Young people bear the burden of South Africa’s severe HIV epidemic, with 10.2% of 15–24 year olds infected with HIV nationally.¹ In response, recent research in South Africa has addressed sexual risk behaviours among young people, particularly women.^2–5 A significant gap in the literature exists, however, with regard to young men: their needs and concerns have received relatively scant attention.⁶

Age at first sex is an important indicator of sexual risk, as it marks the onset of exposure to infection, is a risk factor for HIV infection,⁷ and is a key indicator for monitoring response to the HIV epidemic among youths. However, there is little quantitative information about men’s experiences of sexual debut in sub-Saharan Africa, except a few national surveys that included men.⁸

This report is based on analysis of sexual behaviour data from a cross sectional household survey in rural KwaZulu/Natal, South Africa. We examine young men’s experiences of sexual debut, and the association of early sexual debut with involvement in multiple partnerships.

METHODS

A household census of adults aged 15–49 (n = 2309) was conducted in a rural subdistrict of KwaZulu/Natal province, a severely HIV affected area with prevalence of 14.1% in 15–24 year olds in 2003.¹ Annual per capita income is under $US 1200. The survey included 314 young men aged 15–24, of whom 62.7% were sexually active.

Bivariate associations between age at sexual debut and first sexual experiences were examined. Age at sexual debut was then included in the multivariate analysis to examine associations with the main outcome: ⩾3 partners in the past 3 years. A logistic regression model examined relations between independent and dependent factors after adjustment for all other covariates. Multivariate analysis of the main outcome was restricted to those with sexual debut at least 3 years before the survey (n = 122), as the recently sexually experienced would have lacked sufficient time for partner accumulation. Explanatory variables were limited to more static measures, since sexual debut was a past event. However, some current status variables (education, media exposure, familial circumstances) were included, as relatively recent measures in the 15–24 age group. Main measures were defined as: (1) sexual debut: “at what age did you start your first sexual relationship?” and (2) ⩾3 partners in past 3 years: “how many sexual partners have you had within past three years?” selected as an appropriate period in which to measure multiple partnerships in this age group.

Explanatory factors came from a household schedule administered with the survey. Household wealth was measured from an asset index. Parental residence was defined as living with both, one or no parents, while community participation included membership in one or more of sports, church, singing, dance, or study groups.

Ethical approval to conduct this research was obtained from the ethics review committees of the Faculty of Medicine, University of Natal, Durban, South Africa and the University of London (London School of Hygiene and Tropical Medicine), London, UK.

RESULTS

In all, 13.1% of men 15–24 years of age reported their first sexual relationship before age 15 (range 9–14 years).

Significant associations were found between age at sexual debut and key measures of first sexual experience (table 1). The proportion of men reporting they had been ready and wanted to have sex at that time increased with age at first sex. The proportion that did not know their partner well decreased as age at first sex rose, as did the proportion that had been in love at the time of sexual debut. No respondents indicated they had experienced coercion or had a stranger as their first partner.

Significant differences were observed in condom use at first sex between men who initiated sex at ⩾age 15 compared to those who initiated earlier (19% v 9%, p<0.05), 35.2% reported ⩾3 partners in the past 3 years, while 42.3% and 23.5% reported one and two partners respectively. Only 11.2% reported ⩾5 partners in the past 3 years. In multivariate analysis (table 2), earlier sexual debut was the only significant predictor of ⩾3 partners in the past 3 years (OR = 10.26, p<0.01).

DISCUSSION

Young men in rural South Africa with early sexual debut exhibit a constellation of risk factors for HIV infection, and also later sexual risk.

Young men reporting sexual debut before age 15 were more likely not to use condoms at first sex, to have had multiple and casual partners, and to say they “had not been ready and wanted to have sex.” Young men with early sexual debut were 10 times more likely to have had multiple partners, even after controlling for duration of sexual activity, reinforcing evidence that early sexual experiences may determine sexual risk behaviours throughout the life course.⁹ Importantly, risk behaviours, once established, may be difficult to change.

Yet in-depth interpretation of these findings is difficult. Being “in love” and having been “ready and wanted to have sex at that time” are not mutually exclusive, nor is it clear how these are associated with other risk behaviours. Qualitative findings^6,10 indicate reasons that young men initiate sexual activity: peer pressure, asserting masculinity, relations with older women, and sexual abuse, which is likely under-reported. Although only 3.5% of young men reported sexual debut ⩽age 12, other South African studies indicate high levels of childhood sexual abuse.¹¹

Limitations to this analysis include possible bias in reporting of sexual behaviours, exaggeration of sexual activity and numbers of partners,¹² and the 3 year length of the recall period for number of partners. Further, the definition of “sexual debut” as “age at sexual first sexual relationship” might have neglected some sexual activity outside established relationships.

This study is exploratory, raising questions about the dynamics of young men’s relationships, vulnerability, and sexual risk. What are the pathways to risk among young men with early sexual debut? Are they indeed vulnerable, having had unwanted sexual activity at an early age, and thus forfeiting opportunities to develop a healthy sexuality including protection from risk? Or are they a subgroup of sexually confident young men whose high risk status is reflected in the later accumulation of multiple partners?

Whatever the reasons, these findings highlight a troubling lack of preparation for sexual activity among young men in this setting. Further research should be directed toward better understanding young men’s early relationships and sexual experiences, and to the design of interventions for pre-adolescent young men, before sexual debut. Interventions for young men should emphasise condom promotion, as well as delay in sexual activity.