses: (within contract) $2,000
(outside contract) $1,200
Case management fee: $700
SAVINGS
Hospital costs averted (14 $8,400
days $600)
Home health expenses ($3,200)
Case management fee ($700)
Net savings $4,500

CONFIDENTIAL CASE REPORT 2

Patient: RRRR Client: XXXX
Referred by: SSSS
Prepared by: WWWW Date Opened: 4/15/88 Date closed: 10/88
INITIAL CASE SUMMARY
Patient is a 15-year-old boy with head injury suffered in football game. After surgery,
patient was comatose and was in intensive care unit (ICU) for 1 week before mother's
employer contacted case manager. Attending physician estimates hospital care will be
needed for at least an additional 3 months followed by transfer to a rehabilitation facility
for several weeks.
OUTCOME
Patient spent 3 more weeks in ICU and was then transferred to a rehabilitation unit for an
additional 4 weeks. Home care was provided for 3 months. Family has taken over therapy
and the state is providing a tutor to prepare the boy to return to classes.
COST ANALYSIS ASSUMPTIONS
Patient would have been hospitalized for 3.5 months and would have spent another 3
months in a rehabilitation facility.
Case management permitted earlier transfer to rehabilitation facility and then substitution
of home care.
Hospital days averted: 75 at $1,200/day
Rehabilitation hospital days averted: 21 at $900/day
Actual home care expenses: $8,000
Case management fee: $6,200
SAVINGS
Hospital costs averted: $80,000
Rehabilitation costs averted: $18,900
Home health expenses: ($8,000)
Fee: ($6,200)
Net savings $64,700
to arranging out-of-hospital services that will allow earlier discharge of

patients. In broader programs, case managers may be involved with patients and
their families for many months and may visit them in the hospital, home, or
elsewhere. The activities of these case managers are not confined to facilitating
discharge from the hospital but extend to identifying and coordinating a broad
range of services.
High-cost case management programs also vary in how they identify cases,
who acts as case manager, who the case manager works with, and how directly
case managers are involved with patients and providers. The differences among
case management programs undoubtedly have important implications for the
impact and effectiveness of high-cost case management.

However, these differences have received virtually no empirical study, and

studies of one particular program are often generalized and assumed to apply to
quite different strategies.
How Cases Are Identified and Screened

The case management process may be activated in several different ways.
In many instances, the case manager is notified of a potential case by an
employer, family member, or provider who is aware of the program and knows
a patient who may benefit from it. In such situations, the case manager is likely
to become involved early when the opportunities are greater for influencing the
course of treatment to avoid needless expense. For these reasons, case
management organizations may work with the purchaser to increase employees'
awareness of the program and encourage early contact. Organizations with a
significant presence in a local area may also work with community hospitals
and physicians to bring them into the case identification process. In addition, an
employer's disability insurance plan may refer cases for high-cost case
management
If the organization providing high-cost case management also provides
other utilization management services, it may use its preadmission, admission,
and continued-stay review programs to identify patients that are candidates for
preadmission review. Generally, this involves specifying a set of target
diagnoses such as stroke, ALS, or muscular dystrophy that are flagged by
reviewers or their computer software. The lists of target diagnoses overlap but
are not identical across different companies. Repeated admissions or multiple
extensions of length of stay may also serve as flags. Overall, preadmission
review and admission review may be of limited use, because many patients with
catastrophic illnesses or injuries are admitted on an emergency basis and the
prognosis may be highly uncertain for several days. Also, high-cost case
management is generally handled by a separate staff that may not routinely
communicate with the prior review nurses.
Potential high-cost cases may also be identified through the claims
administration process. Claims may be screened for target diagnoses; for
cumulative payments beyond a certain threshold, for example, $25,000; or for
certain patterns of care, such as repeat admissions. However, by the time a high-
cost case is identified from claims for payment, several months of care may
already have been provided and chances to initiate less costly care missed.
Nevertheless, retrospective analysis of claims caught and missed by a case
management program may be useful in improving the program by suggesting
where opportunities are being overlooked.
To avoid inefficient use of expensive case management resources, the first
step is a review of basic demographic, clinical, and claims information.

Cases with a reasonable potential for case management may then be

subject to a more intensive review, including direct communication with the
patient and family.
Although all case managers make judgments about what cases to accept,
they vary in how they make their decisions. These variations relate either to
client objectives and concerns or to vendor capacities and strategies. Some
organizations, based on their own and their client's objectives, reject cases with
no or little potential for short-term savings. Other organizations reject cases
with complications they feel they are unable to handle (for example, psychosis).
Some clients want intensive case management efforts for AIDS and psychiatric
cases but are less concerned about other problems. If high-cost case
management operates in tandem with per-case payment to hospitals, third-party
case managers may get involved only when hospital staff face an unusual case,
perhaps one that requires exceptions to benefit plan limitations to make more
cost-effective care feasible.3 These sorts of decision rules may or may not be
codified in internal procedure manuals and contracts with clients.
Who Serves as Case Manager?

One of the largest national companies offering high-cost case management
services uses physicians as case managers. In all other programs known to the
committee, nurses are used. The nurse managers always have access to
physicians for consultation but manage most assessment, planning,
coordinating, and monitoring activities themselves. Organizations differ in
whether their nurses handle a wide array of cases or specialize. The most
frequent area of specialization is in psychiatric case management. Some
organizations use psychologists, social workers, and other professionals to
assist case managers or handle certain types of patients or specific problems.
They often speak of a ''team approach'' to case management in which several
types of professionals contribute to patient assessment and recommendations.
How Case Managers Relate to Patients and Providers

Most high-cost case management programs report that they seek to work
with all interested partiesthe patient and family, the doctor, the
3 Blue Cross and Blue Shield plans that pay hospitals on a diagnosis-related group
(DRG) or other per-case basis report some difficulties in serving national employers who
want a uniform program at all company locationseven if it is not cost-effective at some
sites (Michael Cologero, Blue Cross and Blue Shield Association, November 21, 1988).
discharge planners at the hospital, home care agencies, and others. Their initial
contact may be with any of these parties. However, some organizations work
primarily with the patient and family, whereas at least one major national
company works only with the patient's physician.
The emphasis placed on patient and family concurrence with high-cost
case management recommendations varies. HMOs appear to be more directive,
perhaps insisting on a choice between termination of coverage for services
outside the treatment plan or agreement to follow the plan. However, not a lot is
documented about HMO management of high-cost cases. Much of that
management may be invisible to the patient because it occurs when the
physician calls for approval for hospitalization or use of out-of-plan facilities or
services. High-cost case management appears to be integrated into HMO
administration (group, staff, and independent practice associations [IPAs]) more
than in fee-for-service administration.
To date, most utilization management organizations and insurers working
with fee-for-service health plans appear to use a voluntary approach. There are
no statistics on patients who refuse case management, but some patients are
reported to regard coordination with the case managers as a hassle rather than a
help. The committee heard suggestions that some purchasers could become
more insistent on high-cost case management when the likely cost of a patient's
rejecting case management is significant. Some case management organizations
do follow-up surveys of patients and families who have been involved in the
process; others do not.
Case managers generally seek the cooperation of attending physicians. The
attending physicians have essential information about the patient and the course
of treatment, can influence patients' and families' acceptance of case
management recommendations, and can assist in implementing
recommendations by approving hospital discharges, ordering or providing
necessary medical services, or supervising the care provided by others.
Face-to-face work with patients and physicians is most characteristic of
HMOs and utilization management programs that operate in a relatively
confined geographic area. However, at least one national company the
committee visited attempts to do on-site case management through a national
network of offices and employees. Blue Cross and Blue Shield plans in
different states sometimes contract with each other to provide on-site case
management services for patients located outside of their service area. Over 25
Blue Cross and Blue Shield plans are involved in a program for national
employers that includes a utilization management component and involves a
certification process using written application forms, site reviews, and ongoing
guidance groups. Since on-site work is more expensive, particularly if it
involves travel or subcontracting, most organizations target its use carefully.

Some case management firms establish contractual relations with home

health care agencies, nursing homes, rehabilitation centers, durable medical
equipment suppliers, and other institutions. These contracts may specify that
clients of the firm have priority when the health care organization is operating
near capacity. Contracts may also provide for discounts or negotiations on
charges. However, even without such contracts, case managers may negotiate
with provider organizations on charges for particular patients. Centralized firms
without a local presence may be at some disadvantage in developing these kinds
of relationships with providers.
How Much Case Management Costs

Another variable factor with high-cost case management programs is their
cost, whether cost is defined as the price the purchaser of the service pays or the
resources the utilization management organization requires to operate
(Henderson and Wallack, 1987; Cologero, personal communication, 1988).
Some insurers have not explicitly charged for case management services but
have incorporated the cost as part of their premium or administrative services
charge. Other insurers and some independent firms set a flat charge per covered
employee, particularly if high-cost case management is part of a larger package
of utilization management services. Most commonly, programs bill on a flat-fee
or hourly basis for assessment, coordination, and other services. Hourly charges
of $25 to $100 or more have been reported, with charges tending to fall toward
the higher end of this range. For a case involving 40 hours of management
services, the difference in administrative fee could be $3,000 for companies at
the low versus the high end of the range.
IMPACT OF HIGH-COST CASE MANAGEMENT

There seems to be general acceptance that high-cost case management can
save money and improve the quality of care and the quality of life for patients
and families. Results probably vary depending on a large number of factors
the nature of the patient's problem, the family circumstances, physician and
hospital cooperation, community health resources, employer objectives, and
sophistication of the case management program. However, very little systematic
evidence is available to support any specific conclusions about the effects of
high-cost case management. Illustrative reports on case management program
results suggest the kind of information and analysis typically available.
In Oregon, for example, the Bargaining Unit Benefits Board for the
Service Employees International Union (SEIU) decided in 1985 to replace
concurrent review with high-cost case management. The case management

services were provided by the Oregon PRO that also administers preadmission
review and second surgical opinion programs for SEIU. Program savings from
case management were calculated by projecting charges that were averted by
the use of case management (such as additional days of hospital care) and
subtracting the cost of alternative services and the PRO administrative fee. In
the first year, the union reported that more than $13 was saved for each program
dollar spent, but the savings-to-cost ratio subsequently dropped to 4:1 (Service
Employees International Union, 1988; Jo Anne Browne, Service Employees
International Union, personal communication, January 19, 1989).
The Bank of America adopted a case management program for patients
with catastrophic illnesses as one outgrowth of its efforts to deal with its
growing number of employees with AIDS. In the program's first year, the bank
reported savings of $1.2 million compared with costs of $155,000, a savings-to-
cost ratio of more than 7.5:1 (Ricklefs, 1988).
Independence Blue Cross of Philadelphia (formerly Blue Cross of Greater
Philadelphia) began individual case management in 1987. For the 30 cases
actively managed in that year, it reported savings of nearly $2 million, over
$65,000 per case on average (Independence Blue Cross and Pennsylvania Blue
Shield, 1988).
Intracorp, a firm with a history in case management in the workers
compensation arena, has reported savings as high as $75 for every dollar spent
on case management. It reports a savings ratio of 17:1 for a recent review of 14
typical cases (Tonsfelt, 1986).
In a study of an unnamed insurer (and an unnamed separate utilization
management firm), Brandeis University researchers tried to project what would
have happened without high-cost case management. They asked consultant
physicians to examine a sample of the study and project how much hospital care
would likely have occurred without the intervention of the case manager.
Among the small numbers of cases examined initially, savings to cost ratios
ranged from 1.18:1 to 0.42:1. The researchers identified spinal injury cases as
likely to have higher rather than lower costs under case management because of
improved identification and use of rehabilitation options and institutions. In
general, the consultants were considerably more conservative in their
projections of savings than the case managers (Henderson et al., 1987; Mary
Henderson, Brandeis University, personal communication, November 21,
1988). One purchaser of case management services, Caterpillar, Inc., attempts
to compensate for exaggerated estimates of cost savings by making its own
projections of what use and costs would be in the absence of intervention and
then cutting those estimates by one-quarter. Caterpillar uses a PRO to provide
the case management services (Richard Wright, personal communication,
November 22, 1988).

EFFECTS OF HIGH-COST CASE MANAGEMENT ON

SPECIFIC PARTIES
For the most part, high-cost case management appears to have been
favorably received by patients, families, and health care providers. The
committee heard few complaints but numerous anecdotes of positive responses.
The following discussion is based on the committee members' experience,
testimony at the committee's June 1988 hearings, the 12 site visits, the
roundtable discussion among corporate benefits managers in January 1989, and
the six meetings of the committee.
Effects on Enrollees
In discussions with those offering and purchasing high-cost case
management services, most recountedwith some feelingstories of
individuals who have been assisted by case management in confronting
devastating medical problems. Even in the less dramatic cases, the picture
conveyed is of people who are grateful for additional information, guidance,
and support in dealing with stressful and complex situations. The major concern
expressed to the committee involved questions of confidentiality for patients
with AIDS, cancer, or other conditions that might provoke negative reactions at
their place of work.
Three characteristics of case management, as currently practiced, appear to
be responsible for its generally positive reception. First, case management tends
to be quite individualized and cognizant of the special problems created for the
patient and family by the particular nature of the patient's medical problem, the
quality of the home environment, and the limitations of the community's health
care resources. To some degree, then, high-cost case management escapes the
impersonality and rule-bound image that tends to accompany programs dealing
less intensively generally with more patients. Second, most organizations
emphasize cooperative work with patients, families, and physicians and
consensus on the design and implementation of the treatment plan. They do not
put patients in the position of going along or losing benefits. Third, high-cost
case management is often a vehicle for providing a patient and family with
more services and options than would normally be available under the patient's
benefit plan. The patient and family ordinarily are quite aware of these extra
benefits.
To the extent that the administration of high-cost case management departs
from these patterns, this largely positive assessment may not hold. For instance,
if the case manager offers the patient and family a choice between loss of
coverage and implementation of an alternative treatment plan, then the program
may be perceived negatively. Likewise, if the

family must take on substantial extra burdens of care but the assistance of
special home services is denied, then high-cost case management may add
rather than relieve stress. And poor administration of a programby virtue of
poorly selected and trained staff, inadequate documentation of conditions and
decisions, and weak assessment and monitoring of alternative providersan
easily undermine the potential positive features of high-cost case management.
The committee notes these possibilities but does not have evidence that they
now constitute real problems.
Effects on Health Care Providers

Although the pattern of effects on health care providers may be somewhat
more mixed than they are for patients and families, the committee believes that
high-cost case management has probably had a preponderance of positive
effects. For the attending physician and hospital discharge planner, the case
manager can often bring to bear a more specialized knowledge of certain
alternative treatments and providers and a more intensive assessment of patient
circumstances and options. For example, the case manager may know more
about innovative rehabilitation services or specialized home care options and
have established relationships with these providers so that it is easier to tailor
services for particular patients. The committee has no information about the
administrative costs that high-cost case management creates for physicians and
hospitals, but the interactions described by those involved suggest that it
imposes few additional costs and may, in some cases, substitute its resources for
those of the physician or hospital.
Physicians and hospital discharge planners faced with high-cost case
management for the first time may feel somewhat threatened, but familiarity
seems to mitigate initial sensitivities. Most firms say that they try very hard to
select and train case managers to work constructively with physicians and
hospital staffs. One potential source of conflict with providers is the contractual
or other preferential arrangements with particular health care institutions that
some case management organizations have established. Selective arrangements,
per se, may create hard feelings, but the charge that high-cost case management
companies may have a conflict of interest and may be channeling patients to
lower-quality institutions is a more serious charge. The committee was unable
to find specific evidence on this point and heard only a few references to this
possibility during its deliberations.
Effects on Purchasers
Although the benefits of case management can be identified in the abstract,
it seems that purchasers have sometimes been surprised at how

positive the reactions from employees have been. The personalized nature of
most high-cost case management may make the benefits more tangible to both
employers and employees. And it promises more support for access and quality
care than it poses threats. However, the lack of concrete, reliable information on
whether high-cost case management is saving money is an irritant to purchasers.
There is some suspicion that savings are overestimated and that more time is
being charged than is strictly necessary. The committee heard few complaints
about the lack of empirical data on health outcomes, patient satisfaction, or the
impact of alternative program designs. On balance, high-cost case management
seems to be regarded by purchasers as more of a "win/win" program than other
utilization management and cost-containment programs are.
QUESTIONS ABOUT AVAILABLE EVIDENCE

As the study committee discovered for other utilization management
programs, much of the evidence cited in support of high-cost case management
is of questionable valuefor many of the same methodological and
organizational reasons. (Table 5-2 summarizes measures of impact commonly
reported for high-cost case management programs and notes some of their
limitations.) On the other hand, certain characteristics of case management
programs can make their evaluation easier.
One characteristic of high-cost case management that facilitates evaluation
of utilization and savings is the explicit focus on the cost-effectiveness of
proposed alternative care. For each case, an estimate is made of the cost of
alternative care versus that of care that is likely to be provided without case
management. These cost estimates are sounder when they rely on information
about the specific hospitals, home health agencies, and other providers that are
involved or likely to be involved with the patient. Some programs track billing
data during the period of active case management to determine the actual cost
of care. Since alternative treatment plans may be changed, complications may
arise, and other divergences from the projected course of care may occur, this
sort of follow-up is important in assessing cost and other impacts.4 Such
tracking and follow-up may be easier when the high-cost case management and
claims administration functions are managed by the same organization.
Certainly, the case-by-case projections of use and costs incurred and
avoided by high-cost case management may be flawed in various respects.
4 predetermination programs, in contrast, rarely if ever attempt any case-by-case
assessment of whether outpatient surgery will likely be less expensive than inpatient care
in a specific case. That is, the probable expenditure for inpatient care at a particular
hospital is not compared with the probable outlay for a particular alternative site of care.
The assumption has been that, on average, the alternative care will be cheaperan
assumption that is no longer taken for granted.
The program may take credit for treatment strategies that would have been
adopted anyway. For example, home care is often arranged for terminally ill
patients through family, physician, and hospital initiatives without the
involvement of a third-party case manager. In these situations, net savings may
still occur from case management if the program produces earlier and
TABLE 5-2 Measures of High-Cost Case Management Impact
ACTIVITY DATA
Number of cases referred for assessment
Number of cases initiated
Number of cases closed
Number of cases ongoing
Average length of case management
Source
Review organization
Comments
Are not impact measures.
Cannot be used to evaluate efficiency absent additional information on characteristics
of cases, client preferences, cases missed or identified late, etc.
UTILIZATION
Inpatient days of care projected without case management
Alternative services (projected or actual)
Home health
Skilled nursing facility
Durable medical equipment
Other
Source
Claims data
Review organization
Comments
Projected utilization of hospital care is easy to overstate, particularly during periods of
changing practice patterns.
Projected use of alternative services is less satisfactory than are reports based on actual
claims data.
Generally focused on short-term costs and savings.
BENEFIT COSTS
Projected costs of care without case management
Projected or actual cost of substituted services
Cost of services for which contract limitations are waived
Source
Claims data
Bills submitted for special services
Review organization
Comments
See comments on utilization projections.
Cost estimates may not differentiate between more expensive early days of stay and less
expensive (typically) later days and, thereby, may overestimate savings from reduced
length of stay.

ADMINISTRATIVE COSTS
Billed services during specified period
Per contract charge for all contracts
Source
Review organization
Comments
Charges to client may not reflect the real administrative cost of the program,
particularly in a multiservice organization that has not done cost studies for the service.
Do not reflect any costs to patient or providers.
May not always be billed separately from other utilization or benefits management
services.
SAVINGS
Projected cost of care without intervention minus high-cost case management benefit
and administrative costs
Ratio of benefit savings to benefit and administrative costs
Source
Claims data
Review organization
Comments
See comments on use and cost measures.
Needed to assess net impact of high-cost case management on purchaser costs.
Do not include any costs or savings for patient or provider.
more successful transfers (consistent with physician and patient
preferences) and if administrative fees do not exceed savings.
Some projections of averted costs are more plausible than others. To cite
one case, during the process of planning a case management program, staff for
one insurer examined computer printouts to locate patients who had generated
very high costs in recent years. In the process, they discovered a child with
muscular dystrophy who was ventilator dependent and had been hospitalized
continuously for 8 years. The parents' benefit plan had covered the care and
would have covered continued hospitalization. With the family's and physician's
cooperation, the insurer arranged appropriate home care and waived contract
provisions restricting payment for some of these services. The insurer's
intervention clearly precipitated the move to home care for this child and
reduced expenditures for the insurance plan (Thomas, personal communication,
1988).
Nevertheless, trying to assess more generally the savings produced through
case management is difficult without some assessment of comparable groups
not subject to the program. Clearly, practice patterns are changing for many
reasonsnew technologies, more sophisticated resources for out-of-hospital
care, expanded health plan coverage of such

care, and new methods for paying institutions and physicians. In the absence of
comparative data, it is hard to know what savings attributed to high-cost case
management may really be due to other factors.
The committee was unable to discover any studies that included
comparison of a group subject to case management with one having no such
program. Researchers at Brandeis University attempted such comparisons in
their work with one insurer, but limitations in the claims data precluded it
(Henderson, personal communication, 1988).
The relatively small number of cases subject to case management
compared with the large numbers processed through predetermination programs
and the relatively intense nature of the case management process do ease some
burdens of projecting and comparing use and costs. However, the time horizon
for assessment still tends to be constrained, partly for practical reasons, such as
lack of historical data and a short program life span, and occasionally for policy
reasons, in particular, a preference for being conservative in projecting savings.
One example of this conservatism can be found in the Brandeis University
research just mentioned. The researchers concentrated in their initial reports on
short-term savings that "would be realized within six months of implementation
of the case management plan" and that ''were clearly documented and supported
by medical record data and physician input" (Henderson, 1987, p. 44). Longer-
term savings were not studied in the initial assessment because they were both
harder to track and more speculative.5
Although the relatively small number of patients in most high-cost case
management programs has some advantages for evaluation purposes, it also
means that the mix of casesand program effectscan fluctuate for reasons
unrelated to the program or to other systematic influences. Case mix and
program results can also be affected by program design.
For instance, how cases are targeted and screened for acceptance into case
management programs can systematically affect the composition of a group
subject to case management. A program that does not accept a case unless there
is potential for short-term savings could be expected to show higher per-case
savings than one that accepted cases with potential for long-term savings only.
Programs that accept cases with little potential for cost savings but considerable
potential for improving patient and family quality of life would also show lower
per-case savings. Ignoring such program
5 The potential for longer-term savings seems reasonable in some cases, such as
chronically ill patients who, with careful management, can avoid repeated
hospitalizations. The possibility that long-term effects could include higher costs
because of inappropriate short-term treatment under the case management plan does not
seem to be expected. This may be because the intensive nature of the management
process and the communication with the attending physician and the voluntary aspect of
high-cost case management is thought to make this prospect unlikely.
features could lead to incomplete judgments about the effectiveness of

particular programs. However, these details may not be identified explicitly in
the program description or assessment.
In addition, the potential impact of other cost-containment initiatives may
be not be considered. For example, if a purchaser or insurer moved to a per-case
payment method with limited provision for outlier payments at the same time
that it implemented high-cost case management, it might find case management
of little or no marginal value. The results under these circumstances should not
be generalized to programs associated with more traditional provider payment
methods. (Note that Medicare does not have a high-cost case management
feature, although demonstration projects are being planned.)
LEARNING CURVE IN HIGH-COST CASE MANAGEMENT

One reason that relatively little objective evidence is available on the
impact of high-cost case management is that the programs are relatively new.
Most are still evolving. Savings may grow, administrative costs may drop, and
documentation of results may become more specific and realistic. Such
improvements seem likely because of several types of changes in case
management programs. These include
better targeting and earlier identification of cases,
elimination of overlapping cost-containment efforts,
increases in administrative efficiency and computer support,
development of standardized clinical protocols to guide the management of
more cases, and
establishment of more economical and effective arrangements with health
care providers.
As organizations gain familiarity with high-cost case management, they
generally look for ways to improve case identification and screening. In some
cases, this means adding new conditions to their lists of target diagnoses or
problems, for example, AIDS cases, spinal cord injuries, and psychiatric cases.
Or certain target diagnoses may be eliminated, and some diagnoses may be
subject to more up-front screening for associated diagnoses, past
hospitalizations, or other indicators of high-cost potential.
Yet another targeting strategy is to examine whether certain services can
be more effectively managed as part of the basic benefit plan rather than
through an exceptions process. A manager for Caterpillar tractor said that
adding a regular home care benefit in April 1988 ''almost put their case
management program out of business'' (William Beale, Caterpillar, Inc.,
personal communication, November 30, 1985). Patients recuperating from
relatively straightforward acute problems are now rarely seen by

case managers whose smaller caseload consists mostly of chronically and

terminally ill individuals.
Another way organizations try to improve their results is through earlier
identification of potential cases when more options for lower-cost care may be
considered.6 Links to precertification and continued-stay review, informational
mailings and seminars for health care providers, and involvement of employer
personnel offices are common strategies. These steps involve closer integration
of high-cost case management, claims administration, provider relations, and
human resources functions, which can be difficult even when the first three
functions are handled by a single organization.
As links to other aspects of utilization management are developed, it is
sometimes possible to gain administrative efficiencies. For example, continued-
stay review is unnecessary for cases already accepted for case management.
Other administrative efficiencies are possible through refinement of information
collection procedures and routinization of some communications with health
care providers. Better computer support can increase efficiency at all stages in
the case management process, from case identification and screening through
outcome documentation and reporting.
Standardized protocols for managing certain kinds of cases (for example,
substance abuse or AIDS) may minimize the need for intensive involvement of
a case manager in the clinical assessment of the patient and the development of
a treatment plan (Henderson and Wallack, 1987). If methods for identifying
when monitoring or intervention is no longer cost-effective can be improved,
cases may be closed more promptly, resulting in lower administrative and
benefit costs.
Experience can also help case managers develop greater sensitivity and
efficiency in handling certain kinds of difficult cases, for example, AIDS
patients with dementia or nonaccepting families. In addition, as relationships
with providers are established and reinforced over time, the identification of
good alternative sources of care requires less effort. The prospects for
negotiating reduced prices for care or other special arrangements increase.
6 For example, AIDS cases are thought to be good targets for case management to
reduce costs and improve patient quality of life (DiBlase, 1987; Health Insurance
Association of America, 1988; Taravella, 1987). However, the immediate cause of
hospitalization is generally a problem such as pneumonia, and there are understandable
reasons an explicit AIDS diagnosis may not be mentioned in information provided to an
insurer, employer, or utilization management firm. A diagnosis-based method for
identifying probable high-cost cases will not, absent other information, pick up such
cases. Therefore, some case management programs are trying to develop better early
identification strategies such as having predetermination software screen each
pneumonia case for age and sex of patient and any previous admissions. Sensitivity to
confidentiality is a particular concern for this kind of approach, but confidentiality issues
are significant for many high-cost care management cases.
POTENTIAL LEGAL ISSUES

There is virtually no case law, legislation, or regulation that specifically
delineates the responsibilities and liabilities of organizations involved in high-
cost case management. However, three areas of potential concern can be
identified.
First, there are several questions about liability for harm to a patient.
Because high-cost case management can shape the course of treatment for a
patient, the possibility exists that a patient or family who has suffered a bad
outcome will claim that case management contributed, and a jury may agree.
The case of Wickline v. the State of California, discussed in the paper by
William A. Helvestine in Appendix A of this report, makes it clear that
organizations engaged in utilization management may be held liable for their
failure to exercise clue care in designing and administering a program.
Although not mentioned in the Wickline case, a number of cases involving
claims payment administration suggest that an organization would also have to
act in "good faith" in administering a case management program. Some
employers, citing sensitivity about perceived conflicts of interest, are reluctant
to have their own physicians or other employees involved in high-cost case
management and prefer to hire outside firms instead.
Actions that might put an organization at risk are the use of poorly selected
and trained personnel, negligent referrals to particular providers, and poor
protocols for collecting information about patients' conditions. A capricious
refusal to make an exception to a benefit contract might also give rise to a legal
challenge. Good documentation of decisions and rationales, as well as reliance
on written protocols for developing treatment plans and referrals, help
demonstrate the exercise of clue care and good faith (Saue, 1988). The degree
to which the attending physician and the patient are involved in developing and
consenting to the treatment plan may also be relevant in assessing liability.
A somewhat different risk may arise if a case manager becomes aware of
improper treatment by a patient's physician. One commentator believes that the
case manager may be responsible for discussing such problems with attending
physicians and, if that does not work, notifying the quality assurance committee
of the hospital or medical society (Blum, 1989).7
A second legal concern about high-cost case management is whether
exceptions that are made to limits in benefit contracts will set a precedent
7 One question not raised in the site visits or hearings is whether a nurse case manager
might be practicing beyond the scope of his or her nursing license. Blum (1989) raises
this question but suggests that problems are unlikely so long as a nurse manager does not
attempt to control treatment decisions. Blum does, however, raise doubts about whether
the use of licensed practical nurses rather than registered nurses as case managers is
appropriate.
that undermines the contract. For instance, if a patient for whom exceptions
have been made, say, to cover special home care services, reaches the
maximum dollar coverage of the benefit plan, that patient might argue that since
some limits had already been waived, the limit on maximum benefits should not
be enforced.8 Such situations do not appear to be very common, but case
management organizations do report a few sensitive cases. Written agreements
among parties descrying the nature and limits of any benefit exceptions are a
recommended protection (Saue, 1988). Another way organizations can
minimize problems when coverage maximums are being approached is to work
with patients and their families, state Medicaid agencies, and charitable
organizations to arrange a smooth transition to other sources of financing and
avoid disruption in the plan of treatment.
A third concern is that an individual whose request for an exception to
contract restrictions was denied could try to argue unfair treatment because
exceptions had been made for others covered under the same plan. A related
worry is that a claim of discriminatory treatment might be prompted under
Section 89 of the Internal Revenue Service Code, which forbids discrimination
between more and less highly paid employees. In order to protect against a
successful charge of discrimination, some organizations involved in high-cost
case management are improving their case documentation, clarifying contract
provisions, securing written agreements to contract exceptions, and tracking the
characteristics of employees assisted by case management (Cologero, personal
communication, 1988).
In general, reasonable administrative procedures should be in place to
guide any prospective or retrospective case-by-case decisions about benefits.
And these procedures should be implemented responsibly and documented
systematically for both routine and exceptional cases. Evidence of careless,
biased, and uninformed decision making is hazardous whether a complaint
involves a routine payment denial, a bad health outcome, or a charge of
discriminatory treatment.
CONCLUSION
Case management has become a popular utilization management strategy
in a very short period of time. Although it has been the subject of even less
systematic research than prior review techniques, claims of positive impact for
case management appear to be generally accepted. The committee believes that
several characteristics of high-cost case management
8 Some plans set no limit on the maximum total amount that may be paid under the
contract. Where limits exist, they vary from less than $250,000 in some groups to more
than $1,000,000 in others. Many plans set lower limitsoften $50,000on psychiatric
benefits.
account for this acceptance: the relatively personalized and intensive

assessment of patient situations; the explicit attention to the cost-effectiveness
of alternative care; the potential for patients to receive extra benefits; and the
emphasis on consensus in arranging alternative treatment. During the
committee's deliberations, few concerns were raised about high-cost case
management specifically. In the concluding chapter, the committee's concerns
focus less on case management and more on issues involving the less
individualized methods of prior review.
REFERENCES
Alexandre, Leslie M., "Who Are the High Cost Cases in a Health Benefits Plan?," Medical Benefits,
September 15, 1988, p. 7.
Berk, Marc L, Monheit, Alan C., and Hagen, Michael H., "How the U.S. Spent its Health Care
Dollar, 1929-80," Health Affairs, Fall 1988, pp. 46-60.
Blum, John D., "An Analysis of Legal Liability in Healthcare: Utilization Review and Case
Management," Houston Law Review, January 1989, pp. 191-228.
Corporate Health Strategies, The Health Poll, Fall 1988, p. 1.
DiBlase, Donna, "AIDS Claims Hard to Track for Insurers," Business Insurance, September 7,
1987, p. 22.
Equitable Life Assurance Society of the United States, The Equitable Health Care Survey: Options
for Controlling Costs, conducted by Lou Harris and Associates, Inc., New York, August
1983.
Health Insurance Association of America, AIDS Case Management: What Health Insurance
Companies Are Doing, Washington, DC, 1988.
Henderson, Mary, and Collard, Anne, "Measuring Quality in Medical Case Management Programs,"
Quality Review Bulletin, February 1988, pp. 33-39.
Henderson, Mary, and Wallack, Stanley, "Evaluating Case Management for Catastrophic Illness,"
Business and Health, January 1987, pp. 7-11.
Henderson, Mary G., Souder, Barbara A., and Bergman, Andrew, "Measuring Efficiencies of
Managed Care," Business and Health, October 1987, pp. 43-46.
Independence Blue Cross and Pennsylvania Blue Shield, Independence and Leadership in Health
Care: Community Health Care Report 1988, Philadelphia, June 1988.
Mazoway, Jackie M., "Early Intervention in High Cost Care," Business and Health, January 1987,
pp. 12-16.
Merrill, Jeffrey C., "Defining Case Management," Business and Health, July/August 1985, pp. 5-9.
Monroe County Long Term Care Program, Inc., "Direct Assessment vs. Brokerage: A Comparison
of Case Management Models," Final Report for Robert Wood Johnson Foundation, East
Rochester, NY, October 1986.
Ricklefs, Roger, "Firms Turn to 'Case Management' To Bring Down Health-Care Costs," Wall
Street Journal, February 7, 1988, p. D1.
Rosenbloom, David, and Gertman, Paul M., "An Intervention Strategy for Controlling Costly Care,"
Business and Health, July/August 1984, pp. 17-21.
Saue, Jacqueline M., "Legal Issues Related to Case Management," Quality Review Bulletin, August
1988, pp. 239-244.
Service Employees International Union, "Utilization Review and Case Management in Employee
Benefit Plans," Washington, DC, July 1988.
Taravella, Steve, "Coping with AIDS," Business Insurance, September 7, 1987, pp. 1, 20-22.
Tonsfeldt, Lynne, "Using Medical Case Management to Reduce Catastrophic Injury and Chronic
Illness Coverage Costs," FORUM, 1986.

CONCLUSIONS AND RECOMMENDATIONS 143
6
Conclusions and Recommendations
Utilization management is part of a complex balancing act created by

society's struggles with two important questions. First, how do we ensure that
people get needed medical care without spending so much that other social
objectives are compromised? Second, how do we discourage unnecessary and
inappropriate medical services without jeopardizing necessary, high-quality care?
Experience indicates that these questions have no fixed answers. Rather, as
suggested by the first two chapters of this report, we find a series of working
hypotheses and partial solutions that are continually revised, discarded, and
even reinvented as medical technology, social values, economic conditions, and
other circumstances change. In this preliminary report, the committee has
examined one current working hypothesisthat externally applied assessments
(those occurring outside the relationship between the patient and physician) of
the appropriateness of proposed medical services can improve how care is
provided and, as a consequence, help constrain health benefit costs. The validity
of this hypothesis is of interest to everyone involved with health carepatients
and potential patients, providers, employers, unions, insurers, and public
policymakers.
Although attempts to manage the use of health services on a prospective
case-by-case basis are not inventions of the 1980s, the application of utilization
management techniques in private health benefit plans has become widespread
only during the last half-dozen years. Its growth reflects purchasers' dismay
over continuing rapid rises in health care costs and their perception that much
care is unnecessary.

The Institute of Medicine is studying utilization management not only

because the field is expanding but also because it has important implications for
the delivery of medical care that have not been systematically investigated. On
the one hand, utilization management has features that make it an attractive
strategy for managing health benefits. When successfully implemented,
utilization management takes the clinical circumstances of individual patients
into account and gives patients and providers its assessments about whether
services are appropriate before rather than after care is provided. It does not
involve across-the-board limitations on health plan coverage. On the other hand,
utilization management has certain unattractive aspects. It acids to the
administrative demands on both patients and providers and employs criteria for
decision making that are sometimes vague or even secret.
In this first phase of its work, the committee has examined two major
methods of utilization management: (1) prior review of the appropriateness of
proposed medical services and (2) high-cost case management. The first
category includes several techniquespreadmission review, admission review,
continued-stay or concurrent review, discharge planning, and second surgical
opinion.
One of the committee's basic findings is that there is limited empirical
evidence on which to make authoritative conclusions about the impact of
utilization management at this time. Most private sector programs are relatively
new, and the field has been changing rapidly. Therefore, the committee has
tried initially to frame the important questions, offer reasoned judgments about
the state of utilization management, and suggest directions for the future. As
stated in the Preface, the committee has based its judgments and projections on
reviews of published and unpublished reports on the workings and effects of
utilization management, site visits by the committee to several utilization
management organizations, and hearings and discussions involving consumers,
providers, purchasers, and others. Judgments also reflect the committee
members' extensive and diverse experience in health care delivery and financing.
This chapter summarizes the committee's findings about the current status
of utilization management and its evolution. It identifies some issues raised by
that assessment and recommends what prudent decision makers can do to
promote the positive potential of utilization management
CURRENT STATUS OF UTILIZATION MANAGEMENT

Several features of utilization management are important to keep in mind:
(1) its great diversity, (2) its limited scope, and (3) its relative lack of regulatory
oversight. First, utilization management, as it currently operates in the private
sector, is highly variable, which makes generalization about it

difficult. Some variations may reflect the industry's youth, for example, the
growing but uneven extent of computerization. Other variations, such as those
involving links with providers and claims payment arrangements, may depend
on whether the utilization management function is freestanding or integrated
into an insurance or prepaid health plan.
Second, utilization management has, until recently, emphasized the site,
duration, and timing of medical care. The unnecessary and inappropriate use of
the hospital, and not the actual need for a particular procedure, has been the
main focus. Review organizations vary in their willingness to tackle new issues
of appropriate care, but they generally appear to follow the lead of researchers
and medical groups in adopting criteria or selecting targets for review. Also, as
a general rule, prior review programs have not made case-by-case assessments
of the comparative costs of alternative treatments or sites of care. In contrast,
high-cost case management programs have built such assessments into their
evaluations of proposed courses of treatment. The main strategy has been to
discuss and negotiate appropriate care rather than to refuse prior authorization
of benefits explicitly. Denials rates appear to run about 1 to 2 percent of cases
with some later changed upon appeal.
Third, utilization management in the private sector operates under few
explicit regulatory restrictions. There is, however, considerable awareness
among review organizations and major purchasers of the legal risks inherent in
efforts to influence patient care decisions and operationalize the terms of health
benefit plans. And there appears to be growing recognition of the conventional
but not infallibleprotections offered against liability by good management,
good judgment, good faith, and good documentation.
As discussed in Chapters 4 and 5, empirical evidence on the effects of
utilization management is limited and suffers from a number of methodological
weaknesses. Although the number of comparative multivariate analyses is
increasing, there still is little or no evidence about the consequences of specific
differences in the way utilization management is conducted by different
organizations.
Despite these limitations, the committee believes that the available reports
and studies, taken together, suggest that purchasers who have adopted
utilization management have experienced some changes in health care use and
costs. Specifically:
Utilization management has helped to reduce inpatient hospital use and
limit inpatient costs for some purchasers beyond what could be expected
from other factors such as growth in outpatient resources, changes in
benefit plan design, and shifts in methods for paying hospitals. Groups with
higher initial levels of hospital use tend to show more change than groups
with lower initial hospital utilization.

The impact of utilization management on net benefit costs is less clear.

Savings on inpatient care have been partially offset by increased spending
for outpatient care and program administration. Some of this offset is an
expected and acceptable result of utilization management (and other
factors), and some is an unwanted consequence of moving care to
outpatient settings where there are fewer controls on use and price.
Although utilization management probably has reduced the level of
expenditures for some purchasers, itlike other cost-control strategies
does not appear to have altered the long-term rate of increase in health care
costs. Employers who saw some short-term moderation in benefit
expenditures are seeing a return to previous trends.
Systematic evidence about the impact of utilization management methods
on the quality of care and on patient and provider costs (both economic and
noneconomic) is virtually nonexistent. Purchasers have not demanded such
evidence, and researchers have found the measurement of these effects more
costly, time-consuming, and uncertain than measuring effects on purchaser's
costs. During the course of this study, the committee did not locate documented
cases to suggest that prior review programs in the private sector are
jeopardizing patient safety, although some physicians believe they are. They
may also cause anxiety and inconvenience to some patients. Utilization
management does add to the administrative burdens on practitioners and
institutional providers and contributes to resentments about reduced
professional autonomy and satisfaction.1 More positively, the committee has
some confidence that high-cost case management, as currently practiced on a
voluntary basis, is easing some financial and emotional burdens on
catastrophically ill patients and those who care for them.
The lack of good research on the effectiveness and impact of utilization
management is a frequent theme in this report. The gaps and deficiencies in
clinical research on the effectiveness of many medical procedures also must be
emphasized. As utilization management moves toward more review of the
actual need for specific procedures, the clinical bases for these kinds of
assessments become more important. (This is not, however, to assert that
utilization management programs are making sufficient use of clinical
information that is already available.) Later in this chapter, the committee
recommends greater public and private involvement in both program evaluation
and clinical outcomes research.
1 However, the committee found it difficult to be certain (1) when complaints about
utilization management programs stemmed specifically from private programs rather

than Medicare, Medicaid, or internal hospital programs or (2) when they really involved
prior review or case management rather than basic exclusions in benefit plans,
retrospective claims denials, or provider payment methodologies.
HOW UTILIZATION MANAGEMENT IS EVOLVING

The continuing evolution of utilization management is most evident in two
areas: its scope and its operational efficiency. There are several reasons for this
evolution. First, the initial savings from shifting the site and timing of care have
largely been realized in may places, and the survival of utilization management
programs may depend on their ability to continue to have an impact on use and
costs. Second, review organizations are being influenced by researchers' belief
that a significant amount of care is still inappropriate and unnecessary. Third,
the administrative costs of review programs and the provider and patient
relations problems they create make simplification and efficiency important
objectives.
Scope of Review
The emphasis of utilization management is beginning to expand from the
site and duration of care to include more evaluation of the actual need for
specific types of inpatient and outpatient services. The question is not only
whether the hospital is the appropriate site for a proposed procedure but
whether the procedure is clinically indicated. The procedures targeted for this
kind of assessment are a mix of very expensive and relatively high-risk
services, such as coronary artery bypass surgery, and less expensive and less
risky procedures, such as bunionectomy. The assessments themselves tend to
involve more questions about patient history and previously tried treatments
than do site-of-care reviews.
As the focus of utilization management expands, the development of sound
clinical criteria for assessments of medical necessity is becoming increasingly
important. When the question is not where someone will get care but whether or
not a particular treatment is necessary at all, the stakes for the patient, the
practitioner, and the purchaser tend to be higher and the soundness of the
criteria and their application more critical. The generation of useful criteria
requires at least three steps: (1) the development of clinical and other evidence
of treatment effects for particular procedures and services; (2) the formulation
of guidelines for treating specific conditions or using specific procedures; and
(3) the delineation of review criteria based on clinical evidence and practice
guidelines. The research on feedback and education strategies to influence
physician decisions on patient care suggests that the criteria used to make
medical necessity assessments will be more likely to win acceptance and
change behavior if they are based on good clinical evidence from respected
academic and professional sources and if the medical community is involved in
the process of criteria development.

