HIV risk growing among sub-Saharan Africa's least educated

HIV is becoming most prevalent among sub-Saharan Africans at the lowest educational levels, according to a systematic review of published studies. Most data from before1996 indicated that HIV prevalence was either unrelated to education, or higher among more the highly educated. Since 1996, prevalence has been falling among the most educated while rising in the least educated. The review was published in January 30^th edition of AIDS.

James Hargreaves and colleagues at the London School of Hygiene and Tropical Medicine have previously investigated the association between educational level and risk of HIV infection in developing countries. In 2002, they published a systematic review of published studies, predominantly based on epidemiologic data up until 1996. The studies analysed at that point found (with a single exception) that there was either no relation between education and HIV infection risk, or that people at higher educational levels were more likely to be infected.

This earlier Hargreaves study has now been updated to include data published up to July 2006. Articles were drawn from a comprehensive search of Pubmed and Embase and manual searches of the journals AIDS, the Journal of Acquired Immune Deficiency Syndromes, and the International Journal of STD & AIDS. Selected articles were those that:

• Reported original data from at least 300 individuals in groups "broadly representative of the general population" (i.e., not restricted to high-risk groups).
• Reported demographic information on educational attainment.
• Adjusted for at least age, sex, and urban/rural setting, but were not "over-adjusted", by the researchers' criteria, for sexual behaviour or STIs.

The complete analysis included a total of 36 published articles drawn from a review of roughly 1200 published papers and 4000 abstracts. The selected papers reported on 72 discrete populations from eleven sub-Saharan African countries between 1987 and 2003. Populations surveyed included systematic random samples of the general population, and samples of young people, army recruits, sugar estate residents, and prenatal clinics. This analysis was conducted as a "comprehensive review" rather than a formal meta-analysis, as there was no expectation of consistent findings across all the distinct populations studied.

In most of the published papers, educational attainment was grouped into two to five categories. In this review, comparisons were only made between the highest and lowest categories in each study: no attempt was made to find linear trends in the intermediate educational categories, or to further standardise the categories used across the individual studies. Each of the 36 published reports was classified as to whether it showed a statistically significant (p

In general, earlier data (prior to 1996) found a greater risk among the most educated, when a correlation between risk and education was found at all. However, the majority of data from 1996 onward showed that this trend was either weakening or actually reversing, with a higher risk of infection occurring among the least educated. Before 1996, higher risk was seen in the most educated in 15 of 32 populations, and in the least educated in only one. After 1996, higher risk was seen in the most educated in only 5 of 40 populations, and in the least educated in seven of 40. This trend was consistent across studies "conducted in different settings, among different population types, with different study designs and in countries where data were available from both time periods."

While further details of many population-specific findings are discussed in the report, the general trend was that, "across many settings, HIV prevalence fell more consistently among the higher educated than among the less educated groups, in whom prevalence sometimes rose even while overall population prevalence was falling." The paper suggests "that new HIV infections occurring in the latter half of the 1990s and into the 21st century have been occurring disproportionately among the least educated members of society in many sub-Saharan African countries."

Greater changes in risk behaviour among the most educated could explain this change – a suggestion that is supported by other research that has "consistently found higher levels of reported condom use among more educated individuals in a variety of contexts." The researchers believe there is "sufficient evidence to support consideration of policy responses to [this] changing epidemiology of HIV infection."