CClinicalTrials.gg
TerminatedNCT04024488Updated May 22, 2026Results posted

Group-Based Intervention to Improve Mental Health and Adherence Among Youth Living With HIV in Low Resource Settings

An interventional study of Youth TI-CBT Intervention Arm and Youth Discussion Control Arm in HIV-1-infection, sponsored by International Maternal Pediatric Adolescent AIDS Clinical Trials Group. Terminated at 8 sites in 4 countries. Open to participants aged 15 Years and older. Per ClinicalTrials.gov, last updated 2026-05-22.

Sponsored by International Maternal Pediatric Adolescent AIDS Clinical Trials Group · Not applicable, Interventional, and Supportive care

Why this study was terminated
The study was closed early due to funding issue
Phase
Not applicable
Study type
Interventional
Enrollment
484
Allocation
Randomized
Ages
15 Years and older
Sex
All
01

Study summary

IMPAACT 2016 was a multi-site, two-arm, individually randomized, controlled study to evaluate whether an Indigenous Leader Outreach Model (ILOM) of trauma-informed cognitive behavioral therapy (TI-CBT) delivered by Indigenous Youth Leaders (IYL) is associated with improved mental health outcomes and ART adherence among youth living with HIV in resource-limited settings. The intervention was adapted to the local context through advance conduct of focus groups and pilot testing.

Read the detailed description

IMPAACT 2016 was a multi-site, two-arm, individually randomized, controlled study to evaluate whether an Indigenous Leader Outreach Model (ILOM) of trauma-informed cognitive behavioral therapy (TI-CBT) delivered by Indigenous Youth Leaders (IYL) is associated with improved mental health outcomes and ART adherence among youth living with HIV in resource-limited settings. The trial was preceded by an adaption of the TI-CBT Intervention at each site using the "ADAPT-ITT" model, which included focus groups and pilot tests. Participant-level data was not collected during the focus group phase, and the pilot phase consent form specified that collected data would not be analyzed. Per protocol, only data from the randomized trial were analyzed. The study was closed early due to the funding constraints.

In the randomized study, youth were individually randomized to either the TI-CBT Intervention arm or the Discussion Control arm. Youth participants met as a group within their randomized arm and received multiple group sessions within an 8-week period. Each arm featured 15 groups with an average of 8.5 youth per group, or a total of 254 youth in the Randomized Trial. TI-CBT Intervention and Discussion Control groups could be mixed-gender, which required at least two participants of each gender. Caregivers (with consent and youth permission) were assigned to the same study arm as their youth. Caregiver participants met as a group within their arm for two caregiver-specific sessions on two separate weeks, separate from their youth's group sessions. Each arm featured 15 groups with an average of 7.7 caregivers per group, or a total of 230 caregivers in the Randomized Trial. Youth and caregivers completed a follow-up assessment immediately after their last group session and two additional follow-up assessments at 6 and 12 months. Youth and caregivers received 1 two-hour booster group session consistent with their assigned study arm immediately after the 6-month assessments. The booster sessions were intended to enhance treatment effects and increase sustainability.

Following completion of 6-month assessments (primary timepoint), data were analyzed to determine the short-term effects of the interventions. Additional analyses to assess long-term effects of the interventions, including the effects of the booster group session, will be performed after the completion of 12-month assessments.

Results are reported for youth participants in the randomized trial only, because limited sociodemographic characteristics and no primary or secondary outcome measures were collected for caregivers. As specified in the protocol and statistical analysis plan, primary outcomes were analyzed with groups as the analytic units due to the presence of intracluster correlation among participants in the same group.

02

Conditions studied

  • HIV-1-infection

Keywords

  • Mental Health
  • Adherence
  • Trauma-Informed Cognitive Behavioral Therapy
03

Who can participate

Ages eligible
15 Years and older
Sexes eligible
All
Accepts healthy volunteers
No

Eligibility criteria

Inclusion Criteria - For Youth Participants:

  • If of legal age to provide independent informed consent as determined by site Standard Operating Procedures (SOPs) and consistent with site IRB/EC policies and procedures: potential youth participant is willing and able to provide written informed consent for study participation.
  • If not of legal age to provide independent informed consent: Parent or guardian is willing and able to provide written informed consent for study participation and potential youth participant is willing and able to provide written informed assent for study participation.
  • Confirmed HIV-infection based on documented testing of two samples collected at different time points as documented in medical records or by confirmatory testing.
  • Age limit 15-19 years
  • At screening, aware of his or her HIV infection, as confirmed by Investigator of Record or designee.
  • At screening, has been prescribed ART for a minimum of 24 weeks prior to screening based on medical record documentation.
  • At screening, meets at least one of the following indicators of moderate to severe mental health symptomology:

    • Patient Health Questionnaire-9 (PHQ-9) score ≥ 10
    • General Anxiety Disorder-7 (GAD-7) score ≥ 10
    • UCLA Post-Traumatic Stress Disorder-Reaction Index (UCLA PTSD-RI) score >= 35

Inclusion Criteria - For Caregiver Participants:

  • Caregiver, defined as a biological parent, legal guardian, or person who provides emotional, psychological and/or informational care to a youth taking part in the Randomized Trial, as identified by the youth, and for whom the youth has provided written permission to participate in the study.
  • Of legal age to provide independent consent and willing and able to provide written informed consent for study participation.

Exclusion Criteria - For Youth Participants:

  • At entry, participating in a study delivering a mental health or ART adherence intervention.
  • Prior participation in an IMPAACT 2016 Focus Group or Pilot Test
  • Any other condition, adverse social situation or cognitive impairment that, in the opinion of the site investigator, would preclude informed assent and informed consent, make study participation unsafe, complicate interpretation of study outcome data, or otherwise interfere with achieving the study objectives.

