This article was downloaded by: [New York University]

On: 25 May 2015, At: 13:50

Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,
37-41 Mortimer Street, London W1T 3JH, UK

Journal of Clinical Child & Adolescent Psychology

Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/hcap20

The Adolescent Behavioral Activation Program:

Adapting Behavioral Activation as a Treatment for
Depression in Adolescence
a a a b
Elizabeth McCauley , Gretchen Gudmundsen , Kelly Schloredt , Christopher Martell , Isaac
b c d
Rhew , Samuel Hubley & Sona Dimidjian
a
Seattle Children's Research Institute and University of Washington
b
University of Washington
c
University of Colorado School of Medicine
d
University of Colorado at Boulder
Published online: 20 Jan 2015.
Click for updates

To cite this article: Elizabeth McCauley, Gretchen Gudmundsen, Kelly Schloredt, Christopher Martell, Isaac Rhew,
Samuel Hubley & Sona Dimidjian (2015): The Adolescent Behavioral Activation Program: Adapting Behavioral
Activation as a Treatment for Depression in Adolescence, Journal of Clinical Child & Adolescent Psychology, DOI:
10.1080/15374416.2014.979933

To link to this article: http://dx.doi.org/10.1080/15374416.2014.979933

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained
in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the
Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and
are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and
should be independently verified with primary sources of information. Taylor and Francis shall not be liable for
any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever
or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of
the Content.

This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any
form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions
 Journal of Clinical Child & Adolescent Psychology, 0(0), 1–14, 2015
Copyright # Taylor & Francis Group, LLC
ISSN: 1537-4416 print=1537-4424 online
DOI: 10.1080/15374416.2014.979933

The Adolescent Behavioral Activation Program:

Adapting Behavioral Activation as a Treatment
for Depression in Adolescence
Elizabeth McCauley, Gretchen Gudmundsen, and Kelly Schloredt
Seattle Children’s Research Institute and University of Washington

Christopher Martell and Isaac Rhew

University of Washington
Downloaded by [New York University] at 13:50 25 May 2015

Samuel Hubley
University of Colorado School of Medicine

Sona Dimidjian
University of Colorado at Boulder

This study aimed to examine implementation feasibility and initial treatment outcomes
of a behavioral activation (BA) based treatment for adolescent depression, the
Adolescent Behavioral Activation Program (A-BAP). A randomized, controlled trial
was conducted with 60 clinically referred adolescents with a depressive disorder who
were randomized to receive either 14 sessions of A-BAP or uncontrolled evidenced-
based practice for depression. The urban sample was 64% female, predominantly
Non-Hispanic White (67%), and had an average age of 14.9 years. Measures of
depression, global functioning, activation, and avoidance were obtained through clinical
interviews and=or through parent and adolescent self-report at preintervention and end
of intervention. Intent-to-treat linear mixed effects modeling and logistic regression
analysis revealed that both conditions produced statistically significant improvement
from pretreatment to end of treatment in depression, global functioning, and activation
and avoidance. There were no significant differences across treatment conditions.
These findings provide the first step in establishing the efficacy of BA as a treatment
for adolescent depression and support the need for ongoing research on BA as a way
to enhance the strategies available for treatment of depression in this population.

Depression during adolescence is a major public health Klein, & Gotlib, 2003). Depression during adolescence
problem (Lopez, Mathers, Ezzati, Jamison, & Murray, is also linked with increased risk of impaired academic
2006). It is prevalent and associated with high rates of performance, poor peer and family relationship quality,
relapse during adolescence and recurrence during adult- substance abuse, and suicidality (Copeland, Shanahan,
hood (Costello, Egger, & Angold, 2005; Dunn & Costello, & Angold, 2009; Esposito & Clum, 2002;
Goodyer, 2006; Goodyer, Herbert, Tamplin, & Altham, Keenan-Miller, Hammen, & Brennan, 2007; Marmor-
2000; Hankin et al., 1998; Lewinsohn, Rohde, Seeley, stein, 2009; McCarty et al., 2012). These far-reaching
consequences underscore the importance of access to
Correspondence should be addressed to Gretchen Gudmundsen, effective interventions to treat acute symptoms and
Seattle Children’s Hospital, 4800 Sand Point Way NE, M=S OA.5.154, reduce risk of relapse and recurrence. This article
Seattle, WA 98105. E-mail: [email protected]
 2 MCCAULEY ET AL.

describes the development and initial investigation of BA therapy is based on a functional analytic model of
behavioral activation (BA) as a treatment for depression depression (see Figure 1) with two central assumptions
during adolescence. (Ferster, 1973; Lewinsohn, 1974, 2001): (a) the experi-
Both cognitive behavioral therapy (CBT) and inter- ence of response contingent positive reinforcement is
personal therapy (IPT), adapted for use with adoles- necessary to maintain normal mood, and (b) avoidance
cents, have demonstrated efficacy in the treatment of is common and serves as a barrier to engaging in antide-
adolescent depression. However, effect sizes are modest pressant behavior that could naturally be positively rein-
(Weisz, McCarty, & Valeri, 2006) and residual symp- forced. In this model, context is carefully assessed to
toms remain problematic even among responders determine factors that are contributing to and maintain-
(Kennard et al., 2006; Kennard et al., 2009; Mufson, ing depressive behaviors and what behaviors=events are
Dorta, Moreau, & Weissman, 2011; Mufson, Weissman, required to curtail it. Avoidance behavior often offers
Moreau, & Garfinkel, 1999; Rosselló & Bernal, 1999; short-term symptom relief but is maintained via negative
TADS [Treatment for Adolescents with Depression reinforcement and contributes to the persistence of
Study] Team, 2007). Furthermore, important subgroups depression in the long run. Although functional avoid-
of adolescents, such as those exposed to early life ance was recognized as central to depression in the early
adversity, demonstrate poor response to existing treat- 1970s (Ferster, 1973), other treatments for depression
ments (Lewis et al., 2010; Nanni, Uher, & Danese, 2012). have not identified avoidance as a primary treatment
Treatment response for depressed adolescents may be target, making the focus on avoidance in BA unique.
Downloaded by [New York University] at 13:50 25 May 2015

improved by targeting specific functional deficits unique CBT (Beck, Rush, Shaw, & Emery, 1979) includes an
to some adolescents who present with depressive emphasis on scheduling pleasant and mastery activities,
symptoms (Forbes, 2009; Forbes et al., 2006). Neurode- a form of behavioral activation, primarily as a foun-
velopmental data suggest that adolescents are vulnerable dation for cognitive interventions that focus on modify-
to increased sensitivity to social stressors, disruptions in ing distortions in thinking. In BA, activation differs
reward processing, and the tendency to shut down or from simple activity scheduling in that its focus centers
avoid emotional stimuli (Forbes, 2009; Forbes et al., on identifying behaviors that (a) are important within
2006; Hare et al., 2008; Silk et al., 2007; Somerville, the individual’s context, (b) advance the individual’s
Hare, & Casey, 2011; Tottenham, Hare, & Casey, goals and experience of mastery or pleasure, and (c)
2011). Reward processing deficits, a physiological counter avoidance. Although IPT (Mufson et al.,
change associated with risk for depression, may increase 2011) may indirectly address avoidance with its
during adolescence because neural reward systems are attention to improving communication with important
still developing (Davey, Yücel, & Allen, 2008; Forbes, individuals in one’s life, overcoming avoidance is not a
2009). These research findings on disengagement and core treatment target. Studies with adults have
avoidance suggest that treatment for adolescent suggested that BA is an efficacious treatment for
depression may need to target adolescents’ ability both depression (Dimidjian et al., 2006), with positive results
to experience and respond to reward, and overcome across multiple patient populations, settings, and
avoidance (Forbes, 2009). This argument is underscored delivery formats (Dimidjian, Barrera, Martell, Muñoz,
by recent findings that anhedonia is the strongest predic- & Lewinsohn, 2011).
tor, among all depressive symptoms, of increased time to Preliminary findings suggest that the BA approach also
remission and therefore represents an important treat- holds promise as an effective treatment for adolescent
ment target (McMakin et al., 2012). Given that BA depression (Chu, Colognori, Weissman, & Bannon,
focuses on context, and targets environmental stressors 2009; Jacob, Keeley, Ritschel, & Craighead, 2013;
such as social relationships, overcoming avoidance and Ritschel, Ramirez, Jones, & Craighead, 2011). Our study
increasing engagement with reinforcers within the extended this research with a pilot randomized trial com-
environment, it may offer an important approach to paring the Adolescent Behavioral Activation Program
the treatment of depressed adolescents. (A-BAP) to evidence-based practice for depression

FIGURE 1 Behavioral activation model of depression.

