Global HIV prevalence could be "dramatically overstated" because of statistical modeling errors and "undetected" declines in HIV prevalence in several countries, the Boston Globe reports. According to the Globe, some HIV/AIDS experts believe the estimate that there are 40 million people living with HIV worldwide could be inflated by 25% to 50%. Reducing HIV prevalence estimates for several countries would "go against the grain" of years of statements by UNAIDS that said HIV/AIDS was "relentlessly on the rise," according to the Globe. However, U.N. epidemiologists and statisticians who have produced the current estimates say that the estimates for some countries will be "sharply cut" in a report to be released before the XV International AIDS Conference that will be held in Bangkok, Thailand, in July. The UNAIDS report, which will include HIV prevalence estimates for 2003, is expected to say that HIV prevalence is declining in East Africa, "leveling off" in West Africa and still "maintaining a high rate" in Southern Africa, according to the Globe. The UNAIDS report also is expected to reduce the organization's estimate of Rwanda's HIV prevalence to about 5%, compared with 11% in 1999, according to the Globe.

Explanations
AIDS experts said the "major error" in prevalence estimates is that they "relied too heavily" on data from urban areas and did not sufficiently assess rural areas, which have lower HIV prevalence, the Globe reports. Another potential error in estimating prevalence can occur in estimating a country's population, according to the Globe. For example, Nigeria's population has been estimated at 120 million to 160 million, but a census has not been taken in that country for more than 50 years. In addition, most countries do not collect data on deaths, according to the Globe. UNAIDS and World Health Organization officials use HIV prevalence estimates to determine estimates for AIDS-related deaths, AIDS orphan populations, countries' life expectancies and the number of people needing antiretroviral drug treatment, according to the Globe. These other estimates will be "readjusted similarly" if prevalence estimates are reduced, the Globe reports. Reducing prevalence estimates also could affect "[b]illions of dollars" in aid, according to the Globe. In addition, new prevalence estimates could affect how countries approach treating and preventing HIV/AIDS (Donnelly, Boston Globe, 6/20).

Back to other news for June 21, 2004

Search the Newsroom archive