Exclusion Criteria - For Caregiver Participants:

Prior participation in an IMPAACT 2016 Focus Group or Pilot Test.

04

Study design

Phase
Not applicable
Primary purpose
Supportive care
Allocation
Randomized
Intervention model
Parallel assignment
Masking
None (open label)
Enrollment
484 participants (actual)

Study arms

  • Experimental
    Youth TI-CBT Intervention Arm

    For youth, the TI-CBT intervention arm consisted of six 2-hour TI-CBT group sessions led by Indigenous Youth Leaders (IYL) during weeks 1 to 8 and one 2-hour booster group session at 6-months. The TI-CBT Intervention was delivered in a group format. The TI-CBT Intervention arm had 15 groups with an average of 8.5 youth per group (range 6-11).

    Behavioral: Youth TI-CBT Intervention Arm

  • Active comparator
    Youth Discussion Control Arm

    For youth, the Discussion Control arm consisted of six 2-hour discussion group sessions led by IYL during weeks 1 to 8 and one 2-hour booster discussion group session at 6 months. The Discussion Control was delivered in a group format. The Discussion Control arm had 15 groups with an average of 8.4 youth per group (range 5-10).

    Behavioral: Youth Discussion Control Arm

  • Experimental
    Caregiver TI-CBT Intervention Arm

    For caregivers, the TI-CBT Intervention arm consisted of two 2-hour TI-CBT group sessions on two separate weeks within 8 weeks of their youth's first group session and one 2-hour booster group session at 6-months led by adult study staff. The TI-CBT Intervention was delivered in group format. The TI-CBT Intervention arm had 15 groups with an average of 7.7 caregivers per group (range 5-10).

    Behavioral: Caregiver TI-CBT Intervention Arm

  • Active comparator
    Caregivers Discussion Control Arm

    For caregivers, the Discussion Control arm consisted of: two 2-hour discussion group sessions on two separate weeks within 8 weeks of their youth's first group session and one 2-hour booster discussion group session at 6 months led by adult study staff. The Discussion Control intervention was delivered in a group format. The Discussion Control arm had 15 groups with an average of 7.6 caregivers per group (range 3-10).

    Behavioral: Caregiver Discussion Control Arm

Interventions

  • BehavioralYouth TI-CBT Intervention Arm

    For youth, each TI-CBT group session was led by IYL. TI-CBT teaches techniques and ways to manage distress through psychosocial health education, cognitive restructuring, and "mastery of trauma", which refers to the process by which survivors of psychological trauma work through the traumatic experience in a meaningful way, and are able to move on with life. TI-CBT addresses both the trauma of learning one has HIV and the trauma associated with managing a chronic and stigmatized illness. The intervention highlights links between HIV and traditional gender roles, gender inequities, and gender-based violence. Finally, the relaxation training teaches youth strategies to relax and these are integrated at the beginning and end of each session. TI-CBT Youth Intervention Manuals were distributed to sites for translation, backtranslation, and cultural adaptation in preparation for the Randomized Trial.

  • BehavioralYouth Discussion Control Arm

    For youth, each discussion group session was led by IYL. Discussion topics were selected by youth in the group. Discussion Control sessions took place at a separate time from TI-CBT Intervention sessions to minimize contamination. Youth and caregiver group sessions were also held separately.

  • BehavioralCaregiver TI-CBT Intervention Arm

    For caregivers, each TI-CBT group session was led by adult study staff. TI-CBT teaches techniques and ways to manage distress through psychosocial health education, cognitive restructuring, and "mastery of trauma", which refers to the process by which survivors of psychological trauma work through the traumatic experience in a meaningful way, and are able to move on with life. TI-CBT addresses both the trauma of learning one has HIV and the trauma associated with managing a chronic and stigmatized illness. The intervention highlights links between HIV and traditional gender roles, gender inequities, and gender-based violence. Finally, the relaxation training teaches strategies to relax and these are integrated at the beginning and end of each session. TI-CBT Caregiver Intervention Manuals were distributed to sites for translation, backtranslation, and cultural adaptation in preparation for the Randomized Trial.

  • BehavioralCaregiver Discussion Control Arm

    For caregivers, each discussion group session was led by adult study staff. Discussion topics were selected by caregivers in the group. Discussion Control sessions took place at a separate time from TI-CBT Intervention sessions to minimize contamination. Youth and caregiver group sessions were also held separately.

05

What researchers measure

Primary outcomes

  1. Group-level Mean General Anxiety Disorder-7 (GAD-7) at 6 Months

    Average (standard deviation) of the group-level mean General Anxiety Disorder-7 (GAD-7) score within arm. The GAD-7 features 7 items assessing frequency of anxiety symptoms over the past 2 weeks on a scale from 0="Not at all" to 3="Nearly every day". Responses to the 7 items are summed to obtain a GAD-7 score ranging from 0 (least) to 21 (most severe symptoms). A score of 10 or more points indicates moderate to severe anxiety symptoms.

    Time frame: at 6 months

  2. Group-level Mean UCLA Post-Traumatic Stress Disorder-Reaction Index (UCLA-PTSD-RI) at 6 Months

    Average (standard deviation) of the group-level mean UCLA Post Traumatic Stress Disorder-Reaction Index (PTSD-RI) score within arm. The UCLA PTSD-RI assesses DSM-IV criteria for PTSD in youth. If prior trauma is endorsed, assessment asks whether 31 PTSD symptoms have occurred within the past month on a scale from 0="None" to 4="Most of the time". A scoring algorithm is applied to the 31 items to calculate a score ranging from 0 (least) to 80 (most severe symptoms). A score of 35 or more indicates moderate to severe PTSD symptoms. If no trauma is endorsed, no score is calculated.