 ADOLESCENT BEHAVIORAL ACTIVATION PROGRAM 3

(EBP-D) within a university hospital–based community health clinic, referrals from primary care and=or mental
mental health clinic. We addressed three specific aims. health care providers, and flyers placed in health clinics
First, determine the feasibility of (a) adequate participant and school newspapers. Inclusion criteria included (a)
recruitment and enrollment, (b) successful randomization age 12 to 18 with one parent=guardian willing to partici-
across treatment conditions, and (c) retention of parti- pate; (b) primary Diagnostic and Statistical Manual of
cipants during treatment and follow-up. Second, establish Mental Disorders (4th ed.; American Psychiatric Associ-
‘‘proof of concept’’ by determining if the BA approach ation, 1994) diagnosis of Major Depression, Depression
was associated with clinically meaningful reductions in Not Otherwise Specified, or Dysthymia, or a raw score
depressive symptoms and improvements in global func- of 45 (T score of 65) or greater on the CDRS-R; (c)
tioning that were comparable to a stringent comparison self-report score of 11 or greater on the Short Moods
condition. We anticipated that youth in the BA condition and Feelings Questionnaire (SMFQ; Angold et al.,
would demonstrate a significant reduction in depressive 1995) at the start of treatment; and (d) willingness to
symptoms on the Children’s Depression Rating Scale– be randomized to treatment condition. Exclusion cri-
Revised (CDRS-R; Poznanski & Mokros, 2001) and teria included the inability to complete study question-
scores of 1 to 2 on the Clinical Global Impression naires, psychotic or manic symptoms, acute substance
Improvement Scale (CGI; Guy, 1976) and that such gains use, suicidality requiring immediate, intensive treatment,
would be at least as good as those in EBP-D. Thus in and=or acute medical illness. Youth who met inclusion
this context, findings consistent with the null hypothesis criteria but were taking stimulant medications for
Downloaded by [New York University] at 13:50 25 May 2015

(A-BAP ¼ EBP-D) would be acceptable and support ADHD or were on a stable antidepressant medication
further research. Our third specific aim was to conduct regimen were included. The study was approved by the
an initial exploration of change in activation and avoid- Institutional Review Board, and all participants includ-
ance, potential mediators of change in the BA model. ing a parent=guardian completed and signed Insti-
tutional Review Board–approved assent=consent forms
before beginning the initial evaluation process. Youth
Development of the A-BAP Intervention With who were eligible and gave assent, completed a
Adolescents clinic-based, comprehensive intake evaluation. Youth
Based on open trial pilot testing with eight depressed who were ineligible or declined participation in the study
adolescents, we developed a 12-week, 14-session treat- were offered treatment referrals outside the study. The
ment program for depressed youth 12 to 18 years of study was registered with clinicaltrials.gov (ID number
age. This program tested a proposed model of change NCT01137149).
for adolescents diagnosed with depression, as outlined
in Figure 2, which posits that (a) as barriers to acti-
vation are identified and addressed, the adolescent’s Procedures
ability to fully engage in activities and experience Enrolled participants (N ¼ 60) were randomly assigned
response contingent, positive reinforcement will be to A-BAP (n ¼ 35) or EBP-D (n ¼ 25). A computerized
enhanced; (b) repeated trials of activation, despite program was used to randomly assign participants to
negative mood, will increase the likelihood that these treatment group. This program used a stratified rando-
activities will be experienced as positively reinforcing; mization scheme, where randomization was stratified
and (c) when activities are experienced as contingently based on age (categorically 12–14 years of age, 15–18
positively reinforced, this in turn will lead to improved years of age) and gender with weighting to allow
depressive symptoms. We designed our approach to for greater inclusion in the A-BAP group. Youth parti-
include intervention components (e.g., activities cipants received a $25 incentive for completing the
designed to promote competence, psychoeducation baseline and end-of-treatment assessment batteries. In
regarding depression, outline of details of treatment, each treatment condition, participants received up to
self-monitoring of moods and behaviors, etc.) that have 14 sessions of therapy over the course of a 12-week
been found to be elements in beneficial treatments period.
(McCarty & Weisz, 2007).

A-BAP intervention. Our modification of BA for

use with depressed adolescents can best be conceptua-
METHOD
lized as a behavioral treatment based on a functional
conceptualization of each individual case. In an effort
Participants
to adequately engage adolescents in treatment, the pro-
Participants were recruited from a major metropolitan gram utilized a structured psychoeducational format
area on the West Coast via a hospital-based mental early in the treatment process, with a more flexible
 4 MCCAULEY ET AL.

FIGURE 2 Proposed behavioral activation model of change. Note: RCPR ¼ response contingent positive reinforcement.

approach as treatment progressed. Treatment began session are summarized in a treatment manual that is
with two sessions devoted to reviewing the available from the authors.
assessment-based case conceptualization and introdu- Parents participated in at least two of the A-BAP
cing the BA model to the adolescent alone and then in sessions. This allowed (a) parents an opportunity to pro-
the second session with the adolescent and parent vide the therapist with their perspective on the adoles-
together, followed by a series of sessions introducing cent’s depression, (b) adolescents an opportunity to
particular skills (see Table 1). Four additional sessions describe the behavioral model used in therapy to the
Downloaded by [New York University] at 13:50 25 May 2015

were scheduled, either as needed to extend the skill mod- parent as a means of consolidating the adolescent’s
ules or after introduction of all the skills, to allow for grounding in the model, and (c) parents an opportunity
individualized practice and application. The treatment to get ‘‘on board’’ with the adolescent’s work in BA such
ended with two sessions devoted to termination and that they could reinforce appropriate behaviors and sup-
relapse prevention. Although treatment was structured port therapeutic strategies and gains.
to follow the order seen in Table 1, therapists had the
discretion to tailor the sequence as appropriate to the
Evidence-based practice for depression. Our com-
adolescent’s needs. Functional analysis of behavior
parison condition represents standard care offered in an
was introduced in the ‘‘Getting Active’’ sessions. As part
academically affiliated outpatient clinic setting. EBP-D
of this functional analysis the role of reinforcement in
maintaining behavior, including the importance of eval-
uating the payoff versus the price of behavior choices
while considering both short- and long-term conse-
TABLE 1
quences of behavioral choices, was introduced. For the Content of Behavioral Activation (BA) Sessions
associated homework or practice, each adolescent was
asked to identify at least one behavior=activity that they Sessions Content Areas Specific Topics
associated with feeling less depressed, structure time to 1–2 Case . Introduction to the BA
engage in this over the week, and monitor the effect Conceptualization Model (‘‘Getting Started’’)
on his or her mood. In the core skills sessions, each skill And Introduction
was introduced using structured materials that built on to BA Model
and integrated examples drawn from each adolescent’s 3–4 Activation and . The Situation-Activity-Mood
unique circumstances and priorities. Adolescents were Goal-Directed Connection
encouraged to identify their own issues and actively Behavior . Regulating Mood through
engage in elucidating how each skill applied to their situ- Activity (‘‘Goal-Directed vs.
Mood-Directed Behavior’’)
ation. Role-playing was used in some cases, whereas in
. Why Do I Do What I Do?
all cases the skills were presented in the context of an (‘‘Payoff and Price of
example from the adolescent’s life. Homework or prac- Behavior’’)
tice, assigned in each session, reflected the material cov-
5–8 Core Skills . Problem Solving
ered in the session and included monitoring behaviors . Goal Setting
and their association with mood, taking steps toward . Identifying Barriers
goals, identifying barriers, and testing out strategies to . Overcoming Avoidance
overcome avoidance and other barriers. In the less struc-
9–12 Practice and . All of the above
tured sessions each adolescent identified the key area(s) Application
he or she wanted to work on (e.g., spending more time
with friends) and with the therapist’s guidance employed 13–14 Relapse Prevention . Strategies for Preventing
and Termination Relapse
skills from prior sessions to address the identified prob- . Strengths and Challenges
lem or reach the adolescent’s goal. Guidelines for each
 ADOLESCENT BEHAVIORAL ACTIVATION PROGRAM 5

relied on a CBT approach in most cases (n ¼ 21), and an Assessments and Measures
IPT for Adolescents approach in a smaller subset of
Assessments were completed at baseline and end-of-
cases (n ¼ 4). Although no specified manual was pre-
treatment, with follow-up assessments conducted 6 and
scribed, all therapists had prior formal training in one
12 months after study enrollment. All assessments exam-
or both of these therapeutic techniques and routinely
ined psychiatric symptomatology; adaptive functioning;
employed one of these therapies as part of their standard
activation and avoidance; and academic, social, and
care. Therapists were chosen to participate based on
familial functioning. They were conducted by Inde-
experience in providing these treatments. To ensure con-
pendent Clinical Evaluators (ICEs), who were
sistent dose of treatment between conditions, the study
advanced-level clinical psychology graduate students
provided up to 14 sessions of therapy. Both patient
and naive to the adolescent’s treatment condition. ICEs
and therapist were aware of this at the start of therapy
completed a training program before administering
and planned treatment accordingly.
diagnostic interviews, which included (a) an orientation
EBP-D therapists had the option to include parents
to the evaluation tools, (b) the review and rating of two
in treatment ‘‘as needed’’ but could not engage parents
training interview tapes and one in-person interview
in independent treatments (e.g., initiating an additional
conducted by the gold standard clinical evaluator
parent training protocol, working with parents on indi-
(KS), and (c) the administration of an interview that
vidual adult=parent treatment issues). Therapists, in
was co-rated by the gold-standard clinical evaluator to
both conditions, also had the option to offer a limited
Downloaded by [New York University] at 13:50 25 May 2015