Operational Efficiency
With respect to administrative costs, utilization management organizations
have a strong self-interest in measures that can reduce their own costs
particularly since the short-term impact of their programs seems to be
diminishing and the ratio of savings to administrative costs appears to be
dropping. Frequently mentioned priorities include greater computerization,
expanded use of treatment protocols in high-cost case management, and greater
targeting of reviews to high-payoff categories of problems and services. TO
some extent, gains in operational efficiency for review organizations could also
reduce administrative burdens on patients and providers.
In addition, some moves are being made that may standardize certain
aspects of utilization management and reduce the variability and ambiguity now
complained of by both purchasers and providers of care. One example,
described in Chapter 4, is the effort initiated by the American Medical
Association, the Health Insurance Association of America, and the Blue Cross
and Blue Shield Association to develop guidelines for the conduct of prior
review programs. The American Hospital Association also has been working
with insurers to establish guidelines for on-site review of medical records by
insurers, utilization management organizations, and others.
It seems likely that much of the clinical and administrative technology of
utilization management will be increasingly integrated into health benefit
programs, although the details will continue to vary across HMOs, PPOs, and
fee-for-service plans. However, to the extent that clinical standards for medical
decision-making are expanded and then internalized by practitioners, third
parties at risk for the costs of medical care may be able to focus their utilization
management programs more narrowly and reduce the burdens on patients and
providers.
In addition, redesign and more careful specification of health benefits may,
in certain situations, eliminate the need for some case-by-case prospective
review of services and for some forms of case management.
Rationing
At this time, the committee does not see utilization management moving
toward intentional rationing of clinically necessary medical services. A decision
not to approve payment for a medically unnecessary service is not rationing, per
se. Nurse and physician reviewers do not make explicit case-by-case
assessments of whether the expected clinical benefits of a hospital admission or
other proposed service for a specific patient are, in some way, worth not only
the clinical risks but also the economic costs. Reviewers also do not make
decisions, like those sometimes involved in emergency care or transplant
services, to allocate limited medical resources

to those most in need or most likely to benefit. The committee recognizes,

however, that there may be some instances when a review nurse or physician
may apply implicit cost-effectiveness judgments to discourage proposed care
that may have only marginal clinical value in a specific case.
The committee notes that high-cost case management may assess whether
making an exception to routine coverage restrictions will be less expensive for a
specific patient than enforcing the contract provision. This might be considered
a form of rationing. If so, it is a peculiar form in that it adds resources where
they normally would be restricted rather than restricting resources that would
otherwise be available.
It is useful to draw a distinction between micro and macro-level resource
allocation decisions. Micro-level decision-making operates at the individual
level, for example, to determine who gets a heart transplant when organs are in
short supply. Macro-level decision-making operates at an aggregate level, as
when health plans decide to cover only 30 days of inpatient psychiatric care. It
is possible that prior review could become a micro-level rationing tool, but the
committee found no evidence of such a development at this time. In contrast,
employers and others already make a wide range of macro-level allocation
judgments in benefit plan design.
ISSUES FOR THE FUTURE

In Chapters 3, 4, and 5, the committee identified some shortcomings in
utilization management or gaps in our knowledge that raise concerns about
patient protection, particularly given the course now being charted by
purchasers and utilization management organizations. If the positive potential of
utilization management is to be encouraged, then the committee believes that
several issues need attention.
Methods and resources for evaluating the impact of utilization management
on health care costs, use, and quality are limited and must be strengthened.
All those involved in health care delivery and financing need to feel more
confident about what works and what does not under what circumstances.
The criteria used by those engaged in utilization management (including
hospitals and HMOs) should be available for outside scrutiny. Physicians,
purchasers, and others should know the basis for judgments about the site,
timing, and need for care.
Systematic investigation of the effects of utilization management not only
on purchaser costs but also on patient and provider costs and attitudes
should be a higher priority. A more complete picture of costs and benefits
is needed.
The opportunities for patients and providers to appeal utilization

management judgments should be clearly described and free from

unreasonable complexity or other barriers. This is an essential protection
for patients.
The explicit links between utilization management and quality assurance
mechanisms need to be clarified and strengthened. Utilization management
organizations should have standard operating procedures for responding to
the quality problems they uncover.
RECOMMENDATIONS FOR THE NEAR TERM

The committee believes that utilization management has sufficient promise
that a number of short-and long-term efforts should be made to promote its
positive potential and guard against its shortcomings. A prudent course for the
near future is for the parties involved in utilization managementpurchasers,
review organizations, health care providers, and patientsto accept greater
responsibility for the reasonable and fair conduct of utilization management and
the appropriate use of medical care.
The concepts of prudence, fairness, and responsibility described below are
supported by a framework of general rights, duties, and obligations created by
state and federal legislation, regulation, and judicial decisions. The most
important examples include tort and contract law. Depending on the model of
utilization management used and whether it involves a self-insured employer,
an HMO, or other arrangement, different legal rules may apply. Little case law
and not much legislation or regulation apply explicitly to utilization
management. However, it is reasonable to expect that utilization management
will be subject to standards of prudent conduct similar to those affecting other
areas of health plan administration.
Responsibilities of Employers and Purchasers

As financers of both utilization management and health services,
employers are in the best position to exert influence on the conduct of prior
review and high-cost case management programs. They have a direct interest in
programs that work fairly and effectively to ensure the value of their investment
in employee health benefits. The committee, however, recognizes that there are
practical limits to what individual employers can do, particularly small
employers with limited administrative resources, expertise, and leverage.
In general, purchasers should investigate the operating procedures and
capabilities of the organization or organizations from which they purchase
utilization management services. This investigation should include
organizations such as HMOs that provide prior review and high-cost case
management as part of a broader package of services. Benefits managers

can visit review organizations, request detailed information and references from
current and former clients, and seek advice from business coalitions,
consultants, and similar resources.
Table 6-1 lists the types of information a purchaser may want to request.
This kind of information can help purchasers assess in broad terms a review
organization's experience, its investment in staff and information technology,
and its sensitivity to purchaser, patient, and provider concerns. The exercise can
also help purchasers to think through their own preferences for different styles
of utilization management. Unfortunately, this preliminary report cannot offer
purchasers the equivalent for utilization management of the crashworthiness
tests for automobiles or the nutritional quality evaluations for foods that
commonly appear in consumer magazines. Again, the absence of research on
the impact of variations in program structure and operations is a limiting factor.
As prudent managers of a total program of health benefits, employers
should also examine other aspects of their benefit plans for impediments to the
appropriate use of medical services or the rational payment for these services.
For example, if reviewers are judging hospitalization to be unnecessary for
intravenous (IV) therapy, is outpatient IV therapy a covered benefit? Are fees
paid for necessary IV therapy in an outpatient setting subject to fewer controls
than for those paid for inpatient therapy? What is provided to help employees
understand what their benefit plan will pay for or how much they may pay out-
of-pocket for selecting one provider or service rather than another? A benefit
plan's scope of coverage, levels of cost-sharing, provisions for retrospective
claims review and profiling, and methods of paying providers can as affect the
impact of utilization management programs on health care costs and quality.
Although employers have the fight and duty to take vigorous actions to
manage the costs of employee health benefits, they should respect both the
confidentiality of medical information about employees and the primary
obligations of physicians to serve their patients. In their own business interest
and to protect employees, employers should see that workers are informed of
their responsibilities and rights. Key means of reaching employees include
orientation for new employees, clearly written and eye-catching brochures
explaining responsibilities, and health plan identification cards with clear notice
of utilization management requirements. Human resources staff should be
trained to respond to employee questions, assist with problems, and handle
grievances.
Responsibilities of Utilization Management Organizations

Any supplier of services has responsibilities to purchasers that are intrinsic
to the concept of a buyer-seller relationship. It is in the business

TABLE 6-1 Types of Information That May Be Requested by Purchasers of Utilization

Management Services
1. Organization experience and range of services
a. Years in operation
b. Origins/affiliations (insurance company, PRO, etc.)
c. Services offered, for example:
(1) Utilization management
(2) Claims administration
(3) Insurance
(4) Data analysis and consulting
d. Number of clients for utilization management services by category of service
(preadmission review, high-cost case management, etc.)
e. Contract turnover ratenew, renewed, lost
f. Number of individuals covered
g. Other (Medicare contracts, etc.)
h. References from current and past clients
i. Record of lawsuits or judgments
k. Extent of liability insurance
2. Facilities and operations
a. Geographic locations
b. Telephone and other communications capabilities (monitoring and tracking options,
frequency of calls lost, flexibility)
c. Computer hardware and software
(1) On-line access to eligibility, coverage data
(2) Link to claims payment system for monitoring
compliance and reporting data
3. Personnel
a. Primary reviewers or case managers
(1) Number (full-time and part-time employees or
contractors) and location
(2) Education and experience requirements
(3) Responsibilities and resources
(4) Training program and supervision
(5) Ratio of reviewers to covered individuals
b. Physician reviewers (if not primary reviewers)
(1) Number (full-time and part-time employees or
contractors) and location
(2) Clinical qualifications and experience
(3) Responsibilities and resources
(4) Training program and oversight
(5) Ratio to covered individuals
4. Clinical criteria for assessing medical services
a. Primary external sources of criteria
b. Internally developed criteria (request information about development process,
qualifications of clinicians and others involved, updating mechanisms)
c. Availability for review by purchaser, attending physician, other
5. Patient or employee role
a. Responsibilities for initiating review
b. Recommended policy for noncompliance with process
c. Involvement of patient or family in case management
d. Access to utilization management organization for consumer information
e. Information and education provided by organization
f. Survey or other follow-up of employees or patients
g. Complaint handling (volume, process, resolution)

6. Physician or hospital role

a. Responsibilities for initiating review
b. Recommended policy for noncompliance with process
c. Involvement of practitioner in case management
d. Information and education provided by organization
e. Survey or other follow-up of practitioner
7. Preferred employer role
a. Identification of potential high-cost cases
b. Agreement to exceptions to benefit contracts
c. Involvement in appeals of prior review decisions
8. Policies on confidentiality of medical and other data
9. Pattern of prior review decisions: volume or rate of
a. Denials
b. Appeals
c. Appeals approved or denied
10. Pattern of high-cost case management decisions
a. Percentage and types of cases accepted
b. Average cost of cases and benefits exceptions
11. Appeals process
a. Procedure for patient and provider appeals
b. Decision maker and relation to organization and original reviewers
12. Reporting
a. Activity reports (content, frequency, source)
b. Utilization reports (content, frequency, source)
c. Savings reports (content, frequency, source)
13. Basis and level of charges for services
interest of utilization management organizations to anticipate and respond

to purchaser demands for information about the organization, its services, and
its results. Further, it is in the legal interest of these organizations to manage
their activities rationally, to act in good faith, and to maintain careful records.
Although good business and legal judgment should dictate prudent
behavior, those who provide utilization management services also have a moral
obligation not to harm the patients whose medical care they review and
influence. Harm includes discouraging appropriate care and mishandling
confidential information. When organizations perform prior review and high-
cost case management for individually purchased insurance plans (with no
employer sponsorship), they have a particular responsibility to provide good
educational materials and appeals processes for beneficiaries who have no
employer or other sponsor to act as their agent and aid.
With respect to physicians and other providers of care, good business sense
should dictate that review organizations encourage provider acceptance and
cooperation, although their incentives to do so are less strong than the
incentives to respond to client wishes. Chapters 3, 4, and 5

suggest that utilization management organizations can increase physician

acceptance by
involving the medical community in the development and evaluation of
review criteria;
making the bases for review decisions open for examination by purchasers,
physicians, and others;
clarifying and improving the process for discussing cases and appealing
negative decisions; and
minimizing the administrative burdens placed on hospitals and physicians.
Given the diversity and competition now evident in utilization
management programs, significant progress in these two directions will
probably require ongoing cooperation between elements of the utilization
management industry and provider groups to identify problem areas and
propose operating procedures that avoid unreasonable demands on providers.
Currently, there is no natural focal point to coordinate and promote these efforts
or monitor progress. However, the cooperative efforts of organizations like the
American Medical Association, American Hospital Association, Blue Cross and
Blue Shield Association, and Health Insurance Association of America to
develop guidelines for review are encouraging.
The committee is aware that further steps, particularly making clinical
criteria available, raise difficulties given the competitive environment of benefit
plan administration. Organizations that have invested their own resources in
developing criteria will be reluctant, on the one hand, to make them available to
competitors with less initiative and, on the other hand, to reveal some details to
practitioners and institutions for fear of their ''gaming the criteria.'' Although
these are reasonable concerns, on balance, they are outweighed by the need to
move toward open criteria and standards.
It is inevitable that utilization management organizations will occasionally
identify cases of apparent malpractice or other quality-of-care problems. These
cases may involve a proposed treatment for a single patient or a pattern of
behavior. The experience that the Health Care Financing Administration has
had in developing regulations and procedures for PROs that have uncovered
quality problems indicates that balancing patient protection against due process
for providers is a difficult political and legal issue. Nonetheless, review
organizations should prepare guidelines for dealing with clear quality-of-care
problems, at a minimum, having their medical director discuss the situation
with the provider in question and, perhaps, raise the matter with the local PRO.

Responsibilities of Health Care Practitioners and Institutions

The committee found the responsibilities of health care practitioners and
institutions most troublesome to analyze and define. In this respect, utilization
management mirrors the situation with many other questions of provider
responsibility and ethics, whether the questions involve the limits of heroic
efforts to maintain life, the provision of care to those who are unable to pay, or
the recognition of patient preferences about treatment alternatives.
The narrowest area of controversy involves physician self-interest versus
the patient's well-being, where law and ethics have long agreed that the latter
should prevail. Broader controversy and lack of agreement surround the
questions of whether and how physicians should consider societal interests in
making patient care decisions. Intermediate in scope are issues involving the
physician's responsibilities when third parties are paying for patient care. For
example, is it acceptable to shade the truth in order to obtain prior certification
or retrospective payment for services? Is it legitimate to refuse to disclose
information needed by review organizations or claims administrators? When
should a physician challenge a negative payment decision? Is the physician
obligated to provide a service he or she believes is necessary when a review
organization, HMO, or hospital disagrees? What role should patient preferences
play?
With respect to third-party payment, the committee agreed that health care
practitioners and institutions are responsible for
cooperating with payers' reasonable efforts, including utilization
management, to assure that they are paying for appropriate care within the
terms of their benefit plans, but
constructively challenging unreasonable review programs and specific
decisions that threaten patient safety or damage patient privacy.
Although some difficult situations with insurers and review organizations
may be more conveniently and quickly dealt with in the short term by
misrepresenting patient symptoms, diagnoses, or treatments, the committee
believes it is in the patient's, physician's, and society's interest over the long
term for physicians to deal honestly with reviewers and claims administrators
and to challenge questionable criteria, procedures, and decisions directly.
Manipulation and evasion can have serious risks. Specifically, incorrect
information may enter the patient's medical records or insurance history with
later negative consequences for the patient. Moreover, perceptions by
purchasers that physicians are gaming the system undermine professional
credibility and stimulate the sorts of auditing, second-guessing, and external
oversight to which practitioners object.
With respect to patients, the committee sees additional responsibilities for
practitioners. These include informing patients about treatment options,

risks, and benefits and then considering their preferences. Insofar as is feasible
and desired by the patient, the physician should also try to ensure that the
patient gets care when a review organization denies precertification for services
that the physician believes are medically necessary. This may mean locating an
alternative source of care if the patient cannot pay and the physician cannot
provide free care.
In the context of this committee's consideration of appropriate medical
practice, an obvious responsibility for health care practitioners is to stay current
with scientific literature on the necessity and effectiveness of medical services
in their area of practice. (Many researchers would add that practicing physicians
should, if at all possible, participate in the clinical trials and other research on
which this literature rests.) As research results accumulate on the effectiveness
and appropriateness of specific diagnostic and treatment methods, the
committee sees an enormous implementation challenge to make knowledge
more readily usable by busy practitioners. Meeting this challenge will require
continued initiatives in such areas as the development and dissemination of
practice guidelines, continuing medical education, and office-accessible
computer resources for physicians. Utilization management can play a role in
supporting these initiatives.
Responsibilities of Patients
In many respects, patients and potential patients are the weakest strand in
the web of responsibilities for the prudent use of medical services. One person's
view of appropriate patient responsibility can seem like "victim-blaming" to
others. Although it is now a clich among sophisticated consumers that more
medical care is not necessarily better medical care, people may not be able to
act in an informed and prudent way when they are ill. Furthermore, individuals
may find both their benefit plans and their medical care difficult to understand
and evaluate even when they are not stressed by illness. Education levels,
family support systems, employer communication efforts, and other factors may
all affect how much information people can understand and use. Nonetheless,
health plan members should try to understand their responsibilities under their
plan and their responsibilities for protecting their own health.
The challenge for those involved in health care delivery and financing is to
help all kinds of patients make informed decisions about getting or not getting
care. Concepts of probability, differential diagnosis, and risk-benefit trade-offs
are not part of most people's everyday vocabulary. Fortunately, new tools, such
as the video disk recently developed to explain prostatectomy risks, benefits,
and alternatives, have promise in making complex issues more understandable
for patients. It is not inconceivable that protocols used by review organizations
to assess a proposed treatment

might one day ask not only whether certain clinical steps (for example, a trial of
antibiotic therapy) have been taken but whether and how the patient has been
informed of risks, benefits, and alternatives for the proposed treatment.
RECOMMENDATIONS AND QUESTIONS FOR THE LONGER

TERM
Early in this chapter, the committee observed that utilization management,
per se, is essentially a working hypothesis, one of several partial and
overlapping strategies for balancing health care expenditures, access, and
quality. As knowledge advances, economic conditions change, and social values
shift, these partial strategies are revised, integrated, and sometimes discarded.
However, even if some of the techniques now employed by utilization
management organizations are abandoned or the organizations themselves
change, improvements in the criteria for judging appropriate care and for
monitoring the provision of care will continue to be relevant. The longer-term
recommendations of the committee focus on the foundations for such
judgments: knowledge development, knowledge application, and value
clarification.
The following discussion examines three issues:
research on the effectiveness of medical services and the effectiveness of
utilization management and related techniques,
formulation and dissemination of guidelines for medical practice and
criteria for utilization review, and
oversight of utilization management.
Research on Effectiveness
Utilization management can be no better than the clinical evidence and
expertise on which it is based. Although review organizations may not be
effectively using all available research now, they are still constrained by the
large areas of undocumented impact and clinical uncertainty involving many
major medical procedures and services.
Policymakers are increasingly recognizing that the free-market system is
unlikely to invest sufficiently in effectiveness research and data collection
because those making the investment cannot capture all the benefit but must
share much of it with those who have not invested. Since the public gains from
investments in such research, a major public role in financing and priority
setting is appropriate. It should add to, not replace, initiatives being undertaken
by private researchers, health care organizations, and others.
The creation of an agenda to strengthen knowledge of what is effective in
medical care is well under way, and the committee joins the many

other public and private groups that support these efforts. Nonetheless,
compared with the resources invested in biological research and health care
delivery, the promised investment in effectiveness research is still relatively
small.
However, while more investment in research is desirable, it is important to
have realistic expectations. First, there are practical and ethical limits on clinical
researchtoo few researchers, long time horizons, and numerous instances in
which clinicians would balk at research protocols that require withholding
treatments generally thought to be useful or providing treatments generally
thought to be inappropriate.
Second, much research does not really focus on the effectiveness of care in
an average setting or population, nor does it evaluate the impact of care on
quality of life and many other outcomes that are now considered important.
What is needed are longitudinal observations of natural variations in the use and
outcomes of economically and clinically important medical technologies in
different practice situations. Clinical, claims, and other data bases need to be
designed, and parallel attention needs to be given to the methodological issues
in interpreting and disseminating research findings.
Third, effectiveness research that relies on existing claims and other
records, while less expensive and time-consuming than most clinical trials, is
not necessarily quick nor suitable for providing answers to many questions.
Effectiveness research needs to be complemented by judgmental strategies,
such as consensus panels, that use but go beyond scientific literature and
empirical data. Many times the practical results of these processes will be
imperfect. However, in the past 10 years, we have learned that the best
randomized clinical trials have weaknesses and that these weaknesses are
accepted, sometimes too readily. Intelligent compromise with perfection is
necessary.
Fourth, research cannot resolve some questions. For example, researchers
may be able to assess the clinical benefits and risks in using cataract surgery to
correct a 20/40 vision defect. They cannot tell us whether such use of resources
is wise given other uses to which those resources might (but not necessarily
would) be put.
Fifth, sound clinical research does not automatically affect behavior. The
research on what works in medical care should be complemented by research on
how best to ensure that such knowledge is used effectively and efficiently. Such
programmatic research is, for the most part, a low priority today. It is
expensive, methodologically troublesome, and slow to pay off. As part of the
overall strategy for containing total health care costs and improving the
appropriateness of health care for all citizens, the committee urges federal and
private consideration of carefully targeted research

projects to test utilization management strategies and build methodologies for

documenting the effects of existing programs.
Practice Guidelines and Review Criteria

To a very considerable extent in the past, medical practice guidelines have
been implicit and have been linked to the amorphous concept of community
standards of practice. The medical profession has historically resisted setting
standards that are both explicit and national in scope. This resistance has been
challenged by the research on practice variations and inappropriate use
discussed in Chapter 2. The increasing emphasis on effectiveness research is
also adding to the momentum for developing better and more extensive
guidelines for patient care.
Translating clinical and other research into valid, reliable, and usable
guidelines for medical practice and utilization management is a complicated
enterprise that can take many forms. One approach is the development of
clinical protocols that specify the tests, procedures, and other medical services
that should be provided to patients with a specific clinical problem. Another
approach is more proscriptive than prescriptive. It essentially sets boundaries
outside which specific services for specific medical conditions are defined as
inappropriate. Within the boundaries, guidelines may allow considerable room
for an individual practitioner's judgment about the value of alternative courses
of treatment. The two latter approaches have provided the bases for explicit and
implicit review criteria used in utilization management organizations, but
review organizations appear to have made little use of explicit clinical protocols.
The committee has cited analyses that show that some practice guidelines
do not match relevant empirical research very well, and the fit between practice
guidelines and criteria for utilization management may likewise be poor. The
committee has identified a number of questions about the process of
developing, disseminating, and updating practice guidelines and review criteria
that need attention. They include the following:
Should there be some kind of oversight of guidelines developed by different
sourcesa sort of quality review mechanism that assesses both the method
and the substance of specific guidelines? What should happen when
different sources develop conflicting guidelines?
To what extent should patient preferences or cost-effectiveness analyses be
considered in the development of practice guidelines? How should these
issues of value be dealt with in the application of guidelines or in other
strategies?
Should adherence to guidelines provide physicians with protection against
malpractice charges?

Over the long term, should a role for community or local standards continue?
What considerations should apply in translating guidelines into criteria for
prospective or retrospective review programs?
Should private review organizations be encouraged to invest in criteria
development if the results are open to outside examination?
These questions are relevant to much of the Institute of Medicine's agenda.
Further investigation of these issues is now underway. It will draw not only on
the Committee on Utilization Management by Third Parties, but also on the
Committee to Design a Quality Assurance Strategy for Medicare and the
Council on Health Care Technology, as well as outside groups such as the
Health Care Financing Administration and the Physician Payment Review
Commission.
Oversight of Utilization Management

Whether additional mechanisms for oversight of utilization management
are needed to weed out incompetent organizations and programs was a question
the committee discussed at considerable length. The protections offered by
caveat emptor, self-regulation, and tort liability, although important, do not
respond to all concerns about the impact of utilization management on patients,
providers, and overall health care costs. Moreover, the current lack of oversight
poses some risks to utilization management as a promising strategy for
managing benefit costs. Misbehavior by a few organizations could harm
patients, infuriate practitioners, and seriously damage acceptance and
cooperation from patients and providers. On the other hand, premature or
misguided regulation could stifle worthwhile innovations in technique and lock
in inefficient or ineffective methods. Purchasers might then abandon utilization
management for more onerous techniques of cost containment. As this report
has emphasized, utilization management is still relatively new, is evolving quite
rapidly, and works quite differently depending on the delivery and financing
mechanism it serves.
At this time, the committee is not prepared to make recommendations
about where oversight may be appropriate or what oversight strategies may
offer a workable balancing of the conflicting values, interests, and needs of
different parties. Rather, the committee has posed several questions about the
oversight of utilization management that it will be considering further.
How do we decide whether oversight is necessary and feasible?
If an oversight mechanism is necessary and feasible, should it be public or
private? If public, should all 50 states make their own policies or should the
federal government impose uniform standards?2
2 As noted in the discussion of PROs in Chapter 2, federal regulation does not
guarantee uniform
What should be the focus of oversight? Should the clinical criteria for
assessing care and the mechanisms for appealing decisions be the main
concerns? What about staffing, training, outcome reporting, and so on?
Should utilization management conducted by different kinds of
organizations, for example, PPOs versus staff model HMOs, be subject to
different kinds of oversight? Should utilization management of some
servicesfor example, mental health carebe subject to different kinds or
levels of oversight?
Can anything be learned from government oversight of PROs or the
accreditation process managed by the Joint Commission on the
Accreditation of Health Care Organizations?
These are not simple questions. Answering them demands more
information and more thoughtful debate over how to judge the strengths and
weaknesses of utilization management versus other strategies to control costs
and influence patient care decisions and how to weigh the legitimate but
sometimes conflicting needs, interests, and values of the parties involved in
utilization management. At this time, the committee advises caution to public
policymakers currently considering proposals to regulate and limit utilization
management.
CONCLUSION
Utilization management needs to better demonstrate that it reduces the
wasteful use of resources, improves the appropriateness of patient care, and
imposes only reasonable burdens on patients and providers. It is important that
it be given a chance to prove itself because if it works, it offers a means of
accommodating individual patient circumstances that is more sensitive than
many other cost-containment strategies, for example, those that rely on across-
the-board restrictions in benefits.
However, the limits of utilization management and any other single
strategy, even any combination of strategies, also need to be recognized. The
issue is not whether utilization management does everything that needs to be
done but whether it produces desirable results in reasonable ways at an
acceptable cost. Is it, on balance, better to use it than discard it? And if
utilization management is helpful, how can provider payment methods,
retrospective utilization review and feedback techniques, resource planning
mechanisms, and other tools be used to reinforce utilization management and
deal with problemssuch as distorted payment rates or excessive use of low-
cost, high-volume servicesnot effectively addressed by utilization
practice. And as noted in Chapter 3, proposed state regulations vary enormously and,
in some cases, could make prior review and high-cost case management infeasible.
management? This report has provided initial views on some of these questions.
The Institute of Medicine will continue its efforts to better define what role
utilization management might play in helping society find an acceptable balance
of efficiency, access, and appropriateness in health care. This clearly must be a
shared venture. Fortunately, the quest to know what is useful and how to apply
it universally are now central issues in medical research and public policy.

ACKNOWLEDGMENTS 163
Acknowledgments
The participants in an initial meeting to plan this project provided valuable

direction and stimulating ideas for committee investigation. The meeting was
chaired by Stanley B. Jones and included Rhoda Abrahms, U.S. Department of
Health and Human Services; Robert A. Berenson; John M. Burns, Honeywell,
Inc.; E. Langdon Burwell, Falmouth Medical Associates; John W. Bussman,
Oregon Health Sciences University; Priscilla Dasse, Beth Israel Hospital; John
T. Dunlop, Harvard University; Paul Gertman, Health Data Institute; Nathan
Hershey, University of Pittsburgh; William Hoffman, United Auto Workers;
John K. Iglehart, Health Affairs; Karen Ignani, American Federation of Labor
and Congress of Industrial Organizations (AFL-CIO); Arnold Milstein,
National Medical Audit; David Ottensmeyer, Equicor; Robert Patricelli,
ValuCare, Inc.; Mindy Pellissier, Scripps Clinic and Research Foundation; Gail
J. Povar, George Washington University School of Medicine; David
Rosenbloom, Health Data Institute; Richard S. Sharpe, John A. Hartford
Foundation; David N. Sundwall, U.S. Department of Health and Human
Services, and Jeffrey Weiner, U.S. HealthCare.
In June 1988 the committee heard much helpful testimony from a range of
medical, consumer, purchaser, insurer, and other organizations. A brief
summary of that meeting and a list of the participants are included in
Appendix D of the report.
The committee cannot identify the 12 organizations that it visited; it agreed
to keep that information confidential. Many individuals from those
organizations spent considerable time answering questions, showing their

ACKNOWLEDGMENTS 164
facilities, reviewing summaries of the visits, and generally responding to follow-

up inquiries. Several committee members were involved in the site visits: Alice
G. Gosfield, Robert A. Berenson, John M. Burns, John M. Eisenberg, Nathan
Hershey, Carol Ann Lockhart, Cynthia L. Polich, and Bruce S. Wolff.
The authors of three commissioned papers also contributed to the
committee's understanding of issues. Two papers are published in Appendixes
A and B of the report. One other paper, which is not included but is available
from the committee upon request, also provided useful background information
in an area that was not dealt with explicitly by the committee. This paper on
Medicaid cost-containment strategies was prepared by Robert E. Hurley,
Pennsylvania State University, and Deborah Anne Freund, a member of the
committee. During regular meetings, the committee heard presentations on
utilization management from several individuals: Ed Zalta, CaPP Care; Paul
Gertman, ClinMan, Inc.; and Marcia Gold, Group Health Association of
America.
Several committee members drafted materials to assist in developing the
report. Neil Hollander reviewed problems with evaluating the savings claimed
for utilization management programs. Nathan Hershey developed an analysis of
the contents of contracts between utilization management organizations and
their clients. Alice G. Gosfield prepared a background paper on utilization
management in peer review organizations, and Michael Herbert provided
background discussion on utilization management in health maintenance
organizations. Suggestions about what purchasers should look for in utilization
management organizations were contributed by Robert Patricelli, Karen Ignani,
Michael E. Herbert, Alice G. Gosfield, John M. Burns, and Howard L. Bailit.
Alan R. Nelson discussed physician concerns and responsibilities.
For the first year of the project, Richard H. Egdahl chaired the committee.
Dr. Egdahl has since continued to serve as a member of the committee and
kindly extended invitations to committee staff to attend meetings organized for
Pew Foundation Corporate Fellows at Boston University. During the second
year, Jerome H. Grossman chaired the committee as it developed its
conclusions, outlined the report, and planned next steps.
The initial idea for this project came from Bradford H. Gray, who secured
much of its funding, organized the planning session, and served as study
director through December 1988. The project would not have existed without
his initiative. When Dr. Gray left for Yale University, Marilyn J. Field joined
the project as the committee defined its views, and she took major responsibility
for drafting the report. Dr. Field will continue as study director for further work
by the committee. Margaret Walkover was research associate for the early
months of the project. Susan E. Sherman, who served as research associate
during the summer of 1987, went on all the

ACKNOWLEDGMENTS 165
site visits, and assisted in preparing the site visit summaries. Eileen Connor of
Boston University contributed much to the project as a consultant, and Sharon
Rosen of the New England Medical Center assisted Dr. Grossman and the
committee during its final months.
As project secretary, Don Tiller provided much needed continuity as well
as exemplary support for the committee. The ongoing guidance of Karl D.
Yordy, director of the Division of Health Care Services, was important for the
project, particularly given the changes in committee and staff leadership.
A great many people helped the committee with information and insights
about utilization management. Particularly useful was a meeting in January
1989, arranged by Marion Ein Lewin of the IOM, with executives from clients
of review organizations: H. Dean Belk, ALCOA; Jo Anne Brown, Service
Employees International Union; Jerry Clark, UMWA Health and Retirement
Fund; Richard Ekstrom, Westinghouse; Jeanne Kardos, Southern New England
Telephone; Walter B. Maher, Chrysler; William J. Schneider, J.P. Morgan;
Elizabeth Solem, 3-M Company; Julia Smith, Bank of America; and Loring
Wood, NYNEX. Others include Harris Allen, Richard Bondi, and Constance
Winslow (Aetna Life and Casualty); Mark Chassin, Jacqueline Kosecoff, and
Leslie Michaelson (Value Health Sciences, Inc.); Michael Cologero, Robert
Snyder, and Eileen Thomas (Blue Cross and Blue Shield Association); Jon
Gable, Al Minor, and Thomas Muscow (Health Insurance Association of
America); David Bergohlz and Mary Ann Sevick (Pittsburgh Program for
Affordable Health Care); Mary Henderson (Brandeis University); Gail Beiber
(American Medical Association); Howard Birnbaum, Deborah Kottler, and Len
Vernick (Health Data Institute); John Blum (Loyola University, Chicago); Joan
Trauner (coopers & Lybrand); Robert Beale and Richard Wright (Caterpillar,
Inc.); Jill Bernstein (Physician Payment Review Commission); Shan Haley
(American Hospital Association); JoAnne Brown (Service Employees
International Union); Ronald Hurst (Association of Preferred Provider
Organizations); Joseph Restuccia (Boston University), Shelley Greenfield (New
England Medical Center); Steven Sieverts (Blue Cross and Blue Shield of the
National Capital Area); Deborah Chollet (Employee Benefits Research
Institute); and Spencer Vibbert (McGraw-Hill). The library staff of several
organizations provided information or sources on a number of issues: American
Hospital Association, American Medical Association, Blue Cross and Blue
Shield Association, Health Insurance Association of America, and Group
Health Association of America.

ACKNOWLEDGMENTS 166

APPENDIXES 167
Appendixes

APPENDIXES 168

APPENDIX A 169
Appendix A
Legal Implications of Utilization Review
William A. Helvestine*
The recent decade has seen a dramatic increase in the use of utilization
review (UR) programs in an effort to stem the tide of rising health care costs.
UR serves to verify the medical necessity of hospital admissions and specific
medical procedures. However, by intruding into the traditional physician-patient
relationship, UR programs raise a host of liability issues.1
Introduction
This paper reviews the rights and responsibilities of the parties involved in
UR decisions in an effort to determine what legal guidelines exist and which
areas remain unsettled.
The Types of Utilization Review

UR is used in one form or another by government payers such as
Medicare, private insurers, health maintenance organizations (HMOs), and self-
insured employers. Some payers perform the review in-house; others contract
with independent entities to perform all or part of the review. If the physician's
treatment plan is approved, the patient is fully covered
* William A. Helvestine is a Partner at Epstein Becker Stromberg & Green in San
Francisco. The author gratefully acknowledges the assistance of Dena R. Belinkoff, an

attorney with Epstein Becker Stromberg & Green, in the preparation of this paper.
APPENDIX A 170
to the limits of the plan; if not, the patient's financial coverage may be reduced
or denied altogether. Many physicians criticize the intrusion of UR programs,
while most payers claim they reduce health care costs.
Under traditional indemnity plans, UR occurs retrospectively. That is,
claims for payment are reviewed after the treatment is rendered. Recently,
prospective and concurrent review programs have become more prevalent. A
prospective review program requires the physician to obtain prior certification
from the payer before providing treatment. In concurrent review, the payer
monitors a patient's treatment (usually the length of a hospital stay) and
specifies the last clay for which payment is authorized.
Because these programs can directly affect the medical care of the patient,
they increase the potential for harm and, consequently, the potential for liability
arising out of the UR program. Prospective and concurrent UR decisions also
pose dilemmas for the treating physicians and hospitals. If a treatment plan is
disapproved, is the physician free to abandon the plan? To what extent is the
physician or hospital obligated to provide the treatment despite the risk of
nonpayment? How vigorously must the physician pursue any appeal rights to
contest the UR decision?
The patient is also left in a quandary. The patient faces conflicting
judgments by two medical professionals: the treating physician and the UR
consultant. Should the patient rely on the treating physician's opinion, go
forward with treatment and accept the resultant benefit penalties, or should he
limit treatment to that which the payer will cover? Who is ultimately
responsible for such decisions, and where should liability fall if the decision
results in injury to the patient?
There are two leading court decisions regarding liability for UR decisions,
Sarchett v. Blue Shield of California 43 Cal. 3d 1, 233 Cal. Rptr. 76, 729 P. 2d
267, (1987), and Wickline v. California 192 Cal. App. 3d 1630, 239 Cal. Rptr.
810 (1986). Sarchett, a retrospective review case, upheld the fundamental right
of an insurer to challenge the treating physician's determination of medical
necessity. Wickline, a case involving concurrent review, addressed the
reviewer's potential liability to the patient for harm resulting from prospective
or concurrent review decisions.
The Sarchett Decision

The most important aspect of the California Supreme Court's decision in
Sarchett v. Blue Shield of California, supra, was that it affirmed an insurer's
right to disagree with the treating physician's determination of medical
necessity. The decision also made it clear that if coverage is denied, the insurer
must inform the insured of any contractual rights to reconsideration or
independent review, such as by arbitration.
The plaintiff, John Sarchett, was hospitalized for 3 days by his family

APPENDIX A 171
physician. Blue Shield reviewed the hospital records and determined that the
hospitalization appeared to be for diagnostic purposes only. It denied coverage
based on two policy exclusions: first, an exclusion for hospitalization that is
primarily diagnostic, and second, an exclusion for nonmedically necessary
services.
The California Supreme Court upheld Blue Shield's right to challenge the
medical necessity of hospitalization, even though the patient had relied on the
recommendation of the treating physician. The court squarely rejected the
plaintiff's argument that the treating physician is the final arbiter of medical
necessity. It found retrospective review to be an implied right of the insurance
relationship, even though the policy does not expressly state that the insurer
may conduct retrospective review.2
In the Sarchett case, the court also commented favorably on the increasing
practice of health care payers to require preauthorization for elective
procedures. However, lest payers become too aggressive in coverage decisions,
the decision included a reminder that any doubts and uncertainties in an
insurance policy will be construed in favor of coverage for the insured. As a
result, the decision of a treating physician will rarely be reversed as being
unreasonable or contrary to good medical practice, the court predicted.3
The Wickline Decision

Although the decision in Wickline v. California, supra, was only an
intermediate appellate ruling in California, it promises to be a seminal decision
in the area of UR liability. Mrs. Wickline was being treated for problems
associated with her back and legs, and her physicians recommended surgery.
While her hospitalization and treatment were covered by Medi-Cal (California's
Medicaid program), Medi-Cal required precertification for hospital admission
and assigned an approved length of stay for the admission. Any extension of the
approved length of stay had to be authorized. Medi-Cal approved Wickline's
surgery and authorized a 10-day length of stay, with payment approved until
January 17, 1977.
Mrs. Wickline suffered complications after the original surgery, and two
additional surgeries were performed. Her treating physician determined that she
should remain in the hospital 8 clays beyond her scheduled discharge date and
filled out a Medi-Cal form requesting an extension. The Medi-Cal on-site nurse
reviewer, after consulting with a Medi-Cal physician adviser, approved only a 4-
day extension. While there were appropriate spaces on the form for the on-site
nurse's recommendation and the reason for disapproval by the physician
adviser, both were left blank.
The attending surgeon discharged Wickline to her home when the 4-day
extension period expired. All three of her treating physicians were

APPENDIX A 172
aware that there was a process whereby the Medi-Cal decision could be
appealed, but none of them appealed. A week after discharge one of the
physicians examined her and found nothing remarkable. Then, nine days after
discharge, Wickline was readmitted to the hospital with severe pain and
discoloration of her right leg, which eventually had to be amputated at the hip.
Wickline brought an action alleging that her injuries were caused by Medi-Cal's
negligence in failing to authorize the full 8-day extension. A jury awarded her
$500,000.
The Court of Appeal reversed the jury verdict, reasoning that although the
state's preauthorization program played a role in the decision to discharge
Wickline, this role was not determinative. Rather, the decision to discharge
Wickline was made by the attending physicians. The court held that Medi-Cal
was not a party to that medical decision and could not be held liable if that
decision were negligently made. In refusing to find liability for the review
decision, the court placed responsibility for the hospital discharge on the
attending physicians and implicitly criticized them for not appealing the Medi-
Cal denial decision if they disagreed with it.
The fact that the Wickline court based its holding on the lack of causation
by the UR decision, rather than on any other available alternative basis,
suggests that the court was particularly sensitive to the danger of holding the
UR organization liable. Viewed in this light, the Wickline decision is not merely
an ordinary application of the law of causation, but represents a precedent
against UR liability. The Wickline case provides judicial recognition of the need
for and the importance of UR as a check and balance in a system in which
health care costs typically are paid by third parties rather than by the patient.4
At the same time, the court clearly stated that third-party payers could be
held liable for ''defects in the design or implementation of cost containment
mechanisms'' that result in the denial of medically necessary services. The
decision recognizes that negligent UR decisions may result in denial of needed
treatment, thereby causing injury to the patient. It thus sets the stage for further
development of the allocation of responsibilities in this area.
The Elusive Concept of Medical Necessity

A UR decision invariably turns on whether a treatment or service is
"medically necessary." This term is not only difficult to define but it also is
frequently confused with other terms or limitations of the insurance policy.
Failure to isolate the "medical necessity" aspects of a decision can easily result
in wrong decisions, and hence potential liability for the reviewer.
Examples of such confusion are cases in which coverage is denied because
the procedure is "cosmetic" or "experimental."5 Often, a procedure

APPENDIX A 173
will be "cosmetic" only because it is not "medically necessary" under the

circumstances. Plastic surgery that otherwise would be cosmetic becomes
medically necessary following a disfiguring accident. And, yet, a reviewer may
attempt to deny the surgery as cosmetic, without determining whether it is
medically necessary under the circumstances. When the issue is whether
something is ''experimental," it is generally presumed that some medical
treatment is necessary. However, since the concept of "experimental" depends
in part on whether the treatment is customary, the considerations are often
similar to those used in determining medical necessity.
Definitions of "medical necessity" vary greatly.6 Some policies define
medical necessity simply, such as "such services . . . as are reasonably intended,
in the exercise of good medical practice, for the treatment of illness or injury."7
Other policies provide more lengthy criteria.8 Regardless of the definition,
a determination of medical necessity is always a judgment call, and the
reviewer must be careful not to apply an overly restrictive standard.9
An independent review organization is generally able to limit its decisions
to whether something is "medically necessary." Where the review is conducted
in-house by an insurance company or HMO, there may be more of a tendency
to confuse a medical necessity determination with other coverage decisions.
Care should be taken to delineate clearly the different reasons for a denial, even
though the considerations may overlap.
In some cases, the UR entity's contract may require it to apply certain
benefit restrictions, for example, in cases which the health plan covers certain
procedures only if they are provided on an outpatient basis. In that situation, the
UR entity is not making a decision as to medical necessity, but is simply
applying nondiscretionary criteria. Presumably, the benefits restrictions satisfy
any applicable regulatory requirements for minimum plan benefits. In such a
case, the UR entity should not be held responsible for applying nondiscretionary
criteria, since it is not making a medical necessity determination.
In some arrangements, the UR entity may be required to apply criteria, but
has the latitude to deviate from the criteria in cases in which an exception is
medically necessary, for example, when an outpatient procedure would be
particularly risky for an elderly patient. Under this arrangement, the UR entity
is exposed to liability because it must balance individual medical necessity
against the standard criteria in each case. Of course, if such standard criteria
appear in a variety of health plans, the very existence of the criteria may be
evidence of the community standard of care.
Liability of the Review Organization

Potential liability for the review organization exists under various legal
theories, which are discussed in this section. One of the more unsettled

APPENDIX A 174
areas is the extent to which Employee Retirement Income Security Act

(ERISA) preemption applies. In a case in which ERISA applies, none of the
common law claims discussed in this section would be available, and the
plaintiff would be limited to ERISA remedies (see the section ERISA
Preemption below).
Regardless of the legal theory, two factors weigh heavily in any case. The
first factor is whether financial considerations influence the decision on medical
necessity. A payer who conducts a review in-house may be more susceptible to
improper fiscal considerations. For example, in one unreported case there was
evidence that the medical director began systematically denying coverage for a
particular treatment after attending a finance committee meeting where the costs
for this treatment were singled out as being over budget. Independent review
organizations may be less likely to be influenced by financial considerations,
because the relationship between an independent reviewer's UR decision and
the reviewer's compensation is indirect. However, the review organization's
method of compensation may alter this conclusion. Most independent review
organizations are compensated on a per-person or per-review basis. But if the
review organization's compensation is based on the savings it realizes for its
clients, then financial considerations might influence its decisions.
The second important factor in any UR liability case is causation. Wickline
is a prime example of a case in which the plaintiff failed on the issue of
causation. Regardless of the legal theory, causation typically is the biggest
obstacle in the plaintiff's case.
Aside from these factors, the outcome of any liability case against a UR
organization is likely to be influenced by the inherent sympathies of the jury, as
illustrated by the Wickline case. That is, an injured individual generally will
evoke more sympathy than the corporate review organization or insurance
company.
The evidence in Wickline was overwhelming that Mrs. Wickline did not
require continued hospitalization beyond the date of her discharge. Her bodily
functions were all normal. She had commenced rehabilitation treatment. There
was no open infection or open wound. No new symptoms had appeared, and she
did not appear to be in any danger. Her initial length of stay had been extended
for 4 clays for general observation. Her treating physician testified that at the
time of discharge, her condition was neither critical nor deteriorating. If it had
been, he testified, he would have kept her in the hospital regardless of the Medi-
Cal decision. Moreover, she was examined a full week after her discharge by
one of her physicians, who noted nothing remarkable in her condition.
How, then, did the jury assess liability against the utilization reviewer for
half a million dollars? The answer is easythe jury system protects the
individual against faceless bureaucracies. The jurors were presented with a

APPENDIX A 175
tragic situation in which, through no fault of her own, Mrs. Wickline lost her
leg, suffered greatly, and would continue to suffer greatly for the rest of her life.
Yet, the Medi-Cal reviewer could not even remember why he had denied the
hospital stay extension. It should come as no surprise that the jurors who have
the legless and tearful Mrs. Wickline and her family sitting before them day
after day will be inclined to ignore fine distinctions in the evidence and
compensate Mrs. Wickline for her suffering. From this perspective, the
$500,000 verdict does not seem so large.
The Wickline case illustrates the basic difficulty in defending cases of this
nature. Utilization review, when viewed most favorably, solves the broad
societal goal of containing health care costs. But the plaintiff's attorneys are
likely to cast UR as simply a tool of insurance companies to make more money.
The review organization's paper forms, the inability to remember specific cases,
and the inevitable failure to gather every detail and other indicia of bureaucracy,
coupled with the inescapable fact that reducing the amount of medical care
saves money for the insurer, will always embellish the plaintiff's case.
Moreover, unlike a malpractice case in which evidence of a physician's
coverage is carefully shielded from the jury, in a UR case, the insurance
company itself is usually the named defendant. For these reasons, the role of the
jury cannot be ignored in any assessment of utilization review liability.
Negligence
Negligence is likely to be the principal cause of action against both a payer
and an independent review organization for a UR decision. If the case involves
a payer's in-house review, the plaintiff will probably proceed also on theories of
breach of contract and breach of the implied covenant of good faith and fair
dealing. In Wickline, a case involving in-house review, the only cause of action
was negligence, and the court did not discuss any other potential theories. To
establish negligence, the plaintiff must show that the defendant owed the
plaintiff a duty of reasonable care, that the defendant breached the duty, and
that the breach proximately caused the plaintiff's injury.
Existence of an Duty of Care

Whether a duty of care extends from an independent review organization
to the patient depends on a variety of factors, but the courts are likely to find
that a duty exists.10 A primary factor in determining whether a duty of care
exists is foreseeability of harm. When a review organization denies
authorization for a treatment program, it is certainly foreseeable that the patient
may forego treatment. The review organization may argue that its determination
is for payment purposes only and is not to be relied upon as

APPENDIX A 176
medical advice.11 Nevertheless, the UR decision inevitably will have some

affect on the decision whether to proceed with treatment. At the least, the UR
decision may be a factor for the patient to consider and in many cases, it will be
the decisive factor. The argument that the UR decision affects payment only is
more persuasive on the issue of causation than on the issue of duty, as discussed
in the section Causation below.
Wickline clearly recognizes that a duty of care exists for UR decisions
made in-house by a payer.12 However, there is also good reason for finding that
the duty is owed by an independent review organization, given that the potential
harm to the patient is just as foreseeable. The technical distinction that the
independent review organization may not be in direct privity of contract with
the patient should not prevent the imposition of a duty of care for tort law
purposes.
The Standard of Care

The next issue in a negligence action is to define the standard of care.13
There are two subissues involved here. First, the review organization may be
liable if a defect in its procedures resulted in harm to the patient. Second, even
if the organization's procedures were adequate, the duty of care may be
breached if the decision on medical necessity did not meet proper standards.
The standard of care for the procedural aspects of a UR decision is likely
to be based on the standards followed by review organizations generally, that is,
the standard of care in the community of consultants in the same business.
Overall, UR procedures must be sufficient to obtain enough information to
make an informed decision and to enable a timely dialogue and/or appeal if the
treating physician or patient disagrees.
In Wickline the physician consultant reviewed only the Medi-Cal form
completed by the treating physician. He did not review the patient's chart or
consult with the treating physician or a specialist consultant before rejecting the
requested hospital extension.14 The plaintiff attempted to show that these
procedures were insufficient. The Wickline decision did not criticize the review
procedure, thereby implicitly accepting the argument that the reviewer was
entitled to rely on the information on the Medi-Cal form and that the burden
was on the attending physician to justify the request by including all pertinent
information on the form.15 Nonetheless, there is no doubt that the defendant's
sloppy recordkeeping, including the failure to document any reasons
whatsoever for the denial, played a persuasive role in the trial court.
The procedural safeguards for review organizations are increasingly the
subject of articles and seminars. As a result, the industry appears to have
developed minimum standards that are consistently suggested for