    Time frame: at 6 months

  3. Group-level Mean Composite Mental Health Measure at 6 Months

    Average (standard deviation) of the group-level mean Composite score within arm. Composite scores are a study-specific measure of combined mental health symptoms. A participant's PHQ-9, GAD-7, and UCLA PTSD-RI scores at the given timepoint are each standardized according to baseline to get Z scores. The 3 Z scores are summed to obtain a Composite score. Composite scores are not Z scores themselves and can range from -Inf to +Inf, with greater scores reflecting more severe symptoms overall. Composite scores of 0 at baseline represent average symptom levels overall, but otherwise do not have a meaningful central value.

    Time frame: at 6 months

  4. Group-level Mean Patient Health Questionnaire-9 (PHQ-9) at 6 Months

    Average (standard deviation) of the group-level mean Patient Health Questionnaire-9 (PHQ-9) score within arm. The PHQ-9 features 9 items assessing the frequency of depression symptoms over the past 2 weeks on a scale from 0="Not at all" to 3="Nearly every day". Responses to the 9 items are summed to obtain a PHQ-9 score ranging from 0 (least) to 27 (most severe symptoms). A score of 10 or more points indicates moderate to severe depression symptoms.

    Time frame: at 6 months

Secondary outcomes

  1. Group-level Mean General Anxiety Disorder-7 (GAD-7) at IPL (Immediately Post-Last Group Session)

    Average (standard deviation) of the group-level mean General Anxiety Disorder-7 (GAD-7) score within arm. The GAD-7 features 7 items assessing frequency of anxiety symptoms over the past 2 weeks on a scale from 0="Not at all" to 3="Nearly every day". Responses to the 7 items are summed to obtain a GAD-7 score ranging from 0 (least) to 21 (most severe symptoms). A score of 10 or more points indicates moderate to severe anxiety symptoms.

    Time frame: After initial treatment completion (initial treatment lasted an average of 6 weeks)

  2. Group-level Mean Patient Health Questionnaire-9 (PHQ-9) at IPL (Immediately Post Last Group Session)

    Average (standard deviation) of the group-level mean Patient Health Questionnaire-9 (PHQ-9) score within arm. The PHQ-9 features 9 items assessing the frequency of depression symptoms over the past 2 weeks on a scale from 0="Not at all" to 3="Nearly every day". Responses to the 9 items are summed to obtain a PHQ-9 score ranging from 0 (least) to 27 (most severe symptoms). A score of 10 or more points indicates moderate to severe depression symptoms.

    Time frame: After initial treatment completion (initial treatment lasted an average of 6 weeks)

  3. Group-level Mean UCLA Post-Traumatic Stress Disorder-Reaction Index (UCLA PTSD-RI) at IPL (Immediately Post-Last Group Session)

    Average (standard deviation) of the group-level mean UCLA Post Traumatic Stress Disorder-Reaction Index (PTSD-RI) score within arm. The UCLA PTSD-RI assesses DSM-IV criteria for PTSD in youth. If prior trauma is endorsed, assessment asks whether 31 PTSD symptoms have occurred within the past month on a scale from 0="None" to 4="Most of the time". A scoring algorithm is applied to the 31 items to calculate a score ranging from 0 (least) to 80 (most severe symptoms). A score of 35 or more indicates moderate to severe PTSD symptoms. If no trauma is endorsed, no score is calculated.

    Time frame: After initial treatment completion (initial treatment lasted an average of 6 weeks)

  4. Group-level Mean Composite Mental Health Measure at IPL (Immediately Post-Last Group Session)

    Average (standard deviation) of the group-level mean Composite score within arm. Composite scores are a study-specific measure of combined mental health symptoms. A participant's PHQ-9, GAD-7, and UCLA PTSD-RI scores at the given timepoint are each standardized according to baseline to get Z scores. The 3 Z scores are summed to obtain a Composite score. Composite scores are not Z scores themselves and can range from -Inf to +Inf, with greater scores reflecting more severe symptoms overall. Composite scores of 0 at baseline represent average symptom levels overall, but otherwise do not have a meaningful central value.

    Time frame: After initial treatment completion (initial treatment lasted an average of 6 weeks)

  5. Group-level Mean ART Adherence at 6 Months

    Average (standard deviation) of the group-level mean antiretroviral therapy (ART) adherence score within arm. The Wilson 3-Item Adherence Scale assessed self-reported ART adherence. Within the past 30 days, the 3 items correspond to missed daily doses (0-30), adherence to medication instructions ("Very poor" to "Excellent"), and frequency of perfect use ("Never" to "Always"). Responses on the three items are transformed linearly to scores between 0 and 100 points and averaged to obtain a single score. Scores range from 0 (worst) possible adherence to 100 (best possible adherence).

    Time frame: at 6 months

  6. Viral Load at 6 Months

    HIV-1 RNA

    Time frame: at 6 months

  7. Group-level Mean ART Adherence at IPL (Immediately Post-Last Group Session).

    Self-report - Wilson 3-item scale (range 0-100, higher=better)

    Time frame: After initial treatment completion (initial treatment lasted an average of 6 weeks).