ensure appropriate reliability prior to independently

number of additional sessions, if needed, to allow for
conducting interviews. Interrater reliability at the diag-
a smooth termination of treatment.
nostic level was required to move into the role of an
ICE. As the ICEs were only employed as part of this
project, were not working in the clinic on a routine
Therapists basis, and completed most evaluations after regular
clinic hours or on weekends, they had little contact with
Therapists provided treatment in either A-BAP or
the therapists in either condition.
EBP-D but not both, so as not to bias results. A-BAP
therapists included two postdoctoral fellows, two
faculty=staff psychologists, and one MSW—four were Demographics. Participants completed a standard
women and one was a man; one was Asian American demographics form, and the Hollingshead Four Factor
and the remaining were European American. A larger Index of Social Status (Hollingshead, 1975) was used
cadre of therapists delivered the EBP-D intervention, to determine family composite social status. An edu-
including one advanced-level graduate student, three cation score (range ¼ 1–7), and an occupation score
postdoctoral fellows, four faculty=staff psychologists, (range ¼ 1–9) was assigned for each parent=guardian
one faculty=staff psychiatrist, and one MSW. In this based on information provided by the parent reporter.
group four were men, and all were European American. Education and occupation scores were weighted accord-
The A-BAP therapists completed a 1-day training pro- ing to Hollingshead guidelines to obtain a single score
gram and then met weekly for group supervision for each parent=guardian (range ¼ 8–66). For families
throughout the study. The EBP-D therapists were with multiple caregivers, scores for each were averaged
recruited from the general clinic staff and did not receive to obtain a single socioeconomic status score.
additional study training. Because of clinic space limita-
tions, most treatment sessions took place in a research
center outside the clinic. Scheduling for this space pre- Clinical symptoms. The Kiddie-SADS Diagnostic
cluded overlap of therapists from different treatment Interview (KSADS; Kaufman et al., 1997), used to
arms, and none of the clients were included in discus- establish a diagnosis of depression, is a structured
sions at care conferences. For the EBP-D condition, diagnostic interview with established reliability and val-
one of the faculty psychologists participating in that idity. Selected modules of the KSADS were used to
condition was available to clinicians for case consul- assess depression disorders, mania, oppositional defiant
tation and guidance as needed. All participating thera- disorder, generalized anxiety disorder, overanxious
pists were asked to refrain from cross-condition disorder, and conduct disorder.
discussion, which was easily accomplished because most The Children’s Depression Rating Scale–Revised
therapists did not have overlapping clinical days. The (CDRS-R; Poznanski & Mokros, 2001) was utilized
postdoctoral fellows in both conditions came from var- to assess level of depressive symptomatology. The
ied academic training programs, although all had com- CDRS-R is a semistructured interview-based measure
pleted their internship training within the affiliated modeled on the adult Hamilton Rating Scale for
university program but not in the same year. Depression and has been found to be a sensitive and
 6 MCCAULEY ET AL.

reliable severity measure of depression in youth on a scale of 0 (poor) to 100 (excellent) and has been
(Poznanski & Mokros, 2001). ICEs administered the shown to have good psychometric properties.
17-item scale to both the adolescent and the parent. Item The Behavioral Activation for Depression Scale
values range from 1 to 5 or 1 to 7, with higher scores (BADS; Kanter, Mulick, Busch, Berlin, & Martell, 2007;
reflecting more clinically significant difficulties. A raw Kanter, Rusch, Busch, & Sedivy, 2009) is a 25-item
score at or above 45 (T score at or above 65) indicates self-report measure designed to track changes in proposed
a likely depressive disorder. mediators of BA: activation and avoidance. The BADS
The Clinical Global Impression Improvement has four subscales, two of which were analyzed in the cur-
Scale (CGI; Guy, 1976) was used to assess the level of rent study: Activation (focused, goal-directed activation
symptom severity at intake and end of treatment. Scores and completion of scheduled activities) and Avoidance=
on the CGI range from 1 to 7, with lower scores denot- Rumination (avoidance of negative aversive states and
ing less severity=better outcome. The CGI measure has engaging in rumination rather than active problem
been widely utilized in a number of treatment outcome solving). The BADS has been found to have good factor
studies with adolescents (e.g., Mufson et al., 2004; structure, internal consistency, and test–retest reliability
TADS [Treatment for Adolescents with Depression (Kanter et al., 2007). In the current study, the BADS
Study] Team, 2004) to assess treatment response with demonstrated good overall internal consistency (pretreat-
improvement scores of 1 (very much improved) or 2 ment a ¼ .88; end of treatment a ¼ .91), as well as adequate
(much improved), indicating positive response to to good reliability for the subscales examined (Activation:
Downloaded by [New York University] at 13:50 25 May 2015

treatment. pretreatment a ¼ .78; end of treatment a ¼ .77;

The Short Moods and Feelings Questionnaire Avoidance=Rumination: pretreatment a ¼ .84; end of
(SMFQ; Angold et al., 1995) was used to index self- treatment a ¼ .87).
reported distress, including depressive thoughts, feel-
ings, and behaviors. Items are each scored on a 3-point
scale: not true, sometimes true, and true. The 13-item Adherence. BA adherence was rated from video
measure has high internal consistency and has been recordings using items based on the Collaborative Study
noted to successfully distinguish between children and Psychotherapy Rating Scale (Form 6) (Evans, Piasecki,
adolescents with and without depressive disorders Kriss, & Hollon, 1984) and consistent with the Quality
(Angold et al., 1995). In the current sample, internal of Behavioral Activation Scale (Dimidjian, Hubley,
consistency was high (pretreatment a ¼ .84, end of treat- Martell, Herman-Dunn, & Dobson, 2012). Items were
ment a ¼ .95). For inclusion in this study, a categorical rated on a 7-point scale (higher ratings indicating greater
cutoff score of 11 was utilized as an indication of signifi- adherence to BA) and included the following: sets and
cant depressive symptomatology. This cutoff, utilizing follows agenda, maintains an activation focus, reviews
adolescent report only, was selected based on findings homework, assigns homework, attends to client under-
from large-scale epidemiological studies (Angold, standing, plans for preventing relapse, assesses cues and
Erkanli, Silberg, Eaves, & Costello, 2002; McKenzie consequences of behavior, discusses client avoidance,
et al., 2011; Patton et al., 2008). One additional question assesses behavioral patterns, identifies desired behavioral
was added to the SMFQ, but not to the overall score, to changes, generates or evaluates solutions, provides
track level of suicide risk. education about activation principles, encourages
The Multidimensional Anxiety Scale for Children self-monitoring, increases pleasure and mastery, sche-
(MASC; March, Parker, Sullivan, Stallings, & Conners, dules activities, structures activities, manipulates behavior
1997) is a self-report questionnaire used to assess the via cues or consequences, teaches skills, elicits client prac-
nature and severity of anxiety symptoms. Widely uti- ticing new behavior in session, and reestablishes routines.
lized and with adequate psychometric properties (March A team of four raters, including three undergraduate
et al., 1997), the MASC is normed for age and gender students and one advanced graduate student, underwent
and yields raw and T scores. Administered at pretreat- approximately 20 hr of training, which involved an
ment, the MASC Anxiety Disorder Index demonstrated orientation to fidelity monitoring, reviewing instructions
adequate internal consistency (a ¼ .62). for rating BA adherence, and two 5-hr didactic training
sessions that included discussion forums and supervised
group practice. Raters were naı̈ve to treatment
Behavioral function. The Children’s Global assignment, session number, and all study hypotheses.
Assessment Scale (C-GAS; Shaffer et al., 1983) was All ratings were completed while watching a video
administered by ICEs to assess adolescent global func- recording of the session. Each rater watched and rated
tioning and general symptom severity. Based on the each session independently and met biweekly to review
adult GAS and modified by Shaffer et al. (1983) for sessions jointly rated by all raters. A total of 44 sessions
use with children and adolescents, the C-GAS is rated was randomly selected from Sessions 3 to 11 for each
 ADOLESCENT BEHAVIORAL ACTIVATION PROGRAM 7

patient (n ¼ 28 for BA and n ¼ 16 for EBP-D). that missing follow-up data varied by treatment
Reliability was calculated based on 14 sessions rated condition, we also used a multiple imputation approach
jointly by two raters, and using Shrout and Fleiss’s (Raghunathan, 2004). Ten imputed data sets were cre-
(1979) fixed effects model, intraclass correlation coeffi- ated for each outcome using the imputation by chained
cients were computed, yielding an intraclass correlation equations approach with predictors of the missing out-
coefficient of .61. BA therapists received significantly comes including baseline levels of the outcome and up
higher ratings of adherence to BA (65.47, SD ¼ 14.73) to three other covariates found to be highly correlated
than EBP-D therapists (53.55, SD ¼ 10.37), t(42) ¼ with the outcome (Royston, 2005). The linear mixed
2.85, p ¼ .007 suggesting adequate to strong adherence models were run on each imputed data set, and results
to the BA model. were combined across data sets to yield summary esti-
mates and corresponding 95% confidence intervals and
p values according to Rubin’s (1987) rules that account
Data Management and Analyses
for the uncertainty of the imputed values. Results using
Power calculations were done prior to initiating the the multiply imputed data were similar to original
study to determine necessary sample size. Given the models.
fact that the BA treatment approach had never been
applied to this specific population, power calculations
were estimated approximations based on available ado-
Downloaded by [New York University] at 13:50 25 May 2015

