/ 12 October 2001

MRC sticks to its figures

Barry Streek

The controversial Medical Research Council (MRC)report on the extent of HIV/Aids is to be released to the public next Tuesday and the MRC is sticking to its figures and methodology despite reservations expressed by an interdepartmental task team and Statistics SA.

The MRC estimated that between five and seven million South Africans would die from Aids by 2010 and that 40% of deaths of people between the ages of 15 and 49 were due to HIV/Aids.

Statistics SA says there are too many questions about HIV/Aids for adequate answers and “we do not have sufficient information about the probability of transmission of the disease for different groups of people in the country”.

It says that if the Ugandan statistical model is used, up to two million people will die from HIV/Aids by 2010, and as this is substantially different from seven million, it has serious implications for planning. But all models should be treated with caution.

However, health insiders say the dispute about numbers and methodology is basically a side issue because no one disputes that there has been an upsurge in the number of people infected with HIV/Aids.

The director general of Statistics SA, Pali Lehohla, says the Aids epidemic is “serious” and has “profound implications”. This does not need to be debated but the consequences of the differences between the two estimates is serious and has enormous implications.

He denies trashing the MRC report, saying it has made an important contribution by stimulating debate on the changing profile of the disease and the causes of death in South Africa.

Ros Hirschowitz, deputy director general in Statistics SA, also denies trashing the MRC report but says there are problems with all statistical models. The MRC report has not provided any empirical evidence to support its assumptions, one of which is the near-completion in the registration of deaths.

The dispute about the different models, she explains, involves assumptions about the probability of infection through a sexual encounter. The Ugandan model is based on .0011 one out of 1000 sexual encounters will result in HIV infection while the probability model is based on .0026 2,6 out of 1000 sexual encounters will result in infection and the MRC model is based one .0041 4,1 out of 1000 sexual encounters, about one out of 250000 sexual encounters, will result in HIV infection.

Lehohla admits the discussion is “rather academic”, but says it has policy and practical implications.

Health department sources say increased resources are being allocated to fighting HIV/Aids and specific programmes are getting increased budgets, but once again a dispute about the extent of the HIV/Aids crisis is being blurred by messy theoretical debate.

Meanwhile, the United Democratic Movement’s Gerhard Koornhof says his party is “disgusted” by the Afri-can National Congress government’s abuse of Statistics SA in its attempt to discredit the MRC report. He says the MRC is far better qualified to comment on Aids than Statistics SA, whose own statistics have been questioned before.

“This is yet another example of the ANC’s blatant attempts to suppress and discredit statistics or reports that do not suit their often misguided political agenda,” Koornhof says.