In November of last year, Jeanne Marrazzo received some bad news. Marrazzo had helped design the VOICE study, a 5,000-person clinical trial testing three HIV-prevention interventions: a vaginal gel containing the drug tenofovir, a tenofovir pill and a combination pill known as Truvada that also contains emtricitabine. At a routine meeting with an independent review board to discuss the trial's progress, Marrazzo was unexpectedly told to discontinue the arm of the study testing the gel because that particular intervention didn't work at all. The oral tenofovir had been previously discontinued for the same reasons, leaving just the Truvada arm.

Since that day in November, Marazzo has seen the data and says she has made her peace with the decision to end the trial, which stands for Vaginal and Oral Interventions to Control the Epidemic. But the early end of other trials in the preceding months, such as one testing niacin to prevent heart attacks, have not sat as well with other researchers.

One thing that most agree on is that there seems to be no end to the trend of truncated trials: a 2008 paper cited a 50% increase in numbers of stopped cancer drug trials since 2006, compared to the total number of halted trials from 1998 to 2007 (Ann. Oncol. 19, 1347–1353, 2008). And the functions of the independent panels of experts that monitor interim study results, known as data and safety monitoring boards (DSMBs), have prompted recent objections: a review published last June argued that DSMBs, which have the power to stop trials, should consider factors beyond pure statistics—such as previous scientific knowledge of the treatment and impact of the halt—when recommending trial closure (Eur. J. Cancer 47, 2381–2386, 2011).

The authors of the paper say that regardless of whether DSMBs choose to halt a trial due to the overwhelming benefit or futility of an intervention, they sometimes rob the scientific community of valuable clinical data. As a result, “you will never see the big picture,” says review author Som Mukherjee at McMaster University, in Ontario, Canada.

The closely watched AIM-HIGH trial, which tested the benefit of niacin in reducing heart attacks, was stopped last April because the drug seemed ineffective. It also showed a slightly increased risk of stroke at the time of its analysis. Controversially, the final results, published in November, found no increase in risk of stroke—although they also failed to demonstrate a clinical benefit (N. Engl. J. Med. 365, 2255–2267, 2011).

Experts such as cardiologist Steve Nissen from the Cleveland Clinic in Ohio voiced their opposition to the halting of the AIM-HIGH trial. In a story published on theheart.org, Nissen was quoted as saying that the US National Institutes of Health (NIH) had made a panicked decision to end the trial based on a weak statistical signal of strokes. He added that the NIH had squandered an opportunity to learn more about the effects of niacin.

But David Gordon, executive secretary of the trial's DSMB, insists futility of the intervention, and not stroke risk, was the reason for stopping the trial. Nissen did not respond to requests for comment for this article.

To ensure the right decision is made, more transparency is needed in DSMBs' deliberative process, according to Mukherjee, because, ultimately, he says, everyone has the same goal: “We all want to get drugs out to our patients as quickly as possible.”