A Structural Equation Model of the Impact of a Family-Based Economic Intervention on Antiretroviral Therapy Adherence Among Adolescents Living With HIV in Uganda

Purpose: Antiretroviral therapy (ART) adherence among adolescents living with HIV (ALWHIV) is low, with poverty remaining a significant contributor. We examined the mediation pathways between an economic empowerment intervention and ART adherence among ALWHIV. Methods: This cluster-randomized controlled trial (2012–2018) recruited 702 ALWHIV aged 10–16 in Uganda between January 2014 and December 2015. We randomized 39 clinics into the control (n = 344) or intervention group (n = 358). The intervention comprised a child development account, four microenterprise workshops, and 12 mentorship sessions. We used six self-reported items to measure adherence at 24 months, 36 months, and 48 months. We used structural equation modeling to assess the mediation effects through mental health and adherence self-efficacy, on adherence. We ran models corresponding to the 24, 36, and 48 months of follow-up. Results: The mean age of the participants was 12 years, and 56% were female. At 36 (model 2) and 48 months (model 3), the intervention had a significant indirect effect on ART adherence [B = 0.069, β = 0.039 (95% confidence interval [CI]: 0.005–0.074)], and [B = 0.068, β = 0.040 (95% CI: 0.010–0.116)], respectively. In both models, there was a specific mediation effect through mental health [B = 0.070, β = 0.040 (95% CI: 0.007–0.063)], and [B = 0.039, β = 0.040 (95% CI: 0.020–0.117)]. Overall, 49.1%, 90.7%, and 36.8% of the total effects were mediated in models, 1, 2, and 3, respectively. Discussion: EE interventions improve adherence, by improving mental health functioning. These findings warrant the need to incorporate components that address mental health challenges in programs targeting poverty to improve ART adherence in low-income settings.