  8. Viral Load at IPL (Immediately Post-Last Group Session)

    HIV-1 RNA

    Time frame: After initial treatment completion (initial treatment lasted an average of 6 weeks)

06

Results

Posted May 22, 2026
Limitations and caveats
1. Due to gender stratification, 12 youth were assigned five to one arm and seven to another. To meet the requirement of six youth per arm, per protocol, the last youth and their caregiver were reassigned to the Discussion arm 2. Due to funding constraints the study was closed early 3. Thirty-seven percent of youth with a Month 6 assessment and twenty-one percent of youth with an IPL assessment did not endorse a trauma, and subsequently did not receive a UCLA PTSD-RI or Composite score.

Participant flow

Accrual occurred between April 2024 and October 2024 in Botswana, Malawi, South Africa, and Zimbabwe at 8 different sites. The first participant was enrolled on 6 April 2024 and the trial closed to accrual on 7 October 2024.

Participant flow — Overall Study
MilestoneYouth TI-CBT Intervention ArmYouth Discussion Control ArmCaregiver TI-CBT Intervention ArmCaregivers Discussion Control Arm
Started128126116114
Completed128126115113
Not completed0011

Outcome measures

PrimaryGroup-level Mean General Anxiety Disorder-7 (GAD-7) at 6 Months

Average (standard deviation) of the group-level mean General Anxiety Disorder-7 (GAD-7) score within arm. The GAD-7 features 7 items assessing frequency of anxiety symptoms over the past 2 weeks on a scale from 0="Not at all" to 3="Nearly every day". Responses to the 7 items are summed to obtain a GAD-7 score ranging from 0 (least) to 21 (most severe symptoms). A score of 10 or more points indicates moderate to severe anxiety symptoms.

Time frame:
at 6 months
Reported as:
Mean · Score
Group-level Mean General Anxiety Disorder-7 (GAD-7) at 6 Months
ScoreYouth TI-CBT Intervention ArmYouth Discussion Control Arm
Group-level Mean General Anxiety Disorder-7 (GAD-7) at 6 Months4.51 ± 2.14.29 ± 1.7
Statistical analysis
  • Youth TI-CBT Intervention Arm vs Youth Discussion Control Arm · t-test, 2 sided · p = 0.74 · Mean difference (final values): 0.22 · 95% CI -1.23 to 1.68A pair was defined as the TI-CBT Intervention and Discussion Control group within the same site and randomization wave. Youth responses were averaged within-group, and the mean difference in paired group-level means was assessed using a t-test.
  • Youth TI-CBT Intervention Arm vs Youth Discussion Control Arm · t-test, 2 sided · p = 0.71 · Mean difference (final values): 0.36 · 95% CI -1.62 to 2.34A pair was defined as the TI-CBT Intervention and Discussion Control group within the same site and randomization wave. Youth responses were averaged within-group, and the mean difference in paired group-level means was assessed using a t-test.
  • Youth TI-CBT Intervention Arm vs Youth Discussion Control Arm · t-test, 2 sided · p = 0.94 · Mean difference (final values): 0.07 · 95% CI -1.88 to 2.01A pair was defined as the TI-CBT Intervention and Discussion Control group within the same site and randomization wave. Youth responses were averaged within-group, and the mean difference in paired group-level means was assessed using a t-test.
PrimaryGroup-level Mean UCLA Post-Traumatic Stress Disorder-Reaction Index (UCLA-PTSD-RI) at 6 Months

Average (standard deviation) of the group-level mean UCLA Post Traumatic Stress Disorder-Reaction Index (PTSD-RI) score within arm. The UCLA PTSD-RI assesses DSM-IV criteria for PTSD in youth. If prior trauma is endorsed, assessment asks whether 31 PTSD symptoms have occurred within the past month on a scale from 0="None" to 4="Most of the time". A scoring algorithm is applied to the 31 items to calculate a score ranging from 0 (least) to 80 (most severe symptoms). A score of 35 or more indicates moderate to severe PTSD symptoms. If no trauma is endorsed, no score is calculated.

Time frame:
at 6 months
Reported as:
Mean · Score
Group-level Mean UCLA Post-Traumatic Stress Disorder-Reaction Index (UCLA-PTSD-RI) at 6 Months
ScoreYouth TI-CBT Intervention ArmYouth Discussion Control Arm
Group-level Mean UCLA Post-Traumatic Stress Disorder-Reaction Index (UCLA-PTSD-RI) at 6 Months21.72 ± 11.015.88 ± 8.7
Statistical analysis
  • Youth TI-CBT Intervention Arm vs Youth Discussion Control Arm · t-test, 2 sided · p = 0.033 · Mean difference (final values): 5.84 · 95% CI 0.54 to 11.14A pair was defined as the TI-CBT Intervention and Discussion Control group within the same site and randomization wave. Youth responses were averaged within-group, and the mean difference in paired group-level means was assessed using a t-test.
  • Youth TI-CBT Intervention Arm vs Youth Discussion Control Arm · t-test, 2 sided · p = 0.051 · Mean difference (final values): 5.42 · 95% CI -0.04 to 10.88A pair was defined as the TI-CBT Intervention and Discussion Control group within the same site and randomization wave. Youth responses were averaged within-group, and the mean difference in paired group-level means was assessed using a t-test.
  • Youth TI-CBT Intervention Arm vs Youth Discussion Control Arm · t-test, 2 sided · p = 0.32 · Mean difference (final values): 6.08 · 95% CI -6.58 to 18.74A pair was defined as the TI-CBT Intervention and Discussion Control group within the same site and randomization wave. Youth responses were averaged within-group, and the mean difference in paired group-level means was assessed using a t-test.
PrimaryGroup-level Mean Composite Mental Health Measure at 6 Months

Average (standard deviation) of the group-level mean Composite score within arm. Composite scores are a study-specific measure of combined mental health symptoms. A participant's PHQ-9, GAD-7, and UCLA PTSD-RI scores at the given timepoint are each standardized according to baseline to get Z scores. The 3 Z scores are summed to obtain a Composite score. Composite scores are not Z scores themselves and can range from -Inf to +Inf, with greater scores reflecting more severe symptoms overall. Composite scores of 0 at baseline represent average symptom levels overall, but otherwise do not have a meaningful central value.

