Conflict of Interest Is Widespread in Biomedical Research, Study Finds By LILA GUTERMAN
Approximately one-fourth of biomedical scientists have financial affiliations with industry, according to a new review of studies about academic conflicts of interest. The review, conducted by researchers at Yale University School of Medicine and published today in The Journal of the American Medical Association, also concludes that two-thirds of academic institutions hold equity in start-up companies that sponsor research performed at those institutions, and that research financed by industry is more likely to draw commercially favorable conclusions.
"Financial relationships are pervasive and problematic," wrote Justin E. Bekelman, Yan Li, and Cary P. Gross, who are researchers in the medicine and biostatistics departments at Yale. They reviewed the medical literature, looking for studies that analyzed financial relationships between academe and industry.
Their review found remarkable consistency among various studies' results, said Sheldon Krimsky, a professor of urban and environmental policy at Tufts University who is an expert on conflicts of interest. Such consistency, he said, "strengthens the hypothesis that there is a funding effect from private-sector investment in research. It has to be addressed in some fashion to secure the integrity of science."
In seven studies that looked at the extent of individuals' financial relationships to industry, the Yale researchers found that the percentage of researchers with industrial funds, speaking fees, positions as consultants, or other financial ties to industry ranged from 7.6 to 43 percent, with three of the largest studies finding that fraction to be about one-fourth.
Only one study surveyed academic institutions themselves for industry involvement, and it found that 68 percent of universities in the United States and Canada held equity in companies. The review's authors call this finding "particularly concerning" because "ownership carries the responsibility of business stewardship." Mr. Bekelman said that he hoped as a next step to study the impact of equity ownership on biomedical research. "The end goal is improving health. The question becomes, is equity ownership critical to that?"
The researchers found that industrial sponsorship affected the outcome of the studies and even the design of clinical trials. For instance, industrially sponsored studies were more likely to compare new pharmaceuticals with placebos than with currently used medicines, a technique that increases the likelihood of a positive outcome.
Mr. Krimsky said bias stemming from industrial funds is more extensive than is generally believed. The Yale researchers found not only that studies financed by the tobacco industry reported pro-industry results, but that studies on pharmaceuticals were affected by their source of funds as well.
The researchers propose that to manage conflicts of interest, universal policies should be adopted by academic institutions and by journals. They recommend the creation of a public database that would include the results of all clinical trials as well as disclose investigators' financial ties to industry.
But Mr. Krimsky thinks that disclosure may not be enough: "We would not permit a judge, for example, to have equity in a for-profit prison, even if the judge disclosed it. And yet it seems to be that it's OK for scientists to have equity in companies that fund their research as long as they disclose it."