Abstract
In South Africa (SA), depression is common in adolescents and has numerous immediate and long-term
negative consequences. High exposure to childhood adversity contribute to the mental health burden among
youth in SA. Cash transfer (CT) programs to alleviate poverty and its sequela are an essential part of
government social protection programs reaching millions of households in the African region. While there is a
growing body of evidence that CTs are associated with better mental health outcomes in adults, there is less
evidence as to the impacts among adolescents who have experienced childhood adversities. The overall
objective of this proposal is to determine whether, how, and under what conditions cash transfers
work to reduce depression among youth exposed to different adverse childhood experiences and
identify potential combination interventions that could work synergistically to promote health and well-
being among South African adolescent girls and young women (AGYW) with a high burden of
childhood adversity and depression. Using extant data from a rich longitudinal cohort combined with
community data sources, we propose a unique set of analyses to understand how CTs work to reduce
depression for adversity-exposed youth and improve later social outcomes. In Aim 1, we will identify whether
the longitudinal association between childhood adversities and depression differs depending on CT exposure.
We will first use data from a randomized conditional CT (CCT) trial conducted among AGYW in rural SA that
followed a cohort of 2,533 AGYW (HPTN 068, ages 13-20 at baseline) from 2011-2019. Depression was
measured six times over 8 years capturing a critical developmental period across adolescence into early
adulthood. Childhood adversities were assessed through community and family violence exposures, poverty
indicators, and low parental engagement. Using data from the Agincourt Household Socio-Demographic
Surveillance (AHDSS) in which the girls’ cohort is nested, we will also determine the impact of the government
of South Africa’s Child Support Grant (CSG), a monthly unconditional CT (UCT) provided to poor households
provided over time girls in the 068 cohort. In Aim 2, we will examine the processes linking adversities to
depression, including mediation through cognitive, emotional, and stress processes, and use moderated
mediation analysis to determine how CT exposure moderates these pathways. Finally, in Aims 3 and 4 we will
use causal inference models to explore how combining CTs with interventions on mediators of the adversity-
depression pathways could boost promotive impacts on depression and explore how reductions in depression
could reduce longer term social outcomes such as teen pregnancy, intimate partner violence, educational
attainment, and financial well-being. Taken together, we anticipate that findings from this robust set of analyses
will inform our understanding of how CTs may be leveraged to improve adolescent mental health.
