NEWS

Controversial alcohol study cancelled by US health agency

An investigation by the US National Institutes of Health finds missteps that put the industry-funded project’s credibility in doubt.

Search for this author in:

A bartender mixes a drink.

A study that planned to examine the health benefits of drinking alcohol has its last call.Credit: Spencer Platt/Getty

The US National Institutes of Health (NIH) has terminated a controversial US$100-million study examining whether drinking small amounts of alcohol every day can improve health.

The agency's decision, announced on 15 June, came shortly after an NIH advisory council voted unanimously to end the trial. An agency investigation had found that NIH staff and outside researchers acted inappropriately by soliciting industry funding and biasing the grant-review process to favour specific scientists.

Those findings would have undermined the study’s credibility if it had been allowed to proceed, said NIH director Francis Collins at the advisory-council meeting. “Is it even possible at this point that the results of this trial would have the credibility to influence anyone’s decision-making?” he asked. “That does in fact seem quite doubtful.”

The study, which began enrolling participants in February 2018 under the auspices of the National Institute on Alcohol Abuse and Alcoholism (NIAAA), included $67 million from 5 alcohol companies over 10 years. It came under fire in March after the New York Times reported that the study’s lead investigator — cardiovascular researcher Kenneth Mukamal of Beth Israel Deaconess Medical Center in Boston, Massachusetts — and his collaborators had directly courted funding from the liquor industry in 2013 and 2014, before the study’s launch.

Collins put the project, dubbed MACH15, on hold in May and announced two investigations of the matter. He directed a working group within the NIH Advisory Committee to the Director to assess the scientific merits and integrity of the study, and the NIH’s Office of Management Assessment (OMA) to determine whether NIH employees had improper interactions with industry. The findings of the first probe were announced on 15 June; the second probe will be completed later this month.

Industry money

The advisory-council report, which included initial findings from the OMA, found significant problems with the study, beyond those detailed in the media. According to the report, the study's lead scientists and agency employees solicited funds from alcohol-industry representatives in violation of NIH conflict-of-interest rules. The outside researchers also spoke with at least three members of NIAAA leadership for advice on how best to submit a grant proposal for the study.

These interactions could have conferred an unfair advantage to Mukamal and his colleagues in a competitive grant process, according to the report.

Reviews of the proposed study by two NIH staff members had concluded that the study, which aimed to enrol 7,800 people at 16 sites around the world, was too small to draw significant conclusions and that its findings could be misleading. Peer reviewers also mentioned that industry input could bias the study's findings, but a senior NIAAA staff member advised Mukamal and his colleagues to ignore those comments.

E-mails included in the working group’s report suggested that NIAAA staff and the outside researchers worked together to circumvent normal processes and court industry funding. A 2015 e-mail between NIAAA senior staff and the study scientists suggested editing an e-mail message to include “a bullet that states ‘one of the important findings will be showing that moderate drinking is safe’, etc.” The message appeared to be intended for alcohol-industry representatives, the report said. And e-mails between several NIAAA employees in 2014 discussed not mentioning to NIH administrators that a paper published that year showed that moderate drinking was harmful to a person’s health unless specifically asked.

"We are deeply disappointed that issues raised have led to a recommendation to end the trial," Mukamal and his collaborators said in a statement. "As investigators, we stand fully and forcefully behind the scientific integrity of the MACH15 trial protocol and team." They added that the study staff has not discussed the trial with industry since it began.

Public–private partnerships involving the NIH are supposed to be coordinated by a nonprofit known as the Foundation for the NIH (FNIH) in Bethesda, Maryland. The FNIH solicits and distributes industry money to avoid conflicts of interest or the potential for industry influence on a study’s design or results. According to the report, when the scientists involved in the alcohol study submitted the proposal to FNIH in 2015, they included no mention of their prior meetings with liquor-industry representatives or NIAAA staff members.

Next steps

NIAAA director George Koob said he was “disappointed in what transpired” and that he would work with other NIH offices to shut down the trial in an orderly way.

The NIAAA has already spent $4 million on the study, but NIH principal deputy director Lawrence Tabak says that the study seems so compromised that any results would not be credible. “By terminating the study now, we minimize the loss of resources,” he says.

Adriane Fugh-Berman, a pharmacologist at Georgetown University in Washington DC, praises Collins and the NIH for acting so quickly after questions about the study were raised in the media. “These e-mails are a good example of how when you partner with industry, it starts to change the research agenda itself,” she says. “I really hope that this causes NIH to question the benefit of public–private partnership with industry.”

At its meeting, the advisory group also voted to accept the report’s other recommendations, including further measures to prevent NIH employees from soliciting external funding or engaging with applicants inappropriately. Collins said that the NIH will also investigate whether any of its other public–private partnerships have involved inappropriate interactions between staff, applicants and industry, and to ensure that all of the agency’s institutes are adhering to the same principles.

doi: 10.1038/d41586-018-05461-x
Nature Briefing

Sign up for the daily Nature Briefing email newsletter

Stay up to date with what matters in science and why, handpicked from Nature and other publications worldwide.

Sign Up

Updates & Corrections

  • Correction 18 June 2018: The story initially stated that a working group within the NIH Advisory Council to the Director produced the 15 June report. It was the NIH Advisory Committee to the Director. It also said that the MACH15 project began enroling participants in January 2018 — enrolment began in February 2018. The piece has been corrected to reflect this.

  • Update 18 June 2018: This story has been updated with comments from Kenneth Mukamal, a cardiovascular researcher of Beth Israel Deaconess Medical Center in Boston, Massachusetts, and his study collaborators.