How Do We Know the Science Is Conflict-Free?

When an industry group pays for a scientific study, researchers may be tempted to come up with conclusions that advance industry's interests. The mechanisms for preventing this are few and marginally effective. The main one: disclosure.

The issue has been in the news of late — especially the war over research on genetically modified organisms. Now journalist Sara Reardon, in Nature News & Comment, has taken a deep look at disclosure policies at the National Institutes of Health (NIH). NIH, a federal agency, funds some $30 billion in research annually, a portion of which relates to environmental health.

Under Congressional prodding after a conflict scandal, NIH in 2012 tightened its earlier conflict-disclosure rules. "But three years later," Reardon writes, "it is not clear what the costly, cumbersome rules have accomplished."

Reardon filed a Freedom of Information Act request for conflict-of-interest disclosure reports submitted to NIH by the institutions and scientists it funds. Such reports might, for example, include funds received from industries having an interest in the outcome of the research. Many of the amounts involved were relatively small — less than $10,000. NIH lowered the reporting threshold in 2012 from $10,000 to $5,000. But many other disclosed amounts were in the tens or hundreds of thousands.

Researchers disclose outside payments to their host institutions, not to NIH. The host institutions (such as a university), then determine if the payment poses a potential conflict, and work up plans to manage the potential conflict — for example, they might require that payments be disclosed in journal publication of the research.

Reardon reports that universities have spent far more than NIH originally expected to fulfill these requirements: 71 institutions spent some $23 million during the first year. Some university officials said the procedures had no benefit.

However, Paul Thacker, who led the Senate investigation that prompted the reforms, said any benefits were intrinsically tough to measure, and that public confidence in the credibility of some research had been damaged.

Reardon's report cites others who say current NIH conflict-disclosure rules do not go far enough. For example, universities do not have to disclose the "management plans" for handling conflict. Reardon sometimes found delays and obstacles to disclosure in her reporting.

• "NIH Disclosure Rules Falter," Nature (News & Comment), September 15, 2015, by Sara Reardon.
• Previous Story: WatchDog of September 9, 2015.