Research

JAMA Psychiatry | Original Investigation

Effects of a Contextual Adaptation of the Unified Protocol

in Multiple Emotional Disorders in Individuals Exposed
to Armed Conflict in Colombia
A Randomized Clinical Trial
Leonidas Castro-Camacho, PhD; David H. Barlow, PhD; Nicolás García, MA; Todd J. Farchione, PhD;
Fabio Idrobo, PhD; Michel Rattner, MA; Diana M. Quant, MA; Laura González, MA; Julián D. Moreno, MA

Supplemental content
IMPORTANCE A transdiagnostic treatment, the Unified Protocol, is as effective as single
diagnostic protocols in comorbid emotional disorders in clinical populations. However, its
effects on posttraumatic stress disorder and other emotional disorders in individuals living in
war and armed conflict contexts have not been studied.

OBJECTIVE To evaluate the efficacy of a cultural and contextual adaptation of the Unified
Protocol (CXA-UP) on posttraumatic stress disorder, anxiety, and depression compared to
waitlist control in individuals exposed to armed conflict in Colombia.

DESIGN, SETTING, AND PARTICIPANTS From April 2017 to March 2020, 200 participants 18
years and older were randomly assigned to the CXA-UP or to a waitlist condition. CXA-UP
consisted of 12 to 14 twice-a-week or weekly individual 90-minute face-to-face sessions.
Outcomes were assessed at baseline, posttreatment, and 3 months following treatment.
Analyses were performed and compared for all randomly allocated participants
(intent-to-treat [ITT]) and for participants who completed all sessions and posttreatment
measures (per protocol [PP]). The study took place at an outpatient university center and
included individuals who were registered in the Colombian Victims Unit meeting DSM-5
diagnostic criteria for posttraumatic stress disorder, anxiety, or depression or were severely
impaired by anxiety or depression. Individuals who were receiving psychological therapy,
were dependent on alcohol or drugs, were actively suicidal or had attempted suicide in the
previous 2 months, had psychosis or bipolar disorder, or were cognitively impaired were
excluded.

INTERVENTION CXA-UP or waitlist.

MAIN OUTCOMES AND MEASURES Primary outcomes were changes in anxiety, depression, and
somatic scores on the Patient Health Questionnaire and Posttraumatic Stress Disorder
Checklist for DSM-5.

RESULTS Among the 200 participants (160 women [80.0%]; 40 men [20.0%]; mean [SD]
age, 43.1 [11.9] years), 120 were randomized to treatment and 80 to waitlist. Results for
primary outcomes in the ITT analysis showed a significant pretreatment-to-posttreatment
reduction when comparing treatment and waitlist on the posttraumatic stress disorder
checklist for DSM-5 scores (slope [SE], −31.12 [3.00]; P < .001; Cohen d, 0.90; 90% CI, Author Affiliations: Department of
0.63-1.19), 9-item Patient Health Questionnaire (PHQ-9) (slope [SE],−11.94 [1.30]; P < .001; Psychology, Universidad de los
Andes, Bogotá, Colombia
Cohen d, 0.77; 90% CI, 0.52-1.06), PHQ-anxiety (slope [SE], −6.52 [0.67]; P < .001; Cohen d,
(Castro-Camacho, García, Rattner,
0.82; 90% CI, 0.49-1.15), and PHQ-somatic (slope [SE], −8.31 [0.92]; P < .001; Cohen d, 0.75; Quant, González, Moreno); Center for
90% CI, 0.47-1.04). Anxiety and Related Disorders,
Department of Psychological and
CONCLUSIONS AND RELEVANCE In this study, significant reductions and large effect sizes in all Brain Sciences, Boston University,
measures of different emotional disorders showed efficacy of a single transdiagnostic Boston, Massachusetts (Barlow,
Farchione, Idrobo, Moreno);
intervention in individuals exposed to armed conflicts.
Department of Psychology, Palo Alto
TRIAL REGISTRATION ClinicalTrials.gov Identifier: NCT03127982 University, Palo Alto, California
(Rattner).
Corresponding Author: Leonidas
Castro-Camacho, PhD, Department of
Psychology, Universidad de los
Andes, Cra. 1 #18A-24, Of. G-217,
JAMA Psychiatry. 2023;80(10):991-999. doi:10.1001/jamapsychiatry.2023.2392 Bogotá, DC 111711, Colombia (lecastro
Published online July 19, 2023. @uniandes.edu.co).

(Reprinted) 991
© 2023 American Medical Association. All rights reserved.
Downloaded from jamanetwork.com by Andrea Romero Martínez on 10/27/2024
 Research Original Investigation Contextual Adaptation of the Unified Protocol in Emotional Disorders in Individuals Exposed to Armed Conflict in Colombia

