NIH Nixes Knowing Outside Grantees’ Conflicts-of-Interest

After admitting it routinely fails to collect and monitor university conflict-of-interest policies, the National Institutes of Health has rejected an inspector general recommendation that it collect detailed financial information from its extramural grantees and reveal what steps institutions are taking to manage, reduce, or eliminate any conflicts of interest. Federal regulations require the 3,000 universities, medical schools and other research institutions that receive 80 percent of the agency’s $29.2 billion budget manage any financial conflict of interest that “could directly and significantly affect the design, conduct, or reporting” of government-funded research. The grantee institutions must then report those programs to NIH, although they do not have to report individual conflicts of interest. About 325,000 researchers participate in the agency’s 50,000 grants each year.

When asked, NIH could only produce 438 reports filed between 2004 and 2007, according to an investigation released last week by Health and Human Services Department Inspector General Daniel Levinson. Moreover, 89 percent of those reports did not state the nature of the conflicts or how they were managed. The 30 reports that contain that information described cases where “investigators have intellectual property associated with the grant research or financial interests in companies that are subcontractors on the research grants.” The method of managing those conflicts was disclosure in presentations and publications. Knowing such information for all institutions “could assist NIH officials in making a determination as to whether followup with the grantee institutions is necessary,” the report noted.

In the agency response, director Elias Zerhouni said requiring reporting of individual conflicts would put NIH in the business of directly managing researchers at outside institutions. However, it did agree to set up a consolidated reporting system in the Office of Extramural Research, and make the filing of annual reports mandatory as of March 1. The American Association of Medical Colleges agreed that its member institutions shouldn't have to report their faculty members' individual conflicts of interest to NIH.

Meanwhile, The Cancer Letter (subscription required) reports that two leaders of the International Early Lung Cancer Action Project from Weill Medical College of Cornell University in New York failed to disclose patents and patent applications in their published reports claiming success for using CT scans for early detection of lung cancer. The controversial clinical trial reports and related articles have appeared in the New England Journal of Medicine, Journal of the American Medical Association, Archives of Internal Medicine, The Lancet, Chest, Clinical Cancer Research, Natural Clinical Practice Oncology, and Cancer Cytopathology without disclosure. Claudia Henschke and David Yenkelevitz have both received NIH grants in recent years to conduct research involving lung cancer screening, according to the agency’s grants database. The one awarded patent has been licensed by Weill to General Electric, which manufactures CT scanning machinery. The journals are investigating the failures to disclose the patents.

Last October, Congress launched an investigation into a rival NIH-funded group exploring the use of CT scans for early detection of lung cancer after a group called the Lung Cancer Alliance alleged two of its lead investigators took money from the tobacco industry. The Lung Cancer Alliance, a pateitn advocacy group formed to push for lung cancer screening, is funded in part by GE.

This article first appeared in Integrity in Science Watch, a publication of the Center for Science in the Public Interest.