In generalized HIV epidemic settings, anonymous testing of pregnant women attending public sector antenatal clinic (ANC) remain the mainstay for monitoring epidemic trends and are important for measuring the magnitude of the epidemic in the general population [1]. The Joint United Nations Programme on HIV/AIDS (UNAIDS) recommends the importance of understanding the HIV epidemic locally and customizing responses accordingly [2]. As part of the ongoing research conducted by the Centre for the AIDS Programme of Research in South Africa (CAPRISA) in the sub-district of Vulindlela we have monitored the trends in HIV prevalence in this rural community through the annual ANC HIV seroprevalence surveys in the seven primary health care clinics [3]. These surveys have coincided with the national department of health ANC surveys [4] and have revealed a disturbing increase in HIV prevalence among young women below 20 years of age. The overall HIV prevalence in this age group increased from 16.6% in 2006 to 20.8% in 2008 and the HIV prevalence in this age group serves as a reliable proxy marker for incident HIV infections [1, 3, 4].

Whilst South Africa is one of few countries that has a supportive legal framework to enable young children to access sexual reproductive health services autonomously from age 12 [5], there are many ethical and programmatic challenges to HIV testing of young children for care and support or for surveillance. In recognition of the fact that children are sexually active at a very young age the risk of HIV acquisition increases through complex individual behaviours and through sexual networks within a broader context of political, economic and social powers. To understand the evolving HIV epidemic in this rural setting and to establish whether the prevalence in young pregnant girls is similar to that in other non-pregnant young people in this age group we undertook an anonymous survey in high school learners in Vulindlela, KwaZulu-Natal

The HIV prevalence among high-school learners in this rural district is concerning and underscores the importance of targeted HIV risk reduction and sexual reproductive health service efforts in high schools. It is also important to better understand why these young girls are having sex, with whom and when they are having sex to better inform efforts to reduce their HIV risk. Early sexual debut is associated with higher teenage pregnancy rates, sexually transmitted infections and poor school completion rates leading to poor health and economic outcomes in young women [8–12].

South Africa is one of few countries that has a supportive legal framework to enable young children to access sexual reproductive health services autonomously from age 12 [5]. The Children's Act of 2005 (Act No. 38 of 2005) which came into effect in 2007 explicitly allows children 12 years and older access to contraceptives, information on sexuality and reproduction, and the right of consent to HIV/AIDS testing and treatment. This is in recognition of the fact that children are sexually active at a very young age and that there is a high burden of HIV infection in young people [13], despite the legal age of consent for sex being 16 years.

The high HIV prevalence among girls identified through this study suggests that these infections are likely to remain undiagnosed, sustaining the networks for HIV transmission. In spite of high levels of awareness on the benefits of HIV testing, fear of HIV-related stigma from parents, caregivers and teachers could be barriers preventing learners from being tested [14–16]. In high HIV prevalence settings, there is a need for more comprehensive information and support on HIV-related issues, including the importance of prevention.

The large gender disparity in the sample surveyed showed that the HIV prevalence was substantially lower in young boys compared to young girls. Young boys were significantly less likely to be HIV infected than their female counterparts; however, the prevalence in older boys was high. The patterns of infection observed in young girls are similar to trends observed throughout southern Africa, although these are restricted to age groups of 15 years and older [6, 9, 17]. Several large studies focusing on knowledge of HIV and behavioural risk factors for sexually transmitted infections and HIV have been conducted in the absence in biological measurements. These studies have demonstrated increasing risk taking behaviour predisposing to infections [18, 19]. As early adolescent's marks a critical time of physical, developmental and social changes, interventions must focus on the needs of young people from as early as 10 years of age. These interventions need to be effective in shaping healthy attitudes and behaviours while most learners are still at school and less likely to have begun sexual activity.

As young people in this country are becoming sexually active at younger ages, it is important that they are taught that regular HIV testing should be an important part of their routine healthcare. While many social barriers are expected, by engaging each other openly, honestly and directly, young people would have an opportunity to impact on sexual health choices. The HIV and STI risk reduction programmes at schools are crucial in moulding and developing young people's identities and characters to enhance self esteem and thereby reduce or delay early sexual debut. Furthermore these choices would enable young people to reduce their anxiety, fear and stigma associated with HIV testing especially as they move into adulthood and the risk of infection increases, even before they are sexually active.

These results demonstrate the growing need and the opportunities for HIV testing and sexual reproductive health services within the school setting. South Africa has a high teenage pregnancy rate and in 2003 an estimated 12% of teenage girls between the ages of 15 and 19 years had ever been pregnant or were pregnant at the time of the survey [20]. The most recent reports show that the learner pregnancy rate in KwaZulu-Natal was 62.2 per 1000 [21]. Despite a decade of increased spending on sex education and HIV/AIDS awareness campaigns, there has been little impact on pregnancy rates and HIV incidence.

Learners as young as 12 years were included in this study because early age of sexual debut is considered a crucial factor affecting the vulnerability of young people to HIV infection. Several studies have found that sexual debut before the age of 15 years to be approximately 10% for both boys and girls [22–25]. This figure increases substantially following experimentation with alcohol, substance abuse, pressure from mixing with older peer groups, coercion or sexual abuse. Young people initiating sex at an early age has major implications for HIV and STI infection and associated with higher HIV exposure due to it being linked to more frequent sexual intercourse, more lifetime sexually transmitted infections, less consistent contraceptive use, and more sexual partners [26].

The National Minister of Health of South Africa recently announced the prioritization of schools for the HIV counselling and testing (HCT) campaigns [27]. The close partnerships between the government departments and the community that will be necessary to successfully implement this campaign will go a long way towards addressing the challenges of obtaining consent, maintaining confidentiality, and providing the psychological and emotional support for young people to deal with an HIV-positive diagnosis. The magnitude of the epidemic demands more HIV risk reduction efforts in schools with evidence based programs.

There are several limitations to our study. The small sample size limits representativeness and generalizability to the larger school population in the district. Furthermore, the schools learner populations may differ by schools and the geographic variability in the prevalence of HIV infection is likely to be different across the district. Nevertheless this study is the first to report on the prevalence of HIV in high school learners in Vulindlela. These learners represent a population of young people with emerging HIV epidemic. Therefore, surveillance is a necessary tool to understand the transmission dynamics and the evolving epidemic in this vulnerable group. Another limitation was the absence of behavioural data collection which limited our understanding of risk behaviours and on the modes of HIV transmission. Despite these limitations, our findings indicate that additional larger studies are required to determine the full meaning of the alarming HIV prevalence in the sample of young women.

Knowledge of HIV status provides an important gateway to prevention and care efforts and the HCT campaign in schools could provide a narrow window of opportunity to reduce HIV risk in young girls and stem the tide of HIV for this and future generations.