M
ental health problems in vulnerable populations in
low- and-middle-income countries represent a sig- Key Points
nificant challenge for demonstrating efficacy of psy-
Question Is a contextual and cultural adaptation of the Unified
chological treatments. In war zones, refugees and internally Protocol (CXA-UP) more effective than waitlist control in treating
displaced persons present with multiple emotional and be- posttraumatic stress disorder (PTSD) and comorbid emotional
havioral problems and disorders, with posttraumatic stress disorders in individuals exposed to armed conflict?
disorder (PTSD), anxiety, and mood disorders being the most
Findings In this randomized clinical trial of 200 internally
common. These psychological difficulties can contribute to displaced persons, the CXA-UP showed significant decreases and
unemployment, economic hardships, and health problems, in- large effect sizes on all measures of PTSD, anxiety, depression, and
creasing the effects of previous and current traumatic events.1 somatic complaints with effects maintained at 3-month follow-up.
Furthermore, internally displaced persons and refugees in con-
Meaning In this study, when adapted contextually and culturally,
flict zones are at a higher risk of developing mental disorders the CXA-UP significantly improved severe emotional disorders and
because of repeated exposure to traumatic events.2 Research PTSD in internally displaced persons living in violent contexts.
shows that when these disorders are not adequately treated,
they tend to become chronic and lead to other problems.3,4
More than 60 years of armed conflict in Colombia has in the Colombian Victims Unit5 and living in Bogotá by way
left nearly 9 million individuals registered in the Colombian of public announcements, referrals from independent orga-
Victims Unit. 5 Many have been exposed directly or indi- nizations, and word of mouth. Eligible participants were
rectly to extreme violent events or are under continuous adults 18 years and older who met diagnostic criteria for
threats. More than 8 million individuals have been dis- PTSD, generalized anxiety disorder, panic disorder, or major
placed from their lands.6 They live in hostile urban environ- depressive disorder according to the Spanish 5.0.0 version of
ments, where many cannot meet basic needs, obtain the Mini International Neuropsychiatric Interview (MINI)21
employment,7 or secure food and shelter.8 These conditions or who met criteria for significant severity or impairment
have a pronounced negative impact on mental health and from anxiety or depression severity and impairment (scoring
quality of life,9 resulting in a higher incidence of anxiety higher than 7 on the Overall Anxiety Severity and Impair-
disorders, depression, and substance misuse.10-12 PTSD is ment Scale22 or Overall Depression Severity and Impairment
5.1 times more prevalent in internally displaced persons in Scale23). In an initial telephone contact, potential partici-
Colombia than in the general population.13,14 pants were informed about the study, queried for their inter-
Cognitive behavior therapy targeting single emotional dis- est in participating, and screened for alcohol dependency
orders is effective in clinical populations15; however, its effi- or current involvement in psychological therapy. Those
cacy with individuals presenting with multiple diagnoses is not meeting initial exclusion criteria were invited to an
hindered by the costs of training therapists and implementing in-person screening session where all inclusion criteria were
treatments for each disorder. Transdiagnostic interventions, assessed; baseline outcome measures were obtained;
which simultaneously target multiple disorders, may help bridge detailed information about the study, including potential
the science-to-service gap. The Unified Protocol for the trans- risks and unintended outcomes, was provided; and written
diagnostic treatment of emotional disorders16 is a single informed consent was obtained as approved by the Univer-
cognitive-behavior intervention targeting temperamental char- sity of Los Andes Institutional Review Board (code 656-657-
acteristics, particularly neuroticism and resulting emotion dys- 2016). Additional exclusion criteria were active suicidal
regulation, common to multiple emotional disorders. The UP ideation or suicide attempts in the previous 2 months, psy-
has considerable support for treating emotional disorders, in- chosis, bipolar disorder, and cognitive impairment. Partici-
cluding anxiety and depression, and has shown to be as effec- pants who met the inclusion criteria and were willing to par-
tive as single-disorder interventions in clinical populations,17 ticipate were offered free therapy with transportation costs
with large effect size reductions across several common emo- covered. The study followed the Consolidated Standards of
tional disorders.18 Although some case studies and open trials Reporting Trials (CONSORT) reporting guideline.
have shown promising results for UP as a treatment for PTSD,19,20
randomized clinical trials have not been conducted. More- Procedures
over, as this population’s cultural and contextual characteris- Study design and patient flow are summarized in the Figure.
tics differ from the North American sample of the original pro- From April 2017 to March 2020, we recruited 200 partici-
tocol, a cultural adaptation of the UP is warranted. The present pants and randomly allocated 120 to the treatment condition
study evaluates the efficacy of a culturally adapted version of and 80 to the waitlist. The study consisted of 2 arms. The
the UP (CXA-UP) for emotional disorders, disability, and qual- CXA-UP consisted of 12 to 14 individual face-to-face sessions
ity of life in a group of individuals exposed to armed conflict in with outcome measures taken at baseline, end of treatment,
comparison to a waitlist condition. and 3-month follow-up. Participants assigned to the waitlist
condition did not receive any intervention during the 6 weeks
Methods following baseline assessment but were informed that the UP
Participants intervention would commence in 6 weeks. Outcome mea-
A sample of 200 participants was recruited from among sures were taken at the end of the 6-week wait period by
individuals exposed to armed conflict who were registered assessors who were blinded to their group assignment.

992 JAMA Psychiatry October 2023 Volume 80, Number 10 (Reprinted) jamapsychiatry.com

© 2023 American Medical Association. All rights reserved.

Downloaded from jamanetwork.com by Andrea Romero Martínez on 10/27/2024
 Contextual Adaptation of the Unified Protocol in Emotional Disorders in Individuals Exposed to Armed Conflict in Colombia Original Investigation Research

Randomization and Blinding

Figure. Trial Profile
After screening and baseline assessment, a data manager—
not involved as therapist or assessor—randomly assigned each
627 Individuals screened by telephone
new participant to each condition using a simple random for-
mula, which recalculated the assignment independently for 235 Excluded
each participant. As this resulted in an imbalance in partici- 168 Not interested in
the study
pants assigned to CXA-UP, after 20 randomized participants, 67 Met exclusion criteria
subsequent allocation was based on a random number table.
392 Assessed for eligibility

Intervention
192 Excluded
The treatment protocol was culturally adapted24 and translated
132 Declined to participate
from the original UP manuals25,26 into written therapist manu- 60 Did not meet inclusion
criteria
als and participant workbooks in Spanish. Examples of partici-
pants’ own emotional experiences were used to illustrate con-
200 Randomized
cepts within their cultural context and assignments adapted to
their daily activities; text was replaced with graphic material in
the workbook while core elements and order of modules of the
120 Randomized to CXA-UP treatment 80 Randomized to WLC
original protocol were preserved as described in a previous 48 Completed treatmenta 73 Completed WLP
publication.24 Treatment was delivered individually through 12 48 Withdrawn during treatment 7 Withdrawn during WLP
39 External reasonsb
to 14 face-to-face sessions once to twice per week, with each ses- 9 Treatment-related
sion lasting for about 90 minutes over 6 to 12 weeks. For cultural reasonsc
24 Never entered treatmentd
reasons, an initial session was also added to allow participants
to talk freely about their unique experiences, foster trust, and pro-
mote a collaborative relationship with the therapist. A case study 48 Completed posttreatment and 61 Completed post-WLP assessment
3-mo follow-up assessment 12 Incomplete assessments
providing a detailed description of the treatment protocol has
been published previously.27
120 Included in ITT analysis 80 Included in ITT analysis
48 Included in PP analysis 61 Included in PP analysis
Therapists and Treatment Integrity
Therapists were 12 graduate clinical psychology students with
CXA-UP indicates contextual/cultural adaptation of the Unified Protocol;
at least 1 year of supervised experience in cognitive behavior ITT, intent-to-treat; PP, per protocol; WLC, waitlist control; WLP, waitlist period.
therapy for emotional disorders. All therapists undertook 2 in- a
Completed treatment indicates that patients attended 100% of sessions.
tensive training workshops on delivering the UP: the first by 1 b
External reasons included transportation difficulties, moving to another city,
of the developers of the original protocol and the second on or job commitments.
c
the cultural adaptation by the principal investigator. In addi- Treatment-related reasons: 5 mentioned that sessions were too long, 4 felt
tion, video and audio recordings of treatment sessions with a uncomfortable during sessions.
d
study participant were discussed, and role-play was used to Dropped out after assessment and before initial treatment session.