RESULTS
lescent treatment studies. Examining studies of CBT
and IPT utilizing a ‘‘diagnosed’’ sample of adolescents,
Feasibility
effect sizes were reported to range from .07 to 2.02
(Weisz et al., 2006). Assuming a large effect size of .8 Participant recruitment, enrollment, and randomiza-
based on these studies, we determined that a proposed tion. As outlined in Figure 3, phone screens were con-
sample size of 25 and 25 for the two groups would pro- ducted with 124 adolescents: 85 were eligible to
vide adequate power of 80.1% to yield a statistically participate in the intake evaluation, 13 were ineligible,
significant result, which was in the middle of the above and 26 declined the offer to participate further. Reasons
range. for ineligibility included insufficient reading ability
In assessing change over time, consistent with proce- (n ¼ 1); subclinical depression (n ¼ 2); depression sec-
dures used in the TADS Study (TADS [Treatment for ondary to an acute medical condition (n ¼ 1); the
Adolescents with Depression Study] Team, 2003, 2004, absence of depression as the primary disorder, including
2007), we followed the formula described by Rintelmann presence of mania (n ¼ 3), conduct disorder (n ¼ 1), and
et al. (1996) to calculate improvement on the CDRS-R. Asperger’s disorder (n ¼ 1); acute substance use (n ¼ 3);
This procedure corrects for the nonzero minimum score and suicidality requiring immediate, intensive treatment
of the CDRS-R and, in turn is considered to be a more (n ¼ 1). The remaining 26 youth declined participation.
appropriate method for assessing change. Specifically, Eighty-five cases participated in an in-person evaluation
the formula used to calculate percent change was with an ICE, resulting in the identification of 61 indivi-
[baseline CDRS-R score 17] – [endpoint CDRS-R duals who met all eligibility criteria. Reasons for ineligi-
score 17]=[baseline CDRS-R score 17]. Descriptive bility at intake included insufficient reading ability
statistics were used to report on feasibility of recruit- (n ¼ 1); subclinical depression (n ¼ 11); depression sec-
ment, randomization, and retention. ondary to an acute medical condition (n ¼ 1); the
To compare changes in continuous outcomes from absence of depression as the primary disorder, including
pretreatment to end of treatment, linear mixed effects primary diagnoses of bipolar disorder (n ¼ 2), opposi-
models were used with intercepts allowed to vary across tional defiant disorder or conduct disorder (n ¼ 3), an
individual participants. As covariates, treatment eating disorder (n ¼ 2), and posttraumatic stress disorder
condition (0 ¼ EBP-D, 1 ¼ A-BAP), time (0 ¼ pretreat- (n ¼ 1); Asperger’s disorder (n ¼ 1); and suicidality
pretreatment, 1 ¼ end of treatment), and their interac- requiring immediate, intensive treatment (n ¼ 2). One
tion were included in the statistical model. potential participant declined, and 60 agreed to be ran-
Improvement in CGI was modeled as a dichotomous domly assigned to A-BAP or EBP-D.
variable. For analyses of this outcome, we used tra- As summarized in Table 2, the A-BAP and EBP-D
ditional logistic regression with randomized treatment groups did not differ at baseline in relation to gender,
group as the covariate. All analyses were conducted age, ethnicity, family socioeconomic status, or scores
using an intent-to-treat design. on two measures of depression severity—the CDRS-R
Under the assumption of missing-at-random, linear and the SMFQ. Participants had a mean age of 14.9
mixed effects models provide unbiased effects in the (SD ¼ 1.53), and 63.7% were girls. Thirty-seven percent
presence of missing data (Atkins, 2005). However, given of those randomized to A-BAP and 36% of those
 8 MCCAULEY ET AL.
Downloaded by [New York University] at 13:50 25 May 2015

FIGURE 3 Recruitment flow. Note: A-BAP ¼ Adolescent Behavioral Activation Program; EBP-D ¼ evidence-based practice for depression;
FU ¼ follow-up.

randomized to EBP-D were on antidepressant medi- familial move), and two adolescents were unwilling to
cation. One adolescent in each group was on both an continue in treatment; four of the five who dropped out
antidepressant and a stimulant medication; two youth of A-BAP did so within two sessions of initiating therapy.
randomized to A-BAP, and one to EBP-D were on Completers across both treatment groups averaged
stimulant medications only. 13.77 sessions (SD ¼ 1.74), whereas noncompleters aver-
aged 3.0 (SD ¼ 2) sessions. Youth who completed the
Retention and characteristics of completers versus protocol were similar to those who did not on core
noncompleters. Of the 60 cases randomized, 53 com- demographic variables, including gender, age, race,
pleted the treatment protocol (88.3%) with five A-BAP and ethnicity, socioeconomic status, and on baseline
participants (14.1%) and one EBP-D participant (1.7%) depression (CDRS-R and SMFQ) and global function-
dropping out of treatment. An additional A-BAP partici- ing (CGAS). However, compared to youth who com-
pant was transferred to alternative care based on a pleted the protocol, youth who did not complete the
decision made by the treatment team in concert with protocol demonstrated significantly higher anxiety:
the patient and parent that depression was not the core MASC Anxiety Disorders Index, F(1, 54) ¼ 5.10,
problem and that the participant would be better served p < .028. Of the 53 participants who completed treat-
with a different form of treatment. Although higher, the ment, the mean number of sessions attended for youth
drop-out rate for A-BAP did not differ significantly from in the A-BAP condition was slightly higher at 14.44
that in the EBP-D condition (p ¼ .22, two-tailed Fisher’s (SD ¼ 1.78) as compared to 13.13 (SD ¼ 1.30) for the
exact test). Three participants dropped out secondary to EBP-D group, t(51) ¼ 3.01, p ¼ .004. All participants in
transportation or unanticipated family factors (e.g., both groups completed 10 or more sessions.
 ADOLESCENT BEHAVIORAL ACTIVATION PROGRAM 9

TABLE 2
Sample Characteristics at Baseline

Total BA EBP-D
(N ¼ 60) (n ¼ 35) (n ¼ 25) Statistica

Gender N (%), p
Male 22 (36.7) 13 (37.1) 9 (36.0) ns
Female 38 (63.3) 22 (62.9) 16 (64.0) ns
Age M (SD), F 14.90 (1.53) 15.17 (1.52) 14.52 (1.48) ns
Race=Ethnicity Non-Hispanic White N (%), p 40 (66.7) 23 (65.7) 17 68.0) ns
Hollingshead M (SD), F 51.00 (12.65) 52.64 (13.14) 48.66 (11.75) ns
CDRS-R M (SD), F 57.70 (10.39) 57.60 (11.80) 57.84 (8.26) ns
SMFQ M (SD), F 15.87 (6.06) 16.05 (6.05) 15.64 (6.20) ns

Note: BA ¼ behavioral activation; EBP-D ¼ evidence-based practice for depression; CDRS-R ¼ Children’s Depression Rating Scale–Revised;
SMFQ ¼ Short Moods and Feelings Questionnaire.
a
Fisher exact or analysis of variance tests were completed; no differences in means between demographic characteristic levels were statistically
significant using a .05 alpha level.
Downloaded by [New York University] at 13:50 25 May 2015