APPENDIX A 177
review organizations and would be put forth by plaintiff's attorneys in a UR

case.16 These standards include the following:
1. review decisions should be made by qualified medical professionals, and
any denial decisions should be made only by licensed physicians;
2. reviewers should consult with specialist physicians as appropriate;
3. efforts should be made to obtain all necessary information, for example,
by reviewing the patient's charts and consulting with treating physicians
as appropriate;
4. the reasons for decisions should be clearly documented;
5. there should be a well-publicized and readily available appeal
mechanism; and
6. decisions and appeals should be made in a timely manner as required by
the exigencies of the situation.
Finally, the review organization should be careful to follow its own
procedures. The failure to follow one's own procedures exposes the entity to
potential liability, regardless of whether the particular review decision was
correct.17
The other standard of care in a UR liability case is the standard for the
substantive decision on medical necessity. The duty of care for this medical
judgment is likely to be the same as that for physicians generally. By using the
expertise of physicians, UR organizations hold themselves out as having a
special skill in the evaluation of medical treatment. The UR organization,
through its individual physician and nurse review consultants, will likely be
held to a professional standard of care.18 In other words, a UR consultant should
authorize treatment if a physician applying the community standard of care
would recommend this procedure as medically necessary.19
Many review organizations use standardized criteria to determine the
appropriate treatment and length of hospitalization. A statistical methodology
may be used to measure inappropriate treatment, admission, or length of stay.20
Of course, the criteria need to be based on an appropriate standard of care and
should be reviewed periodically to ensure that they do not become outdated.
Some believe that the fear of liability may deter UR organizations from
adopting well-reasoned criteria that deviate from the existing community
standard of care, even if it can be demonstrated that the community standard
may lead to unnecessary care. There is no doubt that because of liability
concerns, close cases tend to be resolved in favor of allowing care based on the
existing community standard. At the same time, it seems clear that UR reduces
the volume of certain types of care, for example, inpatient hospital days for
HMO enrollees. To some extent, the adoption of criteria is serf-validating; that
is, as the use of the criteria becomes more

APPENDIX A 178
widespread, the criteria themselves become evidence of the community

standard of care.
The development and refinement of utilization criteria is a fluid process.
Even though a UR organization is held to the community standard of care, the
community standard is rarely a bright line, but is more often a consensus of
judgment. In a close case, the UR organization is well advised to research the
literature or data on particular criteria before making a final decision, in order to
assess how strong the expert testimony will be if the decision is challenged. If
the literature or data are strong, there is a good likelihood of being able to
defend successfully if the case is taken to trial. On the other hand, if the criteria
are based on weak data, an obscure article, or an article that recognizes that the
procedure is new and unverified, then the UR organization should give the
benefit of the doubt to the patient and await further evidence before adopting a
more restrictive standard.
Standardized criteria have the beneficial effect of providing a uniform
basis for screening all cases initially. Some UR organizations, particularly
payers performing UR in-house, develop criteria from their own enrollment
population and physicians. Such statistical models may provide valid screening
tools for initial decisions, but their use may lead to inappropriate decisions in
individual cases posing variations from the statistical norm. If criteria are used,
there should be procedures available to allow the reviewer to deviate from them
in individual cases of medical necessity.21
Another danger in the use of utilization criteria is the possibility that
overzealous and mechanical application of the criteria may contradict explicit
coverage provisions in the policy. As long as the criteria are limited to
quantitative considerations, such as statistical determinations of lengths of stay
or lists of procedures ordinarily done on an outpatient basis, the criteria serve a
useful purpose. However, a too sweeping exclusion may result in the denial of
services that are medically necessary and, therefore, that are covered under the
basic policy.22
Causation
The third issue in a negligence action against a review organization is
whether the denial of coverage proximately caused the patient's injury.
Causation was the decisive issue in Wickline and promises to be the single
largest hurdle in most UR liability cases.
In Wickline, the treating physician failed to take any steps to contest the
initial denial. In fact, he signed the hospital discharge order and testified at trial
that Mrs. Wickline's condition at the time of discharge was neither critical nor
deteriorating. The fact that the physician may have been intimidated by the
Medi-Cal program did not mean that he was incapable

APPENDIX A 179
of contesting the decision if he disagreed with it. Thus, the court concluded that
Medi-Cal did not participate in the medical decision to discharge Mrs. Wickline
from the hospital and could not be held responsible for that decision.23
An alternative basis for the Wickline decision might have been that, wholly
aside from the hospital discharge, the Medi-Cal decision was too remote from
the infection and gangrene that eventually set in. In other words, the same harm
would have befallen Mrs. Wickline even if she had remained in the hospital for
the additional 4 days. One of her doctors testified that he examined her in his
office a full week after the discharge (that is, 3 days beyond the extension that
Medi-Cal denied) and that he did not note any material or substantial change in
her condition.24
While Wickline provides an excellent description of the facts, its legal
analysis is not as crisp. The holding could have been reached on any number of
points, but the court's case law analysis is limited to a general recitation of
overall negligence principles.25 The court does not specifically discuss
proximate cause, superseding intervening cause, or how the principles of
comparative fault may apply. The court simply concludes as a factual matter
that the Medi-Cal decision had nothing to do with the hospital discharge.
There remains an infinite variety of fact situations in which the causation
issue has yet to be explored. What if Mrs. Wickline's infection had set in only 1
or 2 clays after discharge instead of 10 days? At that time, she still would have
been in the hospital if Medi-Cal had not denied the extension. Based on the
court's analysis, Medi-Cal still would have been absolved of liability because,
according to the court's analysis, Medi-Cal did not participate in the discharge
decision.
Breach of Contract
A UR organization may be liable under contract theories as well as under
tort theories. A contract typically exists between the patient and the third-party
payer to pay for medically necessary services. When the payer performs UR in-
house, an improper review decision that results in nonpayment is a direct breach
of contract.
When the payer delegates UR to an independent review organization and
improperly denies benefits based on the UR decision, a contract remedy against
the payer would still be available to the patient, even though the utilization
review was undertaken by the independent review organization.26 The absence
of a contractual relationship between the patient and the UR organization does
not insulate the payer from contract liability for the UR organization's actions.27
In addition, the patient may assert a contract claim directly against the
independent entity if the patient is a third-party beneficiary of the contract
between the UR entity and the payer. This

APPENDIX A 180
contract often sets forth the scope of services to be rendered and the standards
for review. A third-party beneficiary relationship exists if the contract between
the payer and the UR entity was intended to benefit the patient.28 The review
organization will argue that its contract with the payer was not intended to
benefit the patient, but was simply to provide services to the payer and to
benefit the payer exclusively.29
The measure of damages for breach of contract is all damages reasonably
foreseeable from the breach.30 In prior Or concurrent review, since it is
foreseeable that denying authorization will result in the patient foregoing
medical services, the defendant potentially is liable for injury or death caused to
the patient.
The critical issue in a contract action, just as in a negligence claim, is
causation. The plaintiff still must show that the review decision proximately
caused the harm. Thus, in the final analysis, the pivotal issue in a contract claim
is likely to be the same as in a negligence claim.
Insurance Bad Faith

Many states recognize tort liability against an insurance company for
breach of the implied covenant of good faith and fair dealing. This cause of
action exposes the defendant to punitive damages that ordinarily would not be
available in a simple negligence or breach of contract case.31 Insurance bad
faith theories are also useful to the plaintiff because they inquire directly into
the process used to reach a coverage decision, and not merely the correctness of
the decision itself.32 Similarly, the failure to provide adequate appeal rights may
itself be the basis for bad faith liability.33
Insurers performing UR in-house have direct exposure to this liability.
When the payer delegates the UR function to an independent entity, both the
payor and the reviewer, as its agent, are potentially exposed to bad faith
liability. When the payer ratifies the reviewer's decision, or leaves the decision
primarily in the reviewer's hands, the courts are likely to hold the payer
responsible to the plaintiff for the reviewer's mistakes.34
Whether the independent review organization is directly liable for
insurance bad faith is a more complicated question. The legal basis for bad faith
liability is unique because it contemplates a tortthe breach of a duty of care
arising from a special contractual relationship. In contrast, the tort of negligence
is not based in contract, but arises out of the finding that a duty of care is owed
directly by one party to another. The difficulty in applying the bad faith tort
theory in the UR context is the absence of the direct contractual relationship on
which the tort is based between the plaintiff (the insured) and the independent
UR organization. Furthermore, an independent review entity, unlike an
insurance company, does not stand in an established "special" or fiduciary
relationship to the

APPENDIX A 181
insured. The nature of the relationship between the insurer and the insured is
one reason the courts have sanctioned the development of the bad faith doctrine.
Also, as long as a reviewer's compensation is not incentive-based, the reviewer
is not subject to the insurance company's inherent conflict of interest, namely,
making more money by denying more care.
Despite these differences between the status of the insurance company and
the independent reviewer vis vis the insured, there are arguments that the
reviewer should be subject to the same risk of bad faith liability as the insurer.
First, the review organization makes decisions that directly affect the level of
insurance benefits, and thus cause the same kind of harm as insurance company
decisions do. Moreover, the harm is foreseeable, and the potential magnitude of
the harm is great. Like the relationship between the insurance company and the
insured, the relationship between the review organization and the insured is
reflected in a nonnegotiable "adhesion" contract. As a practical matter, the
insured cannot bargain with the reviewer over the nature of their relationship
and, therefore, is forced to accept the reviewer's terms. Finally, the reviewer is
acting in the role of an insurance company. For these reasons, an independent
reviewer may be exposed to bad faith liability, even though the direct
contractual link is absent.
While the courts generally have not imposed bad faith liability on the
agents and contractors of insurance companies,35 at least one case found
liability despite the lack of contractual privity. In Delos v. Farmers Insurance
Group,36 the management company for a reciprocal insurer was liable under the
implied covenant of good faith and fair dealing for bad faith handling of a
claim. The court found that the management company was acting in the
capacity of an insurer by processing claims. The case may be distinguishable
from other cases because the management company was, in fact, an arm of the
insurer and was the attorney-in-fact and, therefore, a fiduciary with respect to
each insured. Nevertheless, the case is significant because the court wanted to
prevent the insurer from insulating itself from liability by forming a
management company, and consequently, the absence of contractual privity did
not prevent the bad faith tort remedy. In addition, there is a possibility that an
independent reviewer could be sued for conspiracy with the insurer to breach
the implied covenant of good faith and fair dealing.37
As a practical matter, a plaintiff is likely to seek the punitive damages
available for bad faith liability from the "deep pocket," which is the insurer,
rather than the independent reviewer. However, there is a definite risk that an
independent reviewer would be exposed to such liability or forced to defend
against such claims.

APPENDIX A 182
Infliction of Emotional Distress

A review organization is potentially exposed to claims for infliction of
emotional distress. In states where the intentional tort is recognized, proof is
required of (1) extreme and outrageous conduct, (2) intent to cause severe
emotional distress to the insured or reckless disregard of the probability that
such distress would result, and (3) severe emotional distress suffered by the
plaintiff as a proximate result of the defendant's conduct.38 The intentional tort
has been interpreted to require proof that the defendant's acts were ''so extreme
as to exceed all bounds of that usually tolerated in civilized community."39
Since such extreme misconduct is not likely to arise in the UR context, the more
likely cause of action against a UR entity would be negligent infliction of
emotional distress. The negligence tort requires a showing of (1) negligent
conduct by the review organization, (2) severe emotional distress suffered by
the patient, and (3) proximate causation.
Warranty Theories
Consumer materials from the payer or the review organization frequently
will tout the beneficial services performed by UR or emphasize the quality
aspects of the UR procedure. Such materials may support a claim for breach of
warranty. The question generally will be whether the consumer materials were
intended to warrant a specific result as opposed to being generalized sales
puffing.40
An independent UR organization is well advised to specify in its payer
contracts that the payer must include in its consumer documents certain
provisions relating to UR. This is particularly important for UR organizations
contracting with self-insured employers who may not be as sophisticated as
large insurers in preparing consumer documents. At a minimum, the UR
organization will want to make sure that the consumer materials reference the
insured's consent to UR, clarify that the UR system is the payer's enterprise, and
spell out the rights and obligations of the UR process. Another provision would
state that the UR entity does not make medical decisions but only advises on
payment questions, and that the patient and physician are solely responsible for
deciding what medical treatment is provided. Although such statements may not
be strictly enforceable, they can be helpful to a UR organization in defending
against various liability theories.
Products Liability
The increasing use of computer software programs by UR organizations to
determine the appropriateness of admissions may give rise to

APPENDIX A 183
claims focusing on the design of the software. A plaintiff could attempt to

fashion an argument that he or she suffered damages resulting from an
erroneous denial that was made by using a defectively designed software
program. By arguing under a products liability theory that the software is a
defectively designed product, the plaintiff need not prove that any party acted
negligently. Products liability cases require proof only that the product was
defective and that the defect caused the injury.41 In addition, the fact that the
UR organization did not develop the software, but merely acquired it from a
software vendor, may not insulate the UR entity from liability. The products
liability doctrine casts a wide net over prospective defendants, permitting
plaintiffs to sue virtually every party in the chain of the manufacture and
distribution of a product.42
A key issue in a products liability case would be whether the use of the
software program constitutes a product or a service. Products liability theories
do not apply to services, but the distinction between a product and a service is
not clearly defined in the law.43 Moreover, it appears that there is little
precedent for finding that products liability applies to claims arising out of the
use of computer software programs. However, there are cases finding that, for
example, instrument approach charts used by pilots for landing are considered
products for purposes of product liability claims brought in wrongful death
actions against the manufacturers of the charts.44 Plaintiffs may try to extend
these cases by analogy to software programs.
Defamation and Interference with Contractual Advantage

The UR entity's liability is not limited to potential claims by the patient.
Physicians or other providers who are harmed by UR decisions may also bring
suit. Typically, these actions would be for defamation or, potentially, for
interference with the physician's contractual relationship with the patient.45
In Slaughter v. Friedman,46 plaintiff Slaughter was an oral surgeon who
brought an action against a dental insurance company and its dental director for
defamation and intentional interference with prospective advantage. In denying
claims for Slaughter's services, the insurer enclosed a letter to the patients that
described Slaughter's work as ''unnecessary," claimed that Slaughter had been
"overcharging," and stated that the insurance company would report him to a
dental association for disciplinary proceedings. The letters also advised the
patients to make no further payments to Slaughter. The California Supreme
Court upheld Slaughter's right to sue for defamation (the interference claim was
not before the court).
Whenever a UR program informs a patient that the treating physician
rendered, or proposes to render, treatment which is not medically necessary,

APPENDIX A 184
a potential defamation claim exists. The UR decision carries a stamp of

authenticity and may cause a patient to question the doctor's medical judgment
and capability. In addition, a finding that services were not medically necessary
may justify the patient's decision not to pay for the services.
The UR entity would have available in a defamation action the defense of
qualified privilege that applies to communications between persons with a
mutual interest in the subject matter.47 The privilege is not absolute, however,
and may be defeated by showing that the communication was made with
malice. In the Slaughter case, the court found that the pleadings sufficiently
raised the issue of malice to allow the complaint to stand.48 For these reasons, a
UR organization should be careful to limit its communications simply to the
basis for its denial decisions and should avoid unnecessary embellishments or
inflammatory language.
Antitrust
A full discussion of potential antitrust liabilities is beyond the scope of this
paper. However, a few points are worth noting.
For the review organization that does medical necessity determinations,
antitrust is a relatively lesser threat than the other potential liabilities reviewed
above are. Antitrust is a more immediate concern for entities that set uniform
fee schedules or that limit or exclude providers, such as HMOs or preferred
provider organizations (PPOs). When a closed-panel system such as an HMO
excludes a physician because of high utilization, there is some potential for
antitrust liability. As a general rule, the conduct will not be illegal per se but
will be tested under the rule of reason. Under the rule of reason test, the
challenged activity will be upheld when the legitimate, procompetitive interests
outweigh the potential anticompetitive effects.49 As long as the defendant
succeeds in having the court apply the rule of reason test, the beneficial effects
of limiting the provider panel to cost-effective physicians are likely to prevail
over the interest of a single physician in remaining on the panel, at least in cases
in which the defendant's market share is within reasonable bounds.50
The difficulties in defending a provider exclusion case tend to be the
inevitable problems of proving that a physician is a high utilizer. The excluded
physician often attempts to show that he or she was singled out because of some
personal animus by his or her competitors on the review committee. Often, the
defendant's records are not as precise as the attorney would hope and the
statistics are not consistent or are incomplete. Other physicians who have not
been excluded may have equally bad or worse utilization records, thereby
raising the question of unequal treatment. The excluded physician may have an
outstanding reputation for quality, or may

APPENDIX A 185
claim that his or her patients tend to be sicker than the average population. All
of these problems make the defense of the antitrust case more difficult, but there
is nothing inherently illegal in excluding a provider based on utilization data.
For UR entities doing medical necessity determinations, the principal
source of antitrust claims lies in restricting the availability of new or
controversial treatments. Because UR applies a community standard of care,
there is a built-in bias against new treatments that do not have widespread
acceptance in the medical community. For example, various antitrust cases have
been brought by purveyors of dimethyl sulfoxide and laetrile claiming that third-
party payers and their reviewers were unlawfully refusing coverage for these
new "miracle" cures. For a less controversial example, consider heart
transplants. It is likely that utilization review entities continued to deny
transplants as "experimental" or not medically necessary for some period of
time beyond the point when transplants became relatively safe and even
commonplace in sophisticated areas of practice.
In this kind of antitrust case, one of the biggest hurdles is to show that a
combination or conspiracy exists to deny coverage. Unilateral action is
insufficient to establish a violation under Section 1 of the Sherman Act; there
must be a "contract combination or conspiracy in restraint of trade."51 The fact
that several employees or even a committee of reviewers participated in a denial
of coverage generally is not enough to establish antitrust liability because they
all work under a single corporate entity.52 The plaintiff will want to show that a
number of payers are conspiring to deny coverage of a new treatment. While the
plaintiff often can show that a number of payers actually do deny coverage, it is
a rare case when the plaintiff can establish an agreement among the payers to
deny coverage.53
Liability of Consultants and Employees

The consultants and employees of the review organization may be found
liable for torts in which they participated, just as is the case for any other
individuals who work for a corporation. Their protection lies in proper
insurance and in their right to indemnity from the corporation.
Many states have statutes that provide immunity from liability for certain
peer review activities. However, these statutes generally do not protect the
physician consultants of independent review organizations. The statutes
generally are limited to certain types of peer review, such as hospital review
committees, medical society committees, and certain committees of HMOs; and
they generally focus on peer review of quality of care rather than medical
necessity determinations for private payers.54 The immunity

APPENDIX A 186
is usually a qualified rather than absolute immunity, meaning that the

participant who acts with malice is not shielded from liability.
When a physician provides independent consultant services to a UR entity,
the physician's malpractice insurance may or may not cover those services.55
The UR entity's policy probably does not cover independent contractors and
may also exclude claims arising from the provision or failure to provide medical
services. Liability insurance needs to be clarified by both the physician
consultant and the UR entity.
An employee of a corporation who is personally named in a lawsuit and is
successful in the defense of the suit may have the right to indemnity from his or
her employer under the applicable Corporations Code, unless it is shown that he
or she did not act in good faith or acted against the best interests of the
corporation.56 A corporation will generally have the option of indemnifying an
agent for actions taken in good faith and with the reasonable belief that the
actions were in the corporation's best interests As a practical matter, employees
generally expect the corporation to assume any liability in a situation such as
that in the case of Wickline. However, cases will arise in which the corporation
will not indemnify, the corporate insurance is exhausted, or the corporation is
otherwise without assets. A physician consultant, therefore, is well advised to
examine carefully the corporation's liability policy and to review his or her own
consultant contract in light of the corporation's liability policy and the
applicable law on indemnity.
State Regulation
The design of UR procedures is increasingly the subject of state regulation.
The state of Maryland enacted legislation in early 1988 requiring a UR
organization to obtain a certificate from a state agency before conducting UR.58
The application process requires, among other things, submission of a UR plan,
including a description of review standards and procedures, the circumstances
in which UR may be delegated to a hospital UR program; and the appeal
mechanism for patients, providers, and hospitals. The legislation does not
specify any detailed requirements for standards and procedures, and no
regulations have been promulgated as of this writing.
Minnesota also enacted in 1988 a provision requiring prospective
utilization review decisions to be made within ten business days. The time
period does not run until all information reasonably necessary to the decision
has been made available.59
Maine requires insurers with prospective review requirements to file
annual reports setting forth the number and type of such review decisions.

APPENDIX A 187
The reports also must include a summary of denials and any appeals or
lawsuits related to the prospective review.60
Louisiana presently has the most comprehensive legislation regarding
prospective utilization review requirements.61 The Louisiana statute expressly
permits a damages action for "unreasonable delay, reduction or denial" of
services, and sets certain standards for prospective utilization review decisions.
Decisions made within two working days are deemed to be timely; however,
longer time periods may be justified under some circumstances. The statute
further provides that decisions on medical necessity must be based on
"nationally accepted current medical criteria," but the statute fails to give any
further guidance as to what medical necessity means.
While the Louisiana statute expressly authorizes damages actions, it also
limits the potential recovery. Damages for prospective review decisions are
limited solely to physical injuries, which excludes punitive damages and also
may be interpreted to exclude emotional distress damages. However, the
prevailing plaintiff may recover reasonable attorneys' fees. The statute also
imposes on the plaintiff a strict causation requirement, and limits damages
solely to injuries that are the "direct and proximate cause" of the unreasonable
delay, reduction, or denial. Thus, although some may view the Louisiana statute
as unwanted regulation, the statute actually provides certain safeguards to the
insurance industry and limits the amount of recoverable damages.
Erisa Preemption
In Pilot Life Insurance Co. v. Dedeaux,62 the U.S. Supreme Court held that
when insurance is provided through an ERISA plan, common law actions by an
insured for bad faith denial of insurance coverage and related state law tort
claims are preempted by ERISA.63 Most private employers' benefits plans are
covered by ERISA.64 Thus, many of the liability theories discussed above are
subject to an ERISA preemption defense. However, the courts are still
grappling with the scope of preemption, and there remain areas of uncertainty.65
When a utilization review case is brought directly against an ERISA payer
based on a denial of benefits or mishandling of a claim, the action should be
preempted. The alleged wrong at issue is the denial of benefits provided by the
ERISA plan, which is the same as the wrong involved in the Pilot Life case. The
fact that consequential damages flowed from the denial of benefits (for
example, Mrs. Wickline's loss of a leg) should not distinguish the case from that
of Pilot Life.
When the defendant is an independent review entity, the ERISA
preemption issue will turn on whether the review entity is deemed to be a

APPENDIX A 188
"fiduciary" under ERISA. A person or entity performs fiduciary functions, and

is therefore deemed to be a fiduciary under ERISA, if it exercises discretionary
authority or control over the plan.66 When the review entity makes the initial
determination on coverage, for example, by refusing to authorize surgery or
refusing to extend a hospital length of stay, a court is likely to hold that the
review entity is exercising a fiduciary function. If the review entity is held to be
a fiduciary, then ERISA preemption would apply.
However, if the court finds that the review entity is not a fiduciary, it is not
clear whether ERISA preemption would apply.67 This area is one of continuing
development following the Pilot Life decision.
ERISA preemption does not leave the plaintiff entirely without a remedy.
However, the remedies available under ERISA may be severely restricted and
may preclude the plaintiff from recovering consequential damages such as pain
and suffering or, in a situation like that of Wickline, the cost of living a life as
an amputee. The monetary remedy available to a plan participant in an ERISA
action is limited generally to recovery of plan benefits, and consequential
damages are not permitted.68 The court does have discretion under ERISA to
fashion broad equitable relief,69 and given the right set of facts, a creative court
may well attempt to fashion an equitable remedy to provide adequate
compensation for the plaintiff.
Two other features of an ERISA action are worth noting. First, to the
plaintiff's advantage, attorneys' fees may be awarded under ERISA. Second, to
the plaintiff's general disadvantage, there is no right to a jury trial under
ERISA.71
The recognition of ERISA preemption as a defense to state common law
torts may lead plaintiffs to resort to other innovative theories to avoid
preemption. One such theory would rely on a federal racketeer influenced and
corrupt organization (RICO) action,72 which permits a prevailing plaintiff to
recover treble damages and attorneys' fees. Since RICO is a federal action, it is
not preempted by ERISA. Some courts have permitted a RICO cause of action
against an insurance company for denying coverage.73 The plaintiff's theory is
that the insurer misrepresented the coverage at the time the policy was issued.
Since the insurer used the mails in selling the policies, the alleged
misrepresentation constituted mail fraud, which is one of the predicate acts for
RICO liability. Although these cases did not involve health insurance, similar
allegations could arise in the health insurance context.
Liability of the Employer or Payer

If utilization review is performed in-house, then the employer or payer is
directly liable for the consequences of review decisions.74 This section

APPENDIX A 189
addresses the liability of the payer or employer that contracts with an

independent review entity. In that situation, the payer has the potential for direct
liability or vicarious liability.
The principal practical distinction between direct and vicarious liability is
whether the payer will be fully indemnified by the review entity. A principal
who is found to be vicariously liable for the torts of its agent is entitled to full
indemnity from the agent.75 However, if the payer is found to be directly liable
for the harm, the damages will be allocated proportionately between the payer
and the review entity based on their respective faults. Thus, the payer will want
to avoid direct liability, if possible, and shift as much responsibility as possible
to the review entity.
Direct Liability
The employer or payer may be directly liable under several theories. First,
if the payer acts on the reviewer's decision and denies coverage, there is
potential liability for breach of contract. If a bad faith insurance claim is
available, then the payer may also become liable for tort damages in cases
which it adopts or ratifies the reviewer's decision. An insurance company may
not escape liability by delegating its responsibilities to an outside entity.76
When the payer is not an insurance company but is, for example, a self-insured
employer, the employer may still be liable for breach of its contract to pay
benefits (or under ERISA for denial of benefits).
In a contract action, the payer could not avoid liability by showing that the
review entity was an independent contractor. A party to a contract does not
avoid contractual liability by delegating contractual obligations to another
unless the delegate assumes liability and the obligee consents to the
assumption.77 Few, if any, review contracts would qualify as an assignment.
A payer also may be liable directly in tort if it participates in the design of
a faulty UR system or if the payer fails to exercise due care in selecting the
independent review entity. The latter theory, negligent selection, is being
increasingly recognized in a variety of fields.78 For this reason, a payer should
inquire into the reputation of the independent review organization, inquire
directly about any lawsuits or judgments, and document the results of its inquiry
as evidence of the exercise of due care.
Vicarious Liability
A payer also may be found vicariously liable for the torts of the review
entity under traditional principles of agency. It is frequently suggested that a
payer may avoid vicarious liability in cases in which the review organization
serves as an in. dependent contractor, rather than as an agent, as UR contracts
almost universally contain independent contractor provisions.

APPENDIX A 190
While this distinction is theoretically possible, the independent contractor

defense is difficult to sustain in practice. Not only is the independent contractor
doctrine viewed with disfavor today,79 but it also may be difficult to prove
independent contractor status.80
Even if the review entity is an independent contractor, the plaintiff may
still be able to prevail under the doctrine of ostensible agency. Ostensible
agency exists when one either intentionally or negligently allows another to
believe that an agency relationship exists.81 In cases in which the review entity
tells a hospitalized patient that the payer's benefits are terminating or tells the
patient who wants an operation that the procedure is not medically necessary,
the patient is likely to assume that the reviewer is the "agent" of the payer. A
payer might be able to avoid ostensible agency by clearly emphasizing in its
consumer documents that the review entity is entirely separate and independent.
Even so, a court may be reluctant to allow the payer to escape liability entirely.
Indemnity
The most important protection for a payer is to insist that the reviewer
carry adequate liability insurance. Ideally, the payer will be named as an
additional insured on the review entity's policy, and the insurance company will
be required to notify the payer directly of any change in coverage or
cancellation. The payer should also review the terms of the reviewer's policy to
see that it covers the various types of liability that can arise in the utilization
review context.
Many payers also insist on indemnity and hold harmless provisions in the
contract with the review entity. Such clauses are likely to be much less helpful
without an adequate liability policy. Without insurance, the review entity may
not be able financially to satisfy its indemnity obligation. Also, when one
defendant is insured or is otherwise financially viable and the other defendant is
uninsured, the plaintiff will try to tailor the case toward the financially viable
defendant. Thus, the payer who otherwise may be a secondary defendant may
find itself the primary target if the review entity is uninsured.
Liability of the Treating Physician

The impact of UR has fallen most heavily on practicing physicians. While
a physician is accustomed to having some services questioned by retrospective
claims review, the more intrusive forms of prior or concurrent review meddle
directly in the physician's course of treatment of the patient. A review
consultant, who usually and has never seen the patient, is questioning the
physician's judgment as to what is medically necessary. Then, as

APPENDIX A 191
in the Wickline decision, the judicial system blamed the treating physician for
failing to protest strongly enough against the reviewer's decision.
On a philosophical level, many physicians have cried foul. UR usurps the
considered, professional judgment of the physician. It interferes with the
essential relationship of trust between the physician and patient. It is Big
Brother watching. Nevertheless, UR appears to be firmly entrenched in the U.S.
medical system, and the courts will be forced to continue to grapple with the
respective obligations of the parties.
Defining the obligations of treating physicians presents the largest area of
unanswered questions.82 A UR denial places the physician at risk of
nonpayment if he or she proceeds with the proposed treatment. Assuming that
the physician appeals the UR decision, is he or she safe to defer to the reviewer
and withhold treatment if the appeal is denied? Can the physician simply
withdraw from treating the patient if the review decision is adverse and the
patient cannot pay for the treatment? If the physician withholds treatment and
harm results to the patient, does the UR decision provide the physician with a
defense to a malpractice action?
Wickline suggests that a physician may not avoid a malpractice claim
simply by acquiescing in a UR decision. Wickline held that the responsibility
for the hospital discharge was solely that of the physicians. In that case, of
course, the physicians did not protest Medi-Cal's denial, and by their testimony
they agreed that the discharge was appropriate at the time. But Wickline did not
say how far a physician must go in appealing a denial decision. Would the court
have upheld Medi-Cal liability if the treating physicians had simply requested
reconsideration and Medi-Cal had denied the request? Were the treating
physicians required to pursue all available avenues of appeal, including an
administrative review heating and even a court challenge to the decision? Was it
even possible for these avenues of appeal to be resolved in the limited time
before the patient was to be discharged? If, after contesting the decision, the
physician still believed it was unsafe to discharge Mrs. Wickline from the
hospital, were they required to keep her in the hospital and risk nonpayment?
Hard cases will arise when the physician vigorously appeals the denial, to
no avail. In this situation, the risk of nonpayment is even clearer than it is at the
time of the initial decision. At the same time, by protesting the initial decision,
the physician may have created substantial evidence that the treatment is
medically necessaryevidence that will be used against the physician if he or
she then withholds treatment and harm occurs. At some point, if the physician's
appeal is strong enough, a court is likely to hold the review entity responsible
for musing at least part of the harm.83 But the possibility of sharing liability
with another defendant is little solace to the treating physician, who still
remains a clear potential defendant if he or she withholds treatment.

APPENDIX A 192
Most of these questions focus on the physician's obligation to provide

treatment to a patient who cannot pay. When the physician has not commenced
treatment, he or she is relatively free to turn away the case.84 Most UR cases,
however, involve an established physician-patient relationship. Terminating this
relationship may constitute the tort of medical abandonment.
The essence of abandonment is unilateral nonconsensual termination by
the medical practitioner.85 The two elements required to avoid a claim of
abandonment are (1) notice and (2) opportunity to secure alternative medical
services. When a UR denial applies to a truly elective procedure, there is
generally ample opportunity for the physician to satisfy these two elements
before foregoing the course of treatment. However, when the patient is
hospitalized and a certified length of stay expires, as in the Wickline case, there
is little or no practical opportunity to give notice and to seek alternative
treatment. The responsibility remains on the physician to provide care, or to
keep the patient in the hospital, if the physician believes that the community
standard would require such action.86
The physician also has increased exposure to claims of lack of informed
consent.87 When a proposed treatment is denied, the physician must be careful
to explain to the patient the options and the risks. The patient then must decide
whether to go ahead with the treatment and pay for it out of his or her own
pocket. Even if the risks inherent in the treatment are minimal, the financial
pressure on the patient to forego treatment makes it particularly important for
the physician to explain carefully the medical risks of foregoing the treatment.
Many physicians complain that they often are unable to learn the basis for
a UR organization's denial decision. However, the appeals process should
provide a means for obtaining the basis for the decision. In view of the
responsibility placed by Wickline on the physician to appeal adverse decisions,
the physician should demand the basis for the UR decision on the grounds that
he or she needs it to respond adequately in the grievance or appeals process.88
Also, in this writer's opinion, it is generally good practice for the UR appeals
committee to prepare a written statement of its decision. A written statement
demonstrates a good faith, reasoned approach to the problem. Even if the
patient or the physician disagrees with the conclusion, a well-reasoned written
decision serves to dampen the sense of frustration and anger generated by a
terse rejection of the appeal.
In any case in which a reviewer denies care, the physician, of course, is
free to explain the denial and obtain the patient's consent to treatment and to pay
for the care, notwithstanding the denial. However, a physician should take
precautions to ensure that the patient understands fully the financial
implications of such consent.

APPENDIX A 193
Legal Issues For the Patient

The patient, of course, lies at the heart of the controversy. Aside from the
various liability issues discussed above, the most pressing issue for the patient
is how to obtain timely review of a UR denial. The other issue for the patient is
whether he or she must pay for physician services that the reviewer determines
were not medically necessary.
The Need For Expedited Review

Consider Mrs. Wickline's situation when she was lying in the hospital
recovering from multiple surgeries and was told that Medi-Cal would not cover
further hospitalization. She could have either gone home, which would have
been against the initial advice of her physician, or she could have remained in
the hospital and risked the. devastating possibility of having to pay for it herself.
This situation highlights a major weakness in most UR systemsthe
ability to resolve appeals in a timely manner. Most UR appeals systems are
designed for retrospective review. They include multiple levels of in-house
review, followed by an independent decision by an arbitrator, a court, or an
administrative law judge. It may be weeks, months, or even years before a UR
decision is reviewed by an independent third party.
While such a system may suffice for retrospective review, in a prior or
concurrent review situation the patient's medical care is directly at stake. To be
effective, the appeal process must occur within a time frame consistent with the
patient's medical condition. For example, in cases in which the challenged
decision is the discharge of a sick infant from the hospital to the home, it should
be possible to complete the appeal before the infant is sent home and
complications can set in. Another example is when an HMO patient needs
urgent treatment of a specialized nature. If a question arises whether the patient
needs to go out of the HMO plan to secure specialized services, there should be
a means to resolve the question in the short time span dictated by the patient's
medical needs.
One answer may be to seek expedited judicial relief. While immediate
independent review has certain advantages, it also creates a cumbersome burden
on the courts and results in the substitution of judicial judgment for medical
judgment. Although physicians would say that the courts have not shied away
from substituting their own judgment in the past, the prospect of reviewing a
number of UR decisions on an expedited basis is not something the judicial
system will welcome.
An analogous burden was thrust upon the courts in cases involving
decisions to disconnect life support systems. Many courts resisted having to
make these decisions.89 However, there is a difference between court

APPENDIX A 194
review of decisions to forego treatment of terminally ill patients and court

review of UR decisions. Whereas the former decisions frequently border on the
philosophical, decisions in UR cases often simply resolve disputes between two
professionals' judgment of medical necessity. The latter dispute is similar to
what the courts are asked to do in many routine cases. Nevertheless, the courts
can be expected to resist the added burden on the judicial system of having to
review UR denials, particularly if the review is expedited while the patient's
treatment hangs in the balance.
Another means of resolving the problem of expedited review of UR
decisions is to allocate the burden of payment prospectively to the payer. If care
were deemed medically necessary until the conclusion of the review process,
UR organizations would have a significant financial incentive to make careful,
reasoned decisions and to expedite the review process. If a denial were
subsequently upheld, the patient would incur the debt for the treatment rendered.
Payment For Unnecessary Medical Services

Another issue for the patient is whether the physician can collect payment
from the patient for services that are determined to be medically unnecessary.
Or, phrased differently, the issue is who should bear the risk of payment for
claims that the payer denies because they are not medically necessary under the
payer's rubric? One solution would be to allocate this risk by recognizing a new
implied covenant in the physician-patient relationship obligating the physician
to advise the patient on whether the physician's assessment of the medical
necessity of services conforms with, or is defensible under, the insurance
company's definition of medical necessity.90 As the interpreter of ''medical
necessity,'' the physician is placed at financial risk for the cost of unnecessary
services.
The rationale for adopting an implied covenant of medical necessity is to
create a disincentive for physicians to overutilize medical services. It would
also motivate physicians to appeal questionable UR denials out of self-interest,
thereby enhancing the protection of the patient. The concept is supported by the
disparity in knowledge between the physician and patient regarding medical
necessity.91
Theoretically, both the review organization and the treating physician
should operate under the same standard of medical necessity, which, by
definition, should be determined by the community standard of care. There is a
danger, of course, that a review organization would apply a standard in the
middle to low range of the community standard of care. However, the judicial
inclination to construe insurance policies against the insurer, and the general
deference given to the judgment of the treating physician, tend to minimize this
problem. In addition, the effect of the implied covenant

APPENDIX A 195
on the behavior of physicians could alter the community standard of care, as

physicians become more cautious in recommending treatments for fear that they
will be liable for payment to other providers and will not receive their physician
fees.
As a practical matter, and without an implied covenant of medical
necessity, if the physician, patient, and insurer are all joined in a single action in
a case in which medical necessity is the issue, the patient is virtually certain to
escape any payment obligation. The trier of fact can be expected to place the
burden either on the physician or the payer, but not on the patient.92
When the patient is an HMO member, the question of payment for
unnecessary services does not arise, at least for services from a participating
physician, because the physician is prohibited from direct billing for covered
benefits. The HMO system thus protects the patient by excluding him or her
from payment disputes.
It is conceivable that legislation could mandate the same result in the
indemnity insurance context. Such legislation would prohibit physicians from
collecting for services determined to be medically unnecessary, in effect,
codifying the implied covenant in the physician-patient relationship suggested
above. There would be two problems with such legislation. First, it would be
necessary to clarify the point at which there is a final determination that
services were not medically necessary. This issue in itself could increase the
volume of litigation as physicians scramble to obtain final determinations of
medical necessity in order to support their bills. Second, such legislation could
work to the detriment of patients by effectively prohibiting physicians from
rendering care that may ultimately be beneficial, even though it cannot be
shown to be medically necessary under standards of insurance coverage. A
patient who can afford to pay for treatment should not be precluded from using
his or her own funds to purchase the services he or she wants, even if the
insurer will not pay for them.
Conclusion
If the current trends continue, the placement of liability for UR ultimately
will be sorted out by the courts. However, many of the issues raised in this
paper may not be developed thoroughly for judicial review because of the broad
sweep of ERISA preemption of common law actions. Moreover, in cases in
which the allocation of UR liability arises in an HMO context, the issues are
likely to be resolved by arbitration and never reach the courts because of the
binding arbitration clauses that are frequently found in HMO consumer
documents and physician contracts.
When a UR case does reach the courts, the plaintiff has a range

APPENDIX A 196
of legal theories to test and several potential defendants, including the provider,
the payer, and the utilization review organization, to bring into the lawsuit.
Given the increased use of UR by cost-conscious payers, it is likely that
opportunities for testing the various legal theories will be plentiful. In
anticipation of this uncharted development of the law, the health care industry
would do well to begin addressing the liability issues internally.
At the heart of one UR liability debate is the fundamental question of who
should bear the risk of UR decisions and whether and how that risk can be
allocated without resorting to the judicial system. A secondary, but perhaps
more immediate, question is how to provide for expedited review of UR
decisions, so that bad decisions are reversed before they have an immediate
impact on the patient's medical care.
References
1. See generally Byrnes, "Corporation's Institution of Healthcare Utilization Review,"
Medical Trial Technique Quarterly (Spring 1987), at 478; Carabillo, "The Manageable
Risks of Managed Care," Health Cost Management, Vol. 3, No. 6 (Nov./Dec. 1986), at
1; Eisenberg and Rosoff, "Physician Responsibility for the Cost of Unnecessary Medical
Services,'' New England Journal of Medicine, Vol. 299 (July 13, 1978), at 776; Hershey,
"Fourth-Party Audit Organizations: Practical and Legal Considerations," Law Medicine
& Health Care, Vol. 14, No. 2, at 54; Lanzafame, Provider Liability Under Public Law
98-21: The Medicare Prospective Payment System in Light of Wickline v. State, 34
Buffalo L. Rev. 1011 (1985); Jespersen and Kendall, "Utilization Review: Avoiding
Liability While Controlling Health Costs," HEALTHSPAN, Vol. 4, No. 7, at 3 (July 1987).
2. For a contrary decision, see Van Vactor v. Blue Cross Association, 50 Ill. App. 3d 709,
8 I11. Dec. 400, 365 N.E. 2d 638 (1977), which found no justification for the denial of
benefits solely on the ground that the insurer disagrees with the honest judgment of the
treating physician. The court concluded that decisions of medical necessity are "vested
solely and exclusively in the judgment and discretion of the treating physician." Id. 365
N.E. 2d, at 647.
3. The court also concluded that Blue Shield acted in bad faith when it failed to inform
the insured of his fight to impartial review and arbitration as provided in the policy.
Although this aspect of the case turned on the particular conduct of Blue Shield, which
went beyond a simple failure to inform of appeal fights, the decision emphasized the
duty of the insurer to protect the rights of the insured at least equally with its own. 233
Cal. Rptr., at 84-86. Thus, even if the insurance policy contains clear and conspicuous
language regarding remedial rights, the insurer should take affirmative steps to inform
the insured of his or her rights if a denial of coverage is disputed.
4. The decision proceeds to review the beneficial aspects of the Medi-Cal program and
the Medi-Cal regulations providing for prior authorization for hospitalization. Id., 228
Cal. Rptr., at 671. Although the court may have been citing this material in connection
with the alternative defense of immunity raised by the state, the court states that it
declines to address the immunity defense. See 228 Cal. Rptr. 669, at

APPENDIX A 197
672. The purpose of discussing the Medi-Cal regulations appears to be simply to show
that prior review is an integral part of the payer's system.
5. Several courts have held an exclusion for "experimental" procedures to be inherently
ambiguous, and hence unenforceable. See Johnson v. District 2 Marine Engineers
Beneficial Assoc., 857F. 2d 514 (9th Cir., July 11, 1988); DiDomenico v. Employers
Cooperative Industry Trust, 676 E Supp. 903 (N.D. Ind. 1987).
6. Some policies simply use the term but do not define it. These policies run the risk that
a denial based on medical necessity will not be upheld because the term is not adequately
defined. See, e.g., Dallis v. Aetna Life Insurance Co., 100 F.R.D. 765 (N.D. Ga. 1984);
Zuckerberg v. Blue Cross Blue Shield of Greater New York, 119 Misc. 2d 834, 464
N.Y.S. 2d 678 (1983), rev'd on other grounds. 487 N.Y.S. 2d 595, 108 A.D. 2d 56 (1985).
7. Sarchett, supra, 233 Cal. Rptr., at 78.
8. One plan defines "medically necessary services" as those which are:
" (1) Appropriate for the symptoms and diagnosis or treatment of a
condition, illness or injury.
(2) Provided for the diagnosis, or the direct care and treatment of the
condition, illness or injury.
(3) In accordance with the standards of good medical practice.
(4) Not primarily for the convenience of the Member, or the Member's
physician and surgeon, or the provider.
(5) The most appropriate supply or level of service which can safely be
provided to the Member.
When applied to hospitalization, this further means that the member requires acute care
as a bed patient due to the nature of the services rendered or the member's condition, and
the Member cannot receive safe and adequate care as an outpatient."
9. Hughes v. Blue Cross, 199 Cal. App. 3d 318, 245 Cal. Rptr. 273 (1988), upheld
liability against an insurer which applied a standard of medical necessity which was
more restrictive than that of the medical community. Citing Sarchett, the court
emphasized that the term medically necessary must be construed liberally so that
uncertainties are resolved in favor of coverage. The court noted that by employing a
standard of medical necessity significantly at variance with the medical standards of the
community, the insurer places the insured at risk of incurring unforeseen liability, which
is contrary to the insured's reasonable expectations. "[G]ood faith demands a
construction of medical necessity consistent with community medical standards that will
minimize the patient's uncertainty of coverage in accepting his physician's recommended
treatment." Id. 245 Cal. Rptr., at 279.
10. See Wickline, supra, 228 Cal. Rptr., at 669; Rowland v. Christian, 69 Cal. 2d 108, 70
Cal. Rptr. 97, 443 P. 2d 561 (1968). Rowland lists the following factors as the major
considerations in determining whether a duty exists: "The foreseeability for harm to the
plaintiff, the degree of certainty that the plaintiff suffered injury, the closeness of the
connection between the defendant's conduct and the injury suffered, the moral blame
attached to the defendant's conduct, the policy of preventing future harm, the extent of
the burden to the defendant and consequences to the community of imposing a duty to
exercise care with resulting liability for breach, and the availability, cost and prevalence
of insurance for the risk involved." Id. 69 Cal. 2d, at 112. Although the Wickline decision
states that these factors lead it not to find liability, it appears that Wickline's holding is
based on lack of causation rather than on the nonexistence of a duty. This is evident from
a statement later in Wickline to the effect that third-party payers may be held liable for
medically inappropriate decisions resulting from their cost-containment program. 228
Cal. Rptr., at 670.