develop specific skills. Weekly supervision sessions, led by the

principal investigator and 2 senior research associates, were assessors. Secondary outcomes included anxiety severity and
also provided to all therapists. level of interference, measured by the Overall Anxiety Sever-
Therapists were assessed for competence and mastery of ity and Impairment Scale,22 depression severity, and level of
the therapist’s manual at the end of each training workshop. interference as measured by the Overall Depression Severity
All treatment sessions were audio recorded. Ten percent of ses- and Impairment Scale.23 These measures were administered
sions were randomly selected and rated for fidelity to the origi- at baseline, at the beginning of each weekly session, at end of
nal treatment manual and competence by an original UP re- treatment, and at 3-month follow-up. In addition, changes in
search team member using standardized fidelity ratings the Quality-of-Life Enjoyment and Satisfaction Questionnaire30
approved by the protocol developers and used in prior clini- and an adaptation of the World Health Organization Disabil-
cal trials of the UP.17 Specifically, therapists were rated on their ity Assessment Schedule 2.031 were evaluated at baseline,
ability to cover relevant session content, complete in-session posttreatment, and 3-month follow-up.
exercises, and administer core treatment elements. Average
treatment fidelity to the original manual was 86% across all Sample Size Calculation
rated sessions. Power calculations conducted using G-Power32 version 3.1.9.7
with a moderate effect size (f2, 0.15), 2 tails, an α of .05, and
Assessments and Instruments statistical power of 0.95 yielded an estimated total of 90 par-
The primary outcomes were changes in anxiety, depression, ticipants. Since we expected attrition of up to 50% due to chal-
and somatic symptoms as assessed by the Patient Health Ques- lenges often experienced by internally displaced persons (re-
tionnaire (PHQ),28 and changes in symptoms of PTSD, as mea- liance on public transport, long journeys, threats to safety, and
sured by the PTSD checklist for DSM-5 (PLC-5).29 Due to edu- job instability), we aimed to recruit at least 150 participants to
cational barriers, these instruments were read by independent gather pretreatment and posttreatment measurements for 45

jamapsychiatry.com (Reprinted) JAMA Psychiatry October 2023 Volume 80, Number 10 993

© 2023 American Medical Association. All rights reserved.

Downloaded from jamanetwork.com by Andrea Romero Martínez on 10/27/2024
 Research Original Investigation Contextual Adaptation of the Unified Protocol in Emotional Disorders in Individuals Exposed to Armed Conflict in Colombia

Table 1. Demographic and Baseline Characteristics

No. (%)
Total Treatment Control
Variable (N = 200) (n = 120) (n = 80) P valuea
Demographic characteristics
Age, mean (SD), y 43.1 (11.9) 42.6 (11.7) 43.9 (12.3) .22
Female 160 (80.0) 100 (83.3) 60 (75.0)
.50
Male 40 (20.0) 20 (17.7) 20 (25.0)
Marital status
Single 85 (42.5) 56 (46.7) 29 (36.3)
Married 81 (40.5) 42 (35.0) 39 (48.8) .30
Divorced 34 (17.0) 22 (18.3) 12 (15.0)
Have children 164 (82.0) 100 (83.3) 64 (80.0) .36
Education level
No formal education/elementary educationb 49 (24.5) 31 (25.8) 18 (22.5)
High school 85 (42.5) 51 (42.5) 34 (42.5) .97
College/technical school 66 (33.0) 38 (31.7) 28 (35.0) Abbreviations: GAD, generalized
Unemployed 133 (66.5) 79 (65.8) 54 (67.5) .30 anxiety disorder; MDD, major
Diagnoses (MINI) depression disorder; MINI, Mini
International Neuropsychiatric
GAD 25 (12.5) 15 (12.5) 10 (12.5) .97 Interview; OASIS, Overall Anxiety
PTSD 148 (74.0) 89 (74.2) 59 (73.8) >.99 Severity and Impairment Scale;
MDD 129 (64.5) 79 (65.8) 50 (62.5) .86 ODSIS, Overall Depression Severity
and Impairment Scale;
PD 80 (40.0) 50 (41.7) 30 (37.5) .67 PCL-5, posttraumatic stress disorder
≥2 Diagnoses 136 (68.0) 83 (69.2) 53 (66.3) .48 checklist for DSM-5; PD, panic
Outcome measures disorder; PHQ, Patient Health
Questionnaire; PTSD, posttraumatic
PCL-5, mean (SD) 51.1 (15.0) 51.4 (15.5) 50.7 (14.3) .72 stress disorder;
PHQ-somatic score, mean (SD) 22.4 (4.7) 22.7 (4.8) 21.9 (4.5) .23 QLESQ, Quality-of-Life Enjoyment
PHQ-depression score, mean (SD) 24.6 (6.3) 24.9 (6.6) 24.3 (5.7) .49 and Satisfaction Questionnaire;
WHODAS, World Health Organization
PHQ-anxiety score, mean (SD) 16.9 (3.2) 16.8 (3.3) 17.1 (3.0) .51 Disability Assessment Scale.
ODSIS score, mean (SD) 13.2 (4.7) 13.3 (4.6) 13.1 (4.9) .80 a
t Test and χ2 test.
OASIS score, mean (SD) 12.5 (5.0) 12.7 (5.0) 12.3 (5.0) .62 b
No formal education and
QLESQ score, mean (SD) 43.3 (8.4) 43.0 (8.7) 43.6 (8.0) .63 elementary education categories
were combined to prevent
WHODAS score, mean (SD) 1.3 (0.7) 1.3 (0.8) 1.2 (0.7) .19
identifiability of participants.

participants per group. As the attrition was higher than models with time as a fixed effect and individual as a random
expected in the treatment group, we ultimately recruited 200 effect were estimated using data from all participants com-
to attain the desired group sizes. pleting treatment, including those receiving treatment after
the waiting period (waitlist) with complete pretreatment,
Statistical Analysis posttreatment, and 3-month follow-up data to assess out-
Differences between treatment and waitlist groups on pri- comes at follow-up. Effect sizes for the multilevel regression
mary and secondary outcomes and demographic characteris- models were evaluated using the partial η2index and then
tics at baseline were evaluated using t tests and χ2 compari- transformed to Cohen d to facilitate comparison following IBM
sons. Due to unequal group sizes, multilevel regression models guidelines35 based on Cohen.36
were used to estimate the treatment effect, including treat-
ment, time, and their interaction as predictors, with random
intercepts for participants to represent intraindividual vari-
ance across time. Models were estimated using the lmer func-
Results
tion from the lme4 package in R version 4.2.2 (R Foundation),33 Sample Characteristics
P values and degrees of freedom were calculated using the There were 200 participants included (160 women [80.0%];
Satterthwaite method34 using the jtool package in R version 40 men [20.0%]) with a mean (SD) age of 43.1 (11.9) years).
2.2.0 (R Foundation). For treatment effects analysis, results of Table 1 depicts baseline demographic characteristics, diagno-
both intent-to-treat and per-protocol approaches were used and ses, outcome measures, and initial comparisons between treat-
compared to address possible estimation biases generated by ment and waitlist groups. We assessed the between-group
missing data. Intent-to-treat used all randomized partici- equivalence of baseline outcomes using t tests. We did not find
pants, and per-protocol used participants who completed all differences in primary or secondary outcomes, suggesting that
sessions and posttreatment measures. Finally, multilevel both groups were comparable at baseline. We compared

994 JAMA Psychiatry October 2023 Volume 80, Number 10 (Reprinted) jamapsychiatry.com

© 2023 American Medical Association. All rights reserved.

