Scale-up 'education vaccine' in Zambia, concludes HIV prevalence study

A study funded by the Norwegian government and undertaken in Zambia shows a significant reversal of the trends seen previously between higher educational attainment and increased prevalence of HIV infection. The study, published in the April 24^th edition of the journal, AIDS, suggests that HIV prevention campaigns are now reaching higher-educated Zambians in both urban and rural centres, but have not yet reached lower-educated populations. The authors argue that their data suggest current policies need to be reshaped from providing not just universal basic education but rather universal secondary education.

Earlier in the sub-Saharan African HIV/AIDS epidemic, a number of studies found an association between higher educational attainment, higher socio-economic status and travel with a greater risk of being HIV-infected. However, more recent studies have reported a reduced HIV prevalence in higher-educated women undergoing antenatal HIV tests.

It has been argued that educational levels greatly affect HIV knowledge, sexual risk behaviour and behavioural change: the so-called 'education vaccine'. Therefore, understanding the dynamics behind HIV prevalence amongst individuals with differences in educational attainment can help inform the past and future success of health promotion messages.

In order to investigate the change over time in the association between educational attainment and HIV infection in Zambia's general population, researchers from the University of Zambia compared data from population-based surveys conducted during 1995 and 1999 in the rural town of Kapiri Mposhi and in Chelstone, a district of Zambia's capital, Lusaka, to those conducted in 2003.

The surveys were undertaken in clusters of households that were calculated to be representative of the general population, consisted of questions regarding education, sociodemographic characteristics and risk behaviours, and were aimed at all household members aged between 15 and 59.

An anonymous saliva sample was obtained after informed consent, and individuals who wanted to know their HIV status were offered voluntary counselling and testing at home or at a clinic, and an additional blood sample was taken. Full information and saliva samples were obtained from 4442 individuals in 2003, compared with 2989 in 1995 and 3506 in 1999.

The investigators found a universal shift towards reduced HIV prevalence between 1995 and 2003 amongst respondents with higher education. In sharp contrast, HIV prevalence remained stable or even increased amongst respondents with lower education.

For example, urban men aged between 15-49 with had eleven or more years of schooling were half as likely to be HIV-infected than their urban counterpart with seven or fewer years of schooling (Odds Ratio, 0.45; 95% CI, 0.27-0.74). SImilarly, urban women with more than eleven years of schooling were a third less likely to be HIV-infected than their urban counterparts with seven or fewer years of schooling (OR, 0.65: 95% 0.42-0.98).

The odds of being HIV-infected were even lower amongst young urban individuals - those aged between 15 and 24 - who had attended school for more than eleven years. The investigators point out that this younger age group could be seen as marker of HIV incidence, since mortality is low at this age, and HIV infection is likely to have occurred recently.

In fact, HIV prevalence amongst higher educated urban men in 1995 and 1999 was 10.9% and 6.9%, respectively. In contrast, HIV prevalence in this group of individuals in 2003 was 1.4%. This results in a fivefold reduced risk compared to young urban men with less than seven years' education (OR, 0.20; 95% CI, 0.05-.073).

HIV prevalence remains much greater amongst young urban women compared to young urban men, however. Nevertheless, it was still markedly reduced amongst higher educated urban women in 2003 compared with previous years: 21.2% in 1995, 16.1% in 1995 and 8.5% in 2003. This resulted in a threefold reduced risk compared to young urban women with less than seven years' education (OR, 0.33; 95% CI, 0.15-0.72).

In rural Kapiri Mposhi, although there was a significant reduction in prevalence in 2003 amongst higher educated (defined here as greater than eight years of schooling) young men - from 8.9% in 1995 to 3.2% in 2003 - and young women - from 28.1% in 1995 to 5.6% in 2003 - there was only a trend towards a reduced odds of HIV infection amongst the higher educated young women compared with their lower educated counterparts (those with four or fewer years of schooling): OR, 0.77; 95% CI, 0.28-2.1. And young rural men with more than eight years of schooling were actually 28% more likely to be HIV-infected than their lower educated counterparts (OR 1.28; 95% CI, 0.29-5-59).

Nevertheless, these odds were still much reduced compared with previous years, and fit in with the investigators' assertion that HIV prevalence is generally now lower in those with higher educational attainment.

The investigators argue that the main common feature in the reduction of HIV prevalence seen across the genders is higher educational level. Although they do not prove that there is a direct causal link between higher education and reduced HIV prevalence, they suggest that "this is likely to be exerted through mediator factors such as more consistent condom use, lower likelihood of sexually transmitted infections and less number of sexual partners." However, the investigators did address the links between education and these mediator factors in their surveys, and will report in detail on sexual behaviour patterns at a later date.

Although women have been most affected historically, the investigators note that the reduction in HIV prevalence seen in young higher educated women suggests "policies aimed at increasing the proportion of women enrolling in school, and encouraging them to stay on at school, may be bearing fruit and need to be strengthened."

The best way to address hard-to-reach lower educated and predominantly rural populations, argue the investigators, is via better education which "will probably arm the young population with a better requisite tool for behaviour change." They conclude by calling for a change in policy focus from universal basic education to universal secondary education due to the protective effect of higher education.