Studies Call Attention To Ethics Of Industry Support

Sidebar: Fourteen Journals Surveyed on Financial Interest of Authors For More Information On Academic-Industrial Collaborations Two recent studies have renewed attention on the perennial controversy over industry support of academic research. Some observers maintain that conflict-of-interest issues inherent in academic-industrial collaborations compromise scientific integrity and contribute to the public's mistrust of science, while others argue that such relationships benefit society and are an integral part of research. ENSURING HONOR:Lenoard Minsky contends that only compulsory disclosure of financial interests on research papers guarantees integrity. One analysis, led by Sheldon Krimsky, a professor of urban and environmental policy at Tufts University, found that 34 percent of the articles examined had a first or last author from academia with a financial interest in the described research (S. Krimsky et al., Science and Engineering Ethics, 2:396-410, 1996). Virtually none of these papers made note of that relationship. The other study, led by David Blumenthal, chief of the health policy research and development unit at Massachusetts General Hospital, found that faculty members receiving corporate funds had more peer-reviewed articles published-up to a point (D. Blumenthal et al., New England Journal of Medicine, 335:1734-9, 1996). His team also found that academics who received more than two-thirds of their research support from industry published fewer papers than those who received less support from commercial sources. In addition, the team found that articles published by industry-supported researchers were less influential than those published by researchers with no industrial support. This was measured by a set of citation influence weights for 3,100 journals covered by the Science Citation Index. Leonard Minsky, executive director of the Bethesda, Md.-based National Coalition for Universities in the Public Interest, contends that there is reason to be concerned about conflict-of-interest problems in corporate-academic relations. "These papers, from our perspective, are a corroboration for our position on what's going on at universities-that there's a significant amount of financial activity [with industry] and the financial activity is harmful in a way that we predicted it would be." Industry, however, has a different view. Alan Goldhammer, director of technical affairs at the Biotechnology Industry Organization (BIO) in Washington, D.C., reports that the trade organization hasn't taken a specific stance on Blumenthal's findings, although he notes the conclusions do show positive and negative aspects of industrial funding. Adds BIO spokesman Eric Christensen: "The net effect of corporate sponsorship has unquestionably benefited the world. In the last 15 years more than 100 million people around the world have been helped by biotech products, all of which come from a synergy between research and corporate development." That synergy translates into strategic alliances, which means more corporate money in academic research. Corporate support for the research alliances has increased from less than $5 million in 1974 to hundreds of millions in the early 1990s (E. Haber, Nature Biotechnology, 14:441-2, 1996). Krimsky cites technological advances and changes in policy to explain why these collaborations have flourished. "One reason is that in biosciences you can now go from the lab to commerce very quickly because of gene splicing. Secondly, in 1980 the Supreme Court decided that genes, bugs, and cell lines could be patented, that life forms in and of themselves could be patented. That stimulated an enormous amount of investment opportunity." STRICT STANDARDS: To maintain confidence in basc biomedical research, financial interests should be accurately represented, remarks Jonathan King. In the same year, Krimsky adds, the Bayh-Dole Act gave educational institutions ownership of inventions developed with federal dollars. He adds that the perception by university officials that federal research funds were declining and that this support would have to be made up by the private sector also played a role in academics' seeking more money from commercial sources. Krimsky and his coauthors examined nearly 800 articles published in 1992 by life scientists from nonprofit research institutions located in Massachusetts. They derived their database from 14 leading biological and medical journals (see table below). The authors were designated as having a financial interest if they were listed on a patent or patent application; served on a scientific advisory panel of a biotech company developing a related product; or served as an officer or shareholder of a company with commercial ties to the research. Allison Rosenberg, associate executive director of the Government-University-Industry Roundtable, a science policy forum sponsored by the National Academy of Sciences, finds the outcome of Krimsky's study interesting "except for the fact that these are 1992 publications." Five years ago, many journals did not have a policy on disclosure, "so in a certain way, he's demonstrating the obvious," she comments. The lack of a policy in the past is the reason financial disclosure has become a subject of heated debate in the last four years, she believes. Krimsky says that he and his coauthors were surprised by their findings. They didn't think