New infections in South Africa top 500,000 a year, 4.6% of young women infected in 2005

South Africa’s HIV epidemic is continuing unabated, epidemiologists reported at the Third South African AIDS Conference today in Durban. Over 500,000 new infections occurred in 2005, said Dr Olive Shisana, chief executive of the Human Sciences Research Council, with incidence highest in young women aged 15-24 years, at 4.6%.

However a rural study in northern Kwazulu Natal found 8% incidence in men and women aged 25-29%, and researchers from the University of Kwazulu Natal’s Centre for Health and Population Studies estimate that if current trends continue, two-thirds of 15 year olds in the district will have been infected with HIV by the time they reach the age of 35 (Barnighausen).

Similarly a study in high-risk women with multiple partners and women involved in commercial sex work found an incidence of 7.1 infections per 100 person years of follow-up in Kwazulu Natal (Loggerenberg).

HIV incidence in South Africa has been estimated by a variety of methods. The National HIV Prevalence household survey of 2005 used a BED assay, which measures the volume of antibodies that are being produced against a specific HIV protein. International studies have been able to pinpoint the average time it takes for antibody production to reach the maximum level, and also to determine what quantity of antibodies will be present at a given time point after infection. This allows researchers to pick up recent infections that have occurred within the past six months with more accuracy than some other methods.

The 2005 National HIV Prevalence household survey carried out anonymous antibody tests in 15,581 randomly selected individuals recruited to reflect population distribution in South Africa. All HIV-positive samples were tested using the BED assay to determine the incidence of recent infections (Rehle).

The study found an overall incidence of 2.4% in 15-49 year olds, and 2.2% in 15-24 year olds. However women aged 15-24 were more likely to have been infected recently, having an incidence of 4.6%.

The study also found that widowed individuals were at much higher risk of recent infection, with an incidence of 5.8%. Women who were currently pregnant were also at higher risk, having an incidence of 5.2%, and individuals aged 15-24 who had not used a condom on the last occasion of sexual intercourse had a similarly high incidence of 6.1%, compared to incidence of 2.9% in those who did report using a condom.

HIV incidence was highest among those living in informal periurban settlements (5.1%), compared to 0.8% incidence among residents of formal urban areas.

The incidence analysis also found a substantial number of new HIV infections in children above two years of age that cannot be attributed to mother- to-child transmission.

The northern Kwazulu Natal incidence study, carried out by Dr Till Barnighausen, used longitudinal sampling from a household cohort study that carried out two waves of testing, in 2003-2004, and again in 2005-2006. Incidence was calculated by comparing the numbers who tested positive in the first and second waves, although the proportion of local residents who agreed to anonymised HIV tests fell from 60% to 40% between the two waves, and this may have biased the results.

The study found an HIV prevalence of 22% in 2003-2004, which corresponds well with the prevalence found in Kwazulu Natal by the 2005 National Household Survey. However HIV prevalence reached 35% in both men and women who lived near the N2 highway, the main road artery travelling north through the province.

Incidence was calculated from 4040 individuals and 5246 person-years of follow-up. HIV incidence was 4% in women aged 15-19, 1% in men aged 15-19, 8% in women aged 24-29 and 8.8% in men aged 24-29.

“The male risk of infection is the same as women’s, but shifted five years later,” said Dr Barnighausen.

The study also found a much higher risk of infection in the 40% of the population described as have a middle-class income – a 70% higher risk of infection when compared with their poorer or wealthier counterparts.

However, education appeared to be somewhat protective in this rural district: the risk of infection was reduced by 7% for each year of educational attainment.