APPENDIX A 198
11. UR contracts usually expressly distinguish between the role of the UR organization
in determining the availability of coverage, the roles of the payer in making payment
decisions, and the roles of the provider and patient in deciding whether to proceed with
treatment. One UR contract states that the UR provider "shall not determine a
participant's eligibility for benefits under the Group Contract. Group shall have final and
sole authority for all benefit determinations." Another contract states that "the decision or
determination to obtain or deliver any health care service is always made only by the
[patient] and/or his or her physician, and any decisions made by the [UR organization] ...
or the health benefit insurer... shall relate only to the obligation for payment for any such
service under the terms of the group insurance policy "
12. The patient who requires treatment and is harmed when care that should have been
provided is not provided should recover for the injuries suffered from all those
responsible for the deprivation of such care, including, when appropriate, health care
payers. Third-party payers or health care services can be held legally accountable when
medically inappropriate decisions result from defects in the design or implementation of
cost-containment mechanisms, as, for example, when appeals made on a patient's behalf
for medical or hospital care are arbitrarily ignored or unreasonably disregarded or
overridden. Wickline, supra, 228 Cal. Rptr., at 670-671.
13. Every person or entity is expected to exercise the care that the ordinary, reasonable
person of common skill and prudence would use under the circumstances of the case.
The standard of care is heightened if the person causing the injury enjoys some
specialized skill or knowledge. Professionals such as doctors and lawyers are expected to
use the skill and care common to their professions, not merely that of the "ordinary
person." Prosser and Keeton, The Law of Torts, Section 32, at 185-186 (1984).
14. Wickline, supra, 228 Cal. Rptr., at 666.
15. Wickline's suggestion that the burden is on the treating physician to provide the
reviewer with sufficient information is no': a panacea for every utilization review
liability case. In Hughes v. Blue Cross of Northern California, 199 Cal. App. 3d 318,
245 Cal. Rptr. 273 (1988), the reviewing physician consultant testified, in explaining
why his file lacked complete medical information, that he felt it was the responsibility of
the treating doctor or the hospital to submit any information they felt was important. The
court nevertheless upheld liability for bad faith claims denial, noting that the letters sent
to the treating physician did not explain the medical basis for denial and failed to advise
what information the reviewer already had and what additional information would be
useful. Id. 245 Cal. Rptr., at 280.
16. See generally Jesperson and Kendall, supra note 1, at 7.
17. For a discussion of state regulation, see the section State Regulation.
18. See the sections Liability of Consultants and Employees and Liability of the Treating
Physician infra regarding the vicarious liability of the UR organization for the torts of its
agents.
19. See Hughes v. Blue Cross, 199 Cal. App. 3d 318, 245 Cal. Rptr. 273 (1988) (bad
faith verdict upheld in a case in which the insurer applied a standard of medical necessity
that was more restrictive than the community standard).
20. See generally J. Restuccia, "The Appropriateness of Hospital Use," Health Affairs, at
130 (Summer 1984).
21. See the section The Elusive Concept of Medical Necessity for a discussion of
situations in which criteria are expressly included as part of the plan benefits.
22. For example, a criteria that disallows all fertility services under an HMO plan that
covers all medically necessary physician services may be too sweeping an exclusion.
23. Wickline, supra, 228 Cal. Rptr., at 671.

APPENDIX A 199
24. Id. 228 Cal. Rptr., at 668.

25. Id. 228 Cal. Rptr., at 669.
26. The payer's liability for tort damages resulting from the negligent act of an
independent review organization may be more complicated, depending on whether the
UR entity is viewed as an independent contractor or an agent or employee of the payor
(see Liability of the Treating Physician).
27. When a contractual duty is delegated, the obligor remains liable to the obligee, unless
the delegate assumes the obligations and the assumption is accepted by the obligee in
substitution for the original obligor. See Calamari and Perillo, The Law of Contracts,
Sections 277-280 (1970) and Restatement (2d) of Contracts, Section 150(3). There also
would be a significant question of whether the duty to perform UR functions is a
delegable duty under the law. See Calamari and Perillo, supra, Section 278; Hughes v.
Blue Cross, 199 Cal. App. 3d 318, 245 Cal. Rptr. 273, (1988) (an insurance company
generally cannot delegate its responsibilities to the insured).
28. See Calamari and Perillo, Restatement (2d) of Contracts, Section 302. The court will
look to the surrounding circumstances to determine whether the patient is an appropriate
third-party beneficiary. In cases in which the benefit literature describes the UR program
as a beneficial service for the insureds, for example, in helping to avoid unnecessary
hospitalization, the employee has a better chance of establishing a third-party beneficiary
relationship.
29. This issue is likely not to be of much practical importance, since the patient can
probably assert a negligence claim directly against the review organization. Whatever
benefits the patient can obtain from a contract action, such as an action for breach of the
implied covenant of good faith and fair dealing, ordinarily can be obtained in an action
directly against the payor.
30. Restatement (2d) of Contracts, Section 351.
31. See generally Pilot Life Insurance Co. v. Dedeaux, 107 S. Ct. 1549 (1987); Fletcher
v. Western National Life Ins. Co., 10 Cal. 3d 376, 401, 89 Cal. Rptr. 78, 93 (1970).
32. See, e.g., Egan v. Mutual of Omaha, 24 Cal. 3d 809, 157 Cal. Rptr. 482, 598 P. 2d
452 (1979) (bad faith failure to properly investigate claim); Taylor v. Prudential Ins. Co.
of America, 775 E 2d 1457 (11th Cir. 1985) (reversing summary judgment for the insurer
on the issue of bad faith where the insurer relied on a Medicare determination of no
medical necessity without making its own investigation); Hughes v. Blue Cross, 199 Cal.
App. 3d 318, 245 Cal. Rptr. 273 (1988) (bad faith verdict upheld where insurer denied
claims without reviewing all relevant medical records, applied a standard of medical
necessity that was more restrictive than a community standard, failed to explain medical
grounds for denial of coverage, and failed to advise treating physician of what additional
information would be useful for decision); Mordecai v. Blue Cross/Blue Shield of
Alabama, 474 So. 2d 95 (Ala. 1985) road faith allegations arising from failure to
consider portions of nurses' notes and to consult with treating nurses and physicians);
AEtna Life Insurance v. LaVov, 470 So. 2d 1060 (Ala. 1984) (bad faith claim upheld
where insurer misrepresented the extent of its medical review in denying the claim).
33. See generally Sarchett v. Blue Shield, supra; Davis v. Blue Cross of Northern
California, 25 Cal. 3d 418, 158 Cal. Rptr. 828 (1979) (bad faith upheld in a case in
which the insurer failed to inform the insured of rights to appeal an arbitration).
34. In such a case, the payer could seek indemnity from the reviewer, unless prohibited
by some provision of the contract between the payer and reviewer. See Hughes v. Blue
Cross, 199 Cal. App. 3d 318, 245 Cal. Rptr. 273 (1988).
35. Gruenberg v. AEtna, 9 Cal. 3d 566 (1973) (demurrer sustained as to bad faith
liability of insurer's independent contractor adjusters and attorneys who were not party to
the

APPENDIX A 200
insurance contract); Iversen v. Sup. Ct., 57 Cal. App. 3d 168, 127 Cal. Rptr. 49 (1976)
(reversal of judgment against claims supervisor and claims examiner, both found to be
independent contractors, because they were not party to insurance contract); Hale v.
Farmers Insurance, 42 Cal. App. 3d 681, 117 Cal. Rptr. 146 (1974) (insurer not liable
for employee's bad faith handling of claim where insurer has not ratified employee's
acts). See also Reiderscheid v. Comorecare. Inc., 667 P. 2d 766 (Colo. Ct. App. 1983)
("The test of bad faith failure to exercise due care in discharge of a contractual duty and
the granting of damages for mental anguish caused by a willful and wanton breach of
contract are grounded in basic common law, and not solely in the area of insurance
law."); Taylor v. Prudential Ins. Co. of America, 775 F. 2d 1457 (11th Cir. 1985)
(upholding cause of action against insurer for bad faith and emphasizing the weight to be
given to treating physician's opinion); Linthicum v. Nationwide Ins. Co., 723 P. 2d 675
(Sup. Ct. Ariz. 1986) (punitive damages may be recoverable against insurer for failure to
disclose medical basis for denial or failure to seek direct input from treating physician).
36. 93 Cal. App. 3d 642, 155 Cal. Rptr. 843 (1979).
37. See Sprague v. Equifax Inc., 166 Cal. App. 3d 1012, 213 Cal. Rptr. 69 (1985);
Younan v. Equifax Inc., 111 Cal. App. 3d 498, 169 Cal. Rptr. 478 (1980).
38. Fletcher v. Western National Life Ins. Co., 10 Cal. App. 3d 376, 89 Cal. Rptr. 78
(1970).
39. See Schlauch v. Hartford Acc. & Indem. Co., 146 Cal. App. 3d 926, 936, 194 Cal.
Rptr. 658, 665 (1983), quoting Ricard v. Pacific Indem Co., 132 Cal. App. 3d 886, 895,
183 Cal. Rptr. 502, 507 (1982).
40. See Pulvers v. Kaiser Foundation Health Plan, 99 Cal. App. 3d 560, 160 Cal. Rptr.
392 (1979) (advertisement of "high standards" of medical service held not to warrant a
specific result, but was generalized "puffing," to the effect that physicians would
exercise good judgment in care).
41. The Restatement (2d) of Torts, Section 402A, states as follows: "... One who sells any
product in a defective condition unreasonably dangerous to the user or consumer or to
his property is subject to liability for physical harm thereby caused to the ultimate user or
consumer, or to his property, if (a) the seller is engaged in the business of selling such a
product, and (b) it is expected to and does reach the user or consumer without substantial
change in the condition in which it is sold." In California, the plaintiff need not show that
the product was "unreasonably dangerous," but only that the product was "defective."
See Cronin v. J.B.E. Olson, 8 Cal. 3d 121, 104 Cal. Rptr. 433 (1972).
42. See Restatement (2d) of Torts, Section 402A, Comment f.
43. See generally CCH Products Liability Reporter, Section 4235.
44. See AEtna Casualty and Surety Co. v. Jeppesen & Co., 642 E 2d 339 (9th Cir. 1981);
Brocklesby v. United States, 767 F. 2d 1288 (9th Cir. 1985), cert. denied sub nom, 106 s.
Ct. 882 (1986).
45. See Slaughter v. Friedman, 32 Cal. 3d 148, 185 Cal. Rptr. 244, 649 P. 2d 996 (1982)
(defamation; interference with prospective economic advantage); Teale v. American
Manufacturers Mutual Ins. Co., 687 S.W. 2d 218 (Mo. Ct. App. 1985) (tortious
interference); Moore & Assoc. v. Metropolitan Life Ins. Co., 604 S.W. 2d 487 (Tex. Civ.
App. 1980) (claim stated for tortious interference with doctor-patient relationship by
association of anesthesiologists against group medical insurer for insurer's letters to
former patients advising that claims would not be paid in full because association's
charges were excessive).
46. Supra.
47. See, e.g., California Civil Code, Section 47(3).

APPENDIX A 201
48. Slaughter, supra, 185 Cal. Rptr., at 248-249. The court pointed out that defendants
were only required "to inform dental patients of the basis for rejection of their claims;
they were not required additionally to defame plaintiff with accusations regarding his
dental practices."
49. See Chicago Board of Trade v. U.S., 246 U.S. 231 (1981); Dos Santos v. Columbus-
Cuneo-Cabrini Medical Center, 684 F. 2d 1346 (7th Cir. 1982) (under rule of reason
analysis, the hospital was permitted to grant exclusive privileges where policy is
grounded in ensuring quality patient care and necessary hospital services).
50. See generally Northwest Wholesale Stationers. Inc. v. Pacific Stationery Printing,
105 S. Ct. 2613 (1985) (unless an organization possesses market power or controls
access to an element essential for competition, expulsion for failure to follow reasonable
rules is not per se illegal); see generally remarks of Charles E Rule, Assistant Attorney
General, U.S. Department of Justice, March 11, 1988.
51. 15 U.S.C., Section 1.
52. Cf. Copperweld Corp. v. Independent Tube Corp., 467 U.S. 752, 104 S. Ct. 2731, 81
L. Ed. 2d 628 (1984).
53. Plaintiffs often attempt to rely on the doctrine of "conscious parallelism." Conscious
parallelism is easy to allege but exceedingly difficult to prove. It requires proof that the
parallel conduct was against the defendant's self-interest and was not based on good faith
business judgment. See Supermarket of Homes v. San Fernando Valley Board of
Realtors, 786 E 2d 1400 (9th Cir. 1986); Proctor v. State Farm Mutual Ins. Co.. 675 F.
2d 308 (D.C. Cir.), cert. denied, 459 U.S. 839 (1982).
54. See, e.g., California Civil Code, Sections 43.7, 43.8; California Health & Safety
Code, Section 1370.
55. For example, Norcal Mutual Insurance Company's malpractice policy only covers
claims alleging negligence in "direct patient treatment" or involving professional
committee activities (which are limited to hospital staff committees or American
Medical Association or medical society committees). Coverage is specifically excluded
for "the performance of administrative duties, which are not direct patient treatment, as a
medical director." If liability is found against an independent physician reviewer in a
Wickline-type case, it is not clear whether the coverage for "direct patient treatment"
applies. The physician providing consultant services should seek clarification from the
carrier.
56. See, e.g., California Corp. Code, Section 317(d).57.
57. See, e.g., California Corp. Code, Section 317(b).
58. Md. Health Code Ann., Sections 19-1301 et seq.
59. Minn. Stat. 1988, Section 72A.20(4a).
60. Maine Ins. Code Title 24-A, Section 2679. Maine also has legislation pending that, if
enacted, would impose certain criteria on independent review organizations, such as
requiring prospective UR decisions to be made within a set time.
61. La. Revised Stat. 22:657(D).
62. 107 Sup. Ct. 1549 (1987).
63. See 29 U.S.C., Section 1144. See generally Helvestine, "ERISA Preempts Insurance
Bad Faith Actions," HEALTHSPAN, Vol. 4, No. 10, at 8 (Dec. 1987).
64. Significant exceptions to ERISA coverage are government employee plans
(including local government plans, such as school districts), certain church plans, and
certain statutorily required workers' compensation, unemployment, and disability laws.
UR cases will not be preempted when they arise under these kind of plans. See 29
U.S.C., Section 1003(b).
65. One major area of uncertainty has been whether ERISA preempts causes of action
based on state statutes governing unfair insurance practices, such as California

APPENDIX A 202
Insurance Code, Section 790.03(h). In Kanne v. Connecticut General Life Ins. Co., 857
F. 2d 96, No. 85-5641, 5642 (9th Cir., Oct. 4, 1988), the Ninth Circuit held that ERISA
preempts such claims. This issue is of considerably less importance following the
California Supreme Court's decision in Moradi-Shalal v. Fireman's Fund Ins. Co., 46
Cal. 3d 287, 250 Cal. Rptr. 116 (1988), which overturned 10 years of precedent and held
that no private right of action exists under Section 790.03(h). At this time, only Montana
and West Virginia continue to recognize private actions under unfair insurance practices
statutes. See Moradi-Shalal, supra, 250 Cal. Rptr., at 121 and note 6.
66. 29 U.S.C., Section 1002(21)(A); see Stanton v. Shearson Lehman/American Express,
Inc., 631 E Supp. 100, 102 (N.D. Ga. 1986).
67. See Nieto v. Ecker, 945 F. 2d 868 (9th Cir. 1988) (rejecting a theory of liability under
ERISA for aiding and abetting a fiduciary); So. Cal. Meat Cutters Unions v. Investors
Research, 687 F. Supp. 506 (C.D. Cal. 1988) (ERISA only applies to those defendants
against whom ERISA provides a statutory right of action); Munoz v. Prudential Ins. Co.
of America, 633 F. Supp. 564 (D. Colo. 1986) (same findings as previous case).
68. See 29 U.S.C., Section 1132(a)(1)(B); Massachusetts Mut. Life Ins. Co. v. Russell,
473 U.S. 134, 105 S. Ct. 3085 (1985) (emotional distress damages prohibited); Sokol v.
Bernstein, 803 F. 2d 532 (9th Cir. 1986) (same findings as previous case).
69. See 29 U.S.C., Section 1132 (a)(3).
70. Id. Section 1132 (a)(3).
71. See, e.g., Blau v. Del Monte Corp., 748 F. 2d 1348 (9th Cir. 1984).
72. 18 U.S.C., Section 1961 et seq.
73. See Marcial v. Coronet Ins. Co., No. 87C 3072 (N.D. Ill., 1987 WL 19532); Unocal
Corp. v. Superior Court (Harbor Ins. Corp.), 198 Cal. App. 3d 1245, 244 Cal. Rptr. 540
(2d Dist. 1988), decertified (June 2, 1988).
74. Under the theory of respondeat superior, an employer may be held liable for the torts
of its agents or employees acting within the scope of their employment. See Witkin,
Agency and Employment, Sections 113 et seq.; California Civil Code, Section 2338.
75. See Witkin, supra, Agency and Employment, Section 61, at 67.
76. See Hughes v. Blue Cross, 199 Cal. App. 3d 318, 245 Cal. Rptr. 273 (1988).
77. See note 24, supra.
78. See, e.g., Restatement (2d) of Tons, Section 411; Elam v. College Park Hospital, 132
Cal. App. 3d 332, 183 Cal. Rptr. 156 (1982) (hospital negligently allowed podiatrist to
remain on staff despite malpractice complaints). See Burch 122 Mich. App. 798, 333
N.W. 2d 140 (Mich. 1983); Kendall v. Gore Properties, 236 F. 2d 673 (D.C. Cir. 1956);
Giles v. Shell Oil Corp., 487 A. 2d 610 (D.C. 1985) (employer responsible for harm
caused by employee in a case in which the employer negligently failed to screen the
employee's background).
79. The general rule that parties are not liable for the torts of their independent
contractors is fiddled with exceptions. When the plaintiff can show that one party retains
control over the enterprise, benefits from it, selects the independent contractor, and is
free to require indemnity and insurance from the contractor, that party may be found
vicariously liable for the torts of its independent contractor. Also, when the plaintiff can
demonstrate a "special relationship" giving rise to an affirmative duty of care owed by
the defendant, vicarious liability may result. See Witkin, supra, Tons, Section 997. An
insurance company is likely to be found to have a special relationship with its insureds,
thus increasing the likelihood of liability for the acts of independent contractor UR
organizations.
80. The fact that the contract between the payer and reviewer specifies independent
contractor status is not dispositive. The reviewer may be considered an independent

APPENDIX A 203
contractor for purposes of its relationship with the payor, and yet be considered an agent
of the payor in matters involving the patient. Cf. Arthur v. SL Peter's Hospital, 169 N.J.
Super. 575, 405 A. 2d 443 (1979) (physician may be considered an independent
contractor in his relations with the hospital, but be deemed an employee of the hospital in
his relations with the patient). The principal factor in distinguishing an independent
contractor from an employee is the freedom from control by the employer over the
details of the work. See Witkin, supra, Agency and Employment, Sections 12, 14, at
28-31; Prosser and Keaton, The Law of Torts (5th ed., 1984) for a discussion of the
doctrine generally and a suggestion that it is disfavored.
81. See Restatement (2d) of Agency, Sections 8, 159 (apparent authority of agent);
California Civil Code, Section 2300, Quintal v. Laurel Grove Hospital, 62 Cal. 2d 154,
41 Cal. Rptr. 577, 397 P. 2d 161 (1964) (whether the physician was ostensible agent of
the hospital is a jury question); Mduba v. Benedictine Hospital, 384 N.Y.S. 2d 527, 52
App. Div. 2d 450 (1976) (emergency room physicians may be named agents of the
hospital despite independent contractor language in their contracts).
82. Many of the same dilemmas discussed in this section apply to hospitals as well.
Indeed, in a concurrent review situation like Wickline, the hospital has the largest
financial stake in the patient's discharge because it is the hospital bills that would remain
unpaid. The rights and responsibilities of a hospital are similar to those of a physician
and are not treated separately for the purposes of this paper.
83. If the physician is named in a lawsuit, he or she may cross-complain against the
review entity for indemnity or contribution. Likewise, a review entity named in a
Wickline type of case will consider cross-complaining against the treating physician.
84. See, e.g., Goldman v. Ambro, 512 N.Y.S. 2d 636 (1987); Harper v. Baptist Medical
Center-Princeton, 341 So. 2d 133 (Ala. 1976). In Harper, the Alabama Supreme Court
held that a doctor and hospital who rendered emergency treatment to the plaintiff, but
refused to accept him as a patient because he did not have insurance, were not liable for
subsequent injuries because the plaintiff had not been accepted as a patient. Until a
professional doctor-patient relationship is established, the physician's only duty is to
provide emergency care; there is no duty to accept a patient under other circumstances.
85. See generally, Mains, "Medical Abandonment," Medical Trial Technique Quarterly,
at 306 (1985).
86. See, e.g., Meiselman v. Crown Heights Hospital, 285 N.Y. 389, 34 N.E. 2d 367, 217
N.Y.S. 2d 12 (1941) (liability against hospital where the patient was discharged
prematurely because he was unable to pay for further care). See generally Lanzafame,
supra, note 1, at 1023-1030.
87. Informed consent requires disclosure of all information that the patient would
consider material in deciding whether to undergo the treatment. See Canterbury v.
Spence, 464 F. 2d 772 (D.C. Cir. 1972). Presumably, economic consequences would be a
material consideration for the patient.
88. Physicians should be mindful, however, of the UR organization's legitimate concerns
regarding the disclosure of its review criteria and systems, which are usually considered
protectable trade secrets.
89. See In Re Quinlan, 70 N.J. 10, 355 A. 2d 647, 649, cert. denied, 429 U.S. 922 (1976)
(review would be "inappropriate," not only because that would be a gratuitous
encroachment upon the medical profession's field of competence, but because it would
be impossibly cumbersome").
90. For a thoughtful treatment of this issue, see Eisenberg and Rosoff, supra, note 1.
91. The implied covenant of medical necessity should not, however, supplant the patient
as the final decision maker in matters of medical care. Rather, it obligates the

APPENDIX A 204
physician to add a new economic dimension to the physician's determination of medical

necessity, which must be communicated to the patient.
92. In one case, the court apparently refused to relieve the patient of payment
responsibility, even though the insurer denied coverage because the treatment was not
medically necessary. Albert Einstein Medical Center v. Lipoff, No. 3872X (Ct. of
Common Pleas, Phila., Apr. 23, 1973), described in Eisenberg and Rosoff, supra, note 1.
In that case, the patient sought to hold her physician contractually liable for the hospital
bill. The court denied her claim, reasoning that her claim sounded in tort and that she
could recover only by proving that the doctor's treatment was medically unsound. That it
was economically unsound was irrelevant to a tort claim.

APPENDIX B 205
Appendix B
Utilization Management and Quality
Assurance in Health Maintenance
Organizations: an Operational Assessment
Joan B. Trauner and Sibyl Tilson*
In the 1980s, health care payers in the public and private sectors have
relied increasingly on utilization management to help control health care costs.
As a result, the autonomy of health care providershospitals and professionals
has been curtailed as payers have applied predetermined criteria to judge the
appropriateness of care. The medical community, in turn, has expressed
concerns about the design of the clinical standards or algorithms governing
utilization review decisions by health insurers, Blue Cross and Blue Shield
plans, third-party administrators, and health maintenance organizations
(HMOs). In particular, as payers seek to control costs more aggressively,
physician organizations are seeking to review the criteria used for denial of
services under second-opinion and prior authorization programs.1
At the same time, federal and state legislators, health policy analysts, and
consumer groups are concerned about the impact of physician incentive plans
on quality of care and accessibility to services in prepaid delivery systems. This
concern was prompted, in part, by concerns about the quality of care delivered
to Medicare beneficiaries enrolled in International Medical Centers Inc. (IMC),
a for-profit HMO located in south Florida.2 The well-publicized IMC scandals
raised a number of public policy issues, including enrollee understanding of
benefit restrictions in HMOs,3 ineffective
* The authors are members of the accounting firm Coopers & Lybrand, San Francisco,
Calif., and Washington, D.C.

APPENDIX B 206
quality assurance monitoring in HMOs (both internally by HMO staff and

externally by regulatory agencies), and the impact of financial incentives on
ordering of services and specialty referrals by HMO physicians.4 The intent of
this paper is to describe the range of utilization management and quality
assurance strategies that are used by HMOs and then to evaluate their actual
implementation. The paper begins with a brief discussion of the methodology
used in preparing this paper. The second section describes the organizational
structure of existing HMOs, and the third discusses market, structural, and
operational factors affecting HMO performance. The fourth section discusses
the approaches to utilization management and quality assurance, and the fifth
section evaluates the performance of utilization management and quality
assurance programs in HMOs. In the sixth section, there is a review of existing
research on physician risk incentives in HMOs, while the seventh section
addresses some additional policy and research issues. The final section of this
paper contains five condensed case histories.
Except for the discussions on the organizational structure of licensed
HMOs, the term HMO is generically in this paper to cover all closed systems in
which physicians are partly or fully capitated for delivery of care and where
enrollees may receive services only from contracting providers.5 All state-
licensed prepaid medical plans are included in this definition, regardless of
whether they are federally qualified and offer a full range of benefits.
Methodology
This paper reflects the authors' experience as health care consultants for an
international accounting firm. During the past 3 years, we have had the
opportunity to analyze the health benefits offered by numerous employers in
both the public and private sectors and to help design their managed care
programs. (By managed care, we refer to any program that channels patients to
a specific set of health care providers.) During the same time, we and our
associates have conducted operational reviews of over 20 HMOs, ranging from
local to regional and multistate plans. In general, the HMO reviews have
involved an analysis of financial, actuarial, enrollment, and utilization trends in
the context of benefit design, premium pricing, marketing practices, claims
processing procedures, contractual arrangements with providers, utilization
review, and management reporting procedures.
This managed care and HMO consulting experience provides the
background for this paper. To supplement our experience, we reviewed the
health services research literature for HMO-related studies. In describing

APPENDIX B 207
our personal experience, we do not purport to review a statistically

representative sample of HMOs. In fact, because many of the plans that we
studied are in transition, the paper may magnify some of the problems facing
the HMO industry.6 The crucial advantage of an experiential analysis, however,
is that it offers an internal view of HMO operations at all organizational levels
that is not available through standard survey instruments or through a review of
documents filed with regulatory agencies.7
In our HMO discussions, we do not refer to any plans by name, nor do we
provide any information on exact plan size, geographic location, plan age, or
precise sponsorship in order to preserve client confidentiality. The brief case
histories for five plans that appear at the end of this paper do not include any
plan identifiers, but do contain actual descriptions of the utilization
management programs in place at the time that the operational reviews were
conducted. We present these studies to show the range of operational problems
that can undermine the structural design of utilization management and quality
assurance programs.
Hmo Organizational Structure

According to the current InterStudy classification of HMOs, there are four
types: staff, group, network, and individual practice association (IPA) model
plans.8 Staff model HMOs deliver health services through physicians who are
under salary to the plan. Group model HMOs contract with one independent
multispecialty group practice, whereas network model HMOs contract with two
or more multispecialty groups. IPAs contract with physicians in private
practice, either as individuals, through their group practices, or through separate
physician associations.
Under the mandating process established by the HMO Act of 1973, an
employer is required to offer only one group or staff model HMO and one IPA
in a given service area. (The network classification did not exist under the
original HMO Act.) In some geographic areas, HMOs have used a model
designation that allows them to take advantage of the mandating process, rather
than one that strictly corresponds to their underlying structure. For example,
plans that contract almost exclusively with multispecialty groups have sought
IPA designations to compete against long-established group or staff model
plans. Therefore, any analysis of HMOs requires an assessment of the
underlying contractual arrangement with physicians, rather than the plans own
self-designation.9
More recently, ICF, Inc. has categorized HMOs as having two-tier or three-
tier organizational structures.10 In the two-tier structure (Figure B-1), the HMO
contracts directly with individual physicians; in the three-tier structure
(Figure B-2), the HMO makes capitation or fee-for-service payments to the IPA
or group, which, in turn, contracts with individual

APPENDIX B 208
Figure B-1 Organizational structures within HMOs: two-tier structures.
Figure B-2 Organizational structures within HMOs: three-tier structures.

physicians. ICF originally included only staff model plans and direct
contract IPAs in the two-tier classification. In Figure B-1, we have added two
additional categories: HMOs owned by group practices and HMOs in which
physicians contract directly with the HMO but are arbitrarily grouped into risk-
sharing pools or ''pods.''
As described in the literature, a two-tier structure may facilitate contracting
in areas where physicians are geographically dispersed (that is, suburban or
rural areas) or where physicians have not developed their own

APPENDIX B 209
IPAs. According to the literature, some HMOs elect to use a two-tier

structure purposely to forestall physicians from creating IPAs, which, in turn,
can function as bargaining units to provide physicians with higher fees.11
Lately, we have come across a new two-tier contractual arrangement
involving those HMOs that contract directly with individual physicians and then
create arbitrary physician risk pools, or pods. In these arrangements, the
financial risk is shared with other providers with whom the physician may or
may not have direct contact in daily practice.12 Depending upon the local
market, pods may sometimes be organized by group practice or hospital
affiliation; in other areas, physicians may be assigned to a pod solely on the
basis of practice location.
In this paper, we use the two-and three-tier structure as the basis for much
of our analysis. This approach allows us to focus on where responsibility for
utilization management decisions actually rests. For example, in two-tier IPA
arrangement's, most decision-making resides at the HMO level, whereas in
three-tier arrangements those entities that contract with the HMO (for example,
IPAs and multispecialty groups) may take over many of these responsibilities.
In determining how plans are classified, we ignore the self-designated HMO
plan type and instead examine the underlying contractual relationship that exists
with participating physicians. In this way, we can describe more precisely
utilization management activities in those plans that have contractual
arrangements with varying types of provider entities (for example, plans that
have group-, staff-, and hospital-based IPA arrangements).
Market, Structural, and Operational Factors Affecting Hmo

Performance
We believe that any review of HMO performance that focuses exclusively
on structural measures will have serious limitations. The HMO market is in a
state of flux. Not only are many plans reporting financial losses, but they are
having to contend with employer demands regarding the provision of data and
experience rating and with provider concerns about reimbursement rates and
design of utilization review programs. In this section, we briefly describe the
relationship of the health care marketplace to HMO structure and operations.
The intent of the analysis is to show that plan performance, as measured by
financial, utilization, and quality indicators, is a function of multiple variables,
many of which are beyond the control of HMO management.

APPENDIX B 210
The Health Care Marketplace

An HMO both reacts to and helps shape the health care market in which it
operates. For example, the regulatory environment establishes parameters for
benefit design, premium pricing, and marketing practices. The employment
market determines eligibility for health benefits, level of employer
contributions, and design and pricing of indemnity benefits. Demographic and
geographic factors, such as the size of Medicare and Medicaid populations, the
location of local industry, and commuting patterns, affect product development
and the boundaries of HMO service areas.
The use of a two-or three-tier contracting structure by an HMO is
dependent upon existing alternatives to individual physician contracting, such
as the presence of multispecialty group practices or hospital-sponsored IPAs,
and the prior experience of local providers in working with other HMOs or
managed care programs. Negotiation of reimbursement rates, capitation
payments, and risk-sharing arrangements are dependent upon (1) hospital and
physician supply, (2) the presence of competing HMOs and the design of their
payment and risk arrangements, and (3) utilization and cost trends within the
local market. In return, as an HMO becomes established, it will have an impact
on the design and pricing of health benefits and health services in the local
market, as well as on the structure of hospital and physician relationships
(including the emergence of new IPAs, group practices, and joint ventures).
Hmo Structure
HMO structure, in part, reflects plan sponsorship, consumer orientation
date of entry into the market, and short-and long-term financial goals. The
adoption of nonprofit or for-profit status may reflect the history of the
sponsoring entity, the availability of capital, and the marketability of the
program to investors or joint venture partners. Management's perspective on
provider behavior, plus the existence of other HMOs in the local market, help to
determine the operating model (staff, group, network, IPA, or hybrid), the
approach to contracting (direct contract versus three-tier arrangement), and the
extent of any risk-sharing arrangements.
Except in markets with a large oversupply of physicians and hospitals,
there generally is some give and take in contract negotiations between an HMO
and the provider community. While quality of care is always a concern in these
negotiations, actual discussions usually end up focusing on economic
considerations. As noted by Gnessin,13 there is an inherent conflict between the
goals of any HMO's management and those of its participating physicians.
Gnessin defines six goals for HMOs: (1) profit

APPENDIX B 211
maximization; (2) capital appreciation; (3) enrollment growth; (4) shift of

financial risk to third parties; (5) maximum utilization control with minimum
malpractice exposure; and (6) control over all components of HMO operations,
including the actual health care delivery system. For physicians, Gnessin
defines five goals: (1) maximum autonomy in the practice of medicine; (2)
minimal loss of existing patient base or patient volume; (3) expansion of patient
base and volume; (4) minimal financial risk; and (5) increased income from
new revenue sources (other than the direct provision of health services).
It is the tension between these two perspectives that establishes the setting
for the negotiation and implementation of financial incentives. According to
Gnessin's "fairness doctrine," at any given time either HMO management or
HMO providers may be in a superior bargaining position. However, when the
dominant party imposes its will unilaterally on the other, this strategy usually
proves successful only in the short term; in the long term the imbalance in
power will result in acrimony and disintegration of the relationship. In other
words, a plan that transfers risk to providers without reimbursing them
adequately or providing them with data to manage their practices runs the risk
of losing physician participation in the long run. Conversely, a plan that accedes
completely to physician demands and pays them market rates without changing
practice patterns may price itself out of the marketplace or require the plan
sponsors to absorb large operating losses.
Hmo Operations
An HMO cannot transfer risk fairly to providers and deliver quality care in
a cost-effective manner without (1) effective management information systems
for tracking enrollment, premiums, and utilization and for providing usable and
timely reports to management and participating providers; (2) efficient claims
processing procedures to ensure prompt payment to providers and the accurate
capture of procedural and diagnostic data; (3) accurate actuarial and
underwriting procedures for reliable projection of utilization and delivery costs
by provider class (hospital, primary care provider, specialist, and other ancillary
providers); (4) representation of professional staff and/or a medical director at
the HMO's upper management level; and (5) adequate professional staffing to
support a utilization management and quality assurance program.
In Figures B-1 and B-2, we described two-and three-tier physician
contracting arrangements, as first defined by ICF. There is considerable
variation within these two-and three-tier arrangements, however, depending
upon how much control any HMO wishes to retain for itself or to

APPENDIX B 212
delegate to contracting providers. Broadly speaking, there are three issues that
influence the HMO's decision-making process: (1) how physician groups and
IPAs are organized in the local market in terms of the cost-effectiveness of their
delivery networks, the design of their data systems, and their management
capabilities; (2) how institutional and professional providers are reimbursed and
whether the HMO wishes to retain control over the claims payment process; and
(3) whether the HMOs believe that the physicians can discipline themselves
effectively through the utilization review and peer review process.
As shown in Table B-1, which describes operational responsibilities, there
are two control models for two-tier organizations and three control models for
three-tier organizations. Responsibilities for utilization management and
financial operations may rest almost entirely with the HMOs (Model 1), may be
shared between the HMOs and providers (Model 2), or may be assumed largely
by providers (Model 3). In two-tier arrangements, in which the physicians either
are not independent contractors (for example, staff model, group-owned, or
affiliated HMOs) or are fee-for-service providers contracting directly with the
HMO, an arrangement like that in Model 3 is not feasible. Moreover, any
sharing of services with providers is more limited in two-tier plans than it is in
three-tier plans. Typically, sharing of services involves those utilization
management and claims payment activities that revolve around the gatekeeper
role of contracting primary care physicians (for example, preadmission
authorizations and specialty referrals). In pod-type situations in which primary
care physicians are capitated, the physicians may review claims from specialty
providers before authorizing them for payment. In rare cases, when providers
are individually capitated and have direct control over their own specialty
funds, they may actually pay for specialty claims.
In three-tier organizations, HMOs are usually not responsible for physician
selection and peer review, as these responsibilities rest with contracting IPAs or
groups. In Model 1, the IPA or group may share limited utilization review
functions with the HMO. For example, an IPA may receive utilization and
financial reports from the HMO's claims system; the IPA, in turn, uses these
reports as the basis for remedial action against errant providers. In Model 1, all
physician payments and all risk incentives are calculated by the HMO.
In Model 2, the HMO creates a medical (physician) fund, which is then
handled by the IPA or groups; this fund may represent a capitated payment or a
budget allocation based upon anticipated fee-for-service charges. The IPA or
group then processes and pays claims for services rendered by its member
physicians, while the HMO continues to pay for hospital services and
emergency and out-of-area claims. The IPA or group provides feedback on
utilization patterns to contracting providers and provides encounter data

APPENDIX B 213
TABLE B-1 Operational Responsibilities in Two-and Three-Tier Organizational

Structures by Control Model
Two-Tier Organizations (with Pods)
MODEL 1 MODEL 2
HMO MD HMO MD
Marketing X X
Enrollment/ X X
underwriting
Utilization review X X
Claims payment X X X
Encounter data X
Data analysis X X
Provider feedback X X
Peer review X X
Quality assurance X X
Three-Tier Organizations
MODEL 1 MODEL 2 MODEL 3
HMO IPA/Grp HMO IPA/Grp HMO IPA/Grp
Marketing X X X
Enrollment/ X X X
underwriting
Utilization review X X X X X
Claims payment X X X X X
Encounter data X X
Data analysis X X X X
Provider feedback X X X X
Peer review X X X
Quality assurance X X X X X X
NOTE: MD indicates physician; Grp indicates group practice.
on medical services to the HMO itself. In Model 2, the HMO may also be
responsible for utilization review for services covered by the medical fund (for
example, specialty referrals and diagnostic testing). In Model 3, the HMO
largely assumes the role of a broker; it markets benefits to employers, enrolls
members, and pays claims for noncapitated services. Otherwise, the IPA or
group is given complete responsibility for all utilization review and data analysis.
In the case studies, we have provided examples of one two-tier, Model 1
plan (Case 5), three three-tier, Model 1 plans (Case Studies 1, 3 and 4), and one
three-tier, Model 3 plan (Case Study 2).

APPENDIX B 214
Approaches to Utilization Management and Quality

Assurance
As previously noted, responsibility for specific aspects of an HMO's
operations vary by control model. Understanding where the locus of control
rests is particularly important when evaluating those cost-containment
approaches that affect patient care decisions. For this analysis, we have divided
utilization management into its underwriting benefits, delivery of health
services, and quality assurance components.
Underwriting Benefits
Almost always, benefit design and development of underwriting guidelines
remain the responsibility of HMO management, not providers. (An exception
occurs when providers require the use of copayments for office-based care to
control unnecessary demand for services.) These functions can be described as
follows.
Enrollment criteria are underwriting guidelines by industry or service class
and by group size. Premium pricing and marketing practices also affect
enrollment patterns and are included in this subcategory.
There has been a growing body of literature on enrollment patterns in
HMOs to study the issue of selection bias. These analyses have been prompted,
in part, by employer concerns that HMOs have attracted a disproportionate
share of young and presumably healthy employees. In an analysis of 21 studies
on self-selection, Wilensky and Rossiter concluded that most of the recent
studies have shown HMOs to be enrolling a lower risk population. However,
only rarely have any studies examined the impact of enrollee continuity on
utilization rates.15
Benefit design relates to (1) the extent of coverage for specific services,
drugs, and supplies; (2) any required copayments, coinsurance and/or
deductibles; (3) level of services offered (for example, for mental health, the
availability of individual or group therapy); (4) number of visits per condition;
(5) out-of-pocket maximums; and (6) lifetime dollar benefits. Federally
qualified HMOs have historically offered a more comprehensive set of benefits
than have nonqualified prepaid plans or exclusive provider arrangements that
are indemnity-based. With the recent growth of non-qualified plans and the
move away from community rating and defined benefit packages to experience
rating and negotiated benefits, there has been increased emphasis on benefit
design as a cost-containment mechanism within HMOs.

APPENDIX B 215
Delivery of Health Services

The delivery component of utilization management involves seven
subcategories, as follows.
Utilization review is usually administered under the auspices of a medical
director and includes one or more of the following services; (1) preadmission
review; (2) concurrent hospital review; (3) second surgical opinions; (4)
specialty and/or out-of-network referral authorizations; (5) high-cost case
management; (6) retrospective review of services; (7) bill audit; and (8)
remedial action for providers suspected of fraud or noncompliance with
utilization review or referral procedures.
Clinical protocols or guidelines are used for medical decision-making.
Formal protocols exist in writing and define the exact procedures to be followed
for specific clinical conditions. There are also informal guidelines that represent
a consensus of the participating providers about delivery of care. Protocols and
guidelines may define, for example, when it is clinically appropriate for a
patient with a specific condition to undergo a surgical or diagnostic procedure;
they may also define the types of services to be rendered to a patient with a
given medical condition and the appropriate setting where these services should
be provided. The development of formularies to control the dispensing of
pharmaceuticals also falls into this category.
Patient flow procedures are used to control the scheduling and timing of
patient services across the entire HMO physician network. Included in this
category are the assignment of patients to "gatekeepers" for primary care
services, guidelines for scheduling of visits by patient class (for example, acute-
care patients, physical examinations, and follow-up visits), time allocated per
appointment, access to specialists and tertiary providers, maintenance of
telephone lines, office hours, and facility location. Development of staffing
ratios for professional and administrative personnel represents another form of
control over patient flow.
Physician selection involves an assessment of a physician's technical
capabilities and practice style. In medical groups and staff model HMOs, the
selection process begins at the initial time of hire and continues through a
probationary period. In IPAs, the process is more limited and often focuses on
establishing criteria for physician participation, such as a medical license in
good standing, adequate malpractice coverage, local medical staff affiliation,
and board certification or eligibility. For some community-based IPA HMOs,
participation is on an "every willing provider" basis, with the intent of providing
as widespread a geographic distribution of physicians as possible.
Data analysis involves a review of financial and utilization trends by
practice site, by primary care or specialty service, by member class (for

APPENDIX B 216
example, Medicare, Medicaid, commercial group, and individual membership),

and by employer group. Data analysis also involves a review of individual
provider performance by various utilization and clinical measures; feedback of
this information to participating providers allows them to judge their
performance relative to the performance of their peers. In two-tier
arrangements, the data analysis is typically performed by the HMO. In three-
tier, Model 3 IPA or group model HMOs in which providers are capitated for
some or all of their services, the data analysis is usually conducted by the IPA
or the group. In this situation, many HMOs receive only limited encounter
information back from providers; as a result, the HMOs may actually lack the
data to evaluate how providers are performing financially under their capitated
contracts.
Peer review, which can be classified as both a utilization management and
quality assurance tool, occurs on both a formal and an informal level. On a
formal level, peer review may involve provider feedback, performance
evaluations, and disciplinary action. In many plans, the thrust of this activity is
on detecting outlier behavior (that is, poor-quality services, excessive ordering
of services, referral to nonnetwork providers, etc.). For all two-tier HMOs, the
monitoring process is the responsibility of the medical director and/or the peer
review committee at the HMO. Here the onus is on plan management to prove
that a physician is not performing according to preestablished standards.
Typically physicians participating in three-tier IPA or group model HMOs are
not subject to formal performance evaluations by HMO management, as this
responsibility is handed to the contracting physician entity.
There also is the informal aspect of peer review that occurs during rounds,
surgery, case conferences, and consultations. Physicians practicing in group or
staff settings are subject to informal peer review on a daily basis. They can
observe their fellow physicians both in office and hospital settings, whereas in
most IPAs the informal peer review process is limited to the hospital setting.
Physician financial incentives include both the basic method of provider
payment (salary, fee-for-service, or capitation) and the use of bonuses,
withholds, and sharing of surpluses and deficits with the plan.
The delivery components of utilization management vary markedly by
practice setting. In Table B-1, we presented various control models based upon
the design of the contractual arrangement between HMOs and physician
providers (two-and three-tier arrangements). Here, the emphasis is on the locus
of control rather than the underlying organization of the physicians contracting
with the HMO. In Table B-2, we look at one type of HMO (three-tier, Model 3),
in which the HMO has capitated the physician entity (IPA or group) and has
delegated control over the delivery components of utilization management.