Time frame:
at 6 months
Reported as:
Mean · Score
Group-level Mean Composite Mental Health Measure at 6 Months
ScoreYouth TI-CBT Intervention ArmYouth Discussion Control Arm
Group-level Mean Composite Mental Health Measure at 6 Months-3.23 ± 1.8-4.09 ± 1.4
Statistical analysis
  • Youth TI-CBT Intervention Arm vs Youth Discussion Control Arm · t-test, 2 sided · p = 0.089 · Mean difference (final values): 0.86 · 95% CI -0.15 to 1.87A pair was defined as the TI-CBT Intervention and Discussion Control group within the same site and randomization wave. Youth responses were averaged within-group, and the mean difference in paired group-level means was assessed using a t-test.
  • Youth TI-CBT Intervention Arm vs Youth Discussion Control Arm · t-test, 2 sided · p = 0.031 · Mean difference (final values): 0.90 · 95% CI 0.10 to 1.71A pair was defined as the TI-CBT Intervention and Discussion Control group within the same site and randomization wave. Youth responses were averaged within-group, and the mean difference in paired group-level means was assessed using a t-test.
  • Youth TI-CBT Intervention Arm vs Youth Discussion Control Arm · t-test, 2 sided · p = 0.36 · Mean difference (final values): 0.90 · 95% CI -1.14 to 2.95A pair was defined as the TI-CBT Intervention and Discussion Control group within the same site and randomization wave. Youth responses were averaged within-group, and the mean difference in paired group-level means was assessed using a t-test.
SecondaryGroup-level Mean General Anxiety Disorder-7 (GAD-7) at IPL (Immediately Post-Last Group Session)

Average (standard deviation) of the group-level mean General Anxiety Disorder-7 (GAD-7) score within arm. The GAD-7 features 7 items assessing frequency of anxiety symptoms over the past 2 weeks on a scale from 0="Not at all" to 3="Nearly every day". Responses to the 7 items are summed to obtain a GAD-7 score ranging from 0 (least) to 21 (most severe symptoms). A score of 10 or more points indicates moderate to severe anxiety symptoms.

Time frame:
After initial treatment completion (initial treatment lasted an average of 6 weeks)
Reported as:
Mean · Score
Group-level Mean General Anxiety Disorder-7 (GAD-7) at IPL (Immediately Post-Last Group Session)
ScoreYouth TI-CBT Intervention ArmYouth Discussion Control Arm
Group-level Mean General Anxiety Disorder-7 (GAD-7) at IPL (Immediately Post-Last Group Session)4.92 ± 1.84.63 ± 1.6
Statistical analysis
  • Youth TI-CBT Intervention Arm vs Youth Discussion Control Arm · t-test, 2 sided · p = 0.63 · Mean difference (final values): 0.29 · 95% CI -1.0 to 1.59A pair was defined as the TI-CBT Intervention and Discussion Control group within the same site and randomization wave. Youth responses were averaged within-group, and the mean difference in paired group-level means was assessed using a t-test.
SecondaryGroup-level Mean Patient Health Questionnaire-9 (PHQ-9) at IPL (Immediately Post Last Group Session)

Average (standard deviation) of the group-level mean Patient Health Questionnaire-9 (PHQ-9) score within arm. The PHQ-9 features 9 items assessing the frequency of depression symptoms over the past 2 weeks on a scale from 0="Not at all" to 3="Nearly every day". Responses to the 9 items are summed to obtain a PHQ-9 score ranging from 0 (least) to 27 (most severe symptoms). A score of 10 or more points indicates moderate to severe depression symptoms.

Time frame:
After initial treatment completion (initial treatment lasted an average of 6 weeks)
Reported as:
Mean · Score
Group-level Mean Patient Health Questionnaire-9 (PHQ-9) at IPL (Immediately Post Last Group Session)
ScoreYouth TI-CBT Intervention ArmYouth Discussion Control Arm
Group-level Mean Patient Health Questionnaire-9 (PHQ-9) at IPL (Immediately Post Last Group Session)5.81 ± 2.26.13 ± 2.4
Statistical analysis
  • Youth TI-CBT Intervention Arm vs Youth Discussion Control Arm · t-test, 2 sided · p = 0.69 · Mean difference (final values): -0.32 · 95% CI -1.99 to 1.35A pair was defined as the TI-CBT Intervention and Discussion Control group within the same site and randomization wave. Youth responses were averaged within-group, and the mean difference in paired group-level means was assessed using a t-test.
SecondaryGroup-level Mean UCLA Post-Traumatic Stress Disorder-Reaction Index (UCLA PTSD-RI) at IPL (Immediately Post-Last Group Session)

Average (standard deviation) of the group-level mean UCLA Post Traumatic Stress Disorder-Reaction Index (PTSD-RI) score within arm. The UCLA PTSD-RI assesses DSM-IV criteria for PTSD in youth. If prior trauma is endorsed, assessment asks whether 31 PTSD symptoms have occurred within the past month on a scale from 0="None" to 4="Most of the time". A scoring algorithm is applied to the 31 items to calculate a score ranging from 0 (least) to 80 (most severe symptoms). A score of 35 or more indicates moderate to severe PTSD symptoms. If no trauma is endorsed, no score is calculated.