Downloaded from jamanetwork.com by Andrea Romero Martínez on 10/27/2024
 Contextual Adaptation of the Unified Protocol in Emotional Disorders in Individuals Exposed to Armed Conflict in Colombia Original Investigation Research

Table 2. Observed Means for Primary and Secondary Outcomes

Mean (SD)
CXA-UP WL
ITT (n = 120) PP (n = 48) ITT (n = 80) PP (n = 61)
Primary outcomes
PCL-5 score
Pretreatment 51.4 (15.5) 50.4 (17.1) 50.7 (14.3) 51.5 (13.6)
Posttreatment 15.3 (13.8) 15.3 (13.8) 46.4 (14.5) 46.4 (14.5)
3-mo Follow-up 10.8 (14.6) 10.8 (14.6) NA NA
PHQ-depression score
Pretreatment 24.9 (6.6) 24.3 (6.4) 24.3 (5.7) 24.6 (6.8)
Posttreatment 13.1 (6.7) 13.1 (6.7) 24.7 (6.5) 24.7 (6.5)
3-mo Follow-up 12.9 (5.3) 12.9 (5.3) NA NA
PHQ-anxiety score
Pretreatment 16.8 (3.3) 16.7 (3.5) 17.1 (3.0) 17.4 (2.9)
Posttreatment 9.87 (3.6) 9.87 (3.6) 16.9 (3.2) 16.9 (3.2)
3-mo Follow-up 10.0 (3.9) 10.0 (3.9) NA NA
PHQ-somatic score
Pretreatment 22.7 (4.8) 22 (5.1) 21.9 (4.5) 21.8 (4.5)
Posttreatment 14.3 (5.5) 14.3 (5.5) 22.1 (4.8) 22.1 (4.8)
3-mo Follow-up 15.2 (4.6) 15.2 (4.6) NA NA
Secondary outcomes
ODSIS score
Pretreatment 13.3 (4.6) 12.5 (4.4) 13.1 (4.9) 12.9 (5.0)
Posttreatment 1.8 (3.1) 1.8 (3.1) 11.8 (5.4) 11.8 (5.4)
3-mo Follow-up 2.6 (4.3) 2.6 (4.3) NA NA
OASIS score Abbreviations:
Pretreatment 12.7 (5.0) 12.7 (4.1) 12.3 (5.0) 12.1 (4.9) CXA-UP, contextual/cultural
adaptation of the Unified Protocol;
Posttreatment 1.7 (2.7) 1.7 (2.7) 10.7 (4.1) 10.7 (4.1)
ITT, intent-to-treat; NA, not
3-mo Follow-up 3.0 (4.3) 3.0 (4.3) NA NA applicable; OASIS, Overall Anxiety
Q-LES-Q score Severity and Impairment Scale;
ODSIS, Overall Depression Severity
Pretreatment 43.0 (8.1) 43.6 (9.3) 43.6 (8.0) 43.5 (8.2)
and Impairment Scale;
Posttreatment 54.8 (7.2) 54.8 (7.2) 44.3 (8.2) 44.3 (8.2) PCL-5, posttraumatic stress disorder
3-mo Follow-up 58.1 (8.5) 58.1 (8.5) NA NA checklist for DSM-5; PHQ, Patient
Health Questionnaire; PP, per
WHODAS score
protocol; Q-LES-Q, Quality-of-Life
Pretreatment 1.3 (0.8) 1.3 (0.7) 1.2 (0.7) 1.2 (0.7) Enjoyment and Satisfaction
Posttreatment 0.4 (0.4) 0.4 (0.4) 1.3 (0.7) 1.3 (0.7) Questionnaire; WHODAS, World
Health Organization Disability
3-mo Follow-up 1.0 (0.9) 1.0 (0.9) NA NA
Assessment Scale; WL, waitlist.

demographic variables between groups using χ2 and t tests and before initiating treatment, and 48 (24%) discontinued
found that both groups were equivalent on all measures. Ac- treatment. A telephone assessment was conducted to iden-
cording to the MINI, most patients met criteria for PTSD (146 tify possible reasons for dropout. Forty-eight of 61 who were
[73%]) and major depressive disorder (128 [64%]), while 80 reachable (82%) stated that dropout was due to external rea-
(40%) met criteria for panic disorder and 24 (12%) for gener- sons, such as time limitations and transportation difficulties
alized anxiety disorder. In addition, 133 (66%) participants met affecting session attendance. Based on these results, we as-
criteria for 2 or more diagnoses. Finally, 111 participants (58%) sumed that data were missing at random rather than due to
reported being under current life threats. symptom severity, treatment characteristics or therapeutic
Of 200 randomized participants, 109 (54%) completed the relationship. Missing data were addressed using maximum
assessment at the end of treatment. MINI diagnosis, baseline likelihood estimation.37
measures, demographic characteristics, or current threat did Summary of intent to treat and per protocol of observed
not predict missing data in the primary and secondary out- means for primary and secondary outcomes at baseline, post-
comes. The only significant predictor was group assignment, treatment, and follow-up for treatment condition and baseline
with a higher attrition rate for the treatment condition (72 par- and post–wait period for the waitlist condition are described in
ticipants [36%]) compared to waitlist (16 participants [8%]). Table 2. Between-treatment slope differences and effect sizes
Within the treatment group, 24 participants (12%) dropped out for treatment and waitlist comparison for the intent-to-treat and

jamapsychiatry.com (Reprinted) JAMA Psychiatry October 2023 Volume 80, Number 10 995

© 2023 American Medical Association. All rights reserved.

Downloaded from jamanetwork.com by Andrea Romero Martínez on 10/27/2024
 Research Original Investigation Contextual Adaptation of the Unified Protocol in Emotional Disorders in Individuals Exposed to Armed Conflict in Colombia

Table 3. Slope Difference Scores and Between-Condition Effect Sizes for Intent-to-Treat (ITT) and Per-Protocol (PP) Analyses