Follow-up assessments. Completion of the end of completed the end-of-treatment evaluation (45.05,
treatment evaluations was nonsignificantly higher in the SD ¼ 14.23) fell in the lower range of the ‘‘likely’’ depress-
A-BAP group (77%) than in the EBP-D (64%) (p ¼ .38, ive disorder category; 42% had a mean score of 40 or
two-tailed Fisher’s exact test). One A-BAP participant lower at the end of treatment, placing them in the ‘‘poss-
and eight EBP-D participants did not complete an ible’’ or ‘‘unlikely’’ depressive disorder categories. Com-
end-of-treatment evaluation but did complete one of paratively, the mean CDRS-R score for the A-BAP
the subsequent follow-up assessments. Of those who completers (40.18, SD ¼ 13.91) at the end of treatment
did not continue in active treatment, none were willing fell in the ‘‘possible’’ depressive disorder range, with
or able to complete any of the follow-up evaluations. 76% (22 of 29) of treatment completers in this group
having a CDRS-R score of 40 or lower. Of the 27 who
completed the A-BAP treatment protocol and end-of-
Symptom and Function Change
treatment assessment, 21 were rated as no longer meeting
Linear mixed-model analyses were used to examine diagnostic criteria for depression (Major Depression,
changes (improvements) in clinical symptoms and beha- Depressive Disorder NOS, or Dysthymia) on the
vioral function of each group from pretreatment to end K-SADS at the end of active treatment; six continued
of-treatment assessment within each treatment group to meet criteria for a diagnosis of depression. End-of-
(A-BAP and EBP-D; Table 3). Both A-BAP and EBP-D treatment KSADS data were missing for two parti-
treatment groups showed statistically significant cipants: one had a score of 5 on her final session SMFQ,
improvements in their clinical symptoms. There was which is well below the clinical cut off of 11, whereas the
no evidence of statistically significant change over time other completed the 6-month follow-up evaluation and
in Activation, although Avoidance=Rumination signifi- was rated as falling within the ‘‘depressive disorder
cantly decreased over time in both conditions. There likely’’ range on the CDRS-R. Of the EBP-D group, 16
were no statistically significant interactions between completed the end of treatment assessment, four no
treatment condition and time indicating that the longer met criteria for depression, nine were rated as
changes in symptoms and function over time were not improved but with residual symptoms, and three contin-
significantly different between A-BAP and EBP-D. In ued to meet diagnostic criteria for depression.
addition, results from logistic regression of CGI
Improvement between A-BAP and EBP-D were not
statistically significant (odds ratio ¼ 2.08), 95% confi- DISCUSSION
dence interval [0.64, 6.74], p ¼ .220, although both
groups showed significant improvement over time. The last decade has seen a renewed interest in use of
Categorical outcomes were operationalized according behavioral approaches for the treatment of depressive
to CDRS-R categories in which raw scores of 45–55 are disorders. Building on this interest and the promising
defined as ‘‘depressive disorder is likely to be confirmed,’’ preliminary open trial studies testing the BA approach
31–42 as ‘‘possible,’’ and 20–29 as ‘‘unlikely.’’ The mean with depressed adolescents (Chu et al., 2009; Jacob
CDRS-R score for the EBP-D participants who et al., 2013; Ritschel et al., 2011), this pilot study was
 10 MCCAULEY ET AL.

TABLE 3
Means, Standard Deviation of Clinical Symptoms, and Behavioral Function Outcomes and Comparison of Pretreatment and End-of-Treatment
Outcomes for BA and EBP-D Conditions

BA Groupa EBP-D Groupb

End of End of
Pre treatment Treatment Change From Pre treatment Treatment Change From
Pre to Postc Pre to Postc
Measures M (SD) M SD b [95% CI] M SD M SD b [95% CI] p Interactiond

CDRS-R 57.6 (11.8) 40.2 (13.9) 18.6 [23.2, 13.9] 57.8 (8.3) 45.1 (14.2) 13.1 [18.3, 7.8] 0.12
SMFQ 16.1 (6.1) 6.3 (7.4) 9.6 [12.3, 6.9] 15.6 (6.2) 6.5 (6.5) 9.2 [12.7, 5.8] 0.53
CGI Severity 4.4 (0.9) 2.7 (1.3) 1.7 [2.1, 1.3] 4.7 (0.6) 2.8 (1.2) 1.9 [2.5, 1.3] 0.80
CGAS 48.9 (8.9) 68.8 (11.7) 19.9 [14.8, 25.0] 50.4 (7.9) 67.3 (12.8) 16.1 [9.5, 22.7] 0.38
BADS Activation 17.3 (7.6) 21.1 (10.1) 3.5 [0.7, 7.7] 15.5 (9.9) 20.1 (12.1) 4.8 [0.5, 10.2] 0.87
BADS Avoidance 25.1 (9.1) 16.1 (9.9) 9.4 [13.9, 4.8] 21.0 (10.3) 16.9 (11.6) 9.8 [15.7, 3.9] 0.91

Note: BA ¼ behavioral activation; EBP-D ¼ evidence-based practice for depression; CDRS-R ¼ Children’s Depression Rating Scale-Revised;
SMFQ ¼ Short Moods and Feelings Questionnaire; CGI ¼ Clinical Global Impression; CGAS ¼ Children’s Global Assessment Scale;
BADS ¼ Behavioral Activation for Depression Scale.
a
n ¼ 35.
Downloaded by [New York University] at 13:50 25 May 2015

b
n ¼ 25.
c
Indicates the change in mental health outcomes from pretreatment to end of treatment for each study group (BA and EBP-D) and are results
from linear mixed models.
d
p value is the test that compares whether changes from pretreatments to end of treatments are different between BA and EBP-D conditions.

conducted to assess the feasibility of using a BA Adolescents with Depression Study] Team, 2004).
approach to effectively treat depressive symptoms in Although more participants randomized to the BA con-
adolescents as part of a treatment development process. dition dropped out of treatment, factors unrelated to the
Weisz, Jensen, and McLeod (2005) argue that treat- nature of treatment (e.g., burden of a long commute to
ments should be developed and tested through a appointments; family moved out of the area; poor
sequence of studies designed to assure that developing engagement with initial agreement based on need to
treatments work with clinically referred individuals appease their parents) were cited in all cases. Because
being treated in real-life practice conditions; provide evi- it was made clear that the study was designed to test
dence of the treatment’s nature, necessary and sufficient the BA approach, it is possible that these adolescents
components; and explore moderators and mediators or and their parents lacked confidence that BA would be
change processes associated with treatment impact. This effective and therefore left treatment, although this
trial of A-BAP, as just outlined, focused on completing was not stated directly. Anxiety may have been a con-
the first phases outlined in this treatment development tributing factor as well, because the youth who did not
model: (a) manual development, (b) efficacy testing continue in treatment endorsed greater anxiety at base-
under controlled conditions, (c) testing and refinement line than those who completed treatment.
with single cases in the field, and (d) completion of a A rigorous approach was taken in this study to test the
pilot randomized trial. BA treatment model. The BA approach was assessed with
The results of the current pilot study provide support clinically referred individuals who were treated by
for both the feasibility and clinical importance of BA practicing clinicians in an academic outpatient clinic that
strategies in the treatment of adolescents struggling with serves as a regional community mental health clinic.
depression and should encourage continuing research in Furthermore BA was evaluated in comparison to a strong
this area. The goals of the study were to recruit, success- treatment comparison condition in which skilled thera-
fully randomize, retain, and treat 50 participants, with pists provided either CBT- or IPT-based care. Related
the additional goal of demonstrating a significant to this, we anticipated that there would be a positive
reduction in depressive symptoms over the course of response to treatment in the control (EBP-D) condition,
treatment. Recruitment surpassed the initial target and this held true. However, more important for this
allowing us to follow 53 adolescents through treatment. study, the pilot accomplished the goal of demonstrating
Overall retention through the treatment component of that an adequate treatment response could be obtained
the study was 88%, which compares favorably to the using the BA approach. Youth in the A-BAP group
retention rates at the end of 12 weeks of treatment demonstrated statistically significant improvement from
reported in the TADS study (TADS [Treatment for pretreatment to end of treatment as reflected in the
 ADOLESCENT BEHAVIORAL ACTIVATION PROGRAM 11

primary outcome measure, the CDRS-R scores, and in engaged in the overall research project, perhaps
CGI and C-GAS ratings and SMFQ self-report scores. assuming that their input was less valuable or needed
All pretreatment to end-of-treatment outcomes fell because they were not randomized to the condition
within the 95% confidence interval, suggesting that the being ‘‘tested.’’ EBP-D participants were exposed to
estimates of change were reliable, but given the level of general evidence-based treatment approaches, therapists
variance on some outcome measures estimates may not in this condition were not asked to follow a standard
be as precise as might be obtained with a larger sample. treatment manual or given specific guidelines such that
Furthermore although not all participants met criteria the quality and content of intervention in this condition
for remission of depression at the end of treatment, diag- was not controlled. Although an in-depth review of the
nostic data were available for 27 of the 29 participants quality of EBP-D in this pilot is beyond the scope of this
who completed the BA treatment, and 78% of these no article, we do know that participants in our treatment
longer met diagnostic criteria for depression. groups were similar in the severity of depression and ses-
This pilot study also included an initial test of the BA sions completed, and that therapists in both conditions
model by exploring increased activation and reduced were well trained in the approaches offered. However,
avoidance as potential mediators of clinical response. the preliminary analyses of treatment fidelity reported
Although it was premature to actually test a mediation here represent an additional limitation and underscore
model given the small sample size and preliminary nat- the need for a comprehensive comparison of both treat-
ure of this study, we did find positive change on the ment conditions, including more precise specification of
Downloaded by [New York University] at 13:50 25 May 2015

