The long-accepted view that most HIV infections in sub-Saharan Africa are transmitted by sexual activity is being challenged by controversial new research that attributes a greater proportion of the epidemic here to unsafe injections by hospitals and clinics.
The research has caught the attention of some US senators and is galvanizing debate over how Washington should dispense AIDS funds. It's a high-stakes issue, given the toll the disease is taking on the continent and President Bush's pledge of $15 billion over the next five years.
Observers on both sides of the debate say that regardless of their view on the new study's findings, the problem of transmission by needles is significant enough to warrant fresh attention.
"We can all agree that we should work toward ending unsafe injections," says George Schmid, a senior epidemiologist at the World Health Organization (WHO). "The issue is how many resources we devote right now."
The WHO estimates that 90 percent of HIV infections in sub-Saharan Africa are caused by heterosexual sex. It says unsafe injections account for just 2.5 percent of cases here, with the rest spread through mother-to-child transmission, intravenous drug use, homosexual sex, and tainted blood transfusions.
But David Gisselquist, an independent economist and anthropologist based in Hershey, Pa., reviewed past studies and concluded that only one-third of HIV infections in Africa are transmitted sexually, with one-third caused by unsafe injections, and the final third passed on by transfusions and other blood-borne means.
Mr. Gisselquist testified about his research in Senate hearings earlier this year, and last week presented his findings to a packed meeting room here at the International Conference on AIDS and Sexually Transmitted Infections in Africa.
Gisselquist points to studies examining African women's HIV status before and after birth, a time of potentially frequent exposure to injections. "Some of these studies report rates of [HIV] incidence that are just incredible, that can't reasonably be explained by sexual transmission," he says.
"There's a lot of unsafe healthcare in Africa," says Gisselquist, citing an example from Zimbabwe. "In Harare, if you go to the dentist, you're advised to go early because it's cleaner."
Gisselquist's conclusions hinge on the use of a statistical indicator called "crude population attributable fraction," which describes the association of various activities with the virus but doesn't necessarily show causality. It's a measurement that many researchers consider unreliable and that even Gisselquist acknowledges "may overestimate or underestimate the causative association."
His research is coming under fire from top officials at the WHO and the United Nations joint program on AIDS and HIV, who dismiss the conclusions as flawed and his estimates as exaggerated.
The controversy is more than an arcane squabble between opposing scientists: it has implications on how AIDS resources should be allocated. If Gisselquist's claims are correct, the battle against HIV in Africa should place far less emphasis on changing sexual behaviors or promoting condom use, and more on ensuring that healthcare workers always use clean needles.
Dr. Schmid, who recently joined the WHO after 22 years with the Centers for Disease Control in Atlanta, argues that devoting a much greater share of AIDS-prevention spending to ensuring all injections are safe would not have a major impact on reducing the spread of the virus.
Others agree. "As it is now, there's insufficient attention and insufficient resources and effort being given to the prevention of sexual transmission of HIV," adds Stephen Moses, a microbiology professor from the University of Manitoba in Canada. "My concern would be that this debate might distract from those efforts. I hope that it won't."
WHO scientists dispute Gisselquist's claims based on a variety of research, but four arguments are key:
• HIV infection rates in Africa are minuscule for children but rise rapidly after the onset of sexual activity.
• A strong correlation exists between the number of sexual partners over a lifetime and the likelihood of a person testing positive for HIV, a correlation that strengthens in communities where males are uncircumcised and genital herpes is prevalent.
• Countries that have high rates of HIV infection, such as Kenya and South Africa, do not have high rates of hepatitis C, an infection spread by dirty needles.
• Significantly fewer unsafe injections are administered annually per person in sub-Saharan Africa than in such regions as Eastern Europe, South Asia, and the Middle East, which have far lower HIV infection rates.
Holly Burkhalter, policy director for Physicians for Human Rights, based in Boston, says it doesn't matter whether the proportion of Africans who get HIV from dirty needles is 2.5 percent or 33 percent - one infection is too many. Even the conservative estimate would mean that some 75,000 people are likely contract HIV this year in Africa from unsafe injections, based on UN estimates of how fast the virus is spreading.
"That's plenty of deaths to warrant the world's attention," says Dr. Burkhalter. "It's the clearest example of substandard healthcare being tolerated for the poor that rich countries would never tolerate in their own hospitals and clinics."