Sexual violence against children in South Africa: a nationally representative cross-sectional study of prevalence and correlates. Issue 4 (April 2018)

• Record Type: Journal Article
• Authors: Ward, Catherine L; Artz, Lillian; Leoschut, Lezanne; Kassanjee, Reshma; Burton, Patrick
• Is Part Of: Lancet. Volume 6:Issue 4(2018)
• Page Start: e460
• Page End: e468
• Publication Date: 2018-04
• Subjects: World health -- Periodicals; 362.105
• DOI: 10.1016/S2214-109X(18)30060-3
• Languages: English
• ISSNs: 2214-109X

Abstract

Summary: Background: We aimed to complete a nationally representative study of sexual violence against children in South Africa, and its correlates, since we could identify no other such study. Methods: For this nationally representative, cross-sectional study in South Africa, households were selected by use of a multistage sampling frame, stratified by province, urban or rural setting, and race group, and schools were selected on the basis that they were closest to the area in which households were selected. Interviews and self-administered questionnaires in each location inquired into lifetime and last-year prevalence of sexual abuse, and its correlates among children aged 15–17 years, whose parents gave informed consent and they themselves gave informed assent. Findings: The final household sample was 5631 (94·6% participation rate). 9·99% (95% CI 8·65–11·47) of boys and 14·61% (95% CI 12·83–16·56) of girls reported some lifetime sexual victimisation. Physical abuse, emotional abuse, neglect, family violence, and other victimisations were all strongly associated with sexual victimisation. The following were associated with greater risk of sexual abuse (adjusted odds ratio [OR]); school enrolment (OR 2·12, 95% CI 1·29–3·48); rural dwelling (0·59; 0·43–0·80); having a flush toilet (1·43, 1·04–1·96); parental substance misuse (2·37, 1·67–3·36); being disabled (1·42, 1·10–1·82); female (but not male) caregivers' poor knowledge of the child's whereabouts, friends, and activities (1·07, 0·75–1·53) and poor quality of the relationship with the child (ie, poor acceptance; 1·20, 0·55–2·60). The child's own substance misuse (4·72, 3·73–5·98) and high-risk sexual behaviour (3·71, 2·99–4·61) were the behaviours most frequently associated with sexual abuse, with mental health conditions found to be less prevalent than these factors but still strongly associated with sexual victimisation (post-traumatic stress disorder 2·81, 1·65–4·78; depression 3·43, 2·26–5·19; anxiety 2·48, 1·61–3·81). Interpretation: Sexual violence is widespread among both girls and boys, and is associated with serious health problems. Associated factors require multisectoral responses to prevent sexual violence or mitigate consequences. Funding: UBS Optimus Foundation.