/ 20 January 2006

One in nine — it’s official

Two separate studies, using different techniques, have for the first time reached the same conclusion about how many South Africans are infected with HIV. The research has raised hopes of reconciling feuding government departments and eventually of a more effective war on HIV/Aids.

The Actuarial Society of South Africa’s computer model, using data from a wide variety of sources, and a state-of-the-art population survey by the Human Sciences Research Council (HSRC), indicate that five million South Africans carry the virus. That is one in nine people, or about 11%.

”It is reassuring that our results have been confirmed by different approaches,” said Thomas Rehle of the HSRC, which was commissioned by the Nelson Mandela Foundation. ”The research can no longer be denied — and the less confusion you have, the better.”

The matching numbers have raised the possibility that years of bitter arguments over Aids number-crunching will be replaced by something more constructive — but there is a long way to go.

The first step may be to reach consensus with Statistics SA, which said last year that 4,5-million South Africans were infected, and the Department of Health, which puts the number substantially higher at more than 6,3-million infections.

Negotiations are said to be under way to bring all players together to analyse the data and see if agreement can be found.

Not everyone is hopeful that the new surveys will improve policy and planning. ”We have known for years that South Africa has a massive Aids epidemic,” noted Nathan Geffen of the Treatment Action Campaign. ”From a science perspective, it is important to reach consensus on numbers. From a policy perspective, it makes less of a difference.”

Others argue that agreement on the infection rate leaves less room to manoeuvre by Aids denialists in and out of government. ”The consensus is definitely significant,” said Rehle, who is also visiting professor of international health at the University of Cape Town (UCT).

Confusion over South Africa’s infection rate may be rooted in the lack of information available at the start of the epidemic. Often the only reliable data came from government clinic blood tests on generally poor, pregnant women, issued yearly since 1990.

”You have to compare apples with apples,” said Rehle. ”You can’t use only pregnant women using public health facilities to estimate how Aids is affecting an entire nation — especially a complex one such as South Africa.”

The HSRC fieldworkers fanned out across the country, taking HIV blood tests from 15 800 of the nearly 23 300 people they interviewed. Their work reveals widespread misunderstanding of the disease, with many South Africans — particularly those over 50 years of age or in their early teens — either uncertain of or denying any connection between the virus and the disease.

Rob Dorrington, of the Centre for Actuarial Research at UCT, which did much of the time-consuming number-crunching, trawled through surveys of HIV prevalence among pregnant women attending government clinics, but factored in the first HSRC household survey in 2002, a youth survey, data from the most recent population census and death registration records.

The society’s results were singled out for praise by Yale University as ”an excellent model, probably the most sophisticated in the world, because it incorporates data from a wide range of sources”.

The health department issued a press release acknowledging the HSRC results, but it remains unclear what impact they will have in official circles. ”We have had no official or even semi-official responses,” said Dorrington. ”We are not attempting to drive policy, although we would be more than happy to participate in any efforts to reach consensus.”

The need to monitor new infections — the next cycle of the epidemic — is critical, said economist Nicoli Nattrass, who directs the Aids and Society Research Unit at UCT.

Said Nattrass: ”If the South African government successfully intervenes with anti-retroviral drugs, you will see a higher prevalence of people living with Aids, which looks bad but is actually a good sign. For one thing, people aren’t dying, and for another, people on anti-retroviral drugs are less likely to infect others.”