A cross sectional study to examine the association between perceived household educational support and HIV risk in adolescent females in a rural South African community

Introduction and background: South Africa continues to have the highest rates of HIV (prevalence and incidence) worldwide. Young women aged 15-24 appear to be at particular risk, with the rate of new infections in this age group estimated at 2.54% (over four times greater than that of men of the same age). Much research has focussed on trying to better understand the unique risk profile of young women: one such factor is the role of female education in reducing HIV risk. Increasingly evidence suggests that female education (specifically secondary school and improved school performance) reduce HIV risk, teenage pregnancy and risky sexual behaviours (encouraging condom use and delayed sexual debut). This study aims to build on this research considering household interest in education as an additional factor which might reduce HIV risk in adolescent females. Specifically this study proposes that household interest in education would lead to increased educational interest and success, reducing HIV risk, and in addition would build self-esteem and self-efficacy skills, enabling safer sexual practices and empowering young women in the negotiation of sexual relationships. This line of thinking also bridges a gap between research considering the role of female education in reducing sexual risk behaviours, and research focussing on the role of the home environment in promoting academic interest and success. Methods: This was a cross sectional study nested within the Swa Koteka Cash Transfer Trial located in Mpumalanga, South Africa. The study analyses baseline data collected in 2011 from 2533 young women aged 13-20 in grades 8-11. Biological outcomes of interest included HIV and HSV-2 status at baseline, and behavioural outcomes included having had vaginal sex, condom use at last sex and age of sexual debut. Multiple logistic regression was used to determine associations between both biological and behavioural outcomes and the different components of household educational support (as defined by supervision and assistance with homework, discussion of marks and studies, aspirations for further education and degree of disappointment at dropout). Ethical approval of this study was obtained from the UCT Human Research Ethics Committee, recognising approval for the Swa Koteka trial granted by the corresponding Wits committee. Results: HIV prevalence at baseline was 3.13%, increasing significantly with age. A similar pattern was noted in relation to HSV-2 prevalence, where baseline prevalence was 4.86%. 26.6% of young women reported having had vaginal sex: of those 59% reported having used a condom at last sexual act, with the median age of sexual debut being 16 years. Overall young women reported high rates of household educational support across all measures, though lower rates of parental involvement were noted in relation to assistance with homework and discussion of studies (possibly a marker of lower levels of parental education). No association was shown between household interest in education and either HIV or HSV-2 status. Supervision of homework, discussion of marks and greater educational aspirations were significant protective factors in reducing the risk of having had vaginal sex and delaying sexual debut, though not significantly associated with condom use. Greater disappointment at dropout acted as a significant risk factor for vaginal sex (OR=1.29 95%CI 1.14- 1.46, p<0.001) and early sexual debut (HR=1.15 95%CI 1.05-1.27, p=0.004), though again was not significantly associated with condom use. Greater self-efficacy was associated with greater likelihood of having had vaginal sex (OR= 1.14 95%CI 1.11-1.17 p<0.001), and earlier sexual debut (HR = 1.09 95%CI 1.07-1.12, P