CXA-UP vs WL difference slope Treatment effect size CXA-UP vs WL

ITT PP ITT PP
Slope (SE) t Value dfa Slope (SE) t Value dfa Cohen d (90% CI)b Cohen d (90% CI)b
Primary outcomes
PCL-5 score −31.12 (3.00)c −10.35 150 −29.89 (3.24)c −9.21 109 0.90 (0.63-1.19) 0.95 (0.63-1.28)
c
PHQ-depression score −11.94 (1.30) −9.17 146 −11.33 (1.39)c −8.14 107 0.77 (0.52-1.06) 0.82 (0.49-1.15)
PHQ-anxiety score −6.52 (0.67)c −9.71 143 −6.25 (0.72)c −8.61 107 0.82 (0.56-1.09) 0.87 (0.56-1.22)
PHQ-somatic score −8.31 (0.92)c −9.00 150 −7.97 (0.99)c −7.98 108 0.75 (0.47-1.04) 0.80 (0.47-1.12)
Secondary outcomes
ODSIS score −10.02 (0.97)c −10.24 154 −9.63 (1.06)c −9.08 109 0.87 (0.61-1.15) 0.95 (0.61-1.28)
c
OASIS score −9.43 (0.93) −10.12 124 −9.55 (0.94)c −10.10 109 1.01 (0.68-1.32) 1.09 (0.75-1.42)
QLESQ score 10.85 (1.53)c 7.06 140 10.48 (1.63)c 6.41 109 0.54 (0.28-0.80) 0.56 (0.28-0.87)
WHODAS score −1.01 (0.14)c −7.06 119 −1.00 (0.14)c −6.76 100 0.61 (0.32-0.90) 0.65 (0.35-0.98)
a
Abbreviations: CXA-UP, contextual/cultural adaptation of the Unified Protocol; df and P values are derived with the Satterthwaite method.
OASIS, Overall Anxiety Severity and Impairment Scale; ODSIS, Overall b
The Cohen d effect sizes presented are transformations of η2 P to a more
Depression Severity and Impairment Scale; PHQ, Patient Health Questionnaire; common effect size measure (interpretation: 0.2 small, 0.5 medium,
PCL-5, posttraumatic stress disorder checklist for DSM-5; QLESQ, Quality-of 0.8 large).
-Life Enjoyment and Satisfaction Questionnaire; WHODAS, World Health c
P < .001.
Organization Disability Assessment Scale; WL, waitlist.

Table 4. Follow-Up Results for Primary and Secondary Outcomes

Pretreatment slope Pretreatment slope
Pretreatment score vs posttreatment slope vs follow-up slope Effect size
Intercept t Value dfa Slope (SE) t Value dfa Slope (SE) t Value dfa Cohen d (90% CI)b
Primary outcomes
PCL-5 score 50.64 25.13 179 −28.10 (2.56)c −10.96 128 −39.51 (2.56)c −15.40 128 1.28 (1.01-1.56)
c
PHQ-somatic score 21.92 34.69 174 −5.77 (0.79) −7.24 127 −6.57 (0.79)c −8.32 126 0.62 (0.39-0.84)
PHQ-depression score 24.71 29.32 162 −8.41 (1.01)c −8.34 127 −11.64 (1.00)c −11.64 126 0.91 (0.67-1.15)
PHQ-anxiety score 16.84 34.16 172 −5.04 (0.61)c −8.19 128 −6.67 (0.61)c −10.93 127 0.84 (0.60-1.09)
Secondary outcomes
ODSIS score 12.71 21.03 178 −8.43 (0.78)c −10.78 127 −10.40 (0.78)c −13.23 127 1.12 (0.84-1.35)
c
OASIS score 12.57 22.26 185 −8.59 (0.74) −11.51 127 −9.44 (0.74)c −12.59 128 1.09 (0.86-1.39)
QLESQ score 43.16 38.89 157 8.64 (1.27)c 6.79 127 14.07 (1.27)c 11.05 127 0.82 (0.58-1.06)
WHODAS score 1.30 12.72 152 −0.52 (0.13)c −4.02 120 −0.36 (0.12)c −3.02 114 0.29 (0.07-0.55)
b
Abbreviations: OASIS, Overall Anxiety Severity and Impairment Scale; The Cohen d effect sizes presented are transformations of η2 P to a more
ODSIS, Overall Depression Severity and Impairment Scale; PCL-5, posttraumatic common effect size measure (interpretation: 0.2 small, 0.5 medium,
stress disorder checklist for DSM-5; PHQ, Patient Health Questionnaire; 0.8 large).
QLESQ, Quality-of -Life Enjoyment and Satisfaction Questionnaire; c
P < .001.
WHODAS, World Health Organization Disability Assessment Scale.
a
df and P values are derived with the Satterthwaite method.

per-protocol analyses are presented in Table 3. Results for pri- Organization Disability Assessment Schedule, −1.01 (SE, 0.14;
mary outcomes in the intent-to-treat analysis show that the be- P < .001; Cohen d, 0.61; 90% CI, 0.32-0.90). Comparison of
tween-treatment effect for posttraumatic stress disorder check- intent-to-treat and per-protocol analyses found that results and
list for DSM-5 was significant with a change of −31.12 (SE, 3.00; effect sizes for primary and secondary outcomes tended to be
P < .001; Cohen d, 0.90; 90% CI, 0.63-1.19), the estimated ef- close, while intent-to-treat and per-protocol effect sizes and con-
fect for PHQ-9 was −11.94 (SE, 1.30; P < .001; Cohen d, 0.77; 90% fidence intervals overlapped. Although effect sizes for the per-
CI, 0.52-1.06), for PHQ-anxiety was −6.52 (SE, 0.67; P < .001; protocol approach tended to be slightly higher, all effect sizes
Cohen d, 0.82; 90% CI, 0.49-1.15), and for PHQ-somatic was were considered large.
−8.31 (SE, 0.92; P < .001; Cohen d, 0.75; 90% CI, 0.47-1.04). For Longitudinal within-model results for primary and sec-
the secondary outcomes, the treatment effect for Overall ondary outcomes at pretreatment, posttreatment, and
Depression Severity and Impairment Scale was −10.02 (SE, 0.97; follow-up are depicted in Table 4. For all primary and second-
P < .001; Cohen d, 0.87; 90% CI,0.61-1.15); for Overall Anxiety ary outcomes, the pretreatment vs posttreatment slope change
Severity and Impairment Scale, −9.43 (SE, 0.93; P < .001; Cohen tended to be close to the pretreatment vs follow-up slope
d, 1.01; 90% CI, 0.68-1.32); for QLESQ, 10.85 (SE, 1.53; P < .001; change, confirming that the pretreatment-posttreatment
Cohen d, 0.54; 90% CI, 0.28-0.80); and for the World Health effect was maintained over 3-month follow-up.

996 JAMA Psychiatry October 2023 Volume 80, Number 10 (Reprinted) jamapsychiatry.com

© 2023 American Medical Association. All rights reserved.

Downloaded from jamanetwork.com by Andrea Romero Martínez on 10/27/2024
 Contextual Adaptation of the Unified Protocol in Emotional Disorders in Individuals Exposed to Armed Conflict in Colombia Original Investigation Research