Time frame:
After initial treatment completion (initial treatment lasted an average of 6 weeks)
Reported as:
Mean · Score
Group-level Mean UCLA Post-Traumatic Stress Disorder-Reaction Index (UCLA PTSD-RI) at IPL (Immediately Post-Last Group Session)
ScoreYouth TI-CBT Intervention ArmYouth Discussion Control Arm
Group-level Mean UCLA Post-Traumatic Stress Disorder-Reaction Index (UCLA PTSD-RI) at IPL (Immediately Post-Last Group Session)20.32 ± 9.019.0 ± 7.4
Statistical analysis
  • Youth TI-CBT Intervention Arm vs Youth Discussion Control Arm · t-test, 2 sided · p = 0.64 · Mean difference (final values): 1.33 · 95% CI -4.60 to 7.25A pair was defined as the TI-CBT Intervention and Discussion Control group within the same site and randomization wave. Youth responses were averaged within-group, and the mean difference in paired group-level means was assessed using a t-test.
SecondaryGroup-level Mean Composite Mental Health Measure at IPL (Immediately Post-Last Group Session)

Average (standard deviation) of the group-level mean Composite score within arm. Composite scores are a study-specific measure of combined mental health symptoms. A participant's PHQ-9, GAD-7, and UCLA PTSD-RI scores at the given timepoint are each standardized according to baseline to get Z scores. The 3 Z scores are summed to obtain a Composite score. Composite scores are not Z scores themselves and can range from -Inf to +Inf, with greater scores reflecting more severe symptoms overall. Composite scores of 0 at baseline represent average symptom levels overall, but otherwise do not have a meaningful central value.

Time frame:
After initial treatment completion (initial treatment lasted an average of 6 weeks)
Reported as:
Mean · Score
Group-level Mean Composite Mental Health Measure at IPL (Immediately Post-Last Group Session)
ScoreYouth TI-CBT Intervention ArmYouth Discussion Control Arm
Group-level Mean Composite Mental Health Measure at IPL (Immediately Post-Last Group Session)-3.50 ± 1.2-3.76 ± 1.3
Statistical analysis
  • Youth TI-CBT Intervention Arm vs Youth Discussion Control Arm · t-test, 2 sided · p = 0.51 · Mean difference (final values): 0.27 · 95% CI -0.57 to 1.11A pair was defined as the TI-CBT Intervention and Discussion Control group within the same site and randomization wave. Youth responses were averaged within-group, and the mean difference in paired group-level means was assessed using a t-test.
SecondaryGroup-level Mean ART Adherence at 6 Months

Average (standard deviation) of the group-level mean antiretroviral therapy (ART) adherence score within arm. The Wilson 3-Item Adherence Scale assessed self-reported ART adherence. Within the past 30 days, the 3 items correspond to missed daily doses (0-30), adherence to medication instructions ("Very poor" to "Excellent"), and frequency of perfect use ("Never" to "Always"). Responses on the three items are transformed linearly to scores between 0 and 100 points and averaged to obtain a single score. Scores range from 0 (worst) possible adherence to 100 (best possible adherence).

Time frame:
at 6 months
Reported as:
Mean · Score
Group-level Mean ART Adherence at 6 Months
ScoreYouth TI-CBT Intervention ArmYouth Discussion Control Arm
Group-level Mean ART Adherence at 6 Months77.89 ± 5.878.28 ± 4.5
Statistical analysis
  • Youth TI-CBT Intervention Arm vs Youth Discussion Control Arm · t-test, 2 sided · p = 0.84 · Mean difference (final values): -0.39 · 95% CI -4.57 to 3.79A pair was defined as the TI-CBT Intervention and Discussion Control group within the same site and randomization wave. Youth responses were averaged within-group, and the mean difference in paired group-level means was assessed using a t-test.
PrimaryGroup-level Mean Patient Health Questionnaire-9 (PHQ-9) at 6 Months

Average (standard deviation) of the group-level mean Patient Health Questionnaire-9 (PHQ-9) score within arm. The PHQ-9 features 9 items assessing the frequency of depression symptoms over the past 2 weeks on a scale from 0="Not at all" to 3="Nearly every day". Responses to the 9 items are summed to obtain a PHQ-9 score ranging from 0 (least) to 27 (most severe symptoms). A score of 10 or more points indicates moderate to severe depression symptoms.

Time frame:
at 6 months
Reported as:
Mean · Score
Group-level Mean Patient Health Questionnaire-9 (PHQ-9) at 6 Months
ScoreYouth TI-CBT Intervention ArmYouth Discussion Control Arm
Group-level Mean Patient Health Questionnaire-9 (PHQ-9) at 6 Months5.36 ± 2.85.41 ± 1.9
Statistical analysis
  • Youth TI-CBT Intervention Arm vs Youth Discussion Control Arm · t-test, 2 sided · p = 0.95 · Mean difference (final values): -0.06 · 95% CI -1.82 to 1.71A pair was defined as the TI-CBT Intervention and Discussion Control group within the same site and randomization wave. Youth responses were averaged within-group, and the mean difference in paired group-level means was assessed using a t-test.
  • Youth TI-CBT Intervention Arm vs Youth Discussion Control Arm · t-test, 2 sided · p = 0.67 · Mean difference (final values): -0.35 · 95% CI -2.08 to 1.38A pair was defined as the TI-CBT Intervention and Discussion Control group within the same site and randomization wave. Youth responses were averaged within-group, and the mean difference in paired group-level means was assessed using a t-test.
  • Youth TI-CBT Intervention Arm vs Youth Discussion Control Arm · t-test, 2 sided · p = 0.93 · Mean difference (final values): -0.10 · 95% CI -2.51 to 2.30A pair was defined as the TI-CBT Intervention and Discussion Control group within the same site and randomization wave. Youth responses were averaged within-group, and the mean difference in paired group-level means was assessed using a t-test.
SecondaryViral Load at 6 Months

