Recall of early non-fatal suicidality in a nationally representative sample of South Africans

Objectives. Little is known about socio-demographic patterns of non-fatal suicidality in early life in South Africa. We investigated the prevalence of self-reported early suicidality (suicidal ideation, planning and attempts) in a nationally representative sample of South Africans. Design. As part of a larger mental health survey, 3158 individuals aged over 25 years were asked to recall whether they engaged in non-fatal suicidal behaviour in early life (measured from childhood to 25 years). Race-based discrimination institutionalised under Apartheid profoundly influenced delivery and outcomes in health and other social services. Racial categories entrenched during Apartheid were therefore used to analyse data collected from individuals born before 1946, 1947-1956, 1957-1966 and 1967-1976. Results. 3.4% (95% CI = 2.6-4.1) of participants recalled early suicidal behaviour. The youngest group (born 1967-1976) recalled higher rates of early suicidality than older groups in all races. In unadjusted analysis, White people were 2.84 (95% CI = 1.62-4.97) and Coloured people 1.84 (95% CI = 1.15-2.93) times more likely than Black people to recall early suicidality. Individuals growing up in urban and higher socio-economic settings were approximately twice (OR = 2.2; 95% CI = 1.14-4.28 and OR = 1.92; 95% CI = 1.27-2.90) as likely to recall early suicidality as those growing up in rural and lower socio-economic settings. Those with post-primary education were 2.79 (CI = 1.71-4.53) times as likely to recall early suicidality as those with no or only primary education. Racial differences ceased to be significant after adjustment for rural/urban location and other socioeconomic measures estimated for early life. Conclusion. The study provides novel evidence of increasing levels of early non-fatal suicidality recalled by younger South Africans. Levels appeared significantly higher in Whites than in Blacks. Socio-economic contexts in early life were interpreted as mediators rather than confounders of the association between race and recalled early suicidality. The findings for decreased levels of suicidality among participants growing up in lower socio-economic strata and rural settings in South Africa require further investigation. The need for widespread suicide prevention programmes targeting young people at a population level is emphasised.