Skip to main content

HIV prevention trial in Africa is a dud

HIV prevention trial in Africa is a dud


Most women didn't take their doses of preventive medicine

Share this story

Justin Sullivan/Getty Images

A trial among 5,000 women in Zimbabwe, Uganda, and South Africa showed that daily HIV prevention regimens don't work to stop the disease. That may be because many of the women involved in the trial didn't take the medications they were assigned. Today's results illustrate how stigma and fear can waylay disease prevention trials.

The trial, nicknamed VOICE, for Vaginal and Oral Interventions to Control the Epidemic, tested HIV prevention strategies from September 2009 to August 2012 — though it had been planned to run longer, the trial was stopped for futility. There seemed to be no benefit, even though similar preventive tools have worked elsewhere.

One way to prevent HIV from spreading in a high-risk setting is called pre-exposure prophylaxis, or PrEP. In the US, this is typically done with a drug, such Gilead Sciences' Truvada, which a high-risk patient takes every day. The VOICE trial looked at use of Truvada or another Gilead Sciences drug, Viread. The researchers also studied a vaginal gel formulation of the compound in Viread. Even though the drugs' usefulness wasn't in question, they didn't seem to be working.

"It is well-established that medications don't work if they are not taken."So the scientists used blood tests to see if women were using the product they'd been assigned; by three months into the trial, it was clear many women hadn't used any of their preventive medication at all, according to the study results reported in The New England Journal of Medicine. Those least likely to have used their medicine were single women under 25 — those who were at greatest risk. At some study sites, 10 of these women out of 100 were infected every year.

"It is well-established that medications don't work if they are not taken, which probably explains why no difference in efficacy was observed between the active-drug and placebo groups," writes Michael Saag, an HIV researcher at the University of Alabama at Birmingham, in a related editorial.

even though the drugs worked, the human factors didn't In an ancillary study of random participants in VOICE, the researchers found that the poor adherence may have been because people didn't know if the products were effective, because the study drugs are associated with HIV infection rather than prevention, or because of a lack of social support. So even though the drugs worked, the human factors didn't, Saag concludes.

"The study indicates that much more work is needed, not so much in the realm of understanding the biologic basis of pre-exposure prophylaxis as a preventive treatment but rather in the realm of understanding behavioral barriers in the setting of strong social stigma," he writes.

The HIV epidemic in Sub-Saharan Africa is the worst in the world, according to UK charity Avert; there, 25 million people are infected. About 60 percent of people with HIV in Sub-Saharan Africa are female, and sexual intercourse is the main way people are infected. Though condoms would protect these women, many can't insist on condom use.

most women said they were using the treatments, even though blood tests showed otherwise Fewer than 40 percent of the samples taken three months into the trial showed any level of the assigned study drugs. Most of these women didn't have positive samples later, either. But most women — 88 percent to 90 percent —said they were using the treatments, even though blood tests showed otherwise. And women went to great lengths to pretend they were taking their pills, bringing back medicine for a practice called "reconciliation," where researchers count unused pills. In that process, 86 percent of medicine was "taken," though again, that didn't match with the blood plasma samples.

"This means that a large number of participants actively removed unused medications from their allotment before returning to the study site in order to create the appearance of compliance with the protocol," Saag writes. "The question that emerges is this: why did the participants go to such lengths to create the appearance that they were taking medications when they were not?"

One researcher told The New York Times that study participants lied because they were scared — they'd heard rumors the drugs would cause liver damage or hurt their reproductive organs. To compound the issue, the women were paid $10 to $15 per clinic visit, valuable income even if the would-be participants didn't actually intend to participate.

The study results may not mean much for HIV treatment, but they've changed how studies are run. Now, participants' blood tests are run immediately, rather than stored for later.