Previous UNAIDS Figures Overestimated HIV Prevalence Across Africa, U.N., Independent Researchers Say

Recent studies by independent researchers and U.N. officials "raise questions" about whether UNAIDS accurately estimated HIV-prevalence rates in African countries over the last several years, the Washington Post reports. In 2002, the United Nations estimated Zambia to have an HIV-prevalence rate of 21.5%. However, a study done by country officials that same year showed a prevalence rate of 15.6%. In Burundi, UNAIDS estimated a nationwide HIV prevalence rate of 8.3%, but a study done by country officials found a prevalence rate of 5.4%. Previously, UNAIDS estimates often were based on data from young, pregnant women in urban areas with prenatal clinics, the Post reports. More recent studies -- most of which were conducted by ORC Macro, a Calverton, Md.-based research corporation and partially funded by USAID -- examined HIV prevalence in 16 African countries and randomly conducted HIV tests across the countries, the Post reports. Researchers now understand that HIV rates among pregnant women in urban areas tend to be higher than among the general population, according to the Post. The newer studies also often involve two forms of blood testing, which help to avoid false-positive results that also increased previous HIV-prevalence estimates. The new assessments more effectively identify HIV-prevalence disparities between rural and urban populations as well as between men and women, the Post reports. Though research points to a less prevalent epidemic than previously thought in east and west Africa, the disease is "devastating" southern Africa, according to the Post. "What we know now more than ever is southern Africa is the absolute epicenter," said Wilson.

Reaction
Peter Ghys, a UNAIDS epidemiologist since 1999, said HIV prevalence estimates from several years ago were inflated because of the agency's reliance on data from prenatal clinics. However, Ghys said the agency at that time made the most accurate estimates possible with the available figures, and as more data emerged, researchers updated their figures. "What has happened is we have come to realize that indeed we have overestimated the epidemic a bit," he said. He added that he never felt pressured to produce overestimated HIV-prevalence rates. He said, "I can't imagine why UNAIDS or [the World Health Organization] would want to do that. If we did that, it would just affect our credibility." Many other researchers, including David Wilson, a senior AIDS analyst for the World Bank, and two epidemiologists from USAID, reported that the number of new HIV cases in Africa peaked several years ago. Some HIV/AIDS advocates say that finding resources to fight the epidemic is more important than counting the number of cases (Timberg, Washington Post, 4/6).