HIV-1 RNA

Time frame:
at 6 months
Reported as:
Number · copies/mL
Viral Load at 6 Months
copies/mLYouth TI-CBT Intervention ArmYouth Discussion Control Arm
< 40 copies/mL102104
40 - <200 copies/mL75
200 - < 1,000 copies/mL44
1,000 - < 10,000 copies/mL at month 686
>= 10,000 copies/mL55
SecondaryGroup-level Mean ART Adherence at IPL (Immediately Post-Last Group Session).

Self-report - Wilson 3-item scale (range 0-100, higher=better)

Time frame:
After initial treatment completion (initial treatment lasted an average of 6 weeks).
Reported as:
Mean · Score
Group-level Mean ART Adherence at IPL (Immediately Post-Last Group Session).
ScoreYouth TI-CBT Intervention ArmYouth Discussion Control Arm
Group-level Mean ART Adherence at IPL (Immediately Post-Last Group Session).79.09 ± 6.280.81 ± 4.4
Statistical analysis
  • Youth TI-CBT Intervention Arm vs Youth Discussion Control Arm · t-test, 2 sided · p = 0.50 · Mean difference (final values): -1.72 · 95% CI -7.03 to 3.59A pair was defined as the TI-CBT Intervention and Discussion Control group within the same site and randomization wave. Youth responses were averaged within-group, and the mean difference in paired group-level means was assessed using a t-test.
SecondaryViral Load at IPL (Immediately Post-Last Group Session)

HIV-1 RNA

Time frame:
After initial treatment completion (initial treatment lasted an average of 6 weeks)
Reported as:
Number · copies/mL
Viral Load at IPL (Immediately Post-Last Group Session)
copies/mLYouth TI-CBT Intervention ArmYouth Discussion Control Arm
< 40 copies/mL112103
40 - <200 copies/mL69
200 - < 1,000 copies/mL54
1,000 - < 10,000 copies/mL12
>= 10,000 copies/mL46

Adverse events

Collected over All safety events from study entry to the study completion up to month 12.. Non-serious events are listed at a 0% frequency threshold.

Adverse event summary by group
GroupDeathsSeriousOther
TI-CBT Intervention Arm0/128 (0%)2/128 (1.6%)4/128 (3.1%)
Discussion Control Arm0/126 (0%)1/126 (0.8%)1/126 (0.8%)
Most frequent serious events
Most frequent serious events
EventTI-CBT Intervention ArmDiscussion Control Arm
Suicidal ideationPsychiatric disorders0/1281/126
Hand fractureInjury, poisoning and procedural complications1/1280/126
Suicide attemptPsychiatric disorders1/1280/126
Most frequent other events
Most frequent other events
EventTI-CBT Intervention ArmDiscussion Control Arm
Suicidal ideationPsychiatric disorders4/1280/126
Major depressionPsychiatric disorders0/1281/126

Baseline characteristics

Limited sociodemographic characteristics were collected from caregivers at baseline.

Age, Continuous
Age, Continuous(Years)Youth TI-CBT Intervention ArmYouth Discussion Control ArmCaregiver TI-CBT Intervention ArmCaregiver Discussion Control ArmTotal
Mean17.3 ± 1.317.3 ± 1.442.6 ± 11.541.9 ± 11.129.1 ± 14.7
Sex: Female, Male
Sex: Female, Male(Participants)Youth TI-CBT Intervention ArmYouth Discussion Control ArmCaregiver TI-CBT Intervention ArmCaregiver Discussion Control ArmTotal
Female727210495343
Male56541219141
Race (NIH/OMB)
Race (NIH/OMB)(Participants)Youth TI-CBT Intervention ArmYouth Discussion Control ArmCaregiver TI-CBT Intervention ArmCaregiver Discussion Control ArmTotal
American Indian or Alaska Native00000
Asian00000
Native Hawaiian or Other Pacific Islander00000
Black or African American128126116114484
White00000
More than one race00000
Unknown or Not Reported00000
Region of Enrollment
Region of Enrollment(Participants)Youth TI-CBT Intervention ArmYouth Discussion Control ArmCaregiver TI-CBT Intervention ArmCaregiver Discussion Control ArmTotal
Zimbabwe60585958235
Malawi32343128125
Botswana2624212091
South Africa10105833
Group-level mean age
Group-level mean age(Years)Youth TI-CBT Intervention ArmYouth Discussion Control ArmCaregiver TI-CBT Intervention ArmCaregiver Discussion Control ArmTotal
Mean17.3 ± 0.517.3 ± 0.643 ± 4.141.7 ± 4.429.8 ± 13.0
Highest level of education attained
Highest level of education attained(Participants)Youth TI-CBT Intervention ArmYouth Discussion Control ArmCaregiver TI-CBT Intervention ArmCaregiver Discussion Control ArmTotal
Some primary44443928155
Some secondary81786875302
Vocational225312
University124815
Enrolled in school
Enrolled in school(Participants)Youth TI-CBT Intervention ArmYouth Discussion Control ArmCaregiver TI-CBT Intervention ArmCaregiver Discussion Control ArmTotal
Count of participants2017——37
Working outside of home
Working outside of home(Participants)Youth TI-CBT Intervention ArmYouth Discussion Control ArmCaregiver TI-CBT Intervention ArmCaregiver Discussion Control ArmTotal
Count of participants16184744125

22 further baseline measures are reported on the registry.