APPENDIX B 217
TABLE B-2 Provider-Based Utilization Management Activities

Utilization Management Approach (Three-Tier, Model 3 HMO)
Activities Community- Multispeciality Multispeciality
Based IPA (FFS/ Group (FFS/ Group (Capitated)
Capitated) Capitated)
Formal utilization
review
Preadmission *** */** */***
reviewa
Concurrent **/*** */** */***
reviewa
Second surgical */***
opiniona
Referral */*** * *
authorizationa,b
Case managementa */*** */** */***
Bill audit */** * *
Remedial/fraud ***
Remedial/ *
penalties
c
Clinical protocols . In UR Yes Yes
Formularies */*** */***
Patient flow
Primary care */** */** */***
physician
Staff size */** *** ***
Scheduling visits **/*** **/***
Length of *** ***
appointment
Specialty referrals **/*** **/***
Diagnostic **/*** **/***
services
Physician selection */** *** ***
Data analysis */*** **/*** ***
Provider feedback */*** **/*** **/***
Peer review
Formal procedures * *** ***
Financial Yes Yes Yes
incentives
NOTE: *, Occasionally used as a utilization management technique; **, Frequently used as a
utilization management technique; ***, Regularly used as a utilization management technique.
FFS indicates fee-for-service; UR indicates utilization review.
a Includes preestablished utilization guidelines by procedure or diagnosis.
b Includes diagnostic procedures and out-of-network services.
c Includes both protocols and informal clinical guidelines.
In Table B-2, there are three general types of contracting entities: (1) a
community-based IPA whose member physicians continue to practice primarily
in a fee-for-service setting and who are reimbursed by the IPA on a discounted
fee-for-service basis subject to a withhold; (2) a multispecialty group practice
with fee-for-service that has both patients and multiple managed care (HMO,
exclusive provider, and/or preferred provider) contracts; and (3) a
multispecialty group that contracts exclusively with one HMO and that receives
almost all of its income from prepaid services. As shown in Table B-2, the IPA
relies heavily on formal utilization review procedures to

APPENDIX B 218
modify continuing fee-for-service incentives. That is, because participating

physicians continue to be reimbursed on the basis of the number and level of
services provided, the IPA focuses its efforts on controlling unnecessary
utilization and on monitoring billing practices by participating physicians. In
this environment, the decision-making process for high-cost care is transferred
in part from the individual physician to the utilization review team (that is, for
decisions involving certain types of diagnostic testing, inpatient admissions,
referrals to noncontracting facilities, and rehabilitative services).
Depending upon group size, the other two physician entities in Table B-2
may or may not have formal utilization review programs. The larger the
multispecialty group, the more the need for a formal utilization review program
to standardize procedures across sites. As described in the literature, the
utilization review program used by Kaiser Permanente physicians in southern
California includes preadmission and concurrent review; discharge planning;
home health; patient education services; and retrospective review of admissions,
length of stay, and level of care.16 Some groups also have case management
programs to coordinate care for certain categories of high-cost patients. Some
may require authorizations for referrals outside the multispecialty group, and
bill audit, when it occurs, is usually for outside services. Unlike IPAs,
multispecialty groups are generally not concerned with remedial action for
fraud or nonadherence to utilization review protocols, since poorly performing
physicians will not be retained in the practice.
In IPAs, clinical protocols are built into the utilization review algorithms,
with primary emphasis placed on the control of inpatient and high-cost specialty
and diagnostic services. Usually, there is little attempt to influence routine
decision-making in the ambulatory setting other than through the referral
process and benefit design. In multispecialty group practices, physicians may be
encouraged to adopt a specific practice ''style.'' While the group may have
developed formal clinical protocols for specific conditions or procedures,
typically, the process involves little more than senior physicians serving as
mentors for newly hired staff.
In terms of controlling patient flow, IPAs do not generally determine the
scheduling practices of their participating physicians, other than requiring plan
subscribers to sign up with a primary care gatekeeper. Whereas IPAs rely on
formal utilization review programs primarily to control costs, group practices
may control utilization through the patient flow process. As shown in
Table B-2, multispecialty groups rely on scheduling of appointments, the length
of the appointment, and access to specialty providers as utilization management
controls.
Typically, IPAs place less emphasis on staff size and physician selection
than do multispecialty groups. In fact, many IPA HMOs purposely seek

APPENDIX B 219
to develop a large physician base to gain market acceptance. As a result, HMO

patient volume varies significantly among participating providers.
Multispeciality groups, on the other hand, determine staff size for primary care
and specialty services on the basis of preestablished patient-physician or
enrollee-physician ratios. Recruitment of additional staff generally occurs only
when there is sufficient patient or membership volume to provide a patient base
for new physicians.
In terms of data analysis, theoretically, all IPAs and groups develop
utilization statistics. AS will be discussed subsequently, the accuracy and
timeliness of this information and the frequency with which it is supplied to
management and to participating physicians differ markedly across individual
IPAs and groups. One generalization, however, is that groups operating in the
fee-for-service sector tend to have more detailed financial and utilization data,
with procedure codes, than do prepaid group practices that do not routinely
prepare patient bills.
Finally, while the peer review process exists in all three settings described
in Table B-2, the thrust of the activity varies in the IPA versus the group
settings. In the former, the emphasis is on detecting outliers, whereas in the
latter there is additional emphasis placed on how well a given physician adopts
a groups practice style and works cooperatively with his or her peers, as will be
discussed subsequently.
Quality Assurance
Quality assessment is an evaluation of quality of care through structure,
process, and outcome measures, as originally defined by Donabedian.17 Quality
assurance is quality assessment plus correction of identified deficiencies.
Mosser18 has defined quality assurance on three levels. On the first level, and in
its broadest sense, quality assurance includes such diverse activities as
credentialing of providers, evaluating physician performance, disciplining poor
performers, continuing education, tracking malpractice claims, collecting
grievance information, conducting member satisfaction surveys, performing
facilities review, and establishing laboratory quality control. On the second
level, quality assurance is defined as the process of measurement, problem
correction, and reevaluation. On the third and narrowest level, it consists of
focused problem correction in groups, performance feedback to individuals, and
direct surveillance and corrective action.
State HMO licensure requires documentation of a utilization review and
quality assurance program, as does eligibility for a Medicare risk contract and
participation in prepaid Medicaid programs. The actual implementation of these
programs varies widely, with many plans tending to

APPENDIX B 220
focus on utilization review activities and to pay only very limited attention to
quality assurance protocols.
Operational Problems Impairing Utilization Management

and Quality Assurance Programs
Utilization Management Programs

Many HMOs have discovered that their utilization management programs
in concert with risk incentives have had only limited success in modifying
physician practice patterns. The situation is particularly difficult when providers
are reimbursed on a fee-for-service basis, either directly by the HMO or through
the IPA or group (which itself may be capitated). Additionally, there are
specific operational problems that have plagued the HMO industry and that
affect implementation of utilization management programs. These problems can
be broken down into four broad categories.
1. Inadequate management information systems and claims processing
procedures. The first generation of commercially available HMO
software typically handled an enrollment size of no more than
40,000-50,000 members and a limited number of HMO benefit
packages. Software suitable for IPAs contracting with physicians on a
fee-for-service basis (using discounts and/or withholds) often was
inappropriate for group and staff model HMOs that used capitated
arrangements, and vice versa. Moreover, most of the early software had
limited flexibility in terms of report generation, particularly for any
analysis of practice patterns on a provider-specific basis. Since the early
1980s, a number of plans have seen their membership pass the 50,000-
member threshold and have experienced significant claims processing
delays because of system overload and software design. Some HMOs
find that their software cannot accommodate the proliferation of new
benefit packages, including high-and low-option plans, the development
of vision, mental health, dental and drug riders, and so-called indemnity
wraparounds or self-referral options. Existing software also may not
handle various hospital and physician risk-sharing arrangements. At one
carrier-sponsored HMO, for example, the claims processors had to
manually calculate all hospital claims involving per diem payments;
another had to send all hospital claims to an affiliated payer to be
assigned diagnosis-related groups before calculating any payments. In
some cases physician withholds have had to be calculated manually,
while other HMOs cannot handle volume discounts.
As a result of inadequate or outmoded management information
systems (MIS), many HMOs have suffered such problems as buildup in
un-processed claims or encounter forms; inability to implement various
forms

APPENDIX B 221
of reimbursement such as multiple per diems for hospitals and variable

risk withholds for physicians; and failure to provide physician-specific
utilization data on capitated and noncapitated services, on authorized and
nonauthorized admissions and referrals, and on billing practices.
Currently, a number of HMOs across the country are in the process of
defining business requirements for new MIS, while others are installing or
have just installed new systems. To the extent that these systems can
handle a multiplicity of benefit and reimbursement arrangements, some of
the problems described in this section will eventually be resolved.
2. Lack of data integrity. Until recently, the focus of claims processing
operations was on the turnaround time for paying providers rather than
on any statistical review of utilization patterns. In many instances, the
data were grouped so arbitrarily as to eliminate the possibility of
extracting useful information for actuarial or statistical studies. For
example, at one HMO (see Case Study 3), all claims had to be classified
into three categories: inpatient, outpatient, and other. Information on
service location was missing (for example, emergency room, physician's
office, and nursing home). Moreover, when a primary care physician
provided services on both a capitated and self-referral basis during the
same visit, the services had to be recorded as two separate encounters; a
third encounter could result from the same visit if there were any
charges for medical supplies. Because the standard reports from the
claims system showed such visits as multiple encounters, any analysis of
individual physician practice patterns was unreliable. In fact, in Case
Study 3, actuarial projections could not be derived from the claims
system; rather, they had to come from the data base of an outside
consultant. (This consultant provided estimates based upon average
costs and utilization rates across a series of clients.) Some of the data
integrity issues will be obviated with the purchase of new HMO
software; others are a function of data collection and processing
procedures that are independent of software design. To the extent that
HMOs do not audit the performance of claims processors to ensure
accuracy and consistency in data entry, data integrity problems may
continue to go unrecognized in a number of plans.
A related issue concerns the noncomparability of data across HMOs
for standard utilization measures, such as physician encounters per 1,000
members and hospital admissions or days per 1,000 members.
Some group and staff HMOs do not routinely record certain types of
services, such as physician hospital visits, while others do not receive or
record encounter information for capitated primary care services. Other
plans, as previously noted, may have inflated or deflated visit counts or
may arbitrarily collapse service location codes because of the design of
their data systems. Also, data on out-of-plan use is often incomplete.19
Thus,

APPENDIX B 222
any comparison of utilization trends across plans requires a knowledge of

their data classification procedures.20
3. Lack of an integral role for the medical director at upper level HMO
management. Many HMO medical directors are frustrated by their lack
of input into HMO operations and their inability to obtain adequate
budgets for professional support staff with appropriate statistical,
utilization review, and quality assurance experience. In some cases,
medical directors cannot reduce or deny payment to providers who are
suspected of unusual billing practices or provision of unauthorized
services. Depending on the plan, this authority may rest with the chief
financial officer, the controller, or the director of provider relations. In
HMOs with outmoded MIS, medical directors have expressed concern
about their inability to obtain detailed utilization reports because of the
priority attached by plan management to producing financial reports. As
a result, medical directors may be unable to document poor practice
patterns, even though they are aware of their existence; as a result, the
entire peer review process suffers.
4. Lack of provider feedback. Provider relations suffer when there is a
shortfall between the actual claims experience for an IPA and the
projected experience used to calculate the IPA's capitation rate. This
problem worsens when the withholding rate proves insufficient to cover
the deficit. In control models in which the HMO processes all claims,
IPA management cannot anticipate any shortfall and initiate corrective
action unless timely utilization data are provided by the HMO. In Case
Study 1, local physicians threatened to withdraw from the IPA unless
the capitation rate was increased; without appropriate utilization data,
the physicians refused to acknowledge that there had been any
unnecessary provision of services or unusual billing practices.
The first three problems listed above are not unique to the HMO industry;
indemnity carriers and Blue Cross and Blue Shield plans face the same
concerns. Increasingly, the fourth problemlack of provider feedbackis an
issue for carriers and Blue Cross and Blue Shield plans, now that traditional fee-
for-service coverage is giving way to preferred provider and closed panel
arrangements. However, it is the potential for underservice in HMOs that has
led to congressional concern about quality assurance in HMOs. However, as
discussed in the case studies, many HMOs have failed to thrive precisely
because the fee-for-service mentality continues to flourish in the prepaid sector.
Quality Assurance
In our experience, HMO management generally gives the highest priority
to day-to-day marketing, underwriting, and financial operations and the lowest
priority to programs, such as quality assurance, for which an

APPENDIX B 223
immediate impact on plan performance is not evident. Additionally, given the

understaffing and limited MIS capabilities in many HMOs, implementation of
formal quality assurance programs may become problematic. For example, in
certain plans, performance feedback to physicians is uneven because of
inadequate data systems, poor data entry procedures, claims processing lags,
and/or failure to classify claims data in a format useful to participating
providers. Conversely, there may be little incentive for capitated IPAs, groups,
and individual providers to supply timely or complete encounter data back to
the HMOs, since their payment is generally based on the number of covered
enrollees, not on the actual services rendered. As a result, the HMOs may be
unable to determine actual utilization rates or to detect underservice.
Many HMOs have a two-stage approach to quality assurance. Initially,
when signing up physicians (as individuals or through IPAs or groups), they use
eligibility criteria to ensure the adequacy of the provider network. Thereafter,
quality assurance generally consists of problem identification. The process may
be fairly extensive, such as using target medical conditions or types of cases,
such as hospital readmissions, as the basis for a periodic review of medical
records. HMOs that conduct their own utilization review programs may also
obtain feedback on provider performance from their staff. Alternatively, some
plans may limit their quality assurance programs to tracking malpractice claims,
grievance filings, and disenrollment rates by medical group or by primary care
provider if they use individual gatekeepers.
In three-tier model plans, the HMO may require that participating groups
or IPAs assume full responsibility for peer review. Network or group model
HMOs typically require evidence of a functioning peer review program before
they award a contract to a multispecialty practice. HMOs contracting with IPAs
may require that a formal peer review system be in place, but since many IPAs
lack a track record for judging physician performance, the adequacy of these
programs cannot be assured.
Thus, operational reviews frequently disclose that there is a wide gap
between quality assurance programs, as they are described in documents
supplied to federal and state regulators, and as they actually exist within
licensed health plans. Historically, structural measures have been notoriously
poor indicators of both HMO performance and quality of care. For example, in
California during the mid-1970s, licensing of prepaid health plans, with
requisite filing of appropriate quality assurance protocols and provider
contracts, did not prevent abuses in the state's prepaid Medicaid program. More
recently, in the International Medical Centers, Inc. situation in Florida,
structural measures failed to guarantee Medicare beneficiaries adequate access
to medical services. The problem with structural measures was highlighted in a
recent survey of HMOs in Ohio, as conducted

APPENDIX B 224
by the Joint Commission on Accreditation of Health Care Organizations

(JCAHO). In a review of six HMOs with Medicaid contracts, JCAHO surveyors
found that none of the plans had fully implemented its quality assurance
program.21
Because of the limited scope of existing quality assurance programs,
federal legislation now mandates review of inpatient and outpatient services
provided by HMOs with Medicare risk contracts. External peer review
organizations have begun using medical records and target medical conditions
as indicators of the appropriateness of treatment, accessibility to services, and
timeliness of care provided. The possibility of sanctions resulting from these
reviews is now providing the impetus for HMOs to improve their quality
assurance programs. For example, of the 15 HMOs in California, Hawaii, and
Arizona with Medicare risk contracts, only 3 had fully implemented their
quality assurance programs at the time of their initial evaluation by California
Medical Review Inc.22 Some HMOs have also begun to seek voluntary
accreditation from external review organizations, such as JCAHO. However,
because of the limited disclosure associated with mandatory and voluntary
reviews, subscribers signing up with HMOs still have limited information
available for evaluating quality of care within a given program.
Design and Use of Physician Incentives

In the preceding section, we looked at the operational problems impairing
the administration of utilization management and quality assurance programs.
As noted, structural and process measures are often an inadequate measure of
plan performance. A similar situation exists for any evaluation of risk
incentives. In this section we briefly describe the background on current
concerns about risk incentives, the types of risk incentives that have been
identified, and the factors that modify their impact on provider performance.
Background
In a report issued in July 1986, the General Accounting Office (GAO)
identified three types of physician incentive plans with a potential risk of abuse
for Medicare beneficiaries: (1) the Paracelsus or hospital model, in which a
monthly bonus was paid to staff physicians on the basis of a favorable ratio of
total hospital charges for Medicare patients to Medicare payments for those
patients; (2) the Medical Staff-Hospital Joint Venture or the MeSH model, a
theoretical model in which an annual bonus was to be paid to those physicians
with a positive cost performance using targeted costs per discharge; and (3) the
individual practice association model in

APPENDIX B 225
which incentives, including primary care capitation arrangements, cover

inpatient and outpatient services on an annual basis. The GAO report concluded
by recommending that any incentive payment programs should average costs
versus payments over a fairly large patient base to reduce incentives to
undertreat individual patients; the GAO also recommended that the time period
for calculating these incentives should be relatively long (1 year).
The Omnibus Budget Reconciliation Act of 1986 specifically prohibited
hospitals, as well as HMOs with Medicare risk contracts, from using financial
incentives to reduce or limit services provided to Medicare beneficiaries.
Because of the widespread use of physician incentive payments by HMOs and
CMPs, however, Congress delayed application of this provision to prepaid plans
until April 1, 1989, pending further research. Subsequently, the Omnibus
Budget Reconciliation Act of 1987 extended the delay to April 1, 1990, and
new proposals in Congress make implementation unlikely.
To date, much of the published research on physician incentives has
focused on documenting the range of arrangements that exist and the frequency
with which they are used by plan model, plan age, plan size, and geographic
location. With the exception of recent work by Mathematica Policy Research,
Inc.24 there has been little research on the effectiveness of quality assurance
programs or the impact of financial incentives on utilization of services in
HMOs.
Current Surveys Concerning Physician Incentives

Since the GAO report, there have been four additional studies on financial
risk arrangements in HMOs: (1) the Group Health Association of America
(GHAA) Survey mailed to GHAA member organizations in December 1986;25
(2) a Blue Cross and Blue Shield Association Survey, using the GHAA survey
instrument, was mailed in March 1987 to BC/BS plans that sponsored HMOs;26
(3) a March 1987 survey mailed by Alan Hillman, to all 595 HMOs known to
be in operation as of June 1986;27 and (4) a survey of 215 federally qualified
HMOs or CMPs, in combination with a review of documentation in the Office
of Prepaid Health Care of the Health Care Financing Administration, was
conducted by ICF, Inc. under contract with the U.S. Department of Health and
Human Services.28 The GAO has also conducted an additional study of
physician incentives in response to a request by the Ways and Means
Committee of the U.S. House of Representatives.
An analysis of the first three surveys revealed that consistent definitions
were not used across the studies, leading to variations in findings as to the
frequency with which these incentives have been used by various types of
HMOs (that is, staff, group, network, IPA, and mixed or hybrid

APPENDIX B 226
model plans.) In general, financial incentives were broken into four broad
categories, as follows: (1) the basic approach to physician payment (that is,
salary, capitation, and fee-for-service); (2) sharing of plan surpluses and/or
deficits; (3) use of bonuses and/or withholds; and (4) distribution of funds by
individual, group or combined individual-group performance.
While the three studies reported varying use of payment mechanisms and
risk incentives by plan type, the single most important difference between the
three related to the prevalence of financial incentives based on individual
performance factors. The GHAA study found that 140 of 164 responding plans
(85.4 percent) had some form of financial incentive for primary care physicians.
Sixty percent of the plans reporting on the design of these incentives used some
type of individual incentive; 11.2 percent reported exclusive use of individual
incentives. The Hillman survey reported that two-thirds of all contractual
arrangements (232 of 353) withheld a portion of payments to primary care
physicians, but only 18 percent (38 of 211) of the HMOs held physicians at risk
on an individual basis for deficits beyond the withhold. Although the ICF study
did not separate reporting of withholds from other financial incentives, this
study reported that 0.5 percent of responding HMOs relied solely on individual
performance for the distribution of deficits and surplus funds. In a recent
reconciliation of their findings, ICF and GHAA have concluded that in their
common sample of 54 plans, individual physician incentives (including
individual cost-or utilization-based withholds) were used in 20 percent of plans.29
Administering Physician Incentives

Often, there is a discrepancy between risk incentives as defined in
physician contracts and as administered by either the HMO, the IPA, or the
group. The failure to implement risk incentives, according to contract terms,
may reflect a conscious policy decision by management and/or operational
difficulties. For example, one large network model HMO has reported that each
month it withholds 10 percent from "an actuarially generous" capitation
payment to participating multispecialty groups. The groups, however, record
this withhold amount on their accounts receivable and anticipate the return of
the funds. In the event of a deficit in any of the group's accounts because of
unanticipated referral or ancillary service costs, the HMO would still return the
withhold to maintain a good working relationship with the group.
The converse of this situation occurs when physicians automatically
assume that the withhold will not be returned. The ICF study reported on one
IPA HMO in which the withhold amount was seldom returned; the medical
director stated that the withhold was not perceived as a financial incentive by
the physicians but as a discount or as a business cost comparable

APPENDIX B 227
to denied or suspended claims in the fee-for-service environment. We also

found in several of our reviews that physicians respond to loss of the withholds
by changing their billing patterns. For example, physicians may attempt to
compensate for the withhold loss by inflating the CPT codes for consultations
and office and hospital visits and by unbundling routine services, such as
charging for component parts of blood chemistry panels.
In our experience, certain HMOs cannot readily detect inappropriate
billing practices by individual physicians because of inadequate claims
processing systems. In other cases, HMOs may control upcoding and un-
bundling by specifying acceptable CPT codes for specific types of visits and/or
procedures. Alternatively, HMO management may choose to ignore these
billing practices, particularly when taking action may antagonize providers and
lead to loss of their participation.
Sometimes risk incentives are short-circuited when an HMO cannot fulfill
the terms of its contracts because of operational difficulties. For example, if an
HMO has a claims processing backlog and cannot provide requested utilization
reports according to a predefined schedule, the HMO may have to return all or
part of the withholds, despite the existence of deficits in the physician accounts.
For example, one carrier-sponsored HMO, described in Case Study 3, recently
made a conscious decision to distribute 75 percent of all physician withholds
despite deficits in existing risk pools and despite the knowledge that some
primary care providers were self-referring excessively for specialty services not
included in their own capitation rates. This management decision was based, in
part, upon the fact that the HMO was unable to produce its utilization reports
according to schedule and, in part, to placate participating physicians.
Finally, there are cases in which physicians may actually be unaware of
any financial incentives related to effective utilization management. For
example, in the situation described in Case Study 2, the salaried physicians
were not informed during recruitment of a possible bonus based on plan
performance. In this plan, utilization management was more a function of
scheduling rather than of any financial incentives.
Thus, the issues to be addressed when reviewing the financial incentives
within HMOs include the degree to which they have been fully implemented
and the degree to which participating providers are fully aware of them. The
Hillman, GHAA, and ICF survey instruments have focused on the design of
contractual arrangements, but further research is needed to assess the ability of
HMOs and CMPs to support these arrangements.
Policy and Research Issues

There is a whole body of literature involving analysis of utilization rates in
HMO and fee-for-service settings. Another set of studies looks at

APPENDIX B 228
utilization rates in different HMO settings (for example, comparisons of the

performance of totally versus partly prepaid group practices and of IPAs versus
totally or partly prepaid group practices).30 There also has been a growing
interest in utilization rates and outcomes for specific medical conditions in the
two settings.31
To date, health services researchers have assumed that differing financial
incentives are responsible for the lower utilization rates in prepaid versus fee-
for-service settings and have focused little attention on how HMOs achieve
these reductions.32 The literature on the delivery component of utilization
management in HMOs consists largely of anecdotal reports detailing controls in
place at specific plans and case studies involving one or more components of a
utilization management program at a single HMO, IPA, or multispecialty group
practice. For example, a recent report in the New England Journal of Medicine
described one physician's negative experience working in a for-profit, staff
model HMO.33 Other studies have reported on the use of referrals or the impact
of provider feedback on ordering of diagnostic tests in specific plans.34
Our operational reviews suggest that the structure of utilization
management and quality assurance programs and their actual implementation
vary widely across plans. On-site evaluations of these programs are currently
being conducted by mandatory and voluntary review organizations. Until the
findings of these review organizations are made public, it will be difficult to
generalize about which types of HMOs and which control models have
implemented utilization management/quality assurance programs the most
successfully. Moreover, if the findings from the reviews do not address where
the responsibility for utilization management rests, it may be difficult to
develop a policy response that can be used to ensure quality of care in HMOs.
It also should be stressed that a large number of HMOs are not subject to
federally mandated review, as they do not participate in the Medicare risk
program. As already noted, state licensing has tended to focus on structural
measures, with on-site reviews across plans occurring infrequently or restricted
to problem plans. Additionally, none of the voluntary or mandatory reviews is
specifically designed to evaluate the impact of various types of risk incentives
on provider performance.
The Hillman, GHAA, and ICF surveys described above looked at risk
incentives from a structural perspective. Each of the three studies used the
HMO as the unit of observation. The survey information on physician practices
that is currently available from the American Medical Association, from state
medical societies, and from the Medical Economics Company Inc. (Oradell,
New Jersey) does not have sufficient detail to evaluate the financial impact of
risk incentives by HMO plan type or control model. Thus, to better understand
how financial incentives affect practice patterns,

APPENDIX B 229
there may be a need to modify existing survey instruments or to undertake a

separate study that focuses on primary care physicians with and without HMO
affiliations. For HMO physicians, the future studies should evaluate (1) HMO
model type, control model, geographic location, and enrollment; (2) the gross
income of the primary care physician and overall patient volume, as compared
with HMO patient income and volume; (3) the structure of HMO risk incentives
and their financial impact on the responding physicians; and (4) the design and
implementation of utilization management and quality assurance programs, as
perceived by the responding physician.
The results of this survey can then be correlated with findings from
outcome studies, consumer satisfaction surveys, and on-site reviews, as
conducted by various regulatory agencies, peer review organizations, consumer
or employer coalitions, and university-based research teams. Only then will it
be possible to ascertain definitively which types of HMO financial incentives
are the most problematic in terms of their impact upon quality of care. Until
then, it is probably premature to develop legislation that eliminates the use of
any particular form of financial incentives in the HMO setting.
Case Studies Introduction

The following case studies are not based on a representative sample of
HMOs. We present them solely with the intent of showing the range of
problems that may be encountered by HMOs when implementing utilization
management programs. The background material at the beginning of each case
study has been altered to maintain client confidentiality. All plans are defined as
staff, group, or IPA, according to their operational characteristics, rather than
their federal or InterStudy designation. The network category has intentionally
been eliminated, as have any descriptors indicating location of the plan, plan
age, nonprofit, or for-profit status. Size is defined by three categories: under
25,000, 25,000-49,999, and 50,000 and over. Also, any descriptors that might
allow for identification of the plan, such as university affiliation, hospital
ownership, Medicare risk status, or unique organizational characteristics, have
been removed. However, the descriptions of the control type, the financial
incentives, the design of the MIS, the role of the medical director, and the
structure of the utilization management program are reported precisely as they
were found in each of the plans.

APPENDIX B 230
Case Study 1: Hospital-Sponsored Ipa
Background
In the mid-1980s, management of a hospital with 350+ beds decided to
establish an IPA model HMO in order to contract with one of the major
employers in the state; simultaneously, physician leadership at the hospital
helped to set up an IPA, which was open to any active member of the hospital's
medical staff. After the HMO became operational, membership climbed rapidly
to the range of 25,000-49,999, with the growth exceeding original forecasts.
However, the plan posted larger financial losses than anticipated because its
operating and administrative procedures, along with its MIS, could not keep
pace with the enrollment growth. The sponsoring hospital was forced to
subsidize the health plan, both directly and indirectly, by underwriting the
operating losses and by charging the plan a composite per diem that was
actually below its own operating costs.
Eventually, the sponsoring hospital sought financial relief and asked the
hospital system with which it was affiliated to take over the HMO. When the
hospital system took over management of the HMO, it attempted to cut the
financial losses through new reimbursement procedures and risk-sharing
arrangements; as a result, there was a serious breakdown in provider relations.
Control Type
Three-tier, with the HMO controlling most services (Model 1).
Financial Incentives
Under the original risk model, physicians contracted with an IPA that was
fully at risk for all inpatient and outpatient physician services, including
outpatient mental health, ancillary, and in-area emergency services; durable
medical equipment; prosthetic devices; and prescription drugs. The HMO
allocated a flat monthly capitation rate per adult and per child member to the
IPA; physicians submitted their claims directly to the HMO, with payment
based on a predetermined fee schedule subject to a 20 percent percent withhold.
The 20 percent withhold was used to fund an IPA reserve account, as well as to
fund any deficits in a separate hospital fund.
The HMO also allocated a flat per-member, per-month amount to the
hospital fund, which covered inpatient hospital and mental health services,
same-day surgery, skilled nursing and home health care, and out-of-area
emergency services. Providers rendering services covered by the hospital

APPENDIX B 231
fund, including the sponsoring hospital, billed the HMO directly for their
services according to prenegotiated rates. (There was no withhold for any
services covered by the hospital fund.) Any deficits beyond the 20 percent
physician withhold in either the IPA or hospital funds were the responsibility of
the HMO, while any surplus in the hospital fund was to be allocated between
the IPA reserve fund and the HMO on an 80-20 basis. The targeted utilization
rate for all inpatient services was set at 350 days per 1,000 members; in 1985
and 1986, the actual hospital utilization, after adjusting for coordination of
benefits, was approximately 300 days per 1,000 members.
While the IPA had historically operated at a loss because of overutilization
in the outpatient setting, in 1985 and 1986, the plan was able to return the
physician withhold because of the surplus in the hospital fund. Problems
emerged, however, when the plan was taken over in 1987 and new management
wanted to lower the targeted hospital utilization rate, to share surplus and
deficits in the hospital withhold pool on a 50-50 basis, and to increase the per
diem paid to the sponsoring hospital. The physicians recognized that they would
no longer be able to anticipate a return of their withhold and asked the health
plan to modify their capitation to reflect their actual utilization experience. The
net result was a standoff between the doctors and the health plan.
Design of Mis System

The claims processing system was using ''first-generation'' software and
was incapable of handling the volume of claims that were being received. As a
result, there was an emphasis on rapid claims payment, without appropriate
controls on data entry or claims review procedures.
There was no automatic linkage of the authorization files to the claims
history files. Therefore, before any inpatient, emergency, durable medical
equipment, or out-of-area claims were paid, they were sent to a nurse
coordinator for approval. All other claims were entered directly into the claims
system, and unless they were suspended for lack of a provider number or a
duplicate payment, they were paid as billed according to the existing fee
schedule. In other words, outpatient claims were paid without any automatic
review of frequency of service, procedure coding, or consideration of whether
the service was appropriate, given the diagnosis, age, or sex of the patient.
Primary care physicians theoretically were to act as gatekeepers and to
approve all specialty referrals and hospital admissions. In reality, patients

APPENDIX B 232
continued to self-refer themselves to specialists. The primary care providers

(PCPs) were not required to adhere to any protocols for authorizing the use of
specialty or ancillary services, nor had any procedures been established for the
management of high-cost patients.
PCPs were required to notify the HMO's medical director of pending
hospital admissions for his approval. This procedure was generally well
observed, because of the widespread recognition by IPA members of the
financial impact of controlled hospital utilization on the IPA reserve fund. The
surplus in the hospital fund, in effect, had allowed the IPA physicians to
continue practicing on an "as usual" basis in the ambulatory setting.
With the HMO takeover, the physician leadership of the IPA was
vehemently opposed to any changes in allocation of the hospital fund surplus
and to the repricing of the hospital per diem. The IPA leadership believed
firmly that participating physicians were delivering efficient care and did not
want to use the withhold to cover IPA deficits. Meanwhile, HMO management
believed that there was excessive utilization in the ambulatory setting. Neither
side, however, had any documentation to support its argument, as the IPA did
not collect claims data and the HMO had never produced any physician-specific
data or developed any normative standards for evaluating physician
performance in the outpatient setting.
Medical Director and Utilization Management Support Staff

The medical director reported that he had not been an integral member of
top plan management but had been hired only part-time to oversee the nurse
coordinators and to review pending hospital admissions. The utilization review
staff did not have well-defined policies or procedures to guide them,
particularly when adjudicating claims. As a result, claims payment was not
consistent across reviewers (for example, the same services might be paid in
some cases and denied in others).
Although the HMO had a written quality assurance plan, the nurses had
little time to undertake chart audits or to review treatment plans for target
conditions; instead, as much as 50 percent of their time was spent on clerical
work associated with claims processing. While the nurse reviewers recognized
that some IPA physicians regularly did not adhere to utilization review
requirements, they did not have the authority to apply any penalties. Therefore,
of approximately 100 physician infractions identified in 1987 involving the
provision of medically unnecessary services or referrals to nonparticipating
physicians, only two resulted in warning letters from the plan.

APPENDIX B 233
Case Study 2: Group-Model Hmo
Background
In 1985, a multispecialty group practice, which contracts primarily with
one group-model HMO, expanded its operations into two metropolitan areas
where, historically, fee-for-service providers had been highly resistant to
participation in any form of managed care (i.e., HMOs or preferred provider
organizations [PPOs]). The move into the other communities was made at the
request of several large public and private employers anxious to have an
alternative to fee-for-service medicine. Membership at the new sites grew
rapidly. At the end of 2 years, membership size far exceeded HMO projections,
and the medical group found that they had outgrown their facilities at the two
sites. Also, recruitment of additional professional and administrative staff was
not feasible until the practice sites could be expanded (membership size cannot
be specified, per client request).
To ensure appropriate levels of care at the expansion sites, group
physicians began referring patients out on a fee-for-service basis into the
community for primary, specialty, and ancillary services that ordinarily would
have been provided in-house. Lacking contracts with most of the local
specialists, the group was forced to pay for these referrals from its capitation
fund on the basis of billed charges. Because of the lack of cost controls over
these outside referrals, the group sustained operating losses at the expansion
sites in 1986 and 1987.
Control Type
Three-tier, with delivery of services controlled by the medical group
(Model 3).
With the exception of the medical director in each of the two sites and one
practicing physician per site, the rest of the physicians at the expansion sites had
no prior experience working for the group. The newly hired physicians had a 3-
year probationary period before they became eligible for partnership benefits; in
the interim they were paid a fiat salary with no productivity incentives. Salary
levels were based on specialty training, board certification, and prior practice
experience. There also was a profit-sharing program in which partners were
eligible to participate; theoretically, the new hires were eligible for a year-end
bonus, but only if the group had a profitable year. However, knowing the
operating losses sustained by the

APPENDIX B 234
group, the medical directors generally did not include any mention of a bonus in
salary discussions with new recruits.

The MIS in place at the original practice sites had been designed to support
the group's own internal data needs. This system had not been built to track or
adjudicate fee-for-service claims from outside providers, under the assumption
that the amount of care provided in the fee-for-service sector would be very
limited. (At its main site, outside physician bills were reviewed manually, with
payment made according to prenegotiated rates and authorized services.) When
the expansion sites were opened, the group purchased software to allow for
tracking of referral and hospital authorizations and for collection of information
on fee-for-service claims, using standard CPT codes. However, fee caps or fee
schedules were not installed in the system, and clinical edits, which determine
whether procedures had been appropriately billed given the patient's age, sex, or
diagnosis, were not activated. Thus, as long as a preauthorization for a referral
existed in the new MIS, claims from community-based physicians were paid,
without questioning the need for billed services or the use of specific CPT codes.
The primary factor controlling outpatient utilization at each of the practice
sites was the scheduling process. Each salaried physician was scheduled for a
set number of morning and afternoon patient sessions per week, with a
predetermined number of time slots allotted for acute-care patients, continuing
patients, and physical examinations. As membership soared, the primary care
physicians were unable to schedule specialty care within the group or handle
conditions requiring intensive, short-term follow-up care on a timely basis,
given the overbooking that was already taking place. As a result, they began to
refer out into the community cases that ordinarily could have been handled in-
house.
When outside referrals were authorized, there was no dollar limitation
placed on the services to be provided, nor was the level of care specified (that
is, type of consultation or office visit, by CPT code). The only requirement was
that certain laboratory and x-ray tests be performed at the groups facilities;
additionally, any surgical procedures or hospital admissions required another
preauthorization. Aggregate reports summarizing referral rates and costs, by
group physician, were received monthly by the medical directors, as were
summaries on the number of patients seen and billed charges from each outside
provider. However, none of this information

APPENDIX B 235
was provided to the group physicians or their support staffs. Thus, there was no
consensus among the group physicians and the referral clerks as to which local
providers were the most cost-efficient, in terms of future referrals.

Both of the medical directors devoted approximately 50 percent of their
time to their clinical practices and appeared to have taken on the management
role with some reluctance. Neither had worked with their medical staffs to
develop referral protocols or had undertaken any educational efforts to make
them aware of referral costs by specialty. Moreover, both medical directors
were extremely fearful about undertaking any program that would antagonize
local physicians; thus, they did not encourage any claims review that would
result in fee cutbacks by eliminating billing abuses or by using a fee schedule.
Rather, the intent of the medical directors was twofold: to initiate preferred
provider contracting with the specialists identified as being the most cost
efficient and to internalize the high-volume specialties as quickly as possible to
control the dollar outflow. However, the plan's ability to recruit specialty staff
was dependent, in part, on its ability to expand its existing facilities and/or
develop new clinical sites.
Case Study 3: Carrier-Sponsored Hmo
Background
In the mid-1980s, a carrier took over an existing IPA model HMO
operation that had a number of start-up sites. The computer system that was in
place was inadequate to handle the variations in benefit packages across sites,
as well as the different payment mechanisms used to reimburse hospitals and
physicians. Because of the losses sustained at most of the HMO sites and the
costs associated with upgrading the HMO software and developing new
products (that is, triple option or a point-of-service HMO), the carrier decided to
evaluate the future of each HMO site. The following description relates to one
of the sites with fewer than 25,000 enrollees.
Control Type
Three-tier, with control resting with the HMO (Model 1).

APPENDIX B 236
Participating primary care providers received a monthly age-and sex-
adjusted capitation payment for a CPT-coded list of services; this payment was
subject to a 20 percent withhold (that could be increased to 30 percent, under
specific conditions). Primary care physicians retained the right to self-refer for
services not covered under the capitation agreement. The individual PCPs were
then grouped into pods or arbitrary physician incentive pools with a minimum
of at least 1,000 members; the withholds of all PCPs in each pod were then
combined. For each pod, there was a separate referral services budget to cover
the costs for specialty, hospital, and ancillary services; this payment was based
upon the number of members in the pod, adjusted for their age and sex mix.
Specialists were paid out of the referral services budget, according to the
70th percentile of usual and customary rates, subject to a withhold. Hospitals
were paid according to diagnosis-related groups (DRGs) or discounted charges,
but not per diems. During the calendar year, if there were deficits in the referral
services fund, the overrun was to be made up first through withholds in the
specialist pool and then through withholds in the primary care pool, and finally,
any remaining deficits were the responsibility of the carrier. Individual stop-loss
coverage from the carrier protected the PCP from responsibility for charges in
excess of $7,000 per calendar year.
Despite the ability of the HMO to penalize physicians for poor
performance, there was a strong fear by plan management of antagonizing
primary care doctors and losing their participation in the plan. Thus, for
calendar years 1986 and 1987, a corporate decision was made to return 75
percent of the physician withholds at each of the HMO sites, regardless of the
extent of deficits in the referral services funds.

The hardware and software were inadequate to support membership size,
leading to a backlog of claims in excess of 20 days for the HMO site as of
December 31, 1987. Moreover, at any given time, the number of unentered
claims and their dollar value were unknown. Once a claim was entered into the
system, there were minimal edits to allow processors to detect inappropriate or
unnecessary services or unusual billing practices by referral physicians.
From a statistical perspective, the information that came off the claims
system was suspect. For example, if a PCP submitted a bill that included both
capitated services and specialty care (for a self-referral), two encounter numbers
had to be generated; if supplies, reimbursable on a cost basis, were also
included in the billing, a third encounter form was generated. Thus,

APPENDIX B 237
the reports on cost per PCP encounter were unreliable. Another example related
to classification of claims by location of service. All claims had to be grouped
into one of three categories: inpatient, outpatient, or other. As a result, nursing
home expenses could not be easily separated out from hospital expenses, while
home health care expenses, for example, could not easily be separated from
those for physical and occupational therapy. As reported by the executive
director at the local site, the system was designed strictly to allocate costs to the
pods and to calculate withholds.
There were no data by pod on patient demographics and no analysis of
utilization patterns within pods to determine where or why deficits were
occurring. In effect, there was no way to evaluate the performance of individual
PCPs, except in terms of expenditures for self-referrals and outside referrals.
Yet, the physicians in the pod were supposed to be self-disciplining. Their
contract, however, did not specify the mechanics of any review process, nor did
it define how a poorly performing physician could be removed from a pod.

At the local level there was a part-time medical director who largely
handled physician and hospital contracting. Utilization review was handled by
three nurses who functioned as patient care coordinators. They received all
incoming calls for the authorization of hospital admissions, and depending upon
the case and available time, they were responsible for conducting on-site
concurrent review. The three nurses were also supposed to authorize referrals
for some 20 high-cost diagnostic tests and procedures, such as use of magnetic
resonance imagers and lithotripters. However, a number of PCPs regularly
ignored this requirement, knowing that the plan had no intent of closely
monitoring their behavior. In fact, other than the initial credentialing process,
there was no quality assurance program in place.
Case Study 4: Carrier-Sponsored Hmo
Background
A carrier established an IPA model HMO which then contracted with a
series of open panel IPAs. Each IPA was responsible for delivering care in one
or two counties of the state. The HMO provided all marketing, administrative,
enrollment, accounting, financial, claims processing, and

APPENDIX B 238
MIS services for the IPAs. Under its contract, it was also responsible for
supplying the IPAs with necessary data and reports to enable the IPAs to
perform peer review, to operate a quality assurance program, and to determine
the status of withhold accounts.
Problems arose when one of the IPAs discovered that it was operating at a
deficit and requested an increase in its monthly capitation rate. The physicians
in this IPA were convinced that the deficits were attributable to three problems:
(1) the HMO's use of outmoded actuarial information, leading to an
underestimation of outpatient costs and a miscalculation of the capitation
payment to the IPA; (2) the miscoding or misclassification of claims data,
preventing a meaningful comparison between capitated payments and actual
claims costs; and (3) the failure of the HMO to provide the IPA with detailed
financial and utilization reports on a timely basis. This case study examines the
experience of this one IPA.
Control Type
Three-tier, with control resting with the HMO (Model 1).
The HMO allocated to the IPA capitation on a per-member, per-month
basis (not adjusted by age or sex), with the HMO processing all hospital and
professional claims. Claims from participating IPA physicians were to be paid
by the HMO according to the lesser of billed charges or a maximum fee
schedule. This schedule was approximately 10 percent higher than the
sponsoring carrier's schedule for its standard business. Twenty percent of all
physician's fees were to be withheld and placed in a risk account.
The IPA required enrolled members to sign up with a primary care
physician, but did not actively support a gatekeeper system. Patients could self-
refer to participating specialists. Authorizations for hospital admissions could
be requested by primary care physicians or treating specialists. Moreover, there
were no built-in financial incentives for performing the gatekeeper function. For
example, the IPA did not allocate any percentage of funds in the withhold to
primary care physicians, but distributed the funds on a pro-rata basis across the
entire IPA membership.

Only management reports, not claims processing procedures, were
reviewed. By April 30 of each year, an annual financial accounting was due,
showing all capitation payments to the IPA for covered services and all billed
and paid services adjusted for incurred but not reported claims;

APPENDIX B 239
from this accounting, the HMO would determine whether any or all of the
physician withhold would be available to the IPA for distribution to
participating physicians.
The reports received from the HMO in April 1986 indicated that the IPA
did not have an operational deficit, and, therefore, the 1985 withhold was
returned and the amount of the withhold was decreased from 20 to 10 percent.
IPA leadership continued to believe that the IPA was operating profitably until
the end of the year, when the HMO reported a substantial deficit in the IPA
account, requiring an increase in the withhold.
Apparently, during the year the HMO had experienced a buildup in unpaid
claims, resulting in a miscalculation of the incurred but not reported claim and a
restatement of the IPA's financial position. At the same time, the IPA was not
receiving adequate information from the HMO to develop any estimates of its
own performance. For example, while the HMO provided monthly totals on
admissions per 1,000 members and the average length of stay per discharge,
there was no financial information on average cost per patient-day or per
discharge, by type of service, or by diagnosis. Detailed information on cost per
patient visit and average cost per procedure was not provided to the IPA prior to
the announcement of the IPA's operational deficit.
The capitation shortfall was attributed in part to the fact that the HMO was
predicating utilization rates on statistics presented by outside actuarial
consultants; however, the data classification system in place at the HMO did not
correspond to that used by the outside consultants. For example, laboratory and
x-ray costs were allocated differently in the two systems, resulting in an
underestimation of the cost of outpatient services in the capitation payment.
Utilization Management and Support Staff

For all hospital admissions, the admitting physician was to call the
utilization review department at the HMO for preauthorization. The
appropriateness of the admission and the site (inpatient versus outpatient) was
not questioned initially by the staff, but rather was reviewed retrospectively.
Nurses at the HMO were responsible for conducting concurrent utilization
review by phone and/or handling on-site reviews. An additional team of one
physician and two nurses handled all psychiatric admissions.
For outpatient services, the claims were screened by processors at the
HMO by using a series of routine edits to detect duplicate billings, noncovered
services, medically outmoded services, and services covered by total charges.
Otherwise, there were no screens to review the frequency of services or
upgrading of procedure codes. Individual claims were not subject to special
review unless physicians contested not getting paid for all

APPENDIX B 240
services rendered. There were no reports issued on a regular basis (that is,
monthly or quarterly) listing physicians who routinely billed for more services
per visit or had higher costs per visit than the IPA averageor the average
across all participating IPAs. There was no separate tracking of ancillary
services with respect to volume and dollar amounts, by referring or treating
physician.
Approximately every 6 months, on an "as-needed basis," there was a joint
meeting of IPA leadership and the HMO medical director and review staff to
resolve outstanding problems. Actions of that committee typically included a
determination of which physicians were utilizing nonparticipating providers and/
or failing to notify the HMO of inpatient admissions. The IPA was then
authorized to levy a $100 fine against noncomplying physicians; over a 2-year
period, approximately 12 fines were issued. Otherwise, the combined HMO-
IPA review team, which theoretically was responsible for quality assurance
activities, did not undertake any analysis of member grievances, incident
reports, admissions by diagnosis, or mortality rates.

There was a nominal flow of information down to the IPA from the
medical director and his support staff at the HMO. As a result, physicians
participating in the IPA continued to practice under fee-for-service incentives.
The medical director of the IPA, along with the leadership of the IPA, had never
supported an active utilization management program. For example, using
annual information produced by the HMO on the distribution of office visits by
specialty, the medical director had never sought to review the practice patterns
of individual physicians identified as being in the top 25 percent by cost per
visit. He had not worked with IPA leadership to develop protocols to curb
unnecessary ordering of ancillary services and diagnostic testing or to foster the
use of outpatient surgery. Rather, the focus of the IPA's managementand the
medical directorhad been directed to renegotiation of the capitation payment
to eliminate the IPAs operational deficit.
Case Study 5: Physician-Sponsored Ipa
Background
A physician-owned IPA model HMO with under 25,000 members and a
panel of more than 1,000 physicians had consistently reported operating losses
since its founding in the mid-1980s. In 1987, it was reported to be out of
compliance with the capital reserve requirements of the state's insurance
department. Accordingly, the IPA sought to determine how best to obtain

APPENDIX B 241
an immediate infusion of funds. Two primary sources had previously been

identified: additional contributions from participating physicians or a loan from
one or more local hospitals. However, the analysis revealed that the physicians
were quickly becoming disenchanted with the plan and were unlikely to
contribute additional monies, while the hospitals were not anxious to loan
money unless the operational expenses of the plan could be brought under
control.
Control Type
Two-tier, with control resting with the HMO (Model 1).
The physician shareholders in the plan each originally contributed between
$600 and $1,750 to finance the plan's start-up. The HMO has operated on a
discounted fee-for-service basis, with reimbursement predicated upon the 85th
percentile of the customary and reasonable rates used by carriers operating in
the local market. After the first full year's experience, these rates were reviewed
and adjusted downward to control operating costs. In addition, the plan retained
20 percent of paid charges as a withhold; this withhold was not returned after
the initial year of operation, but was returned after the second year as a good
faith effort on the part of the IPA leadership.
There were no other incentives in this plan other than the discounted fees
and the withhold. Participating physicians who were interviewed felt that the
withhold had no impact on utilization patterns in terms of frequency of office
visits, ordering of ancillary services, or use of referrals. In fact, with the
discounted fee arrangement, some primary care physicians noted that it was
more cost-efficient to refer difficult or time-consuming patients to a specialist in
order to maintain the patient flow within their own offices.
Because of the failure of the withhold to control utilization, plan
management is now considering a preestablished budget by practice type or
specialty; under this approach, a percentage of the premium will be designated
to cover specific services (for example, family practice, obstetrical care,
surgery, and allergy), with a reserve fund created from the premium allocation.
Payback of this withhold would be based upon the combined experience of the
specialties covered under each withhold.