Activation and significant improvement on the Avoid- the nature of EBP-D.

ance subscale of the BADS for both treatment groups. Finally, it is important to consider the need for the
It is unclear why A-BAP-exposed youth did not show development of an alternative approach for the treat-
greater changes than controls in this targeted area. As ment of depression in adolescents given the evidence
sustained, observable behavioral changes take time, it in support of both CBT and IPT, particularly in com-
is possible that between group changes will emerge over bination with an antidepressant medication. Although
time. Further research with larger sample sizes will be many youth find CBT, IPT, and medication, alone and
needed to determine whether increased activation and in combination, to be effective treatments, others do
decreased avoidance serve as mechanisms for change not achieve remission of depression and rates of
in depression across therapeutic approaches, or if cogni- relapse remain high (Curry et al., 2011; Emslie et al.,
tive or social support factors contribute differentially 2008). Furthermore, important subgroups of
to change for youth exposed to BA, CBT, or IPT depressed adolescents such as those who have been
interventions. exposed to maltreatment earlier in childhood are not
Although this study provides initial support for the responsive to these treatment approaches (Nanni
value of ongoing investigation of the use of a BA et al., 2012). Thus, innovative treatment development
approach to treat depression among adolescents, a remains critical. The preliminary findings outlined in
number of additional issues need to be taken into con- this article suggest that further development and test-
sideration. This was a small pilot study that was not ing of BA strategies could enhance the armamentar-
adequately powered to fully test the BA model or gener- ium of effective approaches in the field. BA expands
ate an effect size to evaluate change. At the initiation of upon the activation and pleasant events scheduling
this study, the literature supported the assumption of a that is already included as an initial or elective compo-
large effect size (0.8), which determined our sample size. nent of most CBT approaches adapted for adolescents
Recent studies have, however, found more modest effect (Brent & Poling, 1997; Clarke, DeBar, Ludman, Asar-
sizes, underscoring the need for significantly larger now, & Jaycox, 2002). Many of these programs also
samples (e.g., 180–200) to detect differences between build in the option of using only the activation=plea-
treatment conditions. sant events module in recognition that this may be suf-
Additional limitations include the fact that the sam- ficient to effect change for a subset of depressed youth
ple was majority White and female, which could limit (Clarke et al., 2002). Moreover, BA employs func-
the generalizability of the findings to more diverse tional analysis to address issues that get in the way
groups. Also the reliability of the Independent Clinical of change and focuses on overcoming avoidance, both
Evaluators was based on initial training only without of which are factors that may boost and maintain
ongoing adherence ratings as well as the fact that change treatment response. Furthermore BA offers a set of
in medication usage during the trial was not controlled. behaviorally focused strategies that may be more
Another limitation was the poor compliance with acceptable, at least as a starting point, for some ado-
follow-up observed in the EBP-D group. Although we lescents. However, additional research is needed to
were able to accommodate for this in the data analyses, allow us to tailor treatment approach to the needs of
it suggests that these participants=families were less the individual adolescent. Younger adolescents with-
 12 MCCAULEY ET AL.

out the maturity needed to effectively engage in cogni- REFERENCES

tive restructuring (Frankel et al., 2013; Herrington
et al., 2013), youth who have been exposed to early American Psychiatric Association. (1994). Diagnostic and statistical
abuse with residual deficits in executive functioning manual of mental disorders (4th ed.). Washington, DC: Author.
Angold, A., Costello, E. J., Messer, S. C., Pickles, A., Winder, F., &
and decreased sensitivity to reward anticipation
Silver, D. (1995). Development of a short questionnaire for use in
(DePrince, Weinzierl, & Combs, 2009; Gould et al., epidemiological studies of depression in children and adolescents.
2012; Pechtel & Pizzagalli, 2011; Rao et al., 2010), International Journal of Methods in Psychiatric Research, 5, 237–249.
and youth who are less open to ‘‘talking therapy’’ Angold, A., Erkanli, A., Silberg, J., Eaves, L., & Costello, E. J. (2002).
may be more responsive to a BA approach. BA strate- Depression scale scores in 8–17-year-olds: Effects of age and gender.
Journal of Child Psychology and Psychiatry, 43, 1052–1063.
gies, once they are more thoroughly evaluated, may be
doi:10.1111=1469-7610.00232
most useful as a first option in a stepped model of care Atkins, D. C. (2005). Using multilevel models to analyze couple and
or as components in a modular approach to treatment family treatment data: Basic and advanced issues. Journal of Family
of depression, allowing the therapist to draw on Psychology, 19, 98–110. doi:10.1037=0893-3200.19.1.98
evidence-based elements of BA, CBT, and IPT as Beck, A. T., Rush, J. A., Shaw, B. F., & Emery, G. (1979). Cognitive
therapy of depression. New York, NY: Guilford.
needed to provide individually tailored care. Further
Brent, D. A., & Poling, K. (1997). Cognitive therapy treatment manual
research is needed to determine whether there are sub- for depressed and suicidal youth. Pittsburgh, PA: University of
groups of youth that might respond better to one or Pittsburgh Health System Services for Teens at Risk.
the other type of treatment. Chu, B. C., Colognori, D., Weissman, A. S., & Bannon, K. (2009). An
Downloaded by [New York University] at 13:50 25 May 2015

In closing, this study, although limited in size and initial description and pilot of group behavioral activation therapy
for anxious and depressed youth. Cognitive and Behavioral Practice,
scope, provides a first step in establishing the efficacy
16, 408–419. doi:10.1016=j.cbpra.2009.04.003
of BA strategies as either a stand alone or a component Clarke, G., DeBar, L., Ludman, E., Asarnow, J., & Jaycox, L. (2002).
of treatment for adolescent depression. The findings STEADY Project intervention manual: Collaborative care,
summarized in this article, coupled with the significant cognitive-behavioral program for depressed youth in a primary care
effect of BA in the treatment of adult depression and setting. Retrieved from http://www.in.gov/idoc/files/STEADY_
project_Intervention1.pdf
the promising data from other preliminary studies with
Copeland, W. E., Shanahan, L., Costello, E. J., & Angold, A. (2009).
adolescents, support the need for ongoing work to Childhood and adolescent psychiatric disorders as predictors of
establish the efficacy of BA strategies as part of a larger young adult disorders. Archives of General Psychiatry, 66,
effort to improve our ability to effectively manage 764–772. doi:10.1001=archgenpsychiatry.2009.85
depression in young people. Costello, E. J., Egger, H., & Angold, A. (2005). 10-year research
update review: The epidemiology of child and adolescent psychiatric
disorders: I. Methods and public health burden. Journal of the Amer-
ican Academy of Child & Adolescent Psychiatry, 44, 972–986.
ACKNOWLEDGMENTS doi:10.1097=01.chi.0000172552.41596.6f
Curry, J., Silva, S., Rohde, P., Ginsburg, G., Kratochvil, C.,
Simons, A., . . . March, J. (2011). Recovery and recurrence following
We wish to thank Semret Nicodimos and Chuan Zhou treatment for adolescent major depression. Archives of General
for their consultation with data analysis, John Eaton Psychiatry, 68, 263–269. doi:10.1001=archgenpsychiatry.2010.150
for his editorial support, our research staff and the ado- Davey, C. G., Yücel, M., & Allen, N. B. (2008). The emergence of
lescents and parents who participated in the Adolescent depression in adolescence: Development of the prefrontal cortex
and the representation of reward. Neuroscience & Biobehavioral
Behavioral Activation Program. Reviews, 32, 1–19. doi:10.1016=j.neubiorev.2007.04.016
DePrince, A. P., Weinzierl, K. M., & Combs, M. D. (2009). Executive
function performance and trauma exposure in a community sample
of children. Child Abuse & Neglect, 33, 353–361. doi:10.1016=
FUNDING j.chiabu.2008.08.002
Dimidjian, S., Barrera, M., Jr., Martell, C., Muñoz, R. F., &
This research was funded by a grant from the National Lewinsohn, P. M. (2011). The origins and current status of
behavioral activation treatments for depression. Annual Review of
Institutes of Mental Health R34MH076956-03 Beha-
Clinical Psychology, 7, 1–38. doi:10.1146=annurev-clinpsy-032210-
vioral Activation Treatment with Depression Adoles- 104535
cents and was also supported by the University of Dimidjian, S., Hollon, S. D., Dobson, K. S., Schmaling, K. B.,
Washington/Seattle Children’s Hospital ITHS Pediatric Kohlenberg, R. J., Addis, M. E., . . . Jacobson, N. S. (2006).
Clinical Research Center, Grant Number Randomized trial of behavioral activation, cognitive therapy, and
antidepressant medication in the acute treatment of adults with
UL1RR025014 from the National Center for Research
major depression. Journal of Consulting and Clinical Psychology,
Resources (NCRR), a component of the National Insti- 74, 658–670. doi:10.1037=0022-006x.74.4.658
tutes of Health (NIH). Dr. Schloredt was also sup- Dimidjian, S., Hubley, S., Martell, C., & Herman, R. (2012). Quality
ported by a Young Investigator Award from the of behavioral activation scale. Unpublished manuscript, University
Seattle Children’s Research Institute. of Colorado Boulder, Boulder, CO.
 ADOLESCENT BEHAVIORAL ACTIVATION PROGRAM 13