overlapping between intent-to-treat and per-protocol analy-

Discussion ses still suggest that our results were robust under a
missing-at-random assumption. Second, while unequal
Our study evaluated the efficacy of adapting the UP to the groups sizes do not represent a threat to internal validity, as
cultural and contextual characteristics of individuals there was a true randomization, we addressed the possible
exposed to armed conflict and other related events and threat to statistical inference by using multilevel regressions
forced to abandon their lands and live in an urban environ- to assess differences, which addresses problems related to
ment. Previous studies of psychological interventions tar- unequal group sizes. Third , answers to self-report measures
geted to single disorders in low- and-middle-income coun- could have been influenced by having been read to partici-
tries showed moderate treatment effects that were not pants. Fourth, although using a waitlist control might pro-
maintained at follow-up.38 Another transdiagnostic psycho- duce an overestimation bias effect, 45 the effect sizes for
logic al intervention study in Colombia found mixed improvement in the treatment group are comparable to
results.39 In comparison, our randomized waitlist-controlled those observed in previous studies. Additionally, negative
trial yielded significant differences with large effect sizes effects were not evaluated, as they have not been reported
between treatment and waitlist conditions and within all in previous studies with the UP.15,17-19 Fifth, to maximize
participants who completed treatment and 3-month follow- internal validity, we conducted our study under highly con-
up. This result was observed for all measures of PTSD, anxi- trolled conditions with participants who lived in Bogotá.
ety, depression, somatic complaints, and disability and Therefore, the effects of our intervention may not general-
was accompanied by significant improvements in quality-of- ize to similar individuals living in their territorial homeland
life measures. For anxiety and depression, the effect size within armed conflict zones.46
found was comparable to the medium-to-large effect sizes
observed in UP randomized clinical trials with passive
controls 40 across more usual clinical populations, with
similar results maintained over time as seen in previous
Conclusions
studies.41 To our knowledge, our study is the first random- Considering the large number of individuals needing effec-
ized clinical trial to demonstrate the efficacy of the UP in a tive mental health intervention and the scarcity of mental
sample of patients primarily diagnosed with PTSD in violent health professionals in remote regions, especially in low-
contexts with large effect sizes comparable to trauma- and-middle-income countries, demonstrating the efficacy of
focused interventions.42 a brief, evidence-based psychological intervention targeting
These outcomes could be attributed to several factors. First, multiple problems under controlled conditions is a neces-
the specific modules of the protocol target basic psychologi- sary condition prior to developing shorter and more precise
cal processes common to different emotional disorders. scalable interventions available to larger segments of the
Second, we adapted the original protocol maintaining integ- population. Although randomized clinical trials are the gold
rity to core concepts and skills while using culturally appro- standard for establishing the efficacy of interventions, they
priate language, personalized examples, and exercises tai- do not provide information on individual changes over time
lored to cultural practices.43,44 Third, to address possible nor on differential response to treatment.47 As the UP com-
therapist bias and increase quality control and fidelity to origi- prises multiple modules, future research on differential
nal protocols, we conducted extensive training with and response patterns by different individuals to specific inter-
continuous supervision of therapists. vention components through single-case experimental stud-
ies or group latent class trajectory analyses would give rise to
Limitations more precise and efficient targeted treatments. Indeed, as
Nevertheless, this study has limitations suggesting our we obtained repeated measures in all sessions, a subsequent
results should be interpreted cautiously. First, there was step is to evaluate the efficacy of the intervention and indi-
high, albeit expected, attrition before and during treatment. viduals’ change along different points of the treatment.
This can be attributed to several possible causes. Most Identifying subgroups of individuals showing similar pat-
important among them is that internally displaced persons terns of change related to the specific modules of the
comprise a highly mobile population living in remote urban CXA-UP could help clarify not only mechanisms of change
and unsafe locations. Therefore, these individuals experi- but, most importantly, help identify modifiable predictors of
ence public transportation difficulties and unstable job con- specific trajectories based on underlying psychological
ditions leading to frequent lodging changes that constitute a markers. This could lead to development of briefer modular
barrier to treatment engagement and retention. This is sup- interventions targeted to empirically derived individual psy-
ported by our finding that the main predictor of attrition chological markers rather than diagnostic categories.48,49
was group assignment, with higher attrition rates in the Further research on the effectiveness of the CXA-UP pro-
treatment condition where participants had to travel to vided by nonspecialized health workers with individuals
attend sessions, with 82% of dropouts attributing attrition living in their places of origin in community settings would
to these external reasons. Furthermore, although the nearly improve scalability access and have a greater impact on eco-
50% dropout rate could represent a possible estimation bias nomics and public mental health policy in low- and-middle-
of our results, the average treatment effect and effect sizes income countries.

jamapsychiatry.com (Reprinted) JAMA Psychiatry October 2023 Volume 80, Number 10 997

© 2023 American Medical Association. All rights reserved.

Downloaded from jamanetwork.com by Andrea Romero Martínez on 10/27/2024
 Research Original Investigation Contextual Adaptation of the Unified Protocol in Emotional Disorders in Individuals Exposed to Armed Conflict in Colombia