The claims processing section was originally under the jurisdiction of the
chief financial officer, who had no previous experience with a claims

APPENDIX B 242
operation. As a result, the operation was poorly managed find understaffed,

resulting in low morale, high employee turnover, and periodic claims lags. As
for the claims system, the existing hardware was adequate to support the
membership size, but the software lacked the necessary edits to detect unusual
physician billing practices as well as a flexible report writer to produce reports
according to management specifications.
The primary care physician functioned as a gatekeeper and in that capacity
was responsible for completing referral forms for all specialty, ancillary, and
hospital services. Once an authorization was entered into the system, all claims
were automatically paid without review, except for emergency room visits,
mental health services, physical therapy, speech and occupational therapy, out-
of-area claims, and out-of-plan referrals.
Because the IPA was originally established by the leaders of several local
medical societies and/or hospital boards and because these individuals
collectively represented the medical establishment in their communities, there
was a reluctance on the part of the nurse reviewers to challenge participating
physicians regarding their practice patterns. This problem was further
compounded by the fact that there were few or no data with which to document
consistent patterns of overutilization of services or unnecessary admissions.
The plan had two medical directors; one was largely responsible for
utilization review and the other was responsible for physician relations and for
recruiting primary care physicians for outlying areas. To date, the approach to
utilization management has been through gentle persuasion, rather than tightly
written protocols. However, the failure of the plan to bring utilization rates
down (for example, to lower hospital utilization under 375 days per 1,000
members for the under age 65 population) has resulted in a decision to
implement a more stringent review process.
Ipa Management
The same chief executive officer has remained in place since the plan was
founded. However, within the first year of its operations, the plan lost its
original chief financial officer, its director of management information systems,
its director of claims, its nursing supervisor (who was responsible for quality
assurance), and several other key administrative staff. There continues to be a
high degree of tension between the physician leadership of the IPA and its lay
management. One area of major conflict has been over underwriting practices.
Plan management had established a policy of not accepting groups with fewer
than 25 employees because of potential high

APPENDIX B 243
utilization. However, the IPA board of directors overruled management and

decreed that participating physician offices and families were eligible for HMO
membership. The offering was also extended to members of the local medical
societies. Despite evidence to show that the physician offices had been high
utilizers, the IPA physicians continued to overrule plan management.
Notes and References

1. Page, Leigh, ''New Era in Utilization Review,'' American Medical News, Vol. 1,
December 9, 1988, pp. 48-49. (See also Scheier, Ronni, "Medicine by the Book,"
American Medical News, January 6, 1989, pp. 1, 20).
2. U.S. General Accounting Office, Medicare Issues Raised by Florida Health
Maintenance Organization Demonstrations, Report to Congress, GAO/HRD-86-97,
Washington, DC, July 1986.
3. U.S. Congress, House, Maintaining Medicare's HMOs: Problems, Protections, and
Prospects, Hearing before the Select Committee on Aging, 100th Congress, First
Session, Washington, DC, June 11, 1987.
4. U.S. General Accounting Office, Medicare: Physician Incentive Payments by
Hospitals Could Lead to Abuse, Report to the Chairman, Subcommittee on Health,
Committee on Ways and Means, U.S. House of Representatives, GAO/HRD-86-103,
5. Trauner, Joan B., "The HMO Identity Crisis." Best's Review 87, April 1987, pp. 60-70.
6. The Group Health Association of America's 1988 analysis of HMO industry
performance, using 1986 survey data, showed that 39.2 percent of 181 plans submitting
financial data had a profit or surplus. Of the HMOs that we reviewed, 25 percent were
profitable in 1986. See Group Health Association of America, Inc., HMO Industry
Profile: Financial Performance, Vol. 3, Washington, DC, September 1988.
7. In a study of physician incentives in HMOs, ICF researchers noted that they were
unable to check the consistency or accuracy of information supplied by the plans. They
also noted that information on file at the federal Office of Prepaid Health Care (OPHC)
was not always reliable; in some cases, financial incentive arrangements contained in
applications for federal qualification had never become operational, and in other cases,
design of incentives had changed without being reported to OPHC. ICF, Inc., Final
Report: Study of Incentive Arrangements Offered by HMOs and CMPs to Physicians,
Submitted to the Office of the Assistant Secretary for Planning and Evaluation,
Department of Health and Human Services, Washington, DC, May 18, 1988, p. II21.
8. InterStudy, The InterStudy Edge, Excelsior, MN, Spring 1988, p. 54.
9. For a different approach, when traditional community-or foundation-based IPAs are
contrasted with newer group-based IPAs, see Welch, W. P., "The New Structure of
Individual Practice Associations, Journal of Health Politics Policy and Law, Vol. 12,
Winter 1987, pp. 723-739.
10. ICF, Inc., Final Report, p. IV-6 (see note 7 above).
11. Gnessin, Alan M., "Physician Incentive Payment Systems and Risk Sharing
Alternatives," in Group Health Association of America, Inc., Physician Incentive
Programs: Defining the Risk, Washington, DC, October 28-30, 1987.
12. For a description of an HMO in which individual physician incentives are used
within pods, see the statement by Jerome Beloff, Corporate Medical Director, Av-Med
Health

APPENDIX B 244
Plan of Florida before the Physician Payment Review Commission, Washington, DC,
July 15, 1988.
13. Gnessin, Alan M., "Physician Incentive Payment Systems and Risk Sharing
Alternatives," 1987 (see note 11 above).
14. Wilensky, Gall R., and Rossiter, Louis E, "Patient Self-Selection in HMOs," Health
Affairs, Vol. 5, Spring 1986, pp. 66-80. For a more recent study showing the impact of
dependents' claims on health plan selection see Lairson, David R., and Herd, J. Alan,
"The Role of Health Practices, Health Status, and Prior Health Care Claims in HMO
Selection Bias," Inquiry, Vol. 24, Fall 1987, pp. 276-284.
15. Neipp, Joachim, and Zeckhauser, Richard, "Persistence in the Choice of Health
Plans," Advances in Health Economics and Health Services Research, Vol. 6, 1985, pp.
47-72.
16. Borok, Gerald M., "Appropriate Utilization of Resources Program," Quality
Assurance and Utilization Review, Vol. 2, May 1987, pp. 57-61.
17. Donabedian, Avedis, "Explorations in Quality Assessment and Monitoring," The
Definition of Quality and Approaches to Its Assessment, Vol. I, Ann Arbor, MI: Health
18. Mosser, Gordon, "Quality Assurance in HMOs," Presentation to Management and
Physician Orientation Program, Group Health Association of America, Inc., New
Orleans, December 9-11, 1987.
19. For a discussion of 15 ways in which out-of-plan use may occur, see Mott, Peter D.,
"Hospital Utilization by Health Maintenance Organizations: Separating Apples from
Oranges," Medical Care, Vol. 24, May 1986, pp. 398-406.
20. For a discussion of the process used by the Rand Corporation to impute mental health
visits and expenditures at one large HMO (Group Health Cooperative of Puget Sound),
see Wells, Kenneth, Manning, Willard Jr., and Benjamin, Bernadette, "Comparison of
Use of Outpatient Mental Health Services in an HMO and Fee-for-Service Plans:
Sensitivity to Definition of a Visit," Medical Care, Vol. 25, September 1987, pp. 894-903.
21. Joint Commission on Accreditation of Health Care Organizations, Report of the
Findings of the Joint Commission's Quality Assurance Evaluation and Medical Records
Audits of Health Maintenance Organizations in Ohio under the Medical Assistance
Program, Submitted to the Bureau of Alternative Delivery Systems, Ohio Department of
Human Services, December 1987.
22. Conversation with Jo Ellen Ross, Chief Executive Office, California Medical Review
Inc., San Francisco, January 17, 1989.
23. U.S. General Accounting Office, Medicare: Physician Incentive Payments,
24. Mathematica Policy Research, Inc., National Medicare Competition Evaluation,
Final Analysis Report: The Structure of Quality Assurance Programs in HMOs and
CMPs Enrolling Medicare Beneficiaries, Washington, DC, February 1987.
25. Gold, Marsha and Reeves, Ingrid, "Preliminary Results of the GHAA-BC/BS Survey
of Physician Incentives in Health Maintenance Organizations (HMOs)," Group Health
Association of America, Inc., Research Briefs, Vol. 1, November 1987, pp. 1-15.
26. Findings from the BC/BS survey have been used to augment the GHAA survey, with
duplications removed. For the BC/BS results, see Blue Cross and Blue Shield
Association, "A Survey of Physician Financial Payment Arrangements in Blue Cross and
Blue Shield Plan HMOs," Chicago, January 1987.
27. Hillman, Alan, "Sounding Board: Toward Full Disclosure of Referral Restrictions
and Financial Incentives by Prepaid Health Plans," New England Journal of Medicine,
Vol. 317, December 31, 1987, pp. 1743-1748.

APPENDIX B 245
28. ICF, Inc., Final Report (see note 7 above).

29. Memorandum from Lewin/ICF, Inc. and Group Health Association of America to
Chris Bladen, Division of Health Financing Policy, U.S. Department of Health and
Human Services, Washington, DC, June 24, 1988.
30. For a review of early research studies on utilization rates in HMO and fee-for-service
settings and across HMOs, see Luft, Harold S., Health Maintenance Organizations:
Dimensions of HMO Performance, New York: Wiley, 1981.
31. See, for example, Yelin, Edward H., Henke, Curtis J., Kramer, Jane S., Nevitt,
Michael C., Shearn, Martin, and Epstein, Wallace V., "A Comparison of the Treatment
of Rheumatoid Arthritis in Health Maintenance Organizations and Fee-for-Service
Practices," New England Journal of Medicine, Vol. 312, April 11, 1985, pp. 962-967.
Also see Quick, Jonathan D., Greenlick, Merwyn R., and Roghmann, Klaus J., "Prenatal
Care and Pregnancy Outcome in an HMO and General Population: A Multivariate
Cohort Analysis," American Journal of Public Health, Vol. 71, April 1981, pp. 381-390.
32. For a study that assumed that the differential use of diagnostic testing was due to its
increased profitability to fee-for-service providers, see Epstein, Arnold M., Begg, Colin
B., and McNeil, Barbara J., "The Use of Ambulatory Testing in Prepaid and Fee-for-
Service Group Practices: Relation to Perceived Profitability," New England Journal of
Medicine, Vol. 314, April 24, 1986, pp. 1089-1094.
33. Scovern, Henry, "Hired Help: A Physician's Experiences in a For-Profit Staff-Model
HMO," New England Journal of Medicine, Vol. 319, September 22, 1988, pp. 787-790.
34. For an example of a discussion on referral patterns, see Piland, Neil E, White, Robert
E., and Smith, Howard L., "Physician Referral Patterns: Implications for Group
Practice," GHAA Journal, Vol. 7, Winter 1986, pp. 4-12. For an example of a discussion
on provider feedback, see Berwick, Donald M., and Coltin, Kathryn L., "Feedback
Reduces Test Use in a Health Maintenance Organization," Journal of the American
Medical Association, Vol. 255, March 21, 1986, pp. 1450-1454. See also Braham,
Robert L, and Ruchlin, Hirsch S., "Physician Practice Profiles: A Case Study of the Use
of Audit and Feedback in an Ambulatory Care Group Practice,'' Health Care
Management Review, Vol. 12, 1987, pp. 11-16.

APPENDIX C 246
Appendix C
Utilization Management in Peer Review
Organizations*
The utilization management procedures used by federal utilization and

quality control peer review organizations (PROs) have been shaped by the
financial incentives created by the Medicare prospective payment system
(Ermann, 1988; Gosfield, 1989; Physician Payment Review Commission, 1988,
1989). The varieties and complexities of PRO review depart considerably from
the types of prospective and retrospective review typical of the private sector.
First, preadmission review by PROs is much narrower in scope than it is for
most private programs. It focuses on only a handful of procedures rather
than all elective admissions.
Second, PROs perform little continued-stay review because Medicare's
prospective per-case system for paying hospitals is thought to be sufficient
to discourage needlessly long hospital stays.
Third, Medicare does not provide for high-cost case management, although
it is planning four demonstration projects to assess the method's value in
the context of prospective hospital payment.
In addition, compared with most private review programs, PROs are
required to perform a broader array of retrospective reviews of medical records
to assess the quality and necessity of care. For example, they must examine a 3
percent random sample of Medicare admissions and
* The initial draft of this appendix was prepared by Alice G. Gosfield.

APPENDIX C 247
must also review specified samples of other cases: patient transfers among
hospitals, length-of-stay and cost-per-case outliers, and readmissions within 15
or 31 days of a previous discharge. Moreover, if a reviewed provider or
practitioner exceeds a predetermined denial rate (six cases or 5 percent of the
cases reviewed in a quarter, whichever is greater), a more intense level of
review may be initiated. Altogether, the required review activities for PROs
cover about 25 percent of all Medicare discharges, a considerably higher
percentage than the committee was aware of in private programs.
Beyond these differences in review loci in the PRO program, the essential
distinctions between PRO review and private utilization management stem from
two factors: (1) the direct relationship between the provider and payer in
Medicare, and (2) the public law principles that guide PRO procedures.
In Medicare, the responsibility for initiating preadmission review lies with
the hospital or physician, not the patient. Failure to obtain the relevant
certification may lead to a penalty for the provider, not the patient. The provider
is also responsible if the PRO denies payment based on retrospective review of
the necessity of care.1 Generally, in the private sector, a contractual relationship
must exist between the payer and the provider in order for the patient to be
"held harmless" (without financial liability) in similar circumstances.
An even more fundamental difference between PROs and private review is
founded in the public nature of PROs. Their agendas, objectives, criteria, and
results are public. The manner in which they operate is subject to federal
regulation and published guidelines and is informed by public law principles,
including principles of due process. PROs are required by the Health Care
Financing Administration (HCFA) to seek physician comments on their review
criteria, to provide their criteria to state medical organizations, and to send the
criteria to anyone requesting them.
The Congress and the HCFA take a very directive role with the PROs.
Each contract period is covered by an extensive scope of work that lays out the
work expected. This scope of work is accompanied by an array of
1 In 1972, Congress recognized that the burden of a retrospective claims denial under
Medicare fell unfairly on the beneficiary who incurs Medicare costs as a result of orders
and charges generated by the physician and hospital. Therefore, in Section 1879 of the
Social Security Act (1972), Congress established that where services were denied as not
medically necessary or custodial, if the patient did not know and had no reason to know
the care would be denied, Medicare would pay for the unnecessary services, waiving the
patient's liability. The burden would then fall on the hospital and physician, and a
judgment would be made as to whether they knew or should have known that the
services would be deemed medically unnecessary. If they did not have the requisite
knowledge, they too would be indemnified by Medicare. More recently, physicians have
been required to reimburse Medicare patients who have paid for care (for example,
through coinsurance) that was later determined to be inappropriate.
APPENDIX C 248
federal regulations and other rules governing PRO activities. About half of the
original group of PROs failed the evaluation by HCFA for their first contract
period, and only 15 received automatic renewals.
In addition to regular performance monitoring by HCFA, a private
contractorthe so-called Super-PROaudits samples of each PRO's decisions
and compares its judgments of appropriateness with those of the PRO. The
Super-PRO tends to judge more care as being inappropriate than the PROs do.
The Super-PRO also has reviewed a small selection of the review criteria used
by PROs, but there was no provision for more systematic evaluation of the
criteria (Physician Payment Review Commission, 1988). Despite this extensive
oversight, the U.S. General Accounting Office recently issued a report noting
the "extreme variation" in PRO organizational structure and activities (General
Accounting Office, 1988).
The public aspect of PRO operations is also reflected in the legislative
provisions both for physician and patient appeals of PRO determinations. For
the attending physician, the process may include a predenial discussion with a
review physician, a postdenial reconsideration by a specialist (if possible), and,
in limited circumstances, appeals to administrative law judges and even judicial
reviews. Patient access to administrative and judicial review of denials is more
generally available. Federal PRO law establishes that the PRO's determinations
are binding upon claims payment agencies (intermediaries and carriers) on the
matters assigned to PROs. The degree to which a private review entity's
determinations are binding on claims administrators varies, as described in
Chapter 3.
Federal law establishes that PRO norms, criteria, and standards for review
must be public and must reflect at least regional patterns of practice. The basis
of the standards in the community standard of care is so presumed in the law
that a specific malpractice exemption is provided to any practitioner or provider
relying upon the PRO norms and criteria in treating patients or reviewing
utilization. The law states that no practitioner and no provider can be held
civilly liable under any law of the United States or of any state on account of
action taken in reliance upon PRO norms, criteria, and standards, provided that
he or she otherwise exercised due care. Enacted in 1972 and intended to
discourage expensive defensive medicine and encourage cost containment, the
provision was reenacted verbatim in the PRO law. Although this provision has
never been construed by a court and is little known by the provider community,
in the overall design of a public utilization management program it may well be
instrumental in garnering physician conformity. It also sharply distinguishes
PRO review from private sector review.
Another distinguishing feature of PRO review is its range of potential
sanctions. The law provides that each provider and practitioner is obligated to
render medically necessary services in an economical setting, meeting

APPENDIX C 249
professionally recognized standards of quality. Failure to do so can subject an

errant provider or practitioner to fines or exclusion from Medicare. This
sanction authority has been increasingly used over the past few years. There are
two types of sanctions: A substantial failure to comply in a substantial number
of cases presents a basis to penalize a practitioner for persistent overutilization
problems; a single gross and flagrant violation of the statutory requirements
involving quality of care can also trigger a sanction. This type of approach to
patterns of care and inclusion of providers and practitioners in an insurance
program rarely exists even in the context of HMOs.
Finally, it cannot be ignored that since their inception in 1984, PROs have
increasingly focused their attention on quality issues. Moreover, the
Consolidated Omnibus Budget Reconciliation Act of 1985 (P.L. 99-272) gave
PROs the authority, not yet implemented at the time of this writing, to deny
payment when services fail to meet quality standards. PROs or similar
organizations are now to review the quality of care in HMOs, and they are
required to consider whether accreditation agencies and state licensing boards
should be notified when a possible quality problem is identified.
References
Ermann, Danny, "Hospital Utilization Review: Past Experience, Future Directions," Journal of
Health Policy, Politics and Law, Winter 1988, pp. 683-704.
Activities, GAO/PEMD-88-10, Washington, DC, November 1988.
Gosfield, Alice, "PROs: A Case Study in Utilization Management and Quality Assurance," 1989
Health Law Handbook, New York: Clark Boardman Co., Ltd., 1989.

APPENDIX D 250
Appendix D
Summary of Public Hearings*
The Institute of Medicine Committee on Utilization Management by Third

Parties held a public hearing on June 6, 1988, at the National Academy of
Sciences building in Washington, D.C. Speakers from 27 organizations made
presentations to the committee. A question and answer session followed each
panel of three speakers. Eight groups submitted written testimony without any
oral presentation.
Each of the organizations represented fell into one of five categories or
interest groups (Table D-1 lists the organizations by category):
Practitioners and Organized Medicine
Health Care Institutions, Associations, and Suppliers
Patients, Consumers, and Public Health Organizations
Insurers and Utilization Management Firms
Trade Associations and Other Organizations
The testimony reflected diverse sets of interests and perspectives on
utilization management. There were differences of opinion over the appropriate
role of the physician (and/or medical profession) in utilization management; the
validity of the criteria currently being used; the impact of various approaches on
cost, quality, and administration of services; criticisms or shortcomings of
utilization management; and suggestions about what is needed for the future.
* This summary was prepared by Eileen Connor.

APPENDIX D 251
TABLE D-1 Organizations Presenting Testimony at the Public Hearing on Utilization

Management by Third Parties
Practitioners and Organized Medicine
American Academy of Child & Adolescent Psychiatry
American Academy of Otolaryngology
American College of Physicians
American College of Utilization Review Physicians
American Dental Association
American Medical Association
American Psychological Association
American Rheumatism Association
Coalition to Preserve Quality (written only)
Health Care Institutions, Associations, and Suppliers
American Hospital Association
American Pharmaceutical Association (written only)
Hospital Association of Pennsylvania (written only)
National Association of Ambulatory Care
National Association for Private Psychiatric Hospitals
Mayo Clinic (written only)
Patient, Consumer, and Public Health Organizations
American Public Health Association
National Health Law Program
Public Citizen-Health Research Group
Insurers and Utilization Management Firms
ALTA Health Strategies, Inc.
Blue Shield of California (written only)
Celtic Life Insurance
Health Care COMPARE
Health Data Institute
Health Management Strategies International, Inc.
Iowa Foundation for Medical Care
Quality Standards in Medicine, Inc. (written only)
U.S. Administrators
Trade Associations and Other Organizations
American Association of Preferred Provider Organizations
Blue Cross and Blue Shield Association
Group Health Association of America
Healthcare Financial Management Association (written only)
Health Insurance Association of America (written only)
InterQual
Joint Commission on Accreditation of Health Care Organizations
National Association of Quality Assurance Professionals
Despite the differences of opinion, however, there was considerable

agreement on the following:
1. Utilization management is dynamic; it is evolving; studying utilization
management now is like trying to focus on a moving target.
2. There is a proliferation of external review entities in the marketplace
with different criteria and a variety of approaches to managing utilization.
3. There are variations in medical practice.
4. Criteria for appropriate medical care are imperfect.

APPENDIX D 252
5. Resources are limited (There is disagreement as to how limited, that is,

how much the United States is willing or able to devote to health.)
6. Utilization management highlights the quality and cost debate in health
care.
7. There are potential dangers in utilization management by third parties.
8. Utilization management does not seem to influence physician practices.
(There is disagreement on how, why, and if it is good or bad.)
9. Utilization management needs physician involvement. (There is
disagreement on the type and amount of physician involvement.)
10. Current utilization management programs do little or nothing in the
areas of outpatient and office practice and/or monitoring for underservice.

APPENDIX E 253
Appendix E
Summaries of Committee Site Visits To
Utilization Management Organizations*
Members and staff of the Committee on Utilization Management by Third

Parties visited 12 organizations during the summer of 1988. The organizations
visited included three independent companies, three insurance company
subsidiaries, two third-party administrators, two peer review organizations
(PROs) engaged in private review, and two health maintenance organizations
(HMOs) (one staff model and one independent practice association [IPA]). The
sites were selected to convey something of the variety that currently exists in
utilization management. The following summaries and Tables E-1 through E-4
at the end of this appendix briefly describe important features of the
organizations visited. Appendix F provides an analysis of client contracts that
were obtained from seven of the organizations visited.
Organization 1
Organization 1 is a relatively small, independent organization that handles
about 3,000 cases a month. It is a privately held, for-profit organization whose
leaders come from professional standard review organizations
* The summaries and tables in this appendix were originally drafted by Susan Sherman
and edited by Bradford H. Gray. Each organization reviewed its summary for accuracy.
Eileen Connor undertook further editing of the summaries and tables. The data provided
were not independently verified.
APPENDIX E 254
(PSROs) and academic health centers. Clients are third-party administrators and
insurers, and the company sees itself as applying not only its own review
criteria but also as applying the more general coverage provisions of clients'
benefit plans.
A ''standard" list of procedures that normally should be done on an
outpatient basis is used to guide decisions about the appropriateness of
proposed inpatient care. The details of the list, however, may vary by client.
Allowances for preoperative days also vary by client. Criteria are modified on
an ad hoc basis as issues are raised by reviewers. The organization also
performs prior review of the medical need for certain procedures, which are a
mix of inpatient and outpatient services.
The organization's services are limited to utilization management and
include preadmission review, second-opinion screening, high-cost and
psychiatric case management, bill audits, claims review, retrospective
utilization review, and physician adviser services for in-house review programs
operated by insurance companies. Most reviews take place by telephone, and
the information is entered into a computer. Some psychiatric case management
and most retrospective reviews are conducted at the site of service. Data
analysis and program evaluation capabilities are limited because the
organization has only its own activity data, not claims data or medical records.
Although patients are responsible for seeing that the prior review process
occurs, most calls actually come from hospital staff. All required information is
obtained on the first call in an estimated 80 percent of cases. The process works
best when the reviewer calls the physician's office after receiving notification of
an impending admission. Incoming calls are answered by a receptionist who
refers them to review nurses, who collect the information and approve the
admission if the criteria are met. If the criteria are not met, the nurses are
authorized to negotiate changes with attending physicians. Nurse reviewers are
expected to use their own clinical judgment. (It is acknowledged that reviewers
sometimes "feed" the criteria to attending physicians to facilitate admission of a
patient that the reviewer feels should be admitted.) When nurse reviewers find
themselves unable to authorize proposed services, cases are referred to
physician advisers, who serve part-time but work from the organization's office.
Appeals of denied certifications go to a second physician adviser.
Some clients request that all reviews be done by physicians. In these
situations, referrals come to the organization's physician review unit from the
client's nurse reviewers by telephone or facsimile machine. Incoming calls are
answered by an intake coordinator who enters the information into the
organization's data base and prints the referral form for distribution to the
appropriate physician specialist. Faxed referrals are handled in a similar
fashion. The physician reviewer will, if necessary or requested, call

APPENDIX E 255
the client's nurse reviewer to discuss the case and then call the attending
physician to make a certification decision. The physician reviewer will inform
the attending physician of the decision. The final decision is also faxed to the
client's nurse review office.
In the high-cost case management program, the patient calls the insurer or
claims administrator, who validates health plan eligibility and then refers the
case to the case management unit. For those clients where site-of-service case
management is operational, preadmission review is conducted by telephone,
while emergency admission and all continued-stay reviews are conducted at the
site of the service. The monitoring of aftercare or alternative care is done by the
nurse case managers and physician case managers in the office. Obviously, for
those clients for which case management is telephone-based, all levels of
review are conducted directly out of the organization's office. The case
management program uses a team approach, with nurses and physicians
working together in the same unit.
The organization believes in intensive involvement of physicians in the
review process, in the importance of informal communication among physician
advisers, and in seeking cooperative relations with providers. The office is
small, and much of the monitoring, sharing of information, and revision of
procedures or criteria tends to emerge from the informal communication among
nurses and physicians. More formal communication occurs through regular,
scheduled medical review committee meetings that assess the quality of
decisions by physician reviewers and evaluate the need to change review
criteria. The committee is made up of senior physicians who are not involved in
initial reviews.
The staff emphasizes cooperation with attending physicians. Although they
are willing to deny certification, they think that to deny cases ''too readily will
alienate doctors." Rather, they encourage behavior change through effective
negotiating skills by nurse and physician reviewers. And they "will bend over
backwards" to certify a case that they deem a necessary admission, regardless of
criteria. Overall, they find physicians to be generally accepting of their
programs. Most days saved are as a result of persuasion, not denial.
Cost savings are measured by hospital days saved; the organization
believes that it is helping to reduce inpatient days per 1,000 lives for its clients.
Except for the case management program, no data on quality and
appropriateness of service or on patients and families are collected.
Retrospective reviews are done for several clients. The organization believes
that its biggest impact has been in influencing a switch of some inpatient
procedures to outpatient procedures, including cataract surgery, myelogram,
cardiac catheterization, hemorrhoid surgery, hernias, and bronchoscopies, and
in diverting emergency and inappropriate psychiatric and chemical dependency
treatment from inpatient to alternate forms of care.

APPENDIX E 256
The organization expressed concern about the adequacy of the clinical

research base for some review criteria. There was also concern that some
criteria developed by medical specialty societies include too broad a range of
acceptable services. It was noted that refusing to allow a preoperative day may
be self-defeating if the addition of such a day would shorten the overall length
of stay.
Organization 2
Organization 2 is a wholly owned subsidiary of a third-party payer and was
incorporated in 1985. Clients are mainly those of the parent payer, but it is
branching out to others. It covers about 1.12 million lives and reviews about
2,000 cases per month. Services include preadmission review; admission review
and concurrent review for medical, surgical, and psychiatric cases; case
management; second surgical opinion; disability review and workman's
compensation review; and hospital bill auditing.
The organization has developed a detailed set of administrative policies
and procedures and comprehensive clinical screening criteria. Nurse reviewers
are guided through the decision-making process by a criteria manual, which
lists procedures, gives conditions for certifying an admission, and assigns a
rating to conditions denoting the likelihood of an admission for this condition.
Nurses must apply the criteria in all cases and consult a physician adviser if the
information given does not meet the screening criteria. They must document
their reasons for approval or referral to a physician adviser. The organization
emphasizes that it requires its nurses to have at least 5 years of clinical
experience. Physician advisers are local and work from their own Offices.
Advisers may sometimes know the physicians they are reviewing.
The cost-effectiveness of services and the cost-benefit of programs are
emphasized rather than simple cost savings. The staff believes that cost savings
from utilization management will decrease after the initial impact has been
achieved, except in the mental health field where there is a greater potential to
achieve the desired goals. The staff expressed skepticism about most measures
of cost savings used by other companies. In this organization, pricing is based
on the client company's inpatient utilization.
The organization is becoming more concerned about reviewing
appropriateness of care and may call in a physician adviser on cases in which
the quality of treatment is questionable. There is an internal quality assurance
program, and the organization is now undertaking an outcome-based program to
evaluate the appropriateness of decisions made by review coordinators and
physician advisers.
The review process is initiated by the patient, a family member, or the

APPENDIX E 257
provider. The organization expects employers to educate employees about the

program. The organization has a full-time account services representative to
assist employers with that education. When nurse reviewers are notified that an
admission is planned, they call the attending physician to get the reason for
admission, other clinical information, the anticipated length of stay, and the
treatment plan. About 90 percent of the time, the reviewer speaks to the
physician's office staff for medical and surgical admissions. For psychiatric
admissions, the attending physician is almost always called. If the admission is
authorized, the nurse notifies the attending physician, the patient, and the
hospital. At the end of the certified length of stay, the nurse reviewer calls to
verify that the patient is being discharged. If the patient needs additional time in
the hospital, then the continued-stay review process continues until the time of
discharge. If the nurse reviewer cannot approve a case based on the criteria, he
or she will refer the case to a physician adviser, who must call the attending
physician or document that they have made a reasonable attempt to discuss the
case. Most referred cases involve requests for extensions of length of stay.
About 2 percent of the total cases are denied. Appeals can be made to a second
and then a third physician adviser. Of the denied cases, about eight or nine have
been appealed.
Organization 2 does not currently review for quality of care but may refer
problem cases to a physician adviser for review. The company surveys patient
satisfaction by sending out a postcard questionnaire, for which there is a 15
percent return rate. It has observed a sentinel effect on physician requests for
admissions and lengths of stay. The organization believes that it has had its
biggest impact on cutting preoperative days, shortening lengths of stay, and
increasing consumer awareness.
The organization expressed the following concerns about its current
processes. (1) Attending physicians may bill patients for their time spent on the
telephone with reviewers, and these charges are not usually covered by
insurance. (2) All of the physician advisers work out of their own offices. In
addition, the medical director and physician adviser work at the organization's
offices part-time. Physician adviser specialists are reluctant to make calls to
physicians who work in the same metropolitan area. The organization's
management believes it is necessary to have at least one physician adviser on
staff to ensure accessibility and consistency. (3) Patients are often uninformed
about their benefit coverage. It is possible for a reviewer to certify a procedure
as medically necessary, but to have that procedure not be covered by the
patient's benefit package. However, letters sent by the organization to the
patient clearly state that the days certified are certified for medical necessity
only, and there may be contractual exclusions and/or limitations on coverage of
these services in their health plans.

APPENDIX E 258
Organization 3
Organization 3 is a third-party claims administrator that began to offer
utilization management services about 1 year before the site visit. It covers
about 2.3 million lives and has about 700 medium-sized companies (5,000 to
15,000 employees) as clients. Services include preadmission review, continued-
stay review, retrospective review, discharge planning, second-opinion surgery
(focused on specific procedures using criteria-based waivers), case
management, and outpatient surgery review. It also conducts hospital bill audits
and offers health information services and referral services to preferred
providers.
Because this company does claims administration, it has a vast amount of
data on its program and its impact on patients and providers. It can track case
histories of patients, review benefits packages of patients, check outcomes of
care in terms of rehospitalizations, and measure savings in terms of costs. A
variety of reports can be generated on every aspect of its utilization
management activities, including reports on days approved and used by
diagnosis, extensions, averages for days requested and days approved,
variances, estimated savings, and readmissions. Audits are done on both
hospital services and physician services. The company considers data analysis
and reporting one of its four main functions (the others are utilization
management itself, claims administration, and preferred provider referencing).
This company is very client-oriented. It markets its programs as a service
to employers. It believes it offers an added benefit to employees, guiding them
through the maze of health care services. Decisions are rendered on the medical
necessity of services and the reasonableness of provider charges to clients.
To maintain the quality of the review process, Organization 3 monitors
almost every activity of the nurse reviewers, and reviewer comparisons are
made across time. Reviewers have 5 years of clinical experience and some
review experience. They go through a 3-week training program. Physician
advisers are local practitioners who spend about 3 hours a week conducting
reviews out of their own offices. Their decisions are monitored to see how often
they uphold a nurse reviewer's recommendation to deny certification.
Nurse reviewers are allowed to negotiate with providers and may use their
own clinical judgments. The company uses the Health Data Institute Optimed
system, but it believes that the review criteria are too liberal. It plans to switch
to a system of its own. Reviewers are allowed to override criteria with a
supervisor's permission. All of the review information, including the criteria, is
programmed into the organization's computer system.
Patients are expected to trigger the preadmission review process.

APPENDIX E 259
About half of calls come from patients, and about half come from
physicians and hospitals. If the patient or the hospital has made the initial call,
the nurse reviewer will call the physician's office to confirm the diagnosis and
treatment plan. If the nurse reviewer approves the admission, he or she assigns
the case a length of stay (based on the Professional Activity Study Western
Region at the 50th percentile minus 1 day). The reviewer calls back on the day
before the expected discharge date to verify the discharge. If the case is to
involve a surgical procedure, the reviewer determines whether a second surgical
opinion is necessary. The opinion may be waived by the reviewer on the basis
of criteria indicating that there are solid indications for surgery.
If a case does not meet criteria and the reviewer cannot negotiate a change
with the attending physician, he or she refers the case to a physician adviser.
Most referrals concern inpatient versus outpatient decisions. About 75 percent
of the referred cases are denied by the physician advisers. Appeals may be made
for consideration by a second physician adviser. Less than 1 percent of cases are
appealed.
During its 1 year of utilization management experience, the company says
that it has brought about an average 12.5 percent reduction in admissions, a 32
percent reduction in days of hospital care per 1,000 employees, and a 26 percent
reduction in hospital costs per covered person for its clients. The company
monitors quality and appropriateness of care and patient satisfaction by
conducting retrospective chart reviews and by having reviewers make follow-up
calls to all patients after they return home. This follow-up also allows the
organization to confirm information given upon admission about diagnoses and
treatment. The biggest area of impact is believed to have been in moving
tonsillectomies and cataract surgery to outpatient settings.
Staff expressed concerns about how data are used, how impact is
measured, and how savings are calculated. They expect to be able to track the
appropriateness of their decision-making, outcomes of care, and impact in a
much more sophisticated manner as they update and improve their own data
base. A more sophisticated data base is necessary to enable the organization to
modify its review criteria appropriately. It expects to use more restrictive
criteria reviews in the future to further reduce lengths of stay and inpatient
admissions, and the staff wants to use data to justify and monitor these changes.
More retrospective review of cases is planned.
Organization 4
Organization 4 is an independent review organization with origins in a
staff model HMO. It was one of the earliest organizations to apply preadmission
review in an indemnity context. The company views its

APPENDIX E 260
business as a health care cost management business, and it serves a variety of

clientsboth insurers and employers. The goal is to get the most cost-effective
care for the employers. It offers a full range of utilization management services.
The company strives to eliminate all inappropriate and unnecessary care.
Reviews are concerned not only with inpatient versus outpatient surgery,
presurgical days, and length of stay, but also with medical necessity of
admissions, necessity of surgery, and necessity of expensive outpatient testing
and procedures (for example, nuclear magnetic resonance imaging and
lithotripsy). Criteria are developed by panels of outside physicians, who build
protocols using existing criteria sets, specialty society guidelines, and their own
clinical judgment. Some decisions (for example, those regarding outpatient
surgery) are based on clients' benefit plans rather than the company's own
criteria. Information used for medical necessity determination includes all
available clinical data and also the narratives for x-ray and lab work.
This company is at the high-tech end of utilization management in both
computerization and telephone systems. It strives for a paperless process from
initial phone call until discharge, except in those instances in which case
management is involved. The system captures 150 data elements, including
clinical information beyond just diagnoses and procedures.
Calls for review come to an operator who directs each call to one member
of a team of review nurses (the goal is to do this within 30 seconds), who enter
data into the computer. The average call lasts less than 5 minutes. Nurses
collect and enter data but can neither certify nor deny certification for cases. All
cases are transmitted electronically to a member of the full-time medical staff
for review. An estimated 25-30 percent of cases involve virtually automatic
decisions (for example, admission for childbirth), and an estimated 60 percent
present significant complexities for the reviewing physician to evaluate. The
review physician calls the attending physician in virtually all mental health and
catastrophic rehabilitation cases, 65-70 percent of medical cases, 40 percent of
surgical cases, and 15 percent of obstetrical cases. As many as 20-30 percent of
the initial negative decisions are appealed, first to another staff physician and
then to an external consultant. The company reports utilization levels
comparable to those of a staff model HMO.
The organization emphasizes internal quality control. Nurses are
monitored through the computerized telephone system by supervisors and
physician reviewers. Physicians' decisions are reviewed by a team supervisor,
and a random 10 percent of cases are sent each month to members of external
advisory panels for review. They prefer to report to clients on before-and-after
utilization levels rather than on days of care averted.

APPENDIX E 261
Organization 5
Organization 5 is a third-party administrator that has been doing utilization
management since the early 1980s. It has about 40 clients, most with 5,000 or
more employees, and covers about 600,000 lives. Its services include benefit
plan design, preadmission review, concurrent review, discharge planning, case
management, second surgical opinions, and preferred provider organization
(PPO) management and referral, bill audits, and claims payment.
Though committed to cost containment, this organization also takes a
strong pro-employee/patient stance. The staff seeks to eliminate care that is of
questionable medical necessity. It believes that patient cost-sharing is an
important element of cost containment but seeks to have employers also hold
patients harmless for costs incurred when providers do not follow review
requirements or when bill audits identify unnecessary services. It contends that
it is on solid ground in making those decisions and will go to court to defend its
judgments. It considers assisting clients with their benefits packages to be an
integral part of its service. Staff are also directly involved in educating
employees about utilization management. Physician advisers and nurse
reviewers meet with employees, benefits personnel, and representatives of
physicians and hospitals in the communities at key locations of new clients. The
physician or nurse reviewer may spend up to 2 weeks at a client company
holding small group educational sessions. The company's toll-free 800
telephone number is also open to all employees 1 month before services are to
begin, to enable employees to call in for information.
As a claims payer, Organization 5 has extensive data on its review
programs and on actual utilization. It offers clients a wide selection of standard
reports and will prepare ad hoc reports as clients request.
The staff uses internally developed criteria that they base on InterQual and
Professional Activity Study lengths of stay, but nurse reviewers and physician
advisers can make exceptions based on clinical judgment. A medical advisory
committee of physicians from across the country develops and modifies criteria.
They hold that the burden is on the provider to show that services are necessary,
not on the review organization to show that they are unnecessary.
In the preadmission review process, about half of the initial calls come
from patients, and about half come from physicians' offices. Staff members say
that they can get most of the needed information from patients and hospitals
"without disturbing the physician." Nurse reviewers can certify cases
immediately if they meet the criteria, or they can negotiate with providers'
offices. If an attending physician challenges the nurse reviewer, the case goes to
a physician adviser. About I percent of the cases are

APPENDIX E 262
referred. Most referrals are handled by the medical director (who works at the
organization's offices nearly full-time) and four other advisers. The advisers
may call specialists for advice. A majority of the referred cases are denied.
Appeals are made directly to employers through benefit plan grievance
processes.
Savings are measured by days of care averted. The organization also
monitors provider charges to evaluate whether they are within reasonable and
customary rates according to its own and Health Insurance Association of
America data. It can address quality of care to some extent through bill audits
and through second opinions if the proposed service seems unusual. The
organization solicits feedback by sending comment cards to patients who have
gone through utilization management. It believes that its impact has been
greatest from disallowing preoperative days, managing long-term-care cases,
arranging transfers to lower levels of care, and getting concurrence to outpatient
rather than inpatient care.
The leaders of this organization expressed the following concerns. (1) The
preadmission review system is being evaded by patients who are admitted on an
emergency basis for diagnostic workups only. They seek to have these patients
discharged. (2) About half the hospitals they deal with do not cooperate with
concurrent review. Patients may end up paying for services not certified for
payment by the employee's benefit plan. (3) Staff described "rolling"
laboratories, mobile labs that offer complete testing for patients and then submit
big bills for comprehensive, unnecessary tests. Most of these claims are denied.
(4) Many companies lack the baseline data needed to evaluate changes in costs.
The organization plans to do more outpatient reviews and focused reviews
based on historical data. It is developing a data base with claims histories for
each patient.
Organization 6
Organization 6 started doing disability and rehabilitation management and
workman's compensation review in the 1970s. It then moved into medical case
management and, more recently, into prior review services. A subsidiary of an
insurance company, it now offers preadmission review, continued-stay review,
second surgical opinion referrals, discharge planning, disability and
rehabilitation management, vocational rehabilitation, hospital bill and provider
auditing, and case management. It covers 8.5 million lives for large and small
companies, insurers, third-party administrators, and self-insured employers.
Because this company's program grew from case management, it
emphasizes appropriateness of services and review of the entire spectrum of

APPENDIX E 263
services a patient receives. The staff is very patient-oriented. For example,

reviewers will arrange patient appointments for second opinions, and case
managers will help identify new jobs for partially disabled employees through
the vocational rehabilitation program. Case management is done face-to-face
with patients. The company has more than 100 local offices for the case
management program. Case managers typically make several visits to patients
in their homes to monitor the services they are receiving.
The company also emphasizes data collection, auditing, and monitoring of
their services. The size of the quality assurance staff is proportional to the
volume at each service center. One to five quality assurance staff per site is the
range. Every month the quality assurance staff reviews a 2 percent sample of
preadmission review cases, examining 14 different elements such as timeliness,
accuracy of data entry, and application of criteria. They also review a 5 percent
sample of physician advisers' decisions for consistency. Physician advisers also
review a sample of one another's cases each month to monitor accuracy and the
appropriateness of the determination. The quality assurance staff also tracks
second-opinion referrals and provider billing accuracy. The company also has a
sophisticated telephone monitoring system that tracks call volume, duration,
hold time, abandon rates, and turnaround time. Daily reports on telephone
activities are reviewed by a supervisor, who also regularly evaluates reviewers
on their telephone performance.
The company assesses the medical necessity of admissions and the
appropriateness of the level of care. It does not accept responsibility for
determining whether services are otherwise covered under a client's benefit
plan. The staff does, however, remind callers to check the provisions of their
benefit plan. The organization uses a modified version of InterQual to judge the
medical necessity of services. It modifies the criteria on the basis of its own
data and other information data on medical practice. A panel of physicians
approves modifications. The inpatient and outpatient lists vary, depending on
client benefit plans.
Nurse reviewers work in teams to serve particular clients. Nurse reviewers
have hospital and review experience and are allowed to use their own judgment
in conjunction with the organization's written criteria. They can negotiate with
providers and can make exceptions to the guidelines. All nursing judgments
must be accompanied by supportive documentation in the patient's file. An
administrator of the company said "we don't want to keep [the nurses] on a ball
and chain." Physician advisers use their own judgment to decide cases. They
work in the company's office 1 day per week; one physician adviser is on-site
each day. Management suspects that nurse reviewers are more aggressive than
physician advisers in negotiating lower lengths of stays or use of outpatient care.
The organization's review process is intended to be paperless. The

APPENDIX E 264
reviewer enters information into the computer system as it is obtained and as

decisions are made. Paperless processing eliminates the need for participating
employees to submit forms. Through computer linkages to claims entities, the
organization also has the capability of notifying the payers that prior review
obligations have been met.
Patients are expected to trigger the review process. Most calls come from
patients, families, and hospitals, but the nurse reviewer must always contact the
attending physician to obtain information about the admission. About 120 data
elements are collected on each patient. If criteria are met, reviewers may
approve the admission; they call back on the day prior to discharge to verify the
discharge or conduct continued-stay review if further days are requested. About
5 percent of the cases are referred to physician advisers. The most contentious
cases involve preoperative days, overnight stays for outpatient procedures, and
requests for longer stays, in that order. Most referred cases are approved by the
advisers. About 1-2 percent of all cases are ultimately denied. Appealed cases
receive a second review by a specialist in the local area. The company also has
a complaint resolution process.
The company sees an average reduction of 10 percent in their clients'
hospital inpatient costs and a reduction of 18 percent in the ratio of days
requested to days certified after it implemented review. The company does not
review for quality of care, but when quality problems appear in continued-stay
review, the case is referred to a physician adviser, who may counsel the
attending physician. Repeat admissions are referred to the case management
program. Provider profiling is planned for the future. It is difficult for the firm
to measure the impact of its programs because it does not pay claims or have
other access to information about final payment decisions.
The company expressed the concern that current utilization management
programs are only shifting health care costs to the outpatient setting and
suggested that utilization controls on outpatient Services are needed. Products
for outpatient reviews are currently being developed. The staff also believes that
a number of diagnostic tests and procedures are being ''abused" and need to be
monitored. A methodology to track overused tests and procedures is being
developed.
Organization 7
Organization 7 is a subsidiary of a regional third-party payer. Its review
program is based on a model developed by a study committee sponsored by the
Robert Wood Johnson Foundation. It has been engaged in utilization
management since 1985. The organization services 1.7 million subscribers
(including dependents). It offers preadmission review, admission review

APPENDIX E 265
(emergency and maternity cases), home health review, skilled nursing facility
care review, intensive psychiatric review, continued-stay review, second
surgical opinion, medical and psychiatric case management, discharge planning,
medical audits, workman's compensation and disability management, consumer
education, benefits analysis, and hospital bill auditing. It also markets software
products to support health information systems.
The company has a strong local base. More than half of its clients are
located in the area. The company is oriented toward serving the employees of
its client companies. It sees utilization management as a service to employees,
as part of the employees' benefit package. Several of the major figures in the
company have backgrounds in benefits and insurance. The company routinely
conducts employee education programs to describe the details of the review
program. Company representatives attend client meetings and union negotiation
sessions. It has received endorsements from major employee groups.
The company emphasizes service to the employee population and is
concerned that people understand their treatment options. Changing practice
patterns is viewed as an evolutionary process.
An effort is made to minimize interference with individual physician-
patient relationships. Nurse reviewers can certify admission based on
established criteria and accepted medical practice and can negotiate with
attending physicians regarding the appropriate level of care. If that fails, cases
are referred to physician advisers. The physician adviser contacts the attending
physician. The company views one aspect of its patient advocate role as
preventing situations in which patients may see themselves as caught in the
middle between their physician and the review organization.
The review process covers the medical necessity of admissions and
verification that proposed services are covered by the benefit plan. The
company uses a modified version of Appropriateness Evaluation Protocol
criteria as guidelines and negotiating tools, rather than absolute standards.
Company leaders stated that the objective is to affect medical outliers and not to
lead medical practice trends but to follow clinical consensus. The list of
procedures subject to second opinion varies with the client benefit package. The
firm's quality assurance team has developed a rapid assessment tool for selected
procedures and diagnoses. This tool assists the staff in establishing and
maintaining the clinical picture for a patient from admission to discharge.
Nurse reviewers are assigned to teams. They must have 5 years of clinical
experience; most have more. Many continue to work a few hours a week at
local hospitals. Nurse reviewers can certify cases, negotiate with attending
physicians, override existing criteria, and waive second surgical opinions. They
are encouraged to utilize their medical expertise and are provided with
continuing education. The nurse reviewers spend about half

