An evaluation of a psychosocial intervention for orphans on HIV treatment: A phase II RCT of memory work therapy at PASADA, Tanzania

Master Thesis


Permanent link to this Item
Journal Title
Link to Journal
Journal ISSN
Volume Title

University of Cape Town

Aim: Evidence shows high burden of psychosocial problems among orphaned children with HIV. Intervention studies have generally been conducted in developed countries. This mixed methods study aimed to determine whether Memory Work (MWT) Therapy 1 week group intervention improves outcomes for this complex population, in terms of their self-esteem, coping and psychological wellbeing compared to standard care, and to establish mechanisms of action. Method/Issue: An RCT Phase II of 48 Adolescents aged 14-18, full-orphaned AIDS on ART in Tanzania, with a waitlist control. Following (T0) baseline interviews, random allocation was for either intervention experimental condition (EC), or Control Condition (CC). All participated in second round of data collection (T1) at two weeks, and a final (T2) a month later. The EC children and caregivers participated in a focus group at T2. The CC then received the intervention 4 months later. Four questionnaires used, Brief Symptom Inventory (BSI), Rosenberg Self Esteem Scale (SES), Strengths and Difficulties Questionnaire (SDQ), and Self Efficacy Questionnaire (SEQ). Groups were compared at each time-point using multivariable linear regression controlling for baseline characteristics and group as independent variable, with change score as dependent variable. Qualitative data were subjected to thematic analysis to describe the experience of the intervention and how to refine it. Results/Comments: Significant differences favoured the EC at T1: BSI P<0.001, B=44.985, CI=25.5, 64.4), SDQ P=0.010, B=4.811, % CI 1.226, 8.396), SEQ both social scale P=0.015, B=-4.539, % CI-8.161,-0.918) and emotional scale P=0.002, B=-5.803 % CI-9.434, 2.171) (no effect for SES). At T2 we found these effects persisted: BSI P=0.001, B=46.668 %CI=21.541, 71.835), SDQ P=0.002, B=5.218 % CI 1.960, 8.476), SEQ both social scale P=<0.001, B=-7.791, % CI-11.320,-4.262) and emotional scale (P=<0.001, B=-9.007, % CI-12.032,-5.983). Additionally, at T2 found an EC effect for SES P<0.001, B=-4.392, %CI=-6.738,-2.046). The children recalled and described enjoying specific tasks within the intervention memory book, (such as "hero book" and the "tree of life") and described becoming "resilient" people. Discussion: The trial demonstrates improved outcomes in all measures for a complex population, i.e., bereaved children on treatment. Importantly, the effect did not attenuate and indeed self-esteem improved over time. The intervention should now be replicated in a fully powered trial.

Includes bibliographical references