Effects of Adolescent-Focused Integrated Social Protection on Depression: A Pragmatic Cluster-Randomized Controlled Trial of Tanzania's Cash Plus Intervention. Issue 9 (17th May 2022)

• Record Type: Journal Article
• Authors: Prencipe, Leah; Houweling, Tanja A J; van Lenthe, Frank J; Kajula, Lusajo; Palermo, Tia
• Is Part Of: American journal of epidemiology. Volume 191:Issue 9(2022)
• Page Start: 1601
• Page End: 1613
• Publication Date: 2022-05-17
• Subjects: adolescent mental health -- Cash Plus -- depression -- integrated interventions -- social protection -- Tanzania; Epidemiology -- Periodicals; Public health -- Periodicals; 614.4
• DOI: 10.1093/aje/kwac093
• Languages: English
• ISSNs: 0002-9262

Abstract

Abstract: We assessed the impacts of Tanzania's adolescent-focused Cash Plus intervention on depression. In this pragmatic cluster-randomized controlled trial, 130 villages were randomly allocated to an intervention or control arm (1:1). Youth aged 14–19 years living in households receiving governmental cash transfers were invited to participate. The intervention included an intensive period (a 12-session course) and an aftercare period (9 months of mentoring, productive grants, and strengthened health services). We examined intervention impacts on a depressive symptoms scale (10-item Center for Epidemiologic Studies Depression Scale score (range, 0–30)) and rates of depressive symptomatology (score ≥10 points on the scale), recorded at study baseline (April–June 2017), midline (May–July 2018), and endline (June–August 2019). Using intention-to-treat methodology, we employed logistic and generalized linear models to estimate effects for binary and continuous outcomes, respectively. Quantile regression was used to estimate effects across the scale. From 2, 458 baseline participants, 941 intervention and 992 control adolescents were reinterviewed at both follow-ups. At endline, the intervention reduced the odds of depressive symptomatology (adjusted odds ratio = 0.67, 95% confidence interval: 0.52, 0.86), with an undetectable mean scale difference (risk difference = −0.36, 95% confidence interval: –0.84, 0.11). Quantile regression results demonstrated an intervention effect along the upper distribution of the scale. Integration of multisectoral initiatives within existing social protection systems shows potential to improve mental health among youth in low-resource settings.