Dunn, V., & Goodyer, I. M. (2006). Longitudinal investigation into Clinical Psychology & Psychotherapy, 20, 87–96. doi:10.1002=
childhood- and adolescence-onset depression: Psychiatric outcome cpp.774
in early adulthood. The British Journal of Psychiatry, 188, Kanter, J. W., Mulick, P. S., Busch, A. M., Berlin, K. S., & Martell, C.
216–222. doi:10.1192=bjp.188.3.216 R. (2007). The Behavioral Activation for Depression Scale (BADS):
Emslie, G. J., Kennard, B. D., Mayes, T. L., Nightingale-Teresi, J., Psychometric properties and factor structure. Journal of Psychopath-
Carmody, T., Hughes, C. W., . . . Rintelmann, J. W. (2008). Fluox- ology and Behavioral Assessment, 29, 191–202. doi:10.1007=s10862-
etine versus placebo in preventing relapse of major depression in 006-9038-5
children and adolescents. American Journal of Psychiatry, 165, Kanter, J. W., Rusch, L. C., Busch, A. M., & Sedivy, S. K. (2009).
459–469. doi:10.1176=appi.ajp.2007.07091453 Validation of the Behavioral Activation for Depression Scale
Esposito, C. L., & Clum, G. A. (2002). Psychiatric symptoms and (BADS) in a community sample with elevated depressive symptoms.
their relation to suicidal ideation in a high-risk adolescent community Journal of Psychopathology and Behavioral Assessment, 31, 36–42.
sample. Journal of the American Academy of Child & Adolescent Psy- doi:10.1007=s10862-008-9088-y
chiatry, 41, 44–51. doi:10.1097=00004583-200201000-00010 Kaufman, J., Birmaher, B., Brent, D., Rao, U., Flynn, C., Moreci, P.,
Evans, M. D., Piasecki, J. M., Kriss, M. R., & Hollon, S. D. (1984). . . . Ryan, N. (1997). Schedule for affective disorders and schizo-
Raters manual for the collaborative study psychotherapy rating scale, phrenia for school-age children-present and lifetime version
form 6. Minneapolis, MN: University of Minnesota and the St. (K-SADS-PL): Initial reliability and validity data. Journal of the
Paul-Ramsey Medical Center. American Academy of Child & Adolescent Psychiatry, 36, 980–988.
Ferster, C. B. (1973). A functional analysis of depression. American doi:10.1097=00004583-199707000-00021
Psychologist, 28, 857–870. doi:10.1037=h0035605 Keenan-Miller, D., Hammen, C. L., & Brennan, P. A. (2007). Health
Forbes, E. E. (2009). Where’s the fun in that? Broadening the focus on outcomes related to early adolescent depression. Journal of Ado-
reward function in depression. Biological Psychiatry, 66, 199–200. lescent Health, 41, 256–262. doi:10.1016=j.jadohealth.2007.03.015
Downloaded by [New York University] at 13:50 25 May 2015

doi:10.1016=j.biopsych.2009.05.001 Kennard, B. D., Clarke, G. N., Weersing, V. R., Asarnow, J. R.,

Forbes, E. E., Christopher May, J., Siegle, G. J., Ladouceur, C. D., Shamseddeen, W., Porta, G., . . . Brent, D. A. (2009). Effective com-
Ryan, N. D., Carter, C. S., . . . Dahl, R. E. (2006). Reward-related ponents of TORDIA cognitive–behavioral therapy for adolescent
decision-making in pediatric major depressive disorder: An fMRI depression: Preliminary findings. Journal of Consulting and Clinical
study. Journal of Child Psychology and Psychiatry, 47, 1031–1040. Psychology, 77, 1033–1041. doi:10.1037=a0017411
doi:10.1111=j.1469-7610.2006.01673.x Kennard, B., Silva, S., Vitiello, B., Curry, J., Kratochvil, C., Simons,
Frankel, S. A., Herrington, C. G., Evans, L. D., Korelitz, K. E., A., . . . March, J. (2006). Remission and residual symptoms after
Bauman, J. M., & Garber, J. (2013, April). Scientific reasoning short-term treatment in the Treatment of Adolescents with
development and learning cognitive behavioral therapy skills for Depression Study (TADS). Journal of the American Academy of
depression in youth. Poster presented at the meeting of the Society Child & Adolescent Psychiatry, 45, 1404–1411. doi:10.1097=01.chi.
for Research in Child Development, Seattle, WA. 0000242228.75516.21
Goodyer, I. M., Herbert, J., Tamplin, A., & Altham, P. M. E. (2000). Lewinsohn, P. M. (1974). A behavioral approach to depression. In
First-episode major depression in adolescents: Affective, cognitive R. M. Friedman & M. M. Katz (Eds.), The psychology of depression:
and endocrine characteristics of risk status and predictors of onset. Contemporary theory and research (pp. 157–185). New York, NY:
The British Journal of Psychiatry, 176, 142–149. doi:10.1192= Wiley.
bjp.176.2.142 Lewinsohn, P. M. (2001). Lewinsohn’s model of depression. In W. E.
Gould, F., Clarke, J., Heim, C., Harvey, P. D., Majer, M., & Craighead & C. B. Nemeroff (Eds.), The Corsini encyclopedia of psy-
Nemeroff, C. B. (2012). The effects of child abuse and neglect on chology and behavioral science (3rd ed., pp. 442–444). New York,
cognitive functioning in adulthood. Journal of Psychiatric Research, NY: Wiley.
46, 500–506. doi:10.1016=j.jpsychires.2012.01.005 Lewinsohn, P. M., Rohde, P., Seeley, J. R., Klein, D. N., & Gotlib, I. H.
Guy, W. (1976). Early clinical drug evaluation unit (ECDEU) assess- (2003). Psychosocial functioning of young adults who have
ment manual for psychopharmacology (Rev., NIH Publication No. experienced and recovered from major depressive disorder during
76–338). Rockville, MD: US Department of Health, Education, adolescence. Journal of Abnormal Psychology, 112, 353–363.
and Welfare. doi:10.1037=0021-843x.112.3.353
Hankin, B. L., Abramson, L. Y., Moffitt, T. E., Silva, P. A., McGee, R., Lewis, C. C., Simons, A. D., Nguyen, L. J., Murakami, J. L., Reid, M. W.,
& Angell, K. E. (1998). Development of depression from preadoles- Silva, S. G., & March, J. S. (2010). Impact of childhood trauma on
cence to young adulthood: Emerging gender differences in a 10-year treatment outcome in the Treatment for Adolescents with Depression
longitudinal study. Journal of Abnormal Psychology, 107, 128–140. Study (TADS). Journal of the American Academy of Child & Adolescent
doi:10.1037==0021-843x.107.1.128 Psychiatry, 49, 132–140. doi:10.1016=j.jaac.2009.10.007
Hare, T. A., Tottenham, N., Galvan, A., Voss, H. U., Glover, G. H., & Lopez, A. D., Mathers, C. D., Ezzati, M., Jamison, D. T., & Murray,
Casey, B. J. (2008). Biological substrates of emotional reactivity and C. J. L. (2006). Global and regional burden of disease and risk
regulation in adolescence during an emotional go-nogo task. factors, 2001: Systematic analysis of population health data. The
Biological Psychiatry, 63, 927–934. doi:10.1016=j.biopsych.2008.03.015 Lancet, 367, 1747–1757. doi:10.1016=s0140-6736(06)68770-9
Herrington, C. G., Frankel, S. A., Abernathy, E. T., Bauman, J. M., March, J. S., Parker, J. D. A., Sullivan, K., Stallings, P., & Conners, C. K.
Hunt, H., Korelitz, K. E., & Garber, J. (2013, April). (1997). The Multidimensional Anxiety Scale for Children (MASC):
Metacognitive development and learning cognitive behavioral ther- Factor structure, reliability, and validity. Journal of the American
apy skills for depression in children and adolescents. Poster pre- Academy of Child & Adolescent Psychiatry, 36, 554–565. doi:10.1097=
sented at the meeting of the Society for Research in Child 00004583-199704000-00019
Development, Seattle, WA. Marmorstein, N. R. (2009). Longitudinal associations between alcohol
Hollingshead, A. B. (1975). Four factor index of social status (Unpub- problems and depressive symptoms: Early adolescence through early
lished manuscript). New Haven, CT: Yale University. adulthood. Alcoholism: Clinical and Experimental Research, 33,
Jacob, M., Keeley, M. L., Ritschel, L., & Craighead, W. E. (2013). 49–59. doi:10.1111=j.1530-0277.2008.00810.x
Behavioural activation for the treatment of low-income, African McCarty, C. A., & Weisz, J. R. (2007). Effects of psychotherapy for
American adolescents with major depressive disorder: A case series. depression in children and adolescents: What we can (and can’t)
 14 MCCAULEY ET AL.

