AIDS Alert International

Epidemic leveling off globally, with fewer people living with HIV/AIDS

The latest data by UNAIDS of Geneva, Switzerland, show that the global HIV pandemic has leveled off, with an estimated 33.2 million people infected, 2.5 million new infections last year, and 2.1 million AIDS deaths in 2007.1

International media focused on the revisions UNAIDS has made to its annual estimates, a downward projection that eliminates about six million infections. It decreased from the estimate in 2006 of 39.5 million infected worldwide to the new 2006 estimate of 32.7 million infected.

The revised numbers mainly are due to better estimates of infections in six countries, with India topping the list, UNAIDS officials say.

Along with India, 70 percent of the changes were due to surveillance changes in Angola, Kenya, Mozambique, Nigeria, and Zimbabwe.

But the real news, UNAIDS officials say, is that AIDS deaths have declined worldwide, and the number of people living with the disease has leveled off.

"In sub-Saharan Africa, prevalence has declined in recent years in the continent as a whole," says Kevin De Cock, MD, director of the HIV/AIDS department at the World Health Organization (WHO) in Geneva. De Cock and other UNAIDS representatives spoke about UNAIDS' annual report at a media teleconference late last year.

Even with the decline, sub-Saharan Africa is disproportionately impacted by the epidemic, De Cock says.

"Sixty-eight percent of people estimated to be living with HIV are in sub-Saharan Africa, and 76 percent of the estimated deaths occurred there," says De Cock. "Just eight countries of Southern Africa account for almost one-third of all new HIV infections and deaths."

Therefore, the international health community's focus on reducing new infections in sub-Saharan Africa, and the increase in access to antiretroviral therapy among those infected can have reverberations felt across oceans.

"It's not possible from these sorts of global estimates and examination of trends to be absolutely certain about the reasons for some of the trends and the impact of prevention or treatment programs, but I think one can draw some conclusions from specific situations," De Cock says. "There are some countries where we really do believe that the reductions in HIV prevalence are the result of behavioral change and of prevention programs, and such countries would include Kenya, Zimbabwe, Cote d'Ivoire, and some countries of Southeast Asia, such as Cambodia."

Likewise, the decline in AIDS deaths in the past two years can be attributed to the scale-up of antiretroviral therapy, De Cock says.

"There are well over two million, between two and three million people, on antiretroviral therapy in low and middle income countries," De Cock says. "So I think there are some positive elements in this report, some positive elements that are encouraging."

For instance, there are true HIV prevalence declines in Kenya and Zimbabwe, says Karen Stanecki, MD, senior advisor on demographics and related data at UNAIDS' department of social mobilization and information.

"In those two countries, we have done extensive research of the data that's available," Stanecki says. "The work that has been done indicates that this decline in prevalence cannot be attributed solely to deaths, but also it is attributable to behavior changes and the reduction in new infections in those two countries."

The revised estimate numbers represent an improvement in country data collection, including the types of surveys undertaken, says Paul Delay, MD, director of evidence monitoring and policy at UNAIDS.

"One of the big assumptions that has been put into this latest round is increasing survival, which is the amount of time from being infected to living without treatment," Delay says. "This has gone from nine to 11 years, and this changes overall [AIDS death] numbers, based on measured prevalence."

UNAIDS increased the survival time with untreated HIV because of new information, says Peter Gece with UNAIDS.

"There is new information on the survival time of people infected with HIV," Gece says. "This has come from a cohort study in different regions of the world, including from Africa, from the Caribbean, and from Asia."

A recent analysis of that new data indicates that average survival is around 11 years, Gece adds.

"So, the 9-year assumption was based on much less information than this new survival of 11 years, which was formed by a larger group of studies and, within those studies, a larger number of people than we had before," Gece explains.

Delay says it would be absurd to suggest that UNAIDS had previously inflated HIV/AIDS estimates out of a political agenda.

"UNAIDS is not an implementing agency," Delay says. "Our budgets are not really affected by these massive changes in resources that are being devoted to the programmatic response. So the idea that this is somehow self-serving is absurd."

Also, there are too many people working with the data, including ministries of health, statistics bureaus, district-level epidemiologists, and others for the data to be skewed fraudulently without someone blowing a whistle, he adds.

The take-home point is that there is positive news in the revised data, Delay says.

"Though it hasn't really been clearly stated here, these are very encouraging trends — stabilization of prevalence, declines of prevalence in sub-Saharan Africa, and the beginnings of declines in mortality," Delay says. "So it's data that we clearly need to continue to be collecting and making the investments so that the quality continues to improve."


  1. 2007 AIDS Epidemic Update. Produced by UNAIDS/WHO. Available at