The impact of poverty-reduction intervention on child mental health mediated by family relations: Findings from a cluster-randomized trial in Uganda

Abstract

Reviews that synthesize global evidence on the impact of poverty reduction interventions on child and adolescent mental health (CAMH) report inconclusive results and highlight the need to unpack the mechanisms that connect poverty-reduction to CAMH. To address this gap, we examine the proposition that family relations is an important relational factor transmitting effect of poverty on CAMH, and test whether family relations mediate the effect of poverty-reduction intervention on depression, hopelessness, and self-concept among AIDS orphans in Uganda. We use longitudinal data collected over the course of 48 months in a cluster-randomized controlled trial conducted among N = 1410 AIDS orphans from n = 48 schools in Uganda. To examine the relationship between intervention, latent mediator (family relations and support) and CAMH outcomes (Beck Hopelessness Scale (BHS), Tennessee Self-Concept Scale (TSCS), and Depression), we ran structural equation models adjusting for clustering of individuals within schools. Relative to the control group, participants in both treatment arms reported lower levels of hopelessness and depression, and significantly higher levels of self-concept. They also report significantly higher levels of latent family relationship in all three models. In both treatment arms, the direct effect of the intervention on all three outcomes is still significant when the latent family relations mediator is included in the analyses. This suggests partial mediation. In other words, in both treatment arms, the significant positive effect of the intervention on children's depression, hopelessness, and self-concept is partially mediated by their family relationship quality. Our findings support the argument put forward by the Family Stress Model showing that the poverty-reduction program improves children's mental health functioning by improving family relationships. The implications of our study extend beyond the narrow focus of poverty reduction, suggesting that asset-building interventions have broader impacts on family dynamics and child mental health.

Keywords: AIDS orphans; Child and adolescent mental health; Cluster-randomized controlled trial; Mediation models; Pathways from poverty to child and adolescent mental health; Poverty-reduction interventions; Structural equation models (SEM); Sub-Saharan Africa.