A multilevel analysis of the relationship between depression and perceived neighborhood violence and safety: evidence from the South African National Income Dynamics Study

Andrew Tomita, University of KwaZulu-Natal
Charlotte A. Labys, University of KwaZulu-Natal
Jonathan K. Burns, University of KwaZulu-Natal

While South Africa undergoes the gradual economic and political transition from apartheid to non-racial democracy, the troubled history of violence persists in many neighborhoods, which we argue has had an enduring negative effect on mental health. However, the extent to which perceived neighborhood crime and safety issues are associated with depression in South Africa is unknown at the population-level. Data from the second wave of the South African National Income Dynamics Study (SA-NIDS) was used to investigate the aforementioned relationship. The SA-NIDS is the first longitudinal panel survey of a nationally representative sample of households in the country. Approximately 6,800 household were successfully interviewed, with nearly 18,400 adult residents aged 15 years old and older having successfully completed the interview. Depression outcome was assessed using the 10-item four-point Likert version of the Center for Epidemiologic Studies Depression Scale. The main covariate of our study, perceived neighborhood violence and safety, was assessed using 6 items, each scored with a five-point Likert format. A multilevel mixed-effects regression model adjusted for socio-demographic and health status factors was used to examine the association between perceived neighborhood crime/safety and depression. Adjusted regression analysis (n=12,799) indicated that perceptions of poor neighborhood violence and safety were associated with higher depression symptom scores (adjusted OR = 1.04, p<0.001). Female gender, older age, black ethnicity, lower educational attainment, urban informal residence, non-married, and poorer overall health status were also significantly associated with higher depression outcomes. The cross-sectional design of this study limits our understanding of temporality in relation to perceived neighborhood violence and safety and depression; further research utilizing a longitudinal study design is warranted. While individual factors remain relevant, the legacy of violence in South African neighborhoods remains an important social determinant of depression rates in a post-apartheid era.