South Africa: HIV infection rate slowing, study reports

South Africa’s HIV/AIDS epidemic may finally be slowing, according to a new study which found a 35 percent decline in the rate of new HIV infections between 2002 and 2008.

Using prevalence data from national HIV surveys conducted in 2002, 2005 and 2008, researchers from South Africa’s Human Sciences Research Council (HSRC) and Imperial College London applied a mathematical method to estimate HIV incidence (rate of new infections).

They found that 1.3 percent of South Africans in the 15-49 age group were newly infected annually between 2005 and 2008, compared to 2 percent in the years 2002-2005. The decline was mostly due to a 60 percent reduction in incidence among young women, aged 15-24.

The decline in incidence among young women appears to correspond with significant changes in their behaviour relating to condom use and HIV testing. In 2008, 73 percent of young women reported using a condom at last sex compared to 46 percent in 2002, and more than half had tested for HIV compared with only 13 percent six years earlier.

"Young females are the most vulnerable [to HIV infection], so it's good news that we have declines there, but we still have too many new infections," said lead author of the study, Thomas Rehle of the HSRC, adding that the overall incidence rate of 1.3 percent translated to between 350,000 and 400,000 new HIV infections in 2008.

The lack of significant declines in infection rates among older age groups also pointed to the need for prevention campaigns targeting a broader demographic.

"Definitely, one of the findings is that we’re doing the right thing among young people, but we have to get to the older folks," commented Rehle. "The whole of society has to act responsibly."

Prevalence misleading

As anti-retroviral (ARV) treatment has become more widely available in South Africa, HIV prevalence has become an increasingly unreliable measure of how the country’s epidemic is progressing. More HIV-positive people living longer on ARVs can actually increase prevalence, masking the impact of prevention programmes on new infections.

"The estimation of incidence is crucial in this evolving scenario," write the authors of the HSRC study.

Rehle noted that South Africa was in a unique position as the first country in the region to have data from three consecutive national prevalence surveys. “This allows us to compare between two survey intervals. In this way, we could apply this kind of modelling to derive incidence."

Using data from the 2008 prevalence survey, which for the first time tested blood samples for the presence of ARVs, the study was also able to adjust for the impact of ARV treatment. The researchers estimated that an additional 440,000 HIV-positive South Africans were alive in 2008 as a result of treatment.

Other mathematical studies have predicted that achieving universal access to HIV testing and ARV treatment would have a significant impact on HIV incidence, since individuals on ARVs are much less likely to infect others, but South Africa has yet to reach this point.

By the time the next prevalence survey is conducted in 2011, Rehle predicted that treatment coverage may have reached levels sufficient to have an impact on incidence.

In the meantime, national prevention campaigns will need to capitalize on the gains they have already made.

"There’s a change of atmosphere. People are playing it safer and this is bearing fruit," said Rehle. "But people should not be complacent. We have to do better."