South African actuaries and academics have cast doubt on relatively upbeat HIV/Aids statistics released last month by the Department of Health, while the United Nations has warned that the Aids epidemic in sub-Saharan Africa is still in its infancy.
”The scale of the Aids crisis now outstrips even the worst-case scenario of a decade ago. Dozens of countries are in the grip of serious HIV/Aids epidemics,” says a new report by the Joint United Nations Programme on Aids.
The report says that sub-Saharan Africa is the worst-affected region in the world, with 28,5-million people living with HIV. It warns that the epidemic is still in the early phase and is not levelling off.
In June the health department released an HIV prevalence survey among pregnant women attending public clinics, which stated that the epidemic had stabilised in South Africa.
This week the Actuarial Society of South Africa (ASSA) warned that these statistics are open to alternative interpretation. ”Although the results are encouraging, ASSA is warning its members not to jump to conclusions based on this data.”
The society is a professional association for actuaries, with the ASSA Aids Committee made up of 22 actuaries who have refined models to closely fit the data available on HIV prevalence. Predictions are made based on an analysis of past trends.
The national HIV prevalence survey said that prevalence among pregnant women attending antenatal clinics was estimated at 24,8% last year, compared to 24,5% in 2000. It adds: ”Although we see a slight increase, statistically this increase is not significant, and we can confidently say that the prevalence rate has stabilised.”
Sarah Bennett, spokesperson for the society, says: ”There is uncertainty around this number because it was derived from a random sample. According to the report on the antenatal survey, the true value is estimated to lie between 23% and 26%.”
The society’s latest model estimated 2001 prevalence among women at antenatal clinics at 27,3% — indicating an increase in infection rates.
”The model was scientifically adjusted to ensure a close fit to antenatal data in previous years; therefore the 2,5% difference is surprising,” says Bennett.
Rob Dorrington, convenor of the ASSA committee, says the government survey looks at the prevalence rate over a single period, and that care must be taken in predicting a trend from it. The society’s Aids Committee is still waiting for the 2002 antenatal statistics before making major adjustments.
The antenatal prevalence for women aged 15 to 19 was lower than estimated by the society’s model. The government survey finds HIV prevalence among teenagers has dropped for the third consecutive year, and that there has been a 5,6% decrease in HIV prevalence among pregnant women under the age of 20. ”However, the rate has been very level the last three years,” says Dorrington. He says there could be various expla- nations for this, some of them not necessarily implying a fall in prevalence. For example, if fertility in this age group was falling because of oral contraceptives the prevalence measured at antenatal clinics may not reflect the true prevalence.
Quarraisha Abdool Karim, an epidemiologist at the University of Natal, says the survey shows a maturing epidemic. She says antenatal clinic data is a reliable and good marker but has limitations. ”Pregnant women reflect heterosexual transmission of the epidemic. But when fertility rates decline and there is an increase in morbidity and mortality, those estimates become less reliable.”
According to Karim, the main focus in deciding whether the epidemic is levelling off should be on the rate of new infections. The survey does not have this information.
Also sceptical of the government’s statistics is Alan Smith, head of virology at the Nelson R Mandela School of Medicine in Natal. ”We must emphasise that this survey reflects a small category of the population. There is not enough evidence to be optimistic.”
Minister of Health Manto Tshabalala-Msimang said ”second-generation surveillance” — new infection rates and other study groups — would be introduced as a key to determining intervention measures. The government survey does, however, emphasise that the huge challenge is to look at the impact HIV will have on the next generation.
The government statistics found that HIV prevalence increased significantly among women in their thirties. Pregnant women in their late twenties show the highest infection rate, at 31,4%.