ARTICLE INFORMATION Health; the 9th World Congress of Behavioural and prevalence of mood disorders, anxiety, and mental
Accepted for Publication: May 4, 2023. Cognitive Therapies; July 19, 2019; Berlin, Germany. disorders in the Colombian adult population. Article
Data Sharing Statement: See Supplement 2. in Spanish. Rev Colomb Psiquiatr. 2016;45(suppl 1):
Published Online: July 19, 2023. 147-153. doi:10.1016/j.rcp.2016.11.001
doi:10.1001/jamapsychiatry.2023.2392 Additional Contributions: We thank the University
of Los Andes for support throughout the study, the 12. Peevey N, Flores E, Seguin M. Common mental
Author Contributions: Messrs García and Moreno disorders and coping strategies amongst internally
had full access to all of the data in the study and Office of the High Commissioner for Victims of
Armed Conflict, the Office of the Mayor of Bogotá, displaced Colombians: a systematic review. Glob
take responsibility for the integrity of the data and Public Health. 2022;17(12):3440-3454.
the accuracy of the data analysis. and victims’ organizations for support in
participants’ referrals to the study. We especially doi:10.1080/17441692.2022.2049343
Concept and design: Castro-Camacho, Barlow,
Farchione, Idrobo, Rattner, Quant, González. thank therapists and participants who made this 13. Lagos-Gallego M, Gutierrez-Segura JC,
Acquisition, analysis, or interpretation of data: study possible. We thank Iona Naismith, PhD, Lagos-Grisales GJ, Rodriguez-Morales AJ.
Castro-Camacho, García, Idrobo, Rattner, González, University of Los Andes, Bogota, Colombia, for Post-traumatic stress disorder in internally
Moreno. proofreading and her valuable comments on displaced people of Colombia: an ecological study.
Drafting of the manuscript: Castro-Camacho, García, previous drafts. Travel Med Infect Dis. 2017;16:41-45. doi:10.1016/
Idrobo, Quant, Moreno. j.tmaid.2017.02.008
Critical revision of the manuscript for important REFERENCES 14. Richards A, Ospina-Duque J, Barrera-Valencia
intellectual content: Castro-Camacho, Barlow, 1. Miller KE, Rasmussen A. War exposure, daily M, et al. Posttraumatic stress disorder, anxiety and
García, Farchione, Idrobo, Rattner, González. stressors, and mental health in conflict and depression symptoms, and psychosocial treatment
Statistical analysis: García, Idrobo. post-conflict settings: bridging the divide between needs in Colombians internally displaced by armed
Obtained funding: Castro-Camacho, Quant. trauma-focused and psychosocial frameworks. Soc conflict: A mixed-method evaluation. Psychol Trauma.
Administrative, technical, or material support: Sci Med. 2010;70(1):7-16. doi:10.1016/j.socscimed. 2011;3:384-393. doi:10.1037/a0022257
Castro-Camacho, Barlow, Farchione, Rattner, 2009.09.029 15. Hofmann SG, Asnaani A, Vonk IJ, Sawyer AT,
Quant, González, Moreno. 2. Steel Z, Chey T, Silove D, Marnane C, Bryant RA, Fang A. The efficacy of cognitive behavioral
Supervision: Castro-Camacho, Quant, González. van Ommeren M. Association of torture and other therapy: a review of meta-analyses. Cognit Ther Res.
Conflict of Interest Disclosures: potentially traumatic events with mental health 2012;36(5):427-440. doi:10.1007/s10608-012-
Drs Castro-Camacho, García, Rattner, and González outcomes among populations exposed to mass 9476-1
reported grants from Colombian Administrative conflict and displacement: a systematic review and 16. Barlow DH, Allen LB, Choate ML. Toward a
Department of Science, Technology, and meta-analysis. JAMA. 2009;302(5):537-549. unified treatment for emotional disorders. Behav
Innovation, Colciencias; the Office of Research doi:10.1001/jama.2009.1132 Ther. 2004;47(6):838-853. doi:10.1016/j.beth.2016.
Vice-President, Universidad de Los Andes; and the 3. Keller A, Lhewa D, Rosenfeld B, et al. Traumatic 11.005
Office of Research Vice-Dean, Faculty of Social experiences and psychological distress in an urban
Sciences, Universidad de Los Andes during the 17. Barlow DH, Farchione TJ, Bullis, JR, et al. The
refugee population seeking treatment services. unified protocol for transdiagnostic treatment of
conduct of the study. Dr Barlow reported grants J Nerv Ment Dis. 2006;194(3):188-194. doi:10.1097/
from Colciencias during the conduct of the study emotional disorders compared with
01.nmd.0000202494.75723.83 diagnosis-specific protocols for anxiety disorders.
and receives royalties from Oxford University Press
for the Unified Protocol for the Transdiagnostic 4. Siriwardhana C, Adikari A, Pannala G, et al. JAMA Psychiatry. 2017;74(9):875-884. doi:10.1001/
Treatment of Emotional Disorders. Dr Farchione Prolonged internal displacement and common jamapsychiatry.2017.2164
reported personal fees from Oxford University mental disorders in Sri Lanka: the COMRAID study. 18. Sakiris N, Berle D. A systematic review and
Press and grants from the National Institute of PLoS One. 2013;8(5):e64742. doi:10.1371/journal. meta-analysis of the Unified Protocol as a
Mental Health, the National Institute of Alcohol and pone.0064742 transdiagnostic emotion regulation based
Alcohol Abuse, John Templeton Foundation, and 5. Colombian Government, Unit for Victims. intervention. Clin Psychol Rev. 2019;72:101751.
Juvenile Diabetes Research Foundation during the Official Victims’ Registry Bogota D.C.: Colombian doi:10.1016/j.cpr.2019.101751
conduct of the study as well as personal fees from Government; 2022 Updated August 1, 2022. 19. Varkovitzky RL, Sherrill AM, Reger GM.
the MacArthur Foundation, the National Center of Accessed August 1, 2022. https://www. Effectiveness of the unified protocol for
Neurology and Psychiatry in Japan, Brightside unidadvictimas.gov.co/es/registro-unico-de- transdiagnostic treatment of emotional disorders
Health, Tage S.R.L., Allegheny Health Network, and victimas-ruv/37394 among veterans with posttraumatic stress disorder:
Grupo Pacifico outside the submitted work. 6. Shultz JM, Garfin DR, Espinel Z, et al. Internally a pilot study. Behav Modif. 2018;42(2):210-230.
Dr Idrobo reported grants from Minciencias during displaced “victims of armed conflict” in Colombia: doi:10.1177/0145445517724539
the conduct of the study. No other disclosures were the trajectory and trauma signature of forced
reported. 20. Hood CO, Southward MW, Bugher C,
migration. Curr Psychiatry Rep. 2014;16(10):475. Sauer-Zavala S. A preliminary evaluation of the
Funding/Support: This study was funded by grant doi:10.1007/s11920-014-0475-7 unified protocol among trauma-exposed adults
614-2014 from the Colombian Administrative 7. Bozzoli C, Bruck T, Wald N. Self-employment and with and without PTSD. Int J Environ Res Public
Department of Science, Technology, and conflict in Colombia. J Conflict Resolut. 2013;57:117- Health. 2021;18(21):11729. doi:10.3390/
Innovation, Colciencias, by the Office of the 142. doi:10.1177/0022002712464849 ijerph182111729
Research Vice-President of the University of Los
Andes, and by the Office of the Research 8. Ibáñez AM, Moya A. Vulnerability of victims of 21. Sheehan D, Janavs J, Baker R, et al. Mini
Vice-Dean, Faculty of Social Sciences, University of civil conflicts: Empirical evidence for the displaced International Neuropsychiatric Interview. Spanish
Los Andes. Bolivar-Davivienda Foundation funded population in Colombia. World Dev. 2010;38:647- version 5.0.0. Ferrando L, Franco-Alfonso L, Soto
transportation costs of participants of the study. 663. doi:10.1016/j.worlddev.2009.11.015 M, Bobes-García J, Soto O, Franco L, Heinze G,
9. Daniels JP. Mental health in post-conflict trans. Published January 1, 2000.
Role of the Funder/Sponsor: The funders had
no role in the design and conduct of the study; Colombia. Lancet Psychiatry. 2018;5(3):199. 22. Norman SB, Cissell SH, Means-Christensen AJ,
collection, management, analysis, and doi:10.1016/S2215-0366(18)30068-3 Stein MB. Development and validation of an Overall
interpretation of the data; preparation, review, or 10. Bell V, Méndez F, Martínez C, Palma PP, Bosch Anxiety Severity And Impairment Scale (OASIS).
approval of the manuscript; and decision to support M. Characteristics of the Colombian armed conflict Depress Anxiety. 2006;23(4):245-249.
the manuscript for publication. and the mental health of civilians living in active doi:10.1002/da.20182

Meeting Presentation: Preliminary data from this conflict zones. Confl Health. 2012;6(1):10. 23. Bentley KH, Gallagher MW, Carl JR, Barlow DH.
study were presented at the Symposium on doi:10.1186/1752-1505-6-10 Development and validation of the Overall
Cognitive Behavioral Therapy in Global Mental 11. Gómez-Restrepo C, Tamayo-Martínez N, Depression Severity and Impairment Scale. Psychol
Buitrago G, et al. Violence due to armed conflict and Assess. 2014;26(3):815-830. doi:10.1037/a0036216

998 JAMA Psychiatry October 2023 Volume 80, Number 10 (Reprinted) jamapsychiatry.com

© 2023 American Medical Association. All rights reserved.