APPENDIX E 266
their time educating patients about treatment options, resources, and other
health matters as part of a patient consultation module.
The company currently has a contract with an independent physician
adviser network that includes board-certified physicians with utilization
management experience. Only about eight advisers are dealt with on a daily
basis. The private physician group also reviews modifications in the company's
written review criteria. The organization plans to phase out the use of this panel
in favor of a team of on-site physician advisers.
Telephone calls from physicians, hospitals, or patients can trigger the
preadmission review process; no forms are needed. About 3-5 percent of cases
are referred from nurse reviewers to physician advisers. Appeals are handled by
a special review committee.
Through 1987, the company reported a 21 percent reduction in hospital
admissions per 1,000 subscribers and a 20 percent reduction in days per 1,000
subscribers. It collects data on quality of care and checks for provider
noncompliance with review programs. The company has observed that it has
had a sentinel effect on physicians and has virtually eliminated preoperative
days. Client surveys track patient satisfaction, and a combination of
retrospective review and data analysis is used to track outcomes of care.
The organization expressed several concerns. (1) It is sometimes difficult
to explain to employers why patterns of care may differ 'among groups of
employees. Regional differences in demographics, health problems, and other
factors exist, but employers tend to see the lowest reported utilization as the
appropriate utilization for all areas. (2) Physicians need more education on the
availability and efficacy of outpatient care. (3) Some benefit plans do not
include coverage for alternative treatments or treatment settings, and this
inhibits their use.
Organization 8
Organization 8 is an independent review organization that was started in
1982 by a physician with extensive PSRO experience. The organization now
handles about 1 million employees and their dependents and does about 2,000
certifications per day. It offers preadmission review, continued-stay review,
second surgical opinion, case management, disability review, and mental health
services review. It also maintains a medical information telephone line.
The organization claims to have set the pattern for independent utilization
management. As one of the earliest utilization management organizations, it
claims to have developed the basic procedure of having nurses review
admissions and refer more complicated cases on to physicians. However, it
differs from most other organizations in that 40 percent of cases

APPENDIX E 267
are reviewed by physicians. Nurse reviewers may only certify cases that meet
the written criteria. Any cases that vary must be referred to physician advisers,
who work full-time at the organization's office. On receiving a case, the
physician adviser telephones the attending physician to discuss the case.
Physician advisers use their clinical judgment to negotiate with providers. The
organization claims to cut extra days per 1,000 lives for each client because of
this extensive use of physician review. It stresses that the goal of utilization
management is to make physicians accountable for their decisions and believes
that accountability can only be attained through objective peer review.
Physicians must be reviewed by physicians who are affiliated with the review
organization for the program to succeed.
Physician advisers have clinical experience, and the organization sponsors
continuing education. It also has made arrangements with a local hospital to
allow the physician advisers to attend grand rounds at the hospital. Moreover,
the head of the organization notes that physician advisers have contact with
10,000 admissions per year through preadmission and continued-stay reviews
and are continually educated through this process. Nurse reviewers also have
clinical experience. They receive a 6-week training program and continuing
education sponsored by the organization.
The organization claims to be the only review organization that is
operating at a profit. It has several million dollars in billings and has doubled its
business in the past year.
The review process can be triggered by physicians, patients, or hospitals.
Usually, the hospital calls the utilization management organization, and the
nurse reviewer then calls the attending physician's office. Nurse reviewers
approve about 60 percent of the cases. Most of these conform to the
organization's criteria initially; in a small number of instances, even though the
nurse reviewer is not supposed to negotiate, changes occur in the process of
obtaining information. The company's criteria are a modified version of the
InterQual criteria. Forty percent of cases are referred to physician advisers, who
negotiate changes and ultimately approve about 98 percent of the cases that are
referred to them. Sometimes physician advisers can approve a case based on
information provided by a nurse reviewer, but usually the physician adviser
contacts the attending physician to discuss the case. They deny about 1-2
percent of the cases. Appeals can be made to a second physician adviser. About
1 percent of denial cases are appealed.
The organization claims to cut their clients' inpatient health care costs by
20-40 percent. Because they conduct physician-to-physician review, they can
intercede if they perceive quality problems in a case, but they cannot actually
determine care.
The organization expressed the following concerns. (1) Its second opinion
list is too inclusive. They could cut back on waiver rates if the list

APPENDIX E 268
were more selective. Clients need to be educated to the fact that second
opinions have a low return. (2) No information on review organization
experience is getting into the medical literature. There is a lot of information
and expertise that is not being made available to people in the health care field.
(3) The organization does not favor regulation of review organizations because
it feels it is one of the better companies. If all companies were regulated, it
believes that this might cut its effectiveness and it would lose its competitive
edge.
Organization 9
Organization 9 is a staff model HMO. It has been in operation for more
than 15 years and began conducting concurrent review about 1 year ago. The
HMO has over 200,000 members, which is a large share of the HMO
population in its area. It conducts concurrent review, discharge planning, and
case management. Staff physicians authorize admission to the hospital, but this
is an administrative and not a review process. The organization relies on its
physician members to act as gatekeepers. It assumes that the admissions
requested by its physicians are appropriate.
Because it is a staff model HMO, this organization has a great deal of
influence over the utilization practices of participating physicians. Because
physicians who work for the HMO are ''sold on the HMO idea," it is possible to
get physicians to adapt to new patterns of practice. The organization also
emphasizes physician leadership in changing practice patterns. Generally, one
physician moves to a new treatment or diagnostic approach and champions the
idea to other physicians, who eventually adopt procedures reported by peers to
be appropriate. The organization does not use financial incentives to get
physicians to change their practice patterns because it believes that financial
incentives can have adverse effects. It tracks referral patterns and will counsel
physicians who have unusual practice patterns.
The organization reports to have been the first in the community to have
cut preoperative days and the first to have implemented a 2-day length of stay
for normal deliveries. It also asserts that it is the first HMO in its area to offer
managed care services.
The organization contends that concurrent review is more effective with
physician participation and that preadmission review is unnecessary in staff
model organizations. The emphasis is on collaboration with the attending
physician when conducting concurrent review and case management.
For emergency admissions, staff physicians judge whether the admission
has been justifiedusually within 48 hours of the admission. Nurse reviewers
conduct concurrent review. Nurse reviewers work on-site in most hospitals.
Whey review the patient's charts and discuss the patient's care with the
attending physician and the patient. Reviewers use InterQual

APPENDIX E 269
criteria and Professional Activity Study data as guides for their professional
judgments. They communicate closely with the attending physicians. If a
physician disagrees with a shorter length of stay, the reviewer may refer to a
medical adviser from the same specialty group as the attending physician. Each
medical department in the HMO has a utilization review committee; medical
advisers are drawn from each specialty group. The medical adviser confers with
the attending physician. Denial of coverage is rare. It would occur if a patient
goes to a nonplan physician for services or if the patient received services in an
emergency room when clinic services would have been appropriate. Appeals
may be made in the form of a request for a second opinion. Physicians or
patients may also file complaints, and there is an arbitration process through
member services.
The nurse reviewers consider quality of care to be part of the review
process. The organization conducts peer review within each department and has
a quality assurance department that conducts studies on outcomes of care.
Patient satisfaction surveys are also conducted. Although the utilization review
program is "invisible" to patients, patients "appreciate" case management
services. The physicians on the staff are generally cooperative and amenable to
new utilization practices.
Leaders of the organization expressed concern over the difficulty in
measuring cost savings from utilization management because the cost of health
care is generally increasing. They added that effective physician leadership is
essential for changes to be made in practice patterns.
Organization 10
Organization 10 is a private, not-for-profit statewide PRO. About 60
percent of its business involves private contracts; it has 25 private clients,
including third-party payers and self-insured companies. It does Medicaid
review in 6 states, and private review in 50 states. The organization reviews
about 1.5 million cases per year. About 250,000 of the cases are within the
state, a figure that represents 60-80 percent of the discharges within the state. It
offers preadmission review; concurrent and retrospective review; outpatient
review; discharge planning; some quality-of-care reviews and fee reviews;
physician profiling within the state; and long-term-care, chiropractic,
obstetrical, and disability reviews. In addition, it does data processing for 10
other PROs.
The organization has developed a means to focus preadmission review
based on retrospective review experience. On the basis of Medicare data, it has
selected certain procedures for 100 percent review. And it selects certain
physicians for 100 percent review. (If a physician has one case denied
retrospectively, all cases will be reviewed for 3 months. This intensified review
is considered to be a sanction.) The organization believes that

APPENDIX E 270
focused preadmission review would be more efficient and just as effective as

full review of all elective procedures, but finds that most private clients want
100 percent review of all procedures.
The organization reports a high level of cooperation from providers.
Because it is a PRO, the physicians in the state are very familiar with its
methods. It has been in operation since 1972 and operates on a statewide basis.
The organization also claims to have a strong political base and a good
relationship with the state medical society.
The peer review aspect of the program is emphasized. Peer review is
believed to be the only means by which quality-of-care issues and treatment
options can be encompassed in the review process. The organization is planning
to expand its private review, particularly into other states. There is also a desire
to do more outpatient office procedure reviews.
Physicians are responsible for initiating the preadmission review. If a
patient calls, they tell the patient to have their physician call. Nurse reviewers
receive the information, may ask questions, and can certify cases that meet their
criteria. They are not allowed to advise attending physicians. The nurse
reviewers use a modified version of InterQual criteria that was developed and
approved by physician members of the PRO. Cases that fail the screens are
referred to physician advisers. Nurses may choose the physician adviser to
whom they refer a case. The PRO has about 23,000 physician members, but
most reviews are conducted by about 200 physicians, who work out of their
own offices. Physician advisers ordinarily make their decisions based on the
information provided by the nurse reviewer and rarely contact the attending
physician, unless the case has been appealed and is being reconsidered.
Physician advisers make decisions based on their own clinical judgment, not
written criteria. The organization has an internal validity committee that
monitors reviewer and adviser decisions. Reviewers are held to a 95 percent
accuracy standard; that is, the committee must agree with 95 percent of their
decisions.
About 80 percent of the cases are approved by nurse reviewers; 20 percent
get referred. About 1 percent are denied. Appeals take the form of
reconsideration by one or two physicians, depending on the client. A second
formal written appeal process is also available.
The organization believes that it has substantially cut inpatient days for its
clients. It claims to have saved clients a total of $117 million on inpatient days
since 1982. It evaluates quality of care in the review process and believes that
this has had a substantial impact on physician behavior. It reports a 100 percent
compliance rate, and physicians "fight" to stay off of intensified review. Of a
total of 3,600, about 150 physicians are on the list at any one time.
The organization expressed concern about service underuse, which it is
unable to track.

APPENDIX E 271
Organization 11
Organization 11 is an IPA model HMO. It has been operating since 1974
and began preadmission review in 1981. It has 175,000 members and 2,000
participating physicians. Most activity is concentrated in the local area. The
organization offers preadmission certification, preprocedure review, concurrent
review, case management, retrospective claims audit, discharge planning,
second surgical opinions, and some outpatient procedures review. It also
reviews claims and conducts physician profiling.
The organization states its utilization management objectives as follows:
(1) to monitor the medical needs of patients; (2) to monitor the level of care; (3)
to ensure that appropriate resources are used; and (4) to develop and evaluate
utilization data and identify potential utilization problems. It emphasizes
education and physician cooperation but has strict sanctions for noncompliance.
Physicians are fined $25 of the fee the first time, $50 the second time, and $100
the third time they fail to comply with the preadmission review program within
1 calendar year. Sanctions also are imposed for allowing unnecessary days of
care (20 percent of hospital per diem for the first offense, 50 percent for the
second offense, and 100 percent plus an appearance before the membership
committee for the third offense) and referral to nonplan physicians (20, 50, and
100 percent of nonplan charges, respectively). In addition, time limits are set on
referrals. Physicians are paid on a discounted fee-for-service system and split a
surplus risk pool at the end of each year. The pool is accumulated from savings
derived from hospital days saved.
The organization has a utilization review and quality assurance committee
composed of five physicians who hold monthly meetings. They review
physician utilization patterns and assess the utilization review program
annually. They also conduct some special studies.
The preadmission review process can be initiated either in writing or by
telephone. The IPA physician is responsible for beginning the process. Clerks
take the preadmission calls and ask physicians (or their representatives) a series
of questions from a one-page questionnaire. The questionnaires identify
conditions for admission for a variety of disorders. The criteria are used by the
state PRO. If the clerks receive a yes response to any of their questions, they
give the questionnaire to one of two supervisors, who signs the form and
authorizes the admission. The clerks can tell physicians that they will receive
certification. The organization believes that this procedure is more efficient than
having nurses handle the calls, because a vast majority of the admissions are
routine.
Concurrent review begins on admission and continues every 2 days
thereafter. Nurse reviewers work on-site at the hospital, review medical charts,
and may negotiate with the attending physician. They use the

APPENDIX E 272
InterQual criteria and professional activity study lengths of stay for their
region. Standards for any new procedures are developed by the firm's health
care standards committee, which meets every 2 months. Medical necessity
standards for about 12 medical procedures have been developed by the
utilization review and quality assurance committee.
If criteria are not met, the nurse reviewer refers the case to one of the two
supervisors or to the medical director, who is a full-time employee in charge of
the utilization review program. The medical director then discusses the case
with the attending physician. He may also ask two members of the utilization
review and quality assurance committee to review the case. Denials are rare.
Only the medical director can deny a case. Appeals may be made to the
grievance committee. If a physician is sanctioned she or he may appeal to the
utilization review and quality assurance committee and then to the HMO board
of directors.
The organization has decreased its days of care from 560 days per 1,000
members in 1980 to 319 days per 1,000 in 1988. It believes that it has had an
impact on quality of care. Reviewers use a generic quality screen developed by
the utilization review and quality assurance committee on all discharges. The
nurse reviewer conducts a retrospective chart review on cases that do not pass
the screen. Quality problems are categorized into three levels of severity. If
there is any potential harm associated with the quality discrepancy, the
attending physician will be called before the utilization review and quality
assurance committee. The physician is reviewed by three peers. The committee
evaluates the case and recommends corrective action. Corrections can take the
form of continuing education, intensified review of the physician's cases, or
limiting the physician's services within the IPA. This process takes place only
occasionally. The organization also monitors members and has a grievance
committee that receives member complaints.
The organization expressed the concern that assigned lengths of stay can
become a floor and that assigned days will be used even if they are not
necessary. Therefore, reviewers do not assign specific lengths of stay.
Organization 12
Organization 12 is a PRO that was begun by the state medical society in
1970. It started conducting private reviews in 1976; currently, more than half its
business is private review. It covers about 600,000 employees. Reviews are
conducted for 40-50 relatively small companies and for part of the CHAMPUS
program. The organizations offers preadmission review, concurrent and
retrospective review, second surgical opinion, case management, hospital bill
certification, bill audit, claims administration,

APPENDIX E 273
long-term-care review, workman's compensation reviews, discharge planning,

substance abuse review, and quality reviews of hospital care.
Because it was one of the original PSROs, this organization claims to have
set precedents in many programs. One strength is its historical and political base
in the state. Leaders of the organization know the state, and providers in the
state know them. Most nurse reviewers conduct reviews on-site in the hospitals.
They claim to have strong support from the physicians. Ninety-five percent of
the county medical societies participate in the PRO. The organization has a 23-
member board made up of physicians, administrators, business representatives,
and a consumer representative, and it also has five regional councils with the
same composition. These representatives "have made a commitment to make
the system work" and devote a good deal of volunteer time to the program.
The organization emphasizes cooperation with physicians and providers. It
takes an educational approach to solving problems. Written agreements with
each hospital allow private reviews on-site.
Physicians are responsible for initiating reviews. Most are very familiar
with the process. The PRO also accepts information from a physician's
designee. About 15 percent of admissions in the state are reviewed by nurses
operating from the PRO office. These admissions are mostly in small rural
hospitals. The remaining 85 percent of admissions are reviewed by nurse
reviewers on-site. This provides a direct check on the diagnostic and clinical
information reported by physicians.
Nurse reviewers can certify cases that meet the PRO criteria. Criteria were
developed by an internal health care standards committee and cover intensity of
service and severity of illness. Quality of care is considered in each review.
There are guidelines for making exceptions to the criteria. Nurses have some
authority to negotiate.
About 15 percent of the cases are referred to physician advisers. About 200
physicians serve as advisers, working part-time out of their own offices. They
may certify or deny a case based on their own clinical judgment. They may
decide on the basis of the record provided by the nurse reviewer, or they may
call the attending physician. The organization denies slightly more than 3
percent of cases. Denial is a last resort. Appeals go to a second physician
adviser. The patient or physician requesting the appeal may ask for a physician
of the same specialty.
The organization claims to reduce client admissions rates by 10-15
percent. It claims to have had a dramatic effect on preoperative days, which
used to be common practice and are now rare. Reviewers have criteria to screen
cases for certain quality-of-care problem and can refer any case with a potential
quality problem to an internal committee for review.
The organization expressed several concerns. (1) Some hospitals bill

APPENDIX E 274
patients for care for which the PRO has denied certification. (2) Most medical
admissions are emergency admissions, the necessity of which is evaluated only
through concurrent review. (3) Sometimes it is difficult to get physicians to act
as advisers in remote rural areas where the physicians all know each other. (4)
Specialists who develop review criteria tend to be too generous.

TABLE E-1 Summary of Volume of Business at Sites Visited
Reviewer to Population Ratio
Site Number of Covered Lives or Employees Average Number of Cases Reviewed Nurse Physician
1 70,000 lives (comprehensive review) 450 cases/mo 1:15,000 1:70,000
APPENDIX E
2 million lives (case management) 550 cases/mo 1:285,000 1:400,000

3 million lives (independent physician review) 2,000 cases/mo 1:850,000
2 1.1 million lives 2,000 cases/mo 1:16,000 1:16,000
3 2.3 million lives NA 1:7,500 1:365,000
4 2 million lives Up to 4,500 cases/mo 1:32,000 1:200,000
5 600,000 lives NA 1:10,000 1:80,000-100,000
6 8.5 million lives NA 1:20,000 1:180,000
7 1.7 million lives NA 1:25,000-30,000 1:375,000
8 1 million employees 2,000 cases/day 1:5,000 1:125,000
9 235,000 lives NA 1:47,000 1:47,000
10 NA 1.5 million cases/year 1:8,000 1:8,000
11 175,000 lives NA 1:2,500 1:25,000
12 600,000 employees NA 1:15,000 1:3,000
NOTE: NA indicates not available.

275
TABLE E-2 Profile of Nurse Reviewers
Site Experience Training RN Staffing Supervision
1 5 yrs 1:1 supervision 5 RNs; 1 RN director Informal observation
2 5 yrs FTE trainer; 2 wks; review all decisions 25 RNs; 2 supervisors; 1 UM director; 1 Accuracy of work; consistency; number
for 1 month trainer of cases/mo
APPENDIX E
3 5 yrs 3 wks 17 RNs; 2 supervisors; 1 manager Telephone monitoring; listen, review 5

cases/wk; weekly meetings
4 1 yr and some UR 4 wks 81 RNs; 5 LPNs; 1 supervisor per team Telephone monitoring; listen, report
daily on telephone activity, productivity
5 5 yrs 4-6 wits; in-house CE 56 RNs; 7 supervisors Telephone monitoring; listen, review a
percentage of cases weekly
6 Clinical, UR 2 wk class; 2-3 wks in unit 53 RNs; 7 supervisors; 1 director Telephone monitoring; biweekly
evaluation
7 5 yrs 2 FT trainers; 4 wks; CE 130 RNs; 1 supervisor per team Telephone monitoring; random sample of
cases; meetings

276
Site Experience Training RN Staffing Supervision
8 Clinical, several yrs 6 wks; in-house CE 200+ RNs; 1 supervisor per team of 18 Telephone monitoring; referrals
RNs; 1 manager/4 teams; 1 director
9 Hospital experience 1:1 supervision 5 RNs; 1 supervisor NA
10 5 yrs clinical 2 wks 76 RNs; 1 supervisor MD advisers fill out questionnaire reviewing
APPENDIX E
RNs; internal committee monitors decisions

for 95% accuracy of documentation;
application of criteria; referrals; turnaround
time
11 Clinical NA 7 RNs; 1 supervisor; 1 director Medical director does random clerk review
12 3-yr med/surg, ICU or ER Orientation, 1:1 supervision 40 RNs; 1 supervisor; 1 director Telephone calls/hr; number of reviews/hr;
sample of cases; appropriateness of referrals
and consistency, number of cases/reviewer
NOTE: Abbreviations are as follows: RN, registered nurse; FTE, full-time equivalent; UM, utilization management; UR, utilization review; LPN, licensed practical nurse;
FT, full-time; CE, continued education; MD, physician; ICU, intensive care unit; ER, emergency room; NA, not available.

277
TABLE E-3 Profiles of Physician Advisers
Site Experience Training Physician Staffing Payment On-or Off-Site Full-or Part-Time
1 Board-certified On-site Med dir +60a; 7 FTE Per case; incentive On-site; work together PT
MDs for more cases/hr
2 Board-certified; in 1-day orientation; 25 do most reviews; Per call; assume 15 Work in own offices; 1 PT; on call
APPENDIX E
practice ongoing contact with FT med dir min/call MD adviser on-site;

med dir med dir on-site
3 Board-certified; UR In UR 21 MD advisers; 1 By hr Work in own offices PT, 3 hrs/wk
experience; in practice PT director
4 Board-certified; in Several days 18 MDs (10 FTE on Salaried or by hr On-site (consultants off- PT; 60% more
practice med/surg; 2.5 psych); site) than 20 hrs/wk
1 FT dir
5 Experienced and Briefed and observed Med dir; 6 FTE med/ By hr On-site PT; 1; 1 on-site
respected; most board- for a couple days surg; 4 psych at all times
certified
6 Board-certified; UR; in Manual; meet with 6 MD advisers; 40 on Per hr or per case Some on-site; some in PT
practice director call; 1 PT director own office
7 UR; PRO Med dir trains on UR 8 MD advisers Per case 1 on-site; rest offsite; in PT; 1 FT on-site
issues group

278
Site Experience Training Physician Staffing Payment On-or Off-Site Full-or Part-Time
8 Clinical CE in-house; attend grand 24 MD reviewers; 3 MD Annual salary On-site FT
rounds at local hospital middle management; 1 FT
director
9 Clinical 1:1 supervision; CE 1 from each medical Salary On-site PT
APPENDIX E
specialty; clinical
supervisors; dept heads;
med dir
10 In practice 1 day 200 MD advisers; 1 PT $54.00/hr Work in own office PT
med dir
11 In practice NA 5 MD reviewers; 1 med dir $250.00/mo Work in clinic PT
12 Board-eligible; PRO 1 yr of experience in 200 MD advisers; 1 PT By hr Off-site PT
credentialing process hospital UR/QA med dir; 5 regional managers
NOTE: Abbreviations are as follows: med dir, medical director; PT, part-time; FT, full-time; MD, physician; UR, utilization review; FTE, full-time equivalent; CE,
continuing education; QA, quality assurance; NA, not available.
a Most physician review time is not devoted to prior review but to a review contract with an insurer. Most reviews are done by about five physicians.

279
TABLE E-4 Case Managers
Site Staffing Patient Contact Authority Monitoring
1 1 RN By telephone Selects cases; identifies services None
2 3 RNs; 1 supervisor By telephone; site visits every Develops care plan; negotiates for Supervisor oversees
30-90 days services
APPENDIX E
3 7 RNs; 1 manager By telephone; may visit; contract Selects cases; negotiates for Weekly meetings; staff discusses
with home health agency for site services cases
visits
4 9 RNs Contract with home health agency Identifies cases; compares costs None
for site visits
5 2 RNs By telephone Develop care plan; negotiate for Staff meetings and case
services conferences with MD advisers
6 258 nurse coordinators; 25 RNs make site visits, work out of Develop care plan; can negotiate Supervisor evaluates RNs; regional
supervisors; 5 regional supervisors 130 local offices for services supervisors evaluate random
sample of cases; clients audit cases

280
Site Staffing Patient Contact Authority Monitoring
7 8 RNs 1 initial visit; 1 follow-up visit 2-3 Identify cases; negotiate for services Monthly reports on cases
months after case terminated
8 Physicians; 1 director Contact with patients if care (home Negotiates with provider for early None
health) is planned for longer than 1 wk discharge
APPENDIX E
9 10 RNs; 5 discharge planners; 1 Site visits; provide services May negotiate for alternative services Med dir monitors all cases
supervisor
10 None None None None
11 7 RNs who do CM part-time; consider None RNs negotiate with MDs and hospital Med dir oversees
physician as case manager discharge planner for alternative
services
12 1 coordinator who does CM part-time Site visits; provide services Puts together teams for each case; Med dir oversees
team develops care plan; provides care
NOTE: Abbreviations are as follows: RN, registered nurse; MD, Physician; meal dir, medical director; CM, case management

281
APPENDIX F 282
Appendix F
Analysis of Agreements Between
Utilization Management Organizations
and Their Clients*
To secure more information about the operation of prior review programs,
requests were made to several organizations for copies of their contracts or
agreements with their clients. Copies of seven agreements were obtained, and
these were reviewed to assess how they dealt with various matters. The
contracts that were reviewed may or may not be fully representative of those
used in the industry. Also, on certain important matters such as appeals
procedures, some contracts referred to appendixes or attachments that were not
provided for review. Despite these limitations, several observations can be
made about the content and specific approaches used in the contracts that were
reviewed.
Contractual Descriptions of Prior Review Services

Contracts for prior review services vary greatly in length. One reviewed
contract was 28 double-spaced pages long and detailed the various steps that
would be taken in conducting the services covered by the agreement. This
contract, which is a standard form or generic agreement (as were several of the
others that were reviewed), permits the client to select from among the
utilization management company's services, which are described in detail (some
of it repetitive) over 16 pages of the contract. At the other
* The initial draft of this appendix was prepared by Nathan Hershey.

APPENDIX F 283
extreme was a contract that described the company's services in fewer than four
pages.
Notwithstanding the differences from contract to contract in the amount of
detail about the organization's services (for example, the options available when
an initial negative determination of medical necessity has been made), a reading
of the contracts does not reveal substantial differences among organizations in
the services themselves.
Sources of Clinical Criteria and Standards

The review-of contracts suggests that employers and other purchasers
apparently have not required information about the sources of the clinical
criteria and standards used by the organization with which they contract for
review services. Little or nothing in the contracts describes the criteria or
indicates their sources. One contract states that medical review will use
''recognized norms and standards, of the medical need for and appropriateness
of proposed, ongoing or completed medical care.'' Regarding hospital admission
review, the most comprehensive and detailed contract stated only that the
review firm will evaluate proposed admissions to certify "medical necessity and
appropriateness under the [insurer's] benefit plan." Another contract refers to
"guidelines mutually agreed upon" by the contracting parties. The most specific
contractual statement about criteria stated that the review organization would
use the purchaser's "accepted review criteria and guidelines as the framework
for making appropriateness decisions." These guidelines included the 1986
InterQual ISD-A criteria (intensity of service, severity of illness, discharge, and
appropriateness screens).
There may be extracontractual discussions and negotiations between the
review organization and client regarding the criteria used to evaluate the
necessity or appropriateness of care. However, these agreements, like most
commercial contracts, state that the contract is the entire agreement between the
parties and that the agreement cannot be modified, except by action of both of
the parties.
Distinction Between the Review Determination and

Purchaser's Decision To Pay For Services
Most of the contracts that were reviewed contain specific language about
the distinct roles of the review organization and the purchaser. The contractual
agreements generally specify that the review organization makes no decision
respecting benefits, although an adverse review decision often will be the basis
for an adverse decision on benefits. Note the following disclaimer language in
one contract:

APPENDIX F 284
In performing the [review] services the [review organization] shall not

determine a Participant's eligibility for benefits under the Group Contract.
Group shall have final and sole authority for all benefit determinations.
Another contract states:
Unless specifically agreed in writing to the contrary, the [review organization]
shall have no authority to bind you to any of its assessments,
recommendations, findings or certifications in respect to Medical Review, and
you reserve the right to act based upon your own judgment with respect to any
and all claims or issues reviewed hereunder.
Additional provisions in two contracts have relevance to the distinction
between determinations of medical need and decisions about payment of
benefits. In one agreement, stated performance standards for the review
organization implicitly recognize that reviewers may occasionally certify days
of hospitalization that do not meet the guidelines agreed to by the organization
and its client; it states that "95% of all days certified by [the reviewers] shall be
appropriately certified based on guidelines mutually agreed upon." This
provision suggests that the client accepts that a small number of certifications
will not meet the mutually agreed upon criteria in order to reduce the potential
for conflict. It also recognizes implicitly that the exercise of clinical judgment
by the review organization's physicians may sometimes result in decisions that
do not fit the criteria, but it sets up penalties if this happens too often.
Another contractual provision covers a utilization management
organization's obligation to assist its client organizations in specified
circumstances. The organization agreed to
provide [the client] its case records, staff and professional medical consultation
in the event that legal action is taken against [the client] for benefits which
have been denied or reduced when [the client's] determination is supported by
[the review organization] . . . or if [the client] is otherwise sued as a direct
result of [the review organization's] . . . activities.
One review organization's generic contracts, in addition to specifying that
its advice pertains only to medical necessity and appropriateness, not to
payment of benefit claims, further states that
The decision or determination to obtain or deliver any health care service is
always made only by the [health plan participant] and/or his or her physician,
and any decisions made by the [review organization] . . . or the health benefit
insurer... shall relate only to the obligation for payment for any such service
under the terms of the . . . group insurance policy ....
Indemnification and Liability Insurance Coverage

Concern about liability, which is at a high level in virtually all areas of
insurance and health care delivery, is clearly evident in contracts for prior
review services. One manifestation is the indemnification provisions that

APPENDIX F 285
were present in all but one of the contracts reviewed. Generally accompanying
these provisions are statements that the providers and purchasers of review
services are independent contractors. Representative of these provisions is the
following:
No provision of this agreement is intended to create, nor shall it be deemed to
create, a relationship between [the client] and [the review organization] other
than that of independent contractors. In addition, neither of the parties hereto,
nor any of their respective employees, agents, or contractors, shall be deemed
to be, or shall represent themselves as being, the employer, employee, or
representative of the other, except as specifically provided in this Agreement.
Some indemnification provisions are broad or general; others specify the
matters to which they apply. One illustrative indemnification provision states:
[The client] agrees that [the review organization] shall be held harmless by [the
purchaser] from all liability due to decisions made by officers or employees of
[the purchaser] concerning eligibility of any [individual] insured for benefits
under the provisions of the policyholder or contract holder's health benefit
program. [The review organization] agrees that [the purchaser] shall be held
harmless for all liability arising from failure by [the review organization] to
exercise due care in its review procedures, in rendering its decisions, and in
reporting them to [the purchaser].
Another relatively specific provision that is accompanied by an
indemnification clause of a general character applies to breaches of described
confidentiality obligations that would involve liability on theories other than
negligence. It provides, in part:
Should legal action be brought against either party or its affiliates as a result of
a breach of said confidentiality, the party causing said breach agrees to
indemnify and hold harmless the other party and affiliates for any loss, cost, or
damage, including reasonable attorney fees, arising therefrom.
Several of the contracts contain provisions describing or requiring liability
insurance coverage. For example, one contract provides that the review
organization "is responsible for maintaining during the life of this Agreement
liability insurance sufficient to protect it from claims of personal injury or
property damage which may arise from its activities under this Agreement." The
same contract also obliges the organization to require its physician reviewers
"to carry or otherwise be covered by insurance in amounts adequate to protect
them against professional liability claims which may arise hereunder." Only one
of the contracts that contained an insurance requirement specified the amount of
coverage.
Staffing and Performance Criteria

Purchasers of utilization management services seek more appropriate use
of health services by those covered under their benefit plan. This is

APPENDIX F 286
expected to translate into reduced benefit costs in comparison with costs in the
absence of utilization management services. Among matters relevant to the
successful functioning of a review organization are the composition and
qualifications of its staff and the criteria by which its performance is to be
measured.
Most of the contracts reviewed contain little detail regarding staff. Typical
is such language as "appropriate staffing levels" and "services shall be
performed by physicians, registered nurses or record technicians as
appropriate." Generally, references to physicians are preceded by ''licensed,''
and references to board certification are occasionally present in descriptions of
appeal processes.
The contracts usually specify the time periods within which determinations
of medical necessity, length of stay, and other decisions are to be made. These
can become, in effect, implicit staffing requirements (unless the review
organization has flexibility in deciding how many cases to review in detail),
since they require the organization to review, make determinations, and
communicate with benefit plan beneficiaries and/or health care providers.
within stated time periods.
Provisions in several contracts related directly to performance criteria. One
standard contract, in respect to a bill screening and audit service, one of the
services that could be selected by the purchaser, states the following:
[the review organization] guarantees that the savings from the [bill review and
medical audit] will exceed fees paid to [the review organization].
Another, more complicated, provision focused on adherence to numerical
standards in performance under the contract.
[the review organization] shall maintain the following performance standards
for initial certification, recertification and appeal referrals:
(a) 95% of all days certified by [physician reviewers] shall be appropriately
certified based on guidelines mutually agreed upon between [review
organization and purchaser].
(b) 85% of all referrals shall involve direct contact between the [physician
reviewer] and the attending physician.
(c) 85% of all referrals shall be resolved within (12) business hours of the
initial contact by the [review staff].
This contract goes on to provide penalties for failure to meet these
performance standards, based on a reduction in the monthly retainer fee for
each percentage point below those mentioned in the contract, up to a stated
maximum. No other contract reviewed made specific links between the review
organization's activity and its compensation.
Proprietary Information and Competition

Both providers and purchasers of prior review services recognize that they
will, in the operation of a review program, learn much about each

APPENDIX F 287
other's processes and practices. Almost all the reviewed contracts contained
provisions to maintain the confidentiality of proprietary information, with legal
recourse in the event of the breach.
Two contract provisions are particularly noteworthy because they illustrate
the overlap of some review organizations and purchasers in terms of their
knowledge of utilization control procedures. For example, a health insurer that
provides review services for some insured groups but contracts services out for
others may be concerned that its subcontractor will try to solicit the health
insurer's clients. One contract provides in part:
[the review organization] shall use its best efforts to avoid soliciting the sale of
its services to then current [clients] of [the health insurer] except in any cases
where a [client] elects not to obtain from [the health insurer] such services. In
the event [the review organization] inadvertently solicits any party prior to
having actual knowledge that such party is then a current [client] of [the health
insurer], [the review organization] agrees to terminate such solicitation
promptly upon its gaining such knowledge.
The potential clash of interests may run in the opposite direction. For
example, another contract, recognizing that a purchaser of review services can
gain information to enable it to enter the utilization management business,
contains the following limitation:
[the purchaser] covenants and agrees that for the full term of this Agreement,
including any extensions thereof, neither it nor any of its subsidiaries or
affiliates will compete with [the review organization], either directly or
indirectly, by performing or rendering the same or similar services within the
United States of America, as rendered or performed by [the review
organization] pursuant to this agreement.

APPENDIX G 288
Appendix G
Glossary and Acronyms
Glossary
Access: The ability to obtain needed medical care.

Admission Re- Assessment of the appropriateness of urgent or emergency admissions
view: within a limited period after hospitalization.
Ancillary Ser- Supplemental services such as laboratory, radiology, and physical
vices: therapy that are provided in conjunction with medical care.
Appropriate Care which is clinically justified; sometimes used interchangeably with
Care: necessary care and sometimes used only to refer to whether the use of a
particular site of care (for example, hospital) is justified.
Business Coali- Regionally based groups of employers and/or providers, insurers, and
tions: labor representatives who may disseminate information on health care
issues, collect and analyze data, and provide other services for members.
Capitation: A fixed rate of payment to cover a specified set of health services. The
rate is usually provided on a per member per month basis.
Claim: A bill for health services submitted to a health benefits plan for payment.
Coinsurance: The percentage of a covered medical expense that a beneficiary must
pay (after the deductible is paid).
Concurrent See Continued-stay review.
Review:

APPENDIX G 289
Continued- Assessment of the need for continued inpatient care for a hospitalized
Stay Review: patient.
Cost-Sharing: The share of health expenses that a beneficiary must pay, including the
deductibles, copayments, coinsurance, and extra bill.
Criteria: Bases for assessing the necessity or appropriateness of a medical
service; explicit criteria are written.
Current Pro- A listing of descriptive terms and identifying codes for reporting
cedural Termi- physician services and procedures.
nology (CPT):
Deductible: The amount of medical expense that must be incurred and paid by an
individual before a third party will assume any liability for payment of
benefits.
Discharge The process of ensuring that patients are discharged as soon as
Planning: medically appropriate, with follow-up care planned and arranged as
needed.
Effectiveness: Probability of benefit to patients from a specific medical service under
average conditions of use.
Efficacy: Probability of benefit to patients from a specific medical service under
ideal conditions of use.
Efficiency: Level of benefit from a fixed level of input or amount of input cost to
achieve a defined level of benefit.
Encounter: In the health maintenance organization setting, generally refers to an
outpatient visit to a physician or allied health professional.
Enrollee: Individual covered by a health benefit plan.
Feedback Ap- Programs in which physicians' patient care decisions are reviewed based
proaches: on medical records, claims, or other documents of care, with the results
shared with the physician.
Fee-for- A method of paying practitioners on a service-by-service rather than a
Service: salaried or capitated basis.
Gatekeeper: Primary care provider who is responsible for coordinating all medical
treatment rendered to an enrollee of a health plan.
Group Model A health maintenance organization that contracts with a primary care or
Health Main- multispecialty medical practice for delivery of health services.
tenance Orga-
nization:
Health Main- An entity that accepts responsibility and financial risk for providing
tenance Orga- specified services to a defined population during a defined period of
nization time at a fixed price.
(HMO):
High-Cost A process for identifying high-cost patients and facilitating the
Case Manage- development and implementation of less costly appropriate courses of
ment: care.

APPENDIX G 290
Individual A health maintenance organization that contracts with private physicians

Practice Asso- who serve health maintenance organization enrollees in their offices,
ciation (IPA): generally on a fee-for-service basis.
Model Health
Maintenance
Organization:
Insurer: Organization that bears the financial risk for the cost of defined
categories of services for a defined group of enrollees.
Medical Neces- The need for a specific medical service based on clinical expectations
sity: that the health benefits will outweigh the health risks; sometimes used
interchangeably with appropriateness.
Network Mod- A health maintenance organization that contracts with two or more
el Health medical group practices.
Maintenance
Organization:
Outcome: The result of a medical intervention.
Outliers: Cases that are at the extremes of a distribution.
Peer Review A physician-based organization that reviews the medical necessity and
Organization the quality of care provided to Medicare beneficiaries.
(PRO):
Per Diem: A negotiated daily payment for delivery of hospital services, regardless
of the actual services provided; sometimes refers only to "room and
board" charges (meals, routine nursing care, etc.), not ancillary services.
Practice Guide-Clinical recommendations for patient care.
lines:
Practice Pat- Aggregate characteristics of a practitioner's use of medical resources
terns: over time.
Practitioner: One who practices medicine; may include physicians, chiropractors,
dentists, podiatrists, and physician's assistants.
Preadmission Assessment of the clinical justification for a proposed hospital admission.
Review:
Premium: An amount paid periodically to purchase health insurance benefits.
Prepaid A term used before the term health maintenance organization was
Group Prac- coined to refer to multispecialty groups paid on a salaried or capitated
tice: basis.
Preservice Re- Assessment of the clinical justification for a proposed inpatient or
view: outpatient service.
Prior Review/ Prior assessment by a payer or payer's agent that proposed services,
Authorization/ such as hospitalization, are appropriate for a particular patient. Payment
Certification/ for services also depends on whether the patient and the category of
Determination: service are covered by a benefit plan.
Professional Medicare review organization that preceded the peer review organization.
Standard Re-
view Organiza-
tion (PSRO):

APPENDIX G 291
Profile Analy- Use of aggregate statistical data on an institution or practitioner to

sis: compare practice and use patterns, identify inappropriate practices, or
assess other characteristics of practice.
Provider: An individual or organization that provides personal health services.
Quality As- Evaluation of the technical and interpersonal aspects of medical care.
sessment:
Quality Assur- An organized program to protect or improve quality of care by
ance: evaluating medical care, correcting problems, and monitoring corrective
actions.
Referral: An arrangement for a patient to be evaluated and treated by another
provider.
Retrospective Assessment of the appropriateness of medical services on a case-by-case
Utilization Re- or aggregate basis after the services have been provided.
view:
Second- An opinion about the appropriateness of a proposed treatment provided
Opinion Pro- by a practitioner other than the one making the original
gram: recommendation; some health benefit plans require such opinions for
selected services.
Self-Insurance: When an organization bears financial risk for hazards (for example,
medical costs, and property damage) that the organization itself may
experience.
Self-Referral: The process whereby a patient seeks care directly from a specialist
without seeking advice or authorization from the primary care physician.
Site of Service: Location where care is provided, for example, an inpatient facility or
home.
Staff Model A health maintenance organization which provides health services
Health Main- through a multispecialty group practice, usually on a salaried basis.
tenance Orga-
nization:
Third-Party Organization that processes health plan claims without bearing any
Administrator insurance risk.
(TPA):
Third-Party An organization other than the patient (first party) or health care
Payer: provider (second party) involved in the financing of personal health
services.
Triple-Option An experience-rated program for an employer group in which a single
Plan: insurance carrier, Blue Cross and Blue Shield plan, or health
maintenance organization provides indemnity or service benefits in
conjunction with various managed care or health maintenance
organization plans.
Unbundle: Charging for individual services which ordinarily should have been
covered under one procedure code.