07

Study locations

8 sites
  • Gaborone Prevention/Treatment Trials CRS 12701
    Gaborone, Botswana
  • Molepolole Prevention/Treatment Trials CRS 12702
    Molepolole, Botswana
  • College of Medicine CRS 30301
    Blantyre, Malawi
  • University of North Carolina Lilongwe CRS 12001
    Lilongwe, Malawi
  • Soweto IMPAACT CRS 8052
    Soweto, South Africa
  • St. Mary's CRS 30303
    Chitungwiza, Zimbabwe
  • Harare Family Care CRS 31890
    Harare, Zimbabwe
  • Seke North CRS 30306
    Harare, Zimbabwe
08

References and documents

Publications

  • Dow DE, Libous J, Ponatshego PL, Masheto G, Mvalo T, Payesa C, Buckley J, Stranix-Chibanda L, Nematadzira T, Vhembo T, Ace A, Warshaw M, Townley E, Lee S, Allison S, Donenberg G and the IMPAACT 2016 Team. High levels of mental distress among youth with HIV in Southern Africa: A description of enrollment into IMPAACT 2016, a randomized controlled trial of a peer-led, group-based mental health intervention. Presented at IAS, Kigali, Rwanda, 15 July 2025.
  • Libous J, Jeffrey R, George K, Montañez N, Warshaw M, Shakes Raesi M, Ponatshego PL, Maliwichi L, Vhembo T, Mutasa T, Bere CT, Galvin L, Townley E, Varechtchouk O, Dow D, Donenberg G, and the IMPAACT 2016 Team. IMPAACT 2016: The ADAPT-ITT Framework Leveraging Youth and Supporter Voices in the Adaptation of a Trauma-Informed Cognitive Behavioral Therapy Group-Based Intervention for Youth Living with HIV and Mental Health Distress. Presented at Fast Track Cities, Paris, France, 13-15 October 2024.
  • Vhemo T, Mhembere T, Mbengeranwa T, Bere TC, Mutasa T, Katsande R, Gumbo J, Mudzonga N, Nematadzira T, Maturure SJ, Mandima PF, Maonera S, Stranix-Chibanda L. High Prevalence of Common Mental Health Disorders Amongst Adolescents With HIV in Zimbabwe: IMPAACT 2016; Presented at International Workshop on Pediatrics & HIV, Munich, Germany, 19-20 July 2024.
  • Libous JL, Montañez N, Donenberg G, Kapetanovic S, Dow D. IMPAACT 2016: Operationalizing the ADAPT-ITT Model to drive local adaptation of an intervention to improve mental health and medication adherence in youth living with HIV. Presented at Society for Clinical Research Associates 29th Annual Conference, Virtual, 23-26 September 2020.
  • Warshaw M, Dow D, Kapetanovic S, Libous J, Reding C, Montañez , Donenberg G. IMPAACT 2016: Interdisciplinary Collaboration at Multiple Levels. Society for Clinical Trials 41st Annual Meeting, Baltimore, MD (Virtual), 2020.
  • Libous JL, Montanez NA, Dow DE, Kapetanovic S, Buckley J, Kakhu TJ, Kamthunzi P, Maliwichi LA, Vhembo T, Chawana TD, Nematadzira T, Donenberg GR. IMPAACT 2016: Operationalizing HIV Intervention Adaptations to Inform the Science and Outcomes of Implementation. Front Reprod Health. 2021 May 28;3:662912. doi: 10.3389/frph.2021.662912. eCollection 2021. PubMed 36303985 ↗

Study documents

  • Protocol and informed consent form · Jan 13, 2023
  • Statistical analysis plan · Apr 28, 2025

Documents are hosted by the registry — open the source record to download them.

Individual participant data

Plan to share: Yes — Individual participant data that underlie results in the publication, after deidentification.

Supporting information: Study protocol, Sap

09

Registry details

Key details

Study ID
NCT04024488
Lead sponsor
International Maternal Pediatric Adolescent AIDS Clinical Trials Group
Collaborators
National Institute of Allergy and Infectious Diseases (NIAID), Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD), National Institute of Mental Health (NIMH)
Responsible party
Sponsor
First posted
Jul 18, 2019
Start date
Apr 6, 2024
Primary completion
Mar 27, 2025
Completion
Jun 5, 2025
Results posted
May 22, 2026
Last update
May 22, 2026

Study contacts

Dorothy Dow, MD, MSc
study chair · Duke University

Oversight

Data monitoring committee
Yes
FDA-regulated drug
No
FDA-regulated device
No
View the source record on ClinicalTrials.gov ↗

Not currently enrolling

This study is terminated, as verified in Apr 2026. You cannot join it, but the record below documents what was studied.

Follow this study

Get an email when the registry record changes — status, dates, results — or when someone posts here.

Sign in to follow

Discussion

Questions and observations about this study, from anyone following it. Not medical advice, and not a channel to the study team — their contact details are on the registry record.

Sign in to join the discussion. Reading takes no account; posting does. You choose a display name, and a pseudonym is the default.

Nothing here yet. If you are running this trial, taking part in it, or weighing whether to, this is the place to say so.

Start the discussion