Downloaded from jamanetwork.com by Andrea Romero Martínez on 10/27/2024
 Contextual Adaptation of the Unified Protocol in Emotional Disorders in Individuals Exposed to Armed Conflict in Colombia Original Investigation Research

24. Castro-Camacho L, Rattner M, Quant DM, program for the social, behavioral, and biomedical outcome of the Unified Protocol for Transdiagnostic
González L, Moreno JD, Ametaj A. Contextual sciences. Behav Res Methods. 2007;39(2):175-191. Treatment of Emotional Disorders. Compr Psychiatry.
adaptation of the Unified Protocol for the doi:10.3758/BF03193146 2014;55(8):1920-1927. doi:10.1016/j.comppsych.
transdiagnostic treatment of emotional disorders in 33. Bates D, Mächler M, Bolker B, Walker S. Fitting 2014.07.016
victims of armed conflict in Colombia. Cognit Behav linear mixed-effects models using lme4. J Stat Softw. 42. Lewis C, Roberts NP, Andrew M, Starling E,
Pract. 2019;26:351-365. doi:10.1016/j.cbpra.2018. 2015;67(1):1-4. doi:10.18637/jss.v067.i0 Bisson JI. Psychological therapies for
08.002 post-traumatic stress disorder in adults: systematic
34. Luke SG. Evaluating significance in linear
25. Barlow DH, Farchione TJ, Fairholme CP, et al. mixed-effects models in R. Behav Res Methods. review and meta-analysis. Eur J Psychotraumatol.
Unified Protocol for Transdiagnostic Treatment of 2017;49(4):1494-1502. doi:10.3758/s13428-016- 2020;11(1):1729633. doi:10.1080/20008198.2020.
Emotional Disorders: Therapist Guide. Oxford 0809-y 1729633
University Press; 2011. 43. Griner D, Smith TB. Culturally adapted mental
35. IBM. Relationship between partial eta-squared,
26. Barlow DH, Ellard KK, Fairholme CP, et al. Cohen’s F, and Cohen’s D. Accessed February 20, health intervention: a meta-analytic review.
Unified Protocol for Transdiagnostic Treatment of 2023. https://www.ibm.com/support/pages/effect- Psychotherapy (Chic). 2006;43(4):531-548.
Emotional Disorders: Workbook. Oxford University size-relationship-between-partial-eta-squared- doi:10.1037/0033-3204.43.4.531
Press, 2011. cohens-f-and-cohens-d 44. Arundell LL, Barnett F, Buckman JE, Saunders
27. Castro-Camacho L, Moreno JD, Naismith I. 36. Cohen J. Statistical Power Analysis for the R, Pilling S. The Effectiveness of Adapted
Contextual adaptation of the Unified Protocol for Behavioral Sciences. 2nd ed. Lawrence Erlbaum Psychological Interventions for People From Ethnic
the transdiagnostic treatment of emotional Associates, Inc; 1988. Minority Groups: A Systematic Review and
disorders in victims of armed conflict in Colombia: Conceptual Typology. Clin Psych Rev; 2021:88.
a case study. Cognit Behav Pract. 2019;26:366-380. 37. Sullivan TR, White IR, Salter AB, Ryan P, Lee KJ.
Should multiple imputation be the method of 45. Cunningham JA, Kypri K, McCambridge J.
doi:10.1016/j.cbpra.2018.09.002 Exploratory randomized controlled trial evaluating
choice for handling missing data in randomized
28. Kroenke K, Spitzer RL, Williams J, Löwe B. The trials? Stat Methods Med Res. 2018;27(9):2610-2626. the impact of a waiting list control design. BMC Med
Patient Health Questionnaire Somatic, Anxiety, and doi:10.1177/0962280216683570 Res Methodol. 2013;13(1):150. doi:10.1186/1471-
Depressive Symptom Scales: a systematic review. 2288-13-150
Gen Hosp Psychiatry. 2010;32:345-359. 38. Purgato M, Gastaldon C, Papola D, van
Ommeren M, Barbui C, Tol WA. Psychological 46. Asnaani A, Benhamou K, Kaczkurkin AN,
doi:10.1016/j.genhosppsych.2010.03.006 Turk-Karan E, Foa EB. Beyond the constraints of an
therapies for the treatment of mental disorders in
29. Blevins CA, Weathers FW, Davis MT, Witte TK, low- and middle-income countries affected by RCT: naturalistic treatment outcomes for
Domino JL. The Posttraumatic Stress Disorder humanitarian crises. Cochrane Database Syst Rev. anxiety-related disorders. Behav Ther. 2020;51(3):
Checklist for DSM-5 (PCL-5): development and 2018;7(7):CD011849. doi:10.1002/14651858. 434-446. doi:10.1016/j.beth.2019.07.007
initial psychometric evaluation. J Trauma Stress. CD011849.pub2 47. Howard MC, Hoffman ME. Variable-centered,
2015;28(6):489-498. doi:10.1002/jts.22059 person-centered, and person-specific approaches:
39. Bonilla-Escobar FJ, Fandiño-Losada A,
30. Endicott J, Nee J, Harrison W, Blumenthal R. Martínez-Buitrago DM, et al. A randomized where theory meets the method. Organ Res Methods.
Quality of Life Enjoyment and Satisfaction controlled trial of a transdiagnostic 2018;21:846-876. doi:10.1177/1094428117744021
Questionnaire: a new measure. Psychopharmacol Bull. cognitive-behavioral intervention for 48. Insel T, Cuthbert B, Garvey M, et al. Research
1993;29(2):321-326. Afro-descendants’ survivors of systemic violence in domain criteria (RDoC): toward a new classification
31. Üstün TB, Kostanjsek N, Chatterji S, Rehm J; Colombia. PLoS One. 2018;13(12):e0208483. framework for research on mental disorders. Am J
World Health Organization. Measuring Health and doi:10.1371/journal.pone.0208483 Psychiatry. 2010;167(7):748-751. doi:10.1176/
Disability: Manual for WHO Disability Assessment 40. Carlucci L, Saggino A, Balsamo M. On the appi.ajp.2010.09091379
Schedule (WHODAS 2.0). Published June 16, 2012. efficacy of the unified protocol for transdiagnostic 49. Murray LK, Dorsey S, Haroz E, et al. A common
https://www.who.int/publications/i/item/ treatment of emotional disorders: a systematic elements treatment approach for adult mental
measuring-health-and-disability-manual-for-who- review and meta-analysis. Clin Psychol Rev. 2021; health problems in low- and middle-income
disability-assessment-schedule-(-whodas-2.0) 87:101999. doi:10.1016/j.cpr.2021.101999 countries. Cogn Behav Pract. 2014;21(2):111-123.
32. Faul F, Erdfelder E, Lang AG, Buchner A. 41. Bullis JR, Fortune MR, Farchione TJ, Barlow DH. doi:10.1016/j.cbpra.2013.06.005
G*Power 3: a flexible statistical power analysis A preliminary investigation of the long-term

jamapsychiatry.com (Reprinted) JAMA Psychiatry October 2023 Volume 80, Number 10 999

© 2023 American Medical Association. All rights reserved.

Downloaded from jamanetwork.com by Andrea Romero Martínez on 10/27/2024