APPENDIX G 292
Upcode: Using a procedure code which reflects a higher intensity of care than
would normally be used for the services delivered.
Utilization A set of techniques used on behalf of purchasers of health benefits to
Management: manage costs through case-by-case assessments of the clinical
justification for proposed medical services.
Withhold: A portion of a capitated or fee-for-service payment to a contracting
physician withheld by an HMO or similar organization during the year.
Depending on how revenues cover costs, the organization may retain or
return some or all of the amount withheld.
Acronyms
AEP: Appropriateness Evaluation Protocol

BCBSA: Blue Cross and Blue Shield Association
CBO: U.S. Congressional Budget Office
CCMC: Committee on the Costs of Medical Care
CPT: Current Procedural Terminology
DHHS: U.S. Department of Health and Human Services
DRG: Diagnosis-related group
FFS: Fee-for-service
FMC: Foundation for Medical Care
GAO: U.S. General Accounting Office
GHAA: Group Health Association of America
HCFA: Health Care Financing Administration
HIAA: Health Insurance Association of America
HMO: Health maintenance organization
ICD-9: International Classification of Disease
IMC: International Medical Centers Inc.
IOM: Institute of Medicine
IPA: Individual practice association
ISD-A: Intensity of service, severity of illness, discharge, and
appropriateness screens
OBRA-86: Omnibus Budget Reconciliation Act of 1986
PCP: Primary care physician
PPO: Preferred provider organization
PPS: Prospective payment system (for hospitals)
PRO: Peer review organization
TPA: Third-party administrator

APPENDIX H 293
Appendix H
Biographies of Committee Members
JEROME H. GROSSMAN, M.D., is the Chairman and Chief Executive

Officer of New England Medical Center, Inc. He is also Chairman of the
Institute for the Advancement of Health and Medical Care and Associate
Professor of Medicine at Tufts University School of Medicine. He serves as
Trustee/Director of several corporations and institutions, including The Boston
Private Industry Council, Tufts Associated Health Plan, Wellesley College, and
Arthur D. Little, Inc. Dr. Grossman joined the staff of Massachusetts General
Hospital in 1966, where he served in a variety of positions. He came to the New
England Medical Center in 1979. Dr. Grossman was one of the original staff of
the Harvard Community Health Plan, where he developed the world's first
automated medical record system, known today as COSTAR. From 1982 to
1987 Dr. Grossman served as Program Director of the Commonwealth Fund
Task Force on Academic Health Centers. He is a member of the Institute of
Medicine.
HOWARD L. BAILIT, D.M.D., Ph.D., is Vice President for Healthcare
Management at AEtna Life & Casualty. From 1967 to 1983, Dr. Bailit was on
the faculty of the University of Connecticut Health Center, where he served as
Associate Dean and Professor and Head of the Department of Behavioral
Sciences and Community Health. From 1983 to 1986, he was head of the
Division of Health Administration at the School of Public Health, Columbia
University. He was a consultant to the Rand Health Insurance Experiment and
has served on many professional and governmental committees, including the
National Center of Health

APPENDIX H 294
Services Research Study Section. Dr. Bailit has been a member of the Institute
of Medicine since 1984.
ROBERT A. BERENSON, M.D., is a practicing internist with an office in
Washington, D.C., and is cofounder and medical director of the National
Capital Preferred Provider Organization. Prior to entering private practice in
1981, he served on the White House Domestic Policy Staff in the Carter
Administration. Dr. Berenson is on the faculty of the George Washington
University and Georgetown University Schools of Medicine. He has written
extensively about health policy, particularly on issues related to physician
payment and managed care.
JOHN M. BURNS, M.D., is Vice President, Health Management of
Honeywell, Inc. From 1965 to 1981, Dr. Burns was engaged in private practice
in St. Paul, Minn. He is board-certified in internal medicine and nephrology. He
served as medical adviser to Northwestern Bell Telephone Company from 1967
to 1981. He was President of the Ramsey County (St. Paul, Minn.) Medical
Society in 1982. Dr. Burns joined Honeywell in 1981.
RICHARD H. EGDAHL, M.D., Ph.D., is Academic Vice President for
Health Affairs at Boston University, Director of the Boston University Medical
Center, and until May 1989, was an active endocrine surgeon. As Academic
Vice President he serves as the chief academic officer for the university's four
health schools-the Schools of Medicine (including Public Health), Graduate
Dentistry, Social Work, and Sargent School for Allied Health Professions.
Since 1976, Dr. Egdahl has been the director of the Boston University Health
Policy Institute and the Center for Industry and Health Care. He writes and
speaks widely on industry's crucial role in changing the nation's health care
delivery system. Dr. Egdahl is a member of many professional societies
including the American Society for Clinical Investigation, American Surgical
Association, American College of Surgeons, and Society of Medical
Administrators. A member of the Institute of Medicine, he served on its
Governing Council from 1981 to 1985. He serves on the editorial boards of a
number of professional journals including the New England Journal of Medicine.
JOHN M. EISENBERG, M.D., M.B.A., is Sol Katz Professor of General
Internal Medicine and Chief of the Section of General Internal Medicine at the
University of Pennsylvania. He is also a Senior Fellow of the University of
Pennsylvania's Leonard Davis Institute of Health Economics. He has served as
President of the Society for General Internal Medicine and Vice President of the
Society for Medical Decision Making. Dr. Eisenberg is a member of the
Institute of Medicine and a commissioner on the Congressional Physician
Payment Review Commission. He is presently Chairman of the American
College of Physicians' Health Care Financing

APPENDIX H 295
Subcommittee, is a member of its Health and Public Policy Committee, and has
served as a member of the College's Board of Regents. He has served as a
member of the editorial boards of Medical Care, Health Services Research, the
Journal of General Internal Medicine, the Journal of Gerontology, and Medical
Decision Making. He also serves as consultant to the Annals of Internal
Medicine. Dr. Eisenberg has published on topics such as physicians' practices,
test use and efficacy, medical education, and clinical economics. His book,
Doctors' Decision and the Cost of Medical Care, was published by Health
Administration Press in 1986.
DEBORAH ANNE FREUND, Ph.D., is Professor of Health Economics
and Chair of the Health Sciences and Administration Division in the School of
Public and Environmental Affairs and Adjunct Professor of Medicine at Indiana
University. She also serves as Director of Indiana University's Health Services
Research Center. She is an expert on Medicaid case management programs and
has written extensively on this topic as well as on HMOs and alternative
reimbursement strategies. She is the chairperson of the Medical Care Section of
the American Public Health Association, and also serves on the Board of
Directors of the Association of University Programs in Health Administration.
She serves as a member of three editorial boards: Medical Care, Inquiry, and
Health Policy.
PAUL M. GERTMAN, M.D., is Chief Executive Officer of Clinical
Information Advantages, Inc., a new company developing voice-entry, expert
system software for practicing physicians. Dr. Gertman also is Chairman of the
Board of U.S. Quality Systems, Inc., and of Health Systems Advantages, Inc.
Prior to his current position, Dr. Gertman was Associate Professor of Medicine
and Chief of Health Care Research at the Boston University School of
Medicine, then Founder and President of the Health Data Institute, Inc., and
most recently, Vice Chairman of the Board and Chief Scientist at Caremark,
Inc. His principal scientific interests focus on measurement of quality and
efficiency of medical care and on use of information systems technology in
health care.
ALICE G. GOSFIELD, J.D., is an attorney from Philadelphia who has
been working with hospitals and physicians on utilization management, quality
assurance, and peer review issues since 1973. She has been a public member of
the Statewide Professional Standards Review Council and a consultant to state
and federal regulatory agencies on health law issues. She has lectured widely on
PROs, PSROs, utilization management, and quality assurance for various
organizations, including the National Health Lawyers Association, the
American Medical Association, the American Academy of Hospital Attorneys,
the American Medical Peer Review Association, the Blue Cross and Blue
Shield Association, and the American

APPENDIX H 296
Bar Association. A member of the Executive Committee of the National Health

Lawyers Association, she is Program Chairman for their seminar ''Utilization
Management, PROs and Quality Assurance: The Legal Pitfalls'' and on the
Planning Committee for the first National Health Lawyers Association-
American Medical Association program on physician legal issues. Ms. Gosfield
has published a book on PSROs and numerous articles on utilization
management and quality assurance topics and is the editor of the 1989 Annual
Health Law Handbook, published by Clark Boardman, Ltd.
MICHAEL E. HERBERT, M.B.A., is President and Chief Executive
Officer of Physicians Health Services, Inc., the holding company that owns
Physicians Health Services of Connecticut, Inc., and Physicians Health Services
of New York, Inc. Prior to becoming President of Physicians Health Services,
he was Vice President of InterStudy, the national health policy research firm.
Mr. Herbert is Treasurer and Director of the American Medical Care and
Review Association, the national group that represents IPA HMOs, and is
President of the Association of Connecticut HMOs. He is also a consultant to
the federal government's Office of Prepaid Health Care, reviewing HMO
applications for federal qualification and performing HMO compliance reviews.
NATHAN HERSHEY, LL.B., is Professor of Health Law in the Graduate
School of Public Health, University of Pittsburgh, and has been a member of
the Institute of Medicine since 1972. He is counsel to two law firms, Markel,
Schafer & Means, Pittsburgh, Pa., and Post & Schell, Philadelphia, Pa. In 1986,
he authored Fourth-Party Audit Organizations: Practical and Legal
Considerations, which appeared in Law, Medicine & Health Care, and
examined a variety of questions regarding the impact of private utilization
management organizations on relationships among patients, providers, and
payers.
NEIL HOLLANDER is Vice President of Corporate Health Strategies at
Blue Cross of Western Pennsylvania. Prior to joining Blue Cross in Pittsburgh,
Mr. Hollander was a Vice President and Executive Director with the Blue Cross
and Blue Shield Association; Director, Bureau of Program Development,
Division of Medical Assistance, New York State Department of Social
Services; a senior Research Associate at the National Academy of Public
Administration in Washington, D.C.; Assistant to the Representative for the
Ford Foundation in Cairo, Egypt; and Program Planner for North American
Rockwell. He is a member of the Editorial Advisory Board of Same-Day
Surgery; Chairman, Managed Healthcare Services Committee for the Pittsburgh
Program for Affordable Health Care. Mr. Hollander also serves on the boards of
the Alpha Center in Washington, D.C., and

APPENDIX H 297
Keystone Health Plan West, is a Consultant to the World Bank, and is a Panel
Member of the American Arbitration Association.
KAREN IGNANI, M.B.A., directs the activities of the American
Federation of Labor and Congress of Industrial Organizations (AFL-CIO) on
health care, pensions, and social security. Prior to joining the AFL-CIO, Ms.
Ignani served as a professional staff member on the U.S. Senate's Labor and
Human Resources Committee. Ms. Ignani also served as Research Director and
then Assistant Director for the Committee for National Health Insurance and
was a Health Care Research Analyst for the U.S. Department of Health,
Education and Welfare. Ms. Ignani has authored a number of articles about
organized labor's position on employee benefits issues and has represented the
AFL-CIO on numerous panels all over the country.
CAROL ANN LOCKHART, Ph.D., is Executive Director of the Greater
Phoenix Affordable Health Care Foundation. She is a member of the Physician
Payment Review Commission, which is responsible for recommending changes
to the Congress in the way Medicare pays physicians. Dr. Lockhart is a fellow
in the American Academy of Nursing and has been involved in national and
local manpower issues. She is coauthor of two books on labor relations in health
care and is on the editorial boards of Health Care Management, University
Press, West Yorkshire, England, and Home Health Care Nurse. She was
previously a director in the Arizona Department of Health Services and Senior
Research Associate at Boston University School of Public Health, as well as a
Pew Foundation Fellow at Brandeis University.
ARNOLD MILSTEIN, M.D., M.P.H., is a Managing Director of the Wm.
M. Mercer component of Marsh and McLennan and President of its National
Medical Audit (NMA) subsidiary. NMA is a national physician-based
consulting group that specializes in the design and evaluation of medical cost
management programs for Fortune 500 companies and large carriers. Dr.
Milstein is also the medical director of Super-PRO Project of the Health Care
Financing Administration. He has published multiple book chapters and articles
on health care control methods and is the editor of a nationally published
column on utilization review decision-making. In 1987, Business Insurance
selected him as one of 20 people who made a difference in employee benefits
management in the past 20 years. Dr. Milstein is a board-certified psychiatrist
and an associate clinical professor at the University of California, San Francisco.
ALAN R. NELSON, M.D., is a private practitioner of internal medicine
and endocrinology in Salt Lake City, Utah, and was named President of the
American Medical Association in June 1989. A graduate of Northwestern

APPENDIX H 298
University School of Medicine, Dr. Nelson is a Fellow of the American College

of Physicians, and a member of the Endocrine Society. He is also active in
clinical teaching and is Associate Clinical Professor of Medicine at the
University of Utah School of Medicine. Throughout much of his career, he has
been involved in medical peer review and quality assurance. He organized a
statewide peer review program for the Utah Medical Association, and served as
a consultant for the National Center for Health Services Research. In 1975, he
was elected to the Institute of Medicine, and served on its Governing Council
from 1984 to 1987. He was also a Commissioner of the Joint Commission of
Accreditation of Hospitals from 1982 to 1986, and he has authored numerous
papers and several book chapters on medical peer review. He was a member of
the National PSRO Council from 1973 to 1977.
ROBERT PATRICELLI, LL.B., is President of Value Health, Inc., a
managed health care company based in Avon, Conn. Value Health provides a
range of cost-containment and quality assurance services in such areas as
mental health and substance abuse, prescription drugs, podiatry, and review of
high-cost medical and surgical procedures. From 1977 to 1987, he was Senior
Vice President of Connecticut General Life Insurance Company and then
Executive Vice President of CIGNA Corp. and President of its Affiliated
Businesses Group, which included CIGNA's $1 billion health care businesses
(HMOs, utilization review, and rehabilitation services). He also spent 8 years in
the federal government, including service as Deputy Under Secretary for Policy
in the U.S. Department of Health, Education and Welfare from 1969 to 1971.
Mr. Patricelli served from 1984 to 1988 as Chairman of the U.S. Chamber of
Commerce's Health Care Council and is currently Chairman of its
Subcommittee on Mandated Benefits. He is a Member of the Board on Health
Care Services of the Institute of Medicine, the Board of Directors of the
Institute of Living (the nation's largest private psychiatric hospital), and the
Board of the American Pharmaceutical Institute and is on the editorial advisory
boards of Health Affairs, Health Week, and The American Druggist. He is a
graduate of the Harvard Law School.
CYNTHIA L. POLICH, M.A., Vice President, United HealthCare Corp., a
health services and supply company based in Minnetonka, Minn. Until July
1989, she was President of InterStudy, a nonprofit health care research firm in
Excelsior, Minn. Her work has focused on identifying and developing
innovative and effective methods to organize, deliver, and finance health and
long-term care for the elderly. She is the author of many publications in the
areas of long-term care financing, Medicare enrollment in HMOs, case
management, and home health care. She is Past President of The Women's
Health Leadership Trust, Past President of the Minnesota

APPENDIX H 299
Gerontological Society, a Board Member of Quality Quest, and Chair of

Senator David Durenberger's long-term care advisory group.
DONALD M. STEINWACHS, Ph.D., is Director of the Johns Hopkins
Health Services Research and Development Center and is Professor of Health
Policy and Management in the School of Hygiene and Public Health. The
Center is an interdisciplinary research group involving faculty principally from
the School of Hygiene and Public Health and the School of Medicine. Dr.
Steinwachs is also Director of the recently established Center on Organization
and Financing of Care to the Severely Mentally Ill. His research has addressed a
wide range of issues involving the effects of organization, financing,
manpower, and technology on utilization, cost, and patient outcomes from care.
He has a particular interest in the role of management information systems as a
source of data for examining patterns of ambulatory and inpatient care, cost,
and indicators of quality in defined populations. Dr. Steinwachs is currently
President of the Association for Health Services Research and serves as a
consultant to federal agencies and private foundations.
BRUCE S. WOLFF, LL.B., is a partner in the New York and Washington,
D.C., offices of the law firm of Proskauer Rose Goetz & Mendelsohn, where he
specializes in health care transactional and nonprofit institutional matters. From
1977 through 1979, he was Deputy Assistant Secretary for Legislation at the
U.S. Department of Health, Education and Welfare and Special Assistant to the
Secretary. He is a Past Chairman of the Committee on Medicine and Law of the
Association of the Bar of the City of New York and of the Citizens Advisory
Committee to the Medicaid Program of the District of Columbia. He is a
frequent speaker on health care delivery and public policy matters.

APPENDIX H 300

INDEX 301
Index
A variations in use of medical services,

Acquired immune deficiency syndrome 44-45, 46, 48, 65
(AIDS), 119, 121, 128, 131, 138, 139
Admission review, 18, 37 B
Aetna Life and Casualty HEALTHLINE, Back pain, 80-81, 82, 104
97 Bank of America, 131
Allstate, 15 Blue Cross and Blue Shield
American College of Physicians, 36 n.5, development of prior review guidelines,
107 107, 148, 154
American College of Surgeons, 36 n.5 effects of cost-management programs,
American Federation of State, County, 93, 98, 115
and Municipal Employees, 42 employer monitoring of cost-
American Hospital Association, 94, 107, containment activities of, 41-42
148, 154 first plan, 30, 35
American Medical Association, 107, 148, of Greater Philadelphia (Independence),
154 95-96, 131
American Society of Anesthesiologists, health planning efforts, 36
36 n.5 high-cost case management by, 119,
Amyotrophic lateral sclerosis, 121 129, 131
Anderson, Odin, 37 integrated service/insurance products, 50
Assessment of care, see Criteria for losses, 2
assessment of care managed fee-for-service days under, 95
appropriate use of medical services, of Massachusetts, 95
45-46, 48, 52, 83 of Minnesota, 45
data sources for, 49 of North Carolina, 93
extrapolation from rates of nonconfirm- of Northeast Ohio, 36
ing second opinions, 39 operational problems in, 222
methods for assessing illness severity, of Pennsylvania, 43
48-49 small-area study by, 45, 69

INDEX 302
utilization review efforts/programs, 37, Coronary angiographies, 46

59, 60-61, 93, 94, 98, 171 Coronary artery bypass surgeries, 46, 147
Cost management/containment strategies
C basic elements of, 32-33
Cancer, 88, 121, 125 benefit plan design, 20, 34-35, 214
Cardiac pacemakers, 48, 83 control of provider payments, 21, 35
Carotid endarterectomies, 46, 48, 83 early efforts by third parties, 32-38
Case management, see High-cost case effectiveness of, 24-25
management employer interest in, 40-43
Cataract surgery, 49, 158 federal government initiatives, 38-40
Caterpillar, Inc., 131, 138 financial incentives to providers, 20-21
Ciba-Geigy, 122 gatekeeping/triaging, 21
Claims administrators health planning, 35-36
case management by, 123, 125 inappropriate utilization as a target for,
prior review by, 61, 68-69, 258-259, 44-46
261-262 physician education, 21
Claims data and quality assurance, 43
defects in cost and utilization data based and quality of care, 46-47
on, 111, 158 risk pool management, 33-34
high-cost case identification through, 127 by small businesses, 42
HMO processing of, 220-221 utilization review, 36-38
quality-of-care information from, 115 Costs of health care
Codman, Ernest, 36 aging of population and, 24
Commercial health insurance companies, causes of increases in, 23-24, 32, 115
see Health insurance industry; Health and clinical judgments about value of
insurance plans; and individual com- treatments, 23
panies economic shocks and, 40
Commission on Professional and Hospital effects of utilization management on,
Activities, 81-82 1-2, 3-4, 14, 52, 92, 95, 145-146
Committee on the Costs of Medical Care, increases in, 2
28 information resources on, 47-48
Concurrent review, see Continued-stay/ number of providers and, 24, 35-36, 43
concurrent review prior review and, 92, 100-101, 104
Confidentiality of medical information screening for untreatable diseases and, 24
employer respect for, 7, 151 third-party financing and, 28
in high-cost case management, 132, 139 trends in, 3-4
n.6 and U.S. competitiveness, 15
utilization management organization Council on Wage and Price Stability, 40
responsibility for, 7 Criteria for assessment of care
Consumer price index, changes in, 15, 33 adoption and modification of, 83-84,
Continued-stay/concurrent review 177-178
appeals processes, 193 appropriateness evaluation protocol
by Blue Cross plans, 37 (AEP), 80, 82
criteria/standards for, 48, 81-82 availability of, 5
defined, 18, 170 competitiveness and, 22-23, 80, 154, 160
effectiveness of, 98 computerization of, 80, 107 n.3
focus of, 120 diagnosis-specific, 48
high-cost case management and, 127, 139 differentiation by level of review, 84-85
liability for, 180 exceptions to, 3, 72, 79, 173
by PSROs, 39, 66 of HMOs, 215
responsibility for, 66 hospitalization, 79-82, 104
variations in, 65, 66

INDEX 303
ISD-A, 80-81, 82 by retrospective reviews, 87

length-of-stay norms, 48, 81-83, 84 review-versus nonreview groups, 96, 100
and liability, 176-178 see also Studies and surveys of utiliza-
lists of procedures, 79 tion management
medical community involvement in, Expenditures for health care
147, 153, 205 by business/private industry, 15, 16
procedure necessity, 45, 83, 172-173, 177 by health plan members for high-cost
recommendations on, 6, 7, 9, 11, 147, illnesses/injuries, 120-121
149, 154, 159-160 hospital care, 31
reliability of, 80 by major sources of funds, 31
research needs on, 4-5, 159-160 number of physicians and, 24
sources of, 11, 83-84, 283 personal health care, 31
standardizations of, 178 physician services, 31
statistical models for, 178 prior review and, 96
steps in generation of, 147 total, trends in, 14-15
variation in, 89 Experimental Medical Care Review Orga-
nizations, 39
D
Discharge planning, 18, 64, 121-122, 125 F
Dunlop Group of Six, 41 Federal government
Dunlop, John, 40 health care cost-containment initiatives,
38-40
E see also Medicare;
Economic Stabilization Program, effec- Peer review organizations
tiveness of, 30, 40 Fee-for-service
Education high-cost case management and, 129
enrollee/patient, on utilization manage- managed programs for, 50
ment obligations, 7, 103-104, 109, and volume of physician services, 44
151, 156 Foundations for Medical Care, 30, 35, 37,
physician, as cost-containment strategy, 39
21, 48-49, 147, 156
see also Health education G
Employee Retirement and Income Secu- General Motors, cost-containment reports,
rity Act (ERISA), 30, 41, 174, 30, 41-42
187-188, 189 Greater Phoenix Affordable Health Care
Employers, see Private employers Foundation, 42
Evaluation of utilization management Group Health Association of America, 31
barriers to, 110 Group Health Association of Washington,
behavioral biases against, 22 D.C., 30-31, 35
biases in, 96, 101
compared to expected utilization with H
no intervention, 95 Health benefits, see Health insurance plans
competition and, 22-23 Health Care Financing Administration,
conceptual and methodological prob- 121 n.1, 154
lems, 21-23, 86, 93, 100-101, 102, Health education, employer support for, 41
111-116, 145 Health Insurance Association of America,
cross-sectional and longitudinal analy- 107, 148, 154
ses, 97 Health insurance industry
of high-cost case management, 125-126, growth of, 27-32
131, 145 integrated service/insurance products, 50

INDEX 304
losses, 2, 15, 60 defined, 206

premium setting, 100-101 delivery of health services, 215-219
utilization management by, 60 employer support for, 41
withdrawals from group insurance mar- enrollee understanding of benefit restric-
ket, 15 tions in, 205-206
Health Insurance Plan of Greater New enrollment criteria, 214
York, 35 enrollment information, 114
Health insurance plans enrollments, 50, 214, 219, 220
coinsurance rates, 104 factors affecting performance, 209-213
community rating, 33 federally qualified, 214, 219
cost controls in, 20, 34-35 goals of, 210-211
cost sharing in, 96, 109, 115, 151 group-model, 207, 217, 233-235
coverage of alternatives to hospitaliza- growth in, 50
tion, 41 high-cost case management by, 129, 218
derision-making process, 212 hospital-sponsored IPA, 230-232
deductibles, 104 hospital use in, 44, 177
defined contribution programs, 110 integrated service/insurance products, 50
design of benefits, 20, 34-35, 94, 115, losses, 2, 209
148, 151, 214 management information systems,
duplicate coverage, 114 220-221, 222, 223, 231, 234,
enrollee/patient navigation of, 103-104 236-237, 238-239, 241-242
enrollment trends, 30-32 market effects of and on, 209-210
exceptions to limitations in, 19-20, 122, medical director's role, 222, 232, 237, 240
140-141, 173 medical necessity determinations, 173
expenditures by businesses, 14-15 network model, 207
experience rating, 33 open-ended, 51
HMO underwriting practices, 214 operational problems, 220-222
individually purchased, 34, 153 operations, 211-213
information on prior review policies in, organizational structure, 65, 207-209,
79 210-211, 216
limits on coverage, 141 n.8, 173 patient flow procedures, 215, 218
premium increases, 15 patient payment obligations, 195
prior review integrated with, 66, 68-69 peer review in, 216, 219, 222
refusal to certify, 4 physician contracting arrangements
risk pool management for, 33-34 with, 211-212, 217
second-surgical-opinion provisions, 37 physician incentives used by, 47, 211,
for self-employed people, 34 216, 224-227, 230-231, 233-234,
sources of, 32 236, 238, 241
triple option benefit package, 51 physician selection, 215
see also Costs of health care; physician-sponsored IPA, 240-243
Uninsured people policy and research issues, 227-229
Health maintenance organizations (HMOs) quality assurance in, 206, 216, 219-220,
antitrust liability, 184 222-224
benefit design, 214 quality of care in, 50, 205, 210, 211, 223
carrier-sponsored, 235-240 regulation of, 50, 207, 210, 219-220, 223
case studies, 213, 229-243 staff model, 207, 215
claims processing, 220-221 types of, 207
clinical protocols/guidelines, 215, 218 underwriting benefits, 214
congressional support for, 47 utilization management approaches, 6,
data analysis by, 215-216, 219 66, 214-222, 229, 231-232, 234-235,
data integrity in, 221-222 237, 239-240, 242

INDEX 305
see also Independent practice associa- Hill-Burton program, 30, 35

tions Home care, 135, 136, 138, 141
Health planning Hospitalization
cost-containment effects of, 35-36, 40 criteria for determining need for, 79-81,
employer support for, 41 104
High-cost case management for diagnostic purposes, 171
administrative aspects of, 133, 139 effects of utilization management on use
basic process, 3, 123-130 and costs of, 3-4
case manager responsibilities, 121-122, elective, 3
126, 128-130, 139, 140, 280 -281 emergency admissions, 3, 108
confidentiality concerns, 132, 139 n.6 HMO rates, 44
cost-effectiveness judgments in, 5, 66, length-of-stay norms, 81-83
149 prior to day of surgery, 79
costs of, 130 versus outpatient care, 93-94
definition of, 19, 119 Hospitals
denial rates, 145 attitudes on high-cost case management,
employee awareness of, 127 133
employer support for, 41, 133-134 diversification by, 106
enrollee preceptions of, 132-133 incentive payments to physicians, 47
evaluation of, 134-138 multihospital systems, 50
exceptions to limitations in benefit con- occupancy rates, 106
tracts, 19-20, 122 physician relations with, 106
focus and characteristics of, 3, 19, preadmission checks on insurance cover-
120-122, 134, 145 age, 70
freestanding programs, 123-124 responsibilities on utilization manage-
growth of, 119 ment, 155-156
hypothetical case summaries, 123, third-party-payer relationships with,
125-126 106-108
identification and screening of cases, unnecessary ancillary services, 45-46
68-69, 123, 126, 127-128, 137-139, utilization management by, 64-65, 106;
148 see also Discharge planning
impact of, 4, 130-131, 146 Hysterectomy, 46, 88
improvements in, 138-139
legal issues in, 140-141 I
measures of, 135-136 Independent practice associations (IPAs)
of Medicare patients, 138 financial incentives in, 224-225, 226, 241
negative aspects of, 132-133 management information system in, 241
number of patients in programs, 137 model for, 35, 37
objectives of, 122-123 operational responsibilities of, 212-213
on-site, 129 peer review in, 216, 222
operational variations in, 123-127 physician contracting arrangements
positive aspects of, 132, 133, 141-142 with, 217
protocol standardization, 139, 148 physician selection by, 215, 218-219
provider effects, 133 physician-sponsored, 240-243
purchaser effects, 133-134 quality assurance in, 216, 218
purchaser's role in, 122-123 utilization management procedures, 51,
and quality of care, 133, 134 59, 217-218, 242
savings opportunities with, 121-122, 131 Information resources, on cost and use of
sources of, 122 health services, 47-48
team approach to, 128 International Medical Centers Inc., 205
utilization management categories apply- Intracorp, 131
ing to, 123-124

INDEX 306
J of consultants and employees, 185-186

John Deere & Company, 30, 93 for defamation, 183-184
Joint Commission on the Accreditation of direct, 189
Health Care Organizations, 116 and duty of care, 175-177, 180
Joint Commission on the Accreditation of of employer/payer, 109, 188-190
Hospitals, 36 n.5 fiduciary relationship and, 181, 187-188
indemnity, 190
K infliction of emotional distress, 182
and informed consent, 192
Kaiser Foundation Health Plan, 30, 35
for interference with contractual advan-
Kaiser Permanente, 218
tage, 183-184
Kemper, 15
negligence, 175-179, 180, 181, 189
Knee arthroscopy, 83
products, 182-183
L of review organization, 4, 78, 83, 88, 89,
140, 145, 173-188, 284-285
Labor-Management Group, 30, 40-41 and standard of care, 176-178, 185
Legal issues and litigation vicarious, 189-190
appeal rights, 180-181 and warranty theories, 182
award of attorneys' fees, 188 Louisiana, regulation of utilization review
discriminatory treatment claims, 141 organizations, 186
ERISA preemption, 174, 187-188, 189
exceptions to limits in benefit contracts, M
140-141
expedition of judicial review, 193-194 Maine, regulation of utilization review
in high-cost case management, 140-141 organizations, 186-187
independent contractor defense, 189-190 Mammography screening, 23
and jury sympathies, 174-175 Maryland, regulation of utilization review
limits on damages, 187 organizations, 186
medical necessity concept, 171, Massachusetts Business Roundtable, 42
172-173, 174, 176-177, 184, 185, Mayo Clinic, utilization review plans, 59,
187, 194 107
ostensible agency doctrine, 190 McCarthy, Eugene, 37
for patients, 193-195 Medicaid programs
payment for unnecessary medical ser- cost-containment strategies, 38
vices, 194-195 evaluation of cost-containment strate-
qualified privilege defense, 184 gies, 96
racketeer influenced and corrupt organi- geographic data integration on, 69
zation action, 188 liability for cost-containment mecha-
Sarchett v. Blue Shield of California, 88, nisms, 88, 171
170-171 second-opinion programs for, 39
in utilization review, 88-89, 169-196 Medical malpractice, 41, 105, 159, 191;
Wickline v. California, 88, 109, 140, see also Liability
170, 171-172, 174-176, 178-179, Medical organizations, see individual
186, 188, 191, 192 groups
Liability Medical practice
antitrust, 184-185 employer influence on standards for, 109
of attending physician, 88, 105, 190-192 guidelines, 10-11, 23, 49, 107, 147,
bad faith in insurance and, 175, 159-160, 215
180-181, 187, 189 medical necessity concept, 171, 172-173
for breach of contract, 175, 179-180, 189 reasons for changes in patterns of,
causation and, 172, 174, 178-179, 180, 136-137
181, 187 variations in physician styles of, 46

INDEX 307
Medical services/procedures savings calculations, 114-115, 134-137

criteria for determining necessity of a short time series, 100
procedure, 45, 83 Midwest Group on Health, 42
expenditures for, relative to number of Muscular dystrophy, 121, 136
physicians, 24
fee-for-service payment rates and vol- N
ume of, 44 National Conference on Medical Costs, 40
geographic variation in patterns of, 44-45 Nazemetz, Patricia M, 42
inappropriate/unnecessary, 2, 15, 39, Nightingale, Florence, 36 n.5
44-46, 103, 194-195 Nurse reviewers/managers
methods for indentifying inappropriate high-cost case management by, 128, 140
use, 45 n.7
payment for unnecessary services, liability of, 177
194-195 monitoring of, 73
rationing of, 24, 148-149 qualifications, 72-73, 140 n.7, 276-277
variations in utilization, 44-45 responsibilities, 66, 70, 71-73, 77, 84,
Medicare 85, 89, 103-104, 276-277
cost-containment strategies, 38, 43 scope of assessments, 148-149
costs of, 38 training, 73, 104
denials of payments, 77, 102 Nursing homes, prospective payment sys-
economic incentives to minimize days tem and utilization of, 102
of care, 64
high-cost case management in, 138 O
preadmission review for, 14, 102 Office of Personnel Management, 122 n.2
precertification of beneficiaries, 59, 60
prospective payment system, 30, 40, 98, P
101, 102, 106, 115, 116, 121 n.1
Patients/enrollees
quality assurance program for, 116
case manager relationship with, 128-130
quality of care in HMOs, 205
high-cost, types of, 121
responsibility for reviewing care for
legal issues for, 193-195
beneficiaries, see Peer review
medical abandonment of, 192
organizations
payment obligations for unnecessary
second-opinion programs for, 39
services, 194-195
Methodological problems in studies of
penalties for noncooperation in review,
utilization management
18-19, 69, 88
behavioral biases, 22
responsibilities on utilization manage-
claims data, 111, 137, 158
ment, 9, 69-70, 108, 155-156
confounding by other interventions,
Peer review organizations (PROs)
93-94, 96, 102, 115, 138
appeals processes, 85
cost/price change considerations, 100
case management by, 121 n.1
data availability and quality, 48
claims data tracking by, 116
evaluator knowledge of data sources, 100
congressional support for, 47
group data, 94, 111, 114, 137
demonstration project on small-area
high-cost case management, 134-138
analyses for, 45 n.7
measures of impacts and limitations, 22,
denials of Medicare payments, 77, 102
86, 112-113, 134, 135-136
federal oversight of, 11-12, 65, 154
medical care prices, 115
geographic data integration, 69
noneconomic effects, 115-116
high-cost case management by, 131
nonprogram variables, 100
liability of, 185
program data, 100, 114
physician adviser role in, 74-75
in review versus nonreview groups, 100

INDEX 308
preprocedure review by, 83 incentives for acceptance of utilization

private clients, 39, 60 management, 153-154
second opinion programs, 39 information feedback to, 49
utilization management in, 40, 66, 69, interference with contractual advantages
89, 246-249 of, 183-184
variation in, 65, 89 involvement in developmental criteria,
Physician incentives 147, 153, 205
administering, 226-227 liability of, 88, 105, 172, 178-179,
HMO, 205, 211, 216, 224-227, 230-231, 190-192
233-234, 236, 238, 241 number of, and costs of health care, 24
punitive/negative, 49 obligation to provide treatment to
and quality of care, 205, 224-225 patient who cannot pay, 192
to reduce Medicare services, 47 response to prepaid plans, 35
restrictions on, 47, 225 responsibilities on utilization manage-
surveys on, 225-226 ment, 8, 155-156
types of, 224-225 review organization's handling of, 72,
Physician-patient relationship 74, 78
liability and, 183-184, 191, 192, 194 third-party-payer relationships with,
prior review and, 105-106 106-108
Physician Payment Review Commission, variation in practice styles, 46
107 Preadmission review
Physicians, advisers cost assessments during, 66
attitudes of practicing physicians defined, 18
toward, 75 for emergency admissions, 108
bases for decisions, 73-74, 85, 176, 177 employer support of, 41
caseloads, 76, 278-279 focus of, 120
liability of, 176, 177, 185-186 high-cost case screening by, 127
licensing of, 77, 88 by hospitals, 106
local versus nonlocal, 76 impact on Medicare beneficiaries, 102
monitoring performance of, 75, 77 and inpatient hospital days, 93, 98
PRO use of, 74-75, 77 patient initiation of, 87
productivity incentives, 77 physician certification of necessity of
qualifications of, 77, 278-279 admission, 37
role in prior review, 66, 72, 73-78, 83, by PSROs, 39
86, 89 variations in, 65
scope of assessments, 148-149 waiver of second-opinion requirement
training of, 77 because of, 88
Physicians, attending Predetermination programs, 134 n.4, 137
attitudes on high-cost case management, Preferred provider organizations (PPOs),
133 50, 69, 184
attitudes on prior review, 102-103, 105, Prepaid group practices (PGPs), 30-31,
146, 170 35, 51
case manager relationship with, 128-130 Preservice/preprocedure review, 18, 79
contracting arrangements with HMOs, Prior review
211-212 and actuarial estimates of savings, 101
defamation suits by, 183-184 anxiety and inconvenience caused by, 4,
education of, as cost-containment strat- 103, 146
egy, 21, 49, 147, 156 appeals processes, 6, 7, 85-86, 107, 108,
fee-for-service payments and volume of 149-150, 171, 172, 179, 183
services, 44 before-and-after studies of, 91-93, 94
goals of, 211 comparative studies of, 93-96
-hospital relations, prior review and, 106 components of, 3, 18, 66-67, 83

INDEX 309
contingent nature of, 19 see also Admission review;

contractual descriptions of services, Continued-stay/concurrent review;
282-283 Discharge
criteria for assessment of care, 7-8, planning;
79-85, 89, 159-160 Preadmission review;
definition of, 17-18, 170 Second-opinion programs
employer/purchaser reactions to, 42, Private employers
108-110 attitudes on high-coat case management,
enrollee/patient education on, 7, 133-134
103-104, 109 education of employees on utilization
financial penalties for noncooperation management requirements, 7,
in, 18-19, 104 103-104, 109, 151, 182
focus of, 66, 89, 110 effects of prior review on, 108-110
guidelines for conduct of, 108, 148 expenditures for health benefits, 2
handling of attending physicians, 72, 74, factors shaping decisions on prior
78 review, 109-110
impact of programs, 91-116 health care cost-containment initiatives,
initiation of, 42, 69-70 40-43
and inpatient hospital utilization, 92, 93, with insured health benefit plans, 2;
96-97, 110 see also Health insurance plans
integration with benefit plan administra- liability for utilization review, 109,
tion, 66-69 188-190
liability in, 180 population covered by health insurance
limitations of data on, 114 plans through, 32
measures of impact, 112-113 PRO review contracts with, 39
for Medicare recipients, 14 responsibilities in utilization manage-
multivariate studies of, 96-98 ment, 6-7, 69-70, 150-151
nurses' role in, 66, 71-73, 89 self-insurance by, 2, 41, 60, 61, 182, 189
and outpatient/physician office services share of spending for health services, 40
use, 92, 97 Professional Standards Review Organiza-
and patient costs, comfort, and conve- tion (PSROs), 39, 59, 60, 66, 79, 93
nience, 92, 104, 109 Proprietary information, and competition,
physician adviser role in, 66, 72, 73-78, 7-9, 286-287
83, 86, 89 Prospective Payment Assessment Com-
and physician-hospital relations, 106 mission, 102
and physician-patient relationship, Prospective reimbursement
105-106 employer support for, 41
and provider-purchaser relations, see also Medicare
106-108 Prostatectomy, 46, 156
and quality of care, 4, 7, 102-103, 110, Provident Mutual, 15
115-116, 146, 149 Provider payments
refusals to authorize services, 19 controls on, 35
reporting and feedback mechanisms, Psychiatric cases
86-87 health plan coverage for, 141 n.8, 149
responsibility for obtaining, 108 high-cost case management for, 128,
retrospective denial of claims following, 138, 141 n.8
19, 98, 106
spillover effects of, 98, 101, 110 Q
targeting of, 108 Quality assurance
timeliness in handling of requests, 107, defined, 219
108 in high-cost case management, 140
trends in use of, 14 in HMOs, 206, 216, 219-220, 222-224
weaknesses in evidence on, 99-101, 145

INDEX 310
monitoring performance of nurse- Research on variations in utilization man-

reviewers and physician-advisers, 73, agement, 44-45, 46, 48, 65
75, 77 Retrospective utilization review
operational problems impairing, 222-224 advantages and disadvantages of, 20
in utilization management, 6, 73, 75, 77, appeals processes, 193
140, 150 denial of claims following prior certifica-
Quality of care tion, 19, 98, 106
guideline development for, 154 of high-cost cases, 127
high-cost case management and, 133, 134 hospitalization criteria, 79-80
in HMOs, 205, 210, 211 initial focus of, 35
physician incentive plans and, 205 litigation, 88, 170-171
prior review and, 4, 7, 102-103, 110, by PROs, 69
115-116, 146 by PSROs, 39, 69, 79-80
utilization management and, 4, 36 n.5, regulation, 39;
46-47, 146, 154 see also Legal issues and litigation
targeting problem providers, 69
R Risk pools, management of, 33-34
Rand Corporation, 46, 107 Ross-Loos Medical Group, 30, 35
RCA Plan for Health, 95
Recommendations S
appeals of review decisions, 6, 149-150 Second-opinion programs
disclosure criteria, 6, 7, 149, 160 impact of, 98-99
employer/purchaser responsibilities, objectives of, 87
6-7, 150-151 participation rates, 99
long-term, 9-12, 157-161 and patient anxiety, 103
near-term, 6-9, 150-157 penalties for nonparticipation in, 99
oversight of utilization management, prior review combined with, 83, 99
11-12, 160-161 procedures subject to, 68, 79, 87
patient responsibilities, 9, 156-157 requirements of, 87-88
practice guidelines, 10-11, 159-160 resistance from medical community, 37
practitioners/institution responsibilities, screening referrals for, 18
8, 155-156 types of, 87
quality assurance mechanisms, 6, 150 Self-insurance
research on effectiveness, 6, 9-10, 149, advantages of, 41
157-159 claims processing, 61
review criteria, 6, 7, 9, 11, 149, 159-160 and insurance industry structural
utilization management organizations' changes, 60-61
responsibilities, 7-8, 151-154 and liability, 182, 189
Research on appropriate care, 45-46, 48, monthly costs for family coverage, 41
52, 83 Service Employees International Union,
Research on utilization management 95, 130-131
clinical, limits on, 158, 159 Services Interaction Targets for Opportuni-
feedback and education strategies, 4-5, ties program, 120
147 Social/Health Maintenance Organization
impediments to, 10 demonstration projects, 120
programmatic, 158-159 Social Security Amendments of 1972, 39
recommended, 6, 9-10, 157-159 Spinal injury cases, 131, 138
targets of, 89, 158 Store Workers Health and Welfare Fund,
see also Evaluation of utilization man- 37
agement; Studies and surveys of utilization man-
Studies and surveys agement
of utilization management before-and-after, 91-93, 94

INDEX 311
case studies of HMOs, 229-243 industry structure and process, 58-90

case studies of PSROs, 93 information resources for, 47-48
comparative, 93-96, 145 and inpatient days, 95-96
control for systemwide influences, 95 operational efficiency, 148
Health Interview Survey, 94 operational problems impairing, 220-222
Hospital Discharge Survey, 94 origins of, 21-52
multivariate, 96-98, 145 in peer review organizations, 246-249
small-area, 44-45 policy and research issues, 227-229
see also Methodological problems in public hearings on, 250-252
studies of utilization management quality assurance in, 6, 7, 43, 150
and quality of care, 4, 7, 36 n.5, 46-47,
T 102-103, 110, 146, 154
Third-party financing of health care and rationing of medical services,
cost management efforts, 32-38 148-149
defined, 28 n.1 recommendations on, 5-12, 143-162
growth of, 27-32 refusals to certify benefits, 4
see also Health insurance industry regulation/oversight of, 4, 11-12, 73, 89,
Tonsillectomy and adenoidectomy, 68, 144, 150, 160-161
76, 80, 82, 104 research influences on, 5, 43-46, 145, 147
Transamerica Occidental, 15 scope of review, 66, 147
Tympanotomy tube insertion, 83 spillover effects of, 98, 101, 110
standardization of processes, 107
U status of, 144-146
Uninsured people and surgical days/rates, 95-96, 97
access to health care, 24 third-party efforts, 32-38
number of, 31-32 see also Cost management/containment
United Mine Workers Union, 37 strategies;
Utilization management Prior review;
administrative burdens on providers, 4, Second-opinion programs
5, 7, 107, 144, 146, 147, 154 Utilization management organizations
assessment strategies, 48-49, 66; agreements with clients, 282-287
see also Criteria for assessment of care; analysis and reporting of utilization and
Methodological problems in studies of cost data, 109
utilization management antitrust liabilities, 184-185
case studies, 229-243 breach of contract by, 179-180
claims administration integrated with, constraints on growth of, 5
61, 68-69 consultant/employee liability, 185-186
common elements of programs, 66 contact with patients or physicians, 66
computerization of, 5, 70, 71, 73, 139, defamation by, and interference with
145, 148, 182-183, 221 contractual advantage, 183-184
concerns about, 5-6, 149-150 duty of care, 175-176
definition of, 2-3, 17 effects of organizational differences in,
education strategies for, 48-49 65
effects on health care use and cost, 1-2, employer/purchaser investigation of,
3-4, 14, 52, 145-146, 220 150-153
employer initiatives, 40-43 financial incentives offered by, 50, 73
evolution of, 147-149 freestanding services, 51, 62, 63,
federal government initiatives, 38-40 123-124, 145
focus of, 4, 145, 169 geographic data integration in, 69
growth of, 2, 14, 105, 143, 169 growth of, 50, 59-61
guidelines for conduct of, 108, 154
in health maintenance organizations,
205-243

INDEX 312
handling of attending physician, 72, 74, computerized claims screening, 37

78 continued-stay/concurrent review, 3, 18,
improvement of provider relations with, 37, 39, 48, 65, 66, 81-82, 93, 98
153-154 effectiveness of, 38-39
independent review organization, 62, historical background, 36-38
173-174, 179-180, 185, 187-188, model treatment profiles, 37
259-260, 266-268 preadmission, 3, 18, 37, 39, 41, 65, 66,
infliction of emotional distress, 182 87, 88, 93, 98
insurance bad faith liability, 180-181 as a prerequisite to Medicare participa-
insurance company subsidiaries, 62, tion, 38
262-264 retroactive, 41
insurer-based service, 63 timeliness in handling of requests/
insurer-or broker-based service, 63-64 appeals, 107, 108, 186-187, 193 -194
IPA model HMO, 62, 271-272
largest firms, 60 V
liability of, 4, 5, 78, 83, 88, 145, Value Health Sciences, 107 n.3
173-188, 284-285
management priorities, 5 W
moral obligations of, 153 Washington Business Group on Health,
negligence by, 175-179 30, 42
not-for-profit PRO, 269-270 Wellness programs, 45
peer review organizations, 60, 66, Wennberg, John, 30, 44
272-274 Workers compensation, high-cost case
procedural safeguards for, 176-177 management in, 131
proprietary information and competition
among, 7-8, 286-287 Z
provider-based, 64
Zenith, 122
quality assurance mechanisms, 6, 62
regulation of, 50, 65, 75, 88, 94, 145
responsibilities of, 7-9, 151-154
reviews/audits of, 62
role specification in contracts, 283-284
selection of, 6-7
site visit summaries, 62, 253-281
small, private, for-profit, 253-256
staff model HMO, 62, 268-269
staffing and performance criteria, 71-78,
276-281, 285-286;
see also Nurse-reviewers/managers;
Physicians, advisers
state regulation of, 186-187
subsidiary of third-party payer, 256-257,
264-266
telephone call handling, 70-71
third-party claims administrators, 61,
258-259, 261-262
triple option benefit package, 51
types and numbers of, 49-51, 59-62
variability in programs, 4, 65, 89,
144-145
volume of business in, 275
Utilization review
admission review, 18, 37
ambulatory care review, 37

UM Book

Uploaded by

Copyright:

Available Formats

UM Book

Uploaded by

Document Information

Original Title

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

UM Book

Uploaded by

Copyright:

Available Formats

THE NATIONAL ACADEMIES PRESS

This PDF is available at http://nap.edu/1359 SHARE

Controlling Costs and Changing Patient Care?: The Role of

328 pages | 6 x 9 | HARDBACK

FIND RELATED TITLES

Access to free PDF downloads of thousands of scientic reports

Copyright National Academy of Sciences. All rights reserved.

Controlling Costs and

Committee on Utilization Management

Bradford H. Gray and Marilyn J. Field, editors

NATIONAL ACADEMY PRESS

Copyright National Academy of Sciences. All rights reserved.

COMMITTEE ON UTILIZATION MANAGEMENT BY

JEROME H. GROSSMAN,* Chairman, President, New England Medical

* Institute of Medicine member

DONALD M. STEINWACHS, Director, Health Services Research and

Copyright National Academy of Sciences. All rights reserved.

1 Utilization Management: Introduction and Definitions 13

2 Origins of Utilization Management 28

3 The Utilization Management Industry: Structure and Process 58

4 Impact of Prior Review Programs 91

5 High-Cost Case Management 119

6 Conclusions and Recommendations 143

Copyright National Academy of Sciences. All rights reserved.

D Summary of Public Hearings 250

Copyright National Academy of Sciences. All rights reserved.

Copyright National Academy of Sciences. All rights reserved.

discuss the organization and methods of utilization management. This workshop

Copyright National Academy of Sciences. All rights reserved.

With great speed and relatively little public awareness, a significant

Copyright National Academy of Sciences. All rights reserved.

The validity of this hypothesis is of interest to everyone involved with

CURRENT STATUS OF UTILIZATION MANAGEMENT

Copyright National Academy of Sciences. All rights reserved.

to manage health care costs by influencing patient care decision-making

Copyright National Academy of Sciences. All rights reserved.

increase in health care costs. Employers who saw a short-term moderation

Copyright National Academy of Sciences. All rights reserved.

HOW UTILIZATION MANAGEMENT IS EVOLVING

ISSUES FOR THE FUTURE

Copyright National Academy of Sciences. All rights reserved.

specific proposed procedures. If the positive potential of utilization

RECOMMENDATIONS FOR THE NEAR TERM

Responsibilities of Employers and Purchasers

Copyright National Academy of Sciences. All rights reserved.

Responsibilities of Utilization Management Organizations

Copyright National Academy of Sciences. All rights reserved.

benefit plan administration. Organizations that have invested their own

Responsibilities of Practitioners and Institutions

Copyright National Academy of Sciences. All rights reserved.

LONGER-TERM RECOMMENDATIONS AND QUESTIONS

Copyright National Academy of Sciences. All rights reserved.

Practice Guidelines and Review Criteria

Copyright National Academy of Sciences. All rights reserved.

Oversight of Utilization Management

Copyright National Academy of Sciences. All rights reserved.

Can anything be learned from government oversight of PROs or the

Copyright National Academy of Sciences. All rights reserved.

UTILIZATION MANAGEMENT: INTRODUCTION AND DEFINITIONS 13