learn from meta-analysis and component profiling. Journal of the Ritschel, L. A., Ramirez, C. L., Jones, M., & Craighead, W. E. (2011).
American Academy of Child & Adolescent Psychiatry, 46, 879– Behavioral activation for depressed teens: A pilot study. Cognitive
886. doi:10.1097=chi.0b013e31805467b3 and Behavioral Practice, 18, 281–299. doi:10.1016=j.cbpra.2010.
McCarty, C. A., Wymbs, B. T., King, K. M., Mason, W. A., Stoep, A. V., 07.002
McCauley, E., & Baer, J. (2012). Developmental consistency in Rosselló, J., & Bernal, G. (1999). The efficacy of cognitive-behavioral
associations between depressive symptoms and alcohol use in early and interpersonal treatments for depression in Puerto Rican adoles-
adolescence. Journal of Studies on Alcohol and Drugs, 73, 444–453. cents. Journal of Consulting and Clinical Psychology, 67, 734–745.
McKenzie, D. P., Toumbourou, J. W., Forbes, A. B., Mackinnon, A. J., doi:10.1037==0022-006x.67.5.734
McMorris, B. J., Catalano, R. F., & Patton, G. C. (2011). Predicting Royston, P. (2005). Multiple imputation of missing values: Update of
future depression in adolescents using the short mood and feelings ice. Stata Journal, 5, 527.
questionnaire: A two-nation study. Journal of Affective Disorders, Rubin, D. B. (1987). Multiple imputation for nonresponse in surveys.
134, 151–159. doi:10.1016=j.jad.2011.05.022 New York, NY: Wiley & Sons.
McMakin, D. L., Olino, T. M., Porta, G., Dietz, L. J., Emslie, G., Shaffer, D., Gould, M. S., Brasic, J., Ambrosini, P., Fisher, P., Bird, H.,
Clarke, G., . . . Brent, D. A. (2012). Anhedonia predicts poorer & Aluwahlia, S. (1983). A Children’s Global Assessment Scale
recovery among youth with selective serotonin reuptake inhibitor (CGAS). Archives of General Psychiatry, 40, 1228–1231. doi:10.1001=
treatment–resistant depression. Journal of the American Academy archpsyc.1983.01790100074010
of Child & Adolescent Psychiatry, 51, 404–411. doi:10.1016= Shrout, P. E., & Fleiss, J. L. (1979). Intraclass correlations: Uses in
j.jaac.2012.01.011 assessing rater reliability. Psychological Bulletin, 86, 420–428.
Mufson, L., Dorta, K. P., Moreau, D., & Weissman, M. M. (2011). doi:10.1037==0033-2909.86.2.420
Interpersonal psychotherapy for depressed adolescents. New York, Silk, J. S., Dahl, R. E., Ryan, N. D., Forbes, E. E., Axelson, D. A.,
NY: Guilford Press. Birmaher, B., & Siegle, G. J. (2007). Pupillary reactivity to emotional
Downloaded by [New York University] at 13:50 25 May 2015

Mufson, L., Dorta, K. P., Wickramaratne, P., Nomura, Y., Olfson, M., information in child and adolescent depression: Links to clinical
& Weissman, M. M. (2004). A randomized effectiveness trial of and ecological measures. American Journal of Psychiatry, 164,
interpersonal psychotherapy for depressed adolescents. Archives of 1873–1880. doi:10.1176=appi.ajp.2007.06111816
General Psychiatry, 61, 577–584. doi:10.1001=archpsyc.61.6.577 Somerville, L. H., Hare, T., & Casey, B. J. (2011). Frontostriatal
Mufson, L., Weissman, M. M., Moreau, D., & Garfinkel, R. (1999). maturation predicts cognitive control failure to appetitive cues in
Efficacy of interpersonal psychotherapy for depressed adolescents. adolescents. Journal of Cognitive Neuroscience, 23, 2123–2134.
Archives of General Psychiatry, 56, 573–579. doi:10.1001= doi:10.1162=jocn.2010.21572
archpsyc.56.6.573 TADS (Treatment for Adolescents with Depression Study) Team.
Nanni, V., Uher, R., & Danese, A. (2012). Childhood maltreatment (2003). Treatment for Adolescents with Depression Study (TADS):
predicts unfavorable course of illness and treatment outcome in Rationale, design, and methods. Journal of the American Academy
depression: A meta-analysis. American Journal of Psychiatry, 169, of Child & Adolescent Psychiatry, 42, 531–542. doi:10.1097=
141–151. doi:10.1176=appi.ajp.2011.11020335 01.chi.0000046839.90931.0d
Patton, G. C., Olsson, C., Bond, L., Toumbourou, J. W., Carlin, J. B., TADS (Treatment for Adolescents with Depression Study) Team.
Hemphill, S. A., & Catalano, R. F. (2008). Predicting female (2004). Fluoxetine, cognitive-behavioral therapy, and their combi-
depression across puberty: A two-nation longitudinal study. Journal nation for adolescents with depression: Treatment for Adolescents
of the American Academy of Child & Adolescent Psychiatry, 47, with Depression Study (TADS) randomized controlled trial. Journal
1424–1432. doi:10.1097=chi.0b013e3181886ebe of the American Medical Association, 292, 807–820. doi:10.1001=
Pechtel, P., & Pizzagalli, D. A. (2011). Effects of early life stress on jama.292.7.807
cognitive and affective function: An integrated review of human TADS (Treatment for Adolescents with Depression Study) Team.
literature. Psychopharmacology, 214, 55–70. doi:10.1007=s00213- (2007). The Treatment for Adolescents with Depression
010-2009-2 Study (TADS): Long-term effectiveness and safety outcomes.
Poznanski, E., & Mokros, H. (2001). Children’s depression rating Archives of General Psychiatry, 64, 1132–1144. doi:10.1001=
scale-revised (CDRS-R). Los Angeles, CA: Western Psychological archpsyc.64.10.1132
Services. Tottenham, N., Hare, T. A., & Casey, B. J. (2011). Behavioral assess-
Raghunathan, T. E. (2004). What do we do with missing data? Some ment of emotion discrimination, emotion regulation, and cognitive
options for analysis of incomplete data. Annual Review of Public control in childhood, adolescence, and adulthood. Frontiers in
Health, 25, 99–117. doi:10.1146=annurev.publhealth.25.102802. Psychology, 2, 39. doi:10.3389=fpsyg.2011.00039
124410 Weisz, J. R., Jensen, A. L., & McLeod, B. D. (2005). Development and
Rao, U., Chen, L.-A., Bidesi, A. S., Shad, M. U., Thomas, M. A., & dissemination of child and adolescent psychotherapies: Milestones,
Hammen, C. L. (2010). Hippocampal changes associated with methods, and a new deployment-focused model. In E. D. Hibbs &
early-life adversity and vulnerability to depression. Biological Psy- P. S. Jensen (Eds.), Psychosocial treatments for child and adolescent
chiatry, 67, 357–364. doi:10.1016=j.biopsych.2009.10.017 disorders: Empirically-based approaches (2nd ed., pp. 9–39).
Rintelmann, J. W., Emslie, G. J., Rush, A. J., Varghese, T., Gullion, C. Washington, DC: American Psychological Association.
M., Kowatch, R. A., & Hughes, C. W. (1996). The effects of Weisz, J. R., McCarty, C. A., & Valeri, S. M. (2006). Effects of
extended evaluation on depressive symptoms in children and adoles- psychotherapy for depression in children and adolescents: A
cents. Journal of Affective Disorders, 41, 149–156. doi:10.1016= meta-analysis. Psychological Bulletin, 132, 132–149. doi:10.1037=
s0165-0327(96)00084-5 0033-2909.132.1.132