the proportion of researchers would be that high because they were looking only for patents that were directly related to the research described in the papers. Counters Tom Desrosier, vice president and chief patent counsel at Genetics Institute in Cambridge, Mass., "to draw any conclusion just because a person's name is on a patent application is unfair." To impute financial interest to academic researchers simply because their university made them file a patent application is unjust, he says, because often university technology transfer offices file patent applications on behalf of scientists whether they want them or not. Krimsky believes that all scientists should declare their financial interest in their published research and that all journals should have disclosure policies. "There's this tremendous loss of trust at a time when people are trying to sort out competing views in medicine and health care. We're talking about the integrity and the confidence that the public has in science. The question is: How do you get this information to the individuals that would benefit from it? The only way this can be done fairly is if people are up front with any financial interest they have." Jonathan King, a professor of molecular medicine at the Massachusetts Institute of Technology, agrees with Krimsky's call for a standard disclosure policy. "The great majority of the research that was in [Krimsky's study] depended on research funded by the National Institutes of Health and the National Science Foundation. Scientific productivity depends on public investment. Biotechnology is skimming off the profitable cream of this basic core, and in order to keep confidence in the basic biomedical endeavor in the presence of these influences, you have to maintain the strictest standards of disclosure and accurate representation of interests involved." Rosenberg, who works with scientists as well as representatives from industry and government, thinks the private sector has "absolutely no objection" to academic researchers' disclosing their financial interests on peer-reviewed literature. She agrees that there should be disclosure policies. Jack DeForrest, director of contract and license management at New York-based Bristol-Myers Squibb Co., says that it is rare for academic researchers to have a financial interest-as defined in the Krimsky paper-in investigations related to his company. He contends that this is more of an issue for biotech firms. D. Blumenthal, N.Causino, E.G. Campbell, K.S. Louis, "Relationships between academic institutions and industry in the life sciences-an industry survey," New England Journal of Medicine, 334:368-73, 1996. D. Blumenthal, E.G. Campbell, N. Causino, K.S. Louis, "Participation of life-science faculty in research relationships with industry," N. Engl. J. Med., 335:1734-9, 1996. E. Haber, "Industry and the university," Nature Biotechnology, 14:441-2, 1996. S. Krimsky, L. S. Rothenberg, P. Stott, G. Kyle, "Financial interests of authors in scientific journals: A pilot study of 14 publications," Science and Engineering Ethics, 2:396-410, 1996. (Published by Opragen Publications, Gilford, United Kingdom; http://www.cableol.co.uk/opragen/) K.J. Rothman, "Conflict of interest: The new McCarthyism in science," JAMA-Journal of the American Medical Association, 269:2782-4, 1993. Harvard Medical School Faculty Policies on Integrity in Science http://www.med.harvard.edu/programs/integrity/ In the past several years, a handful of U.S. journals have begun to call for disclosure. The latest was Proceedings of the National Academy of Sciences in May of last year. But some editors decry the mandatory declaration of financial interests by some journals, claiming it is a misplaced policing mechanism. In an editorial, the journal Nature stood by its policy to not require disclosure: "[The Krimsky paper] makes no claim that the undeclared interests led to any fraud, deception, or bias in presentation, and until there is evidence that there are serious risks of such malpractice, this journal will persist in its stubborn belief that research as we publish it is indeed research, not business" (Nature, 385:469, 1997). "That's a naïve point of view," responds Krimsky. "It assumes the kind of 19th-century/early 20th-century view that science has its own internal ethics-that somehow or other scientific financial interests could never distort the interpretation of data because the higher calling of science is above it all." Kenneth J. Rothman, a professor of public health at Boston University and editor of the journal Epidemiology, wrote in a 1993 editorial in JAMA-Journal of the American Medical Association (K.J. Rothman, 269:2782-4, 1993) that "while disclosure may label someone as having a conflict of interest, it does not reveal whether there actually is a problem with the work or whether the implicit prediction is a 'false positive.'" He called this the "new McCarthyism in science." Rothman remarks that he's not against disclosure. He says his journal is "happy to publish disclaimers, and we do publish them, but we don't think that our role is to be police. I think there's a tremendous detriment to putting all that emphasis on it. You see it more and more in science and the way it gets reported. What everyone wants to know is: 'Who funded this work?', as if that explains the results. For many people, once they hear that, they're not interested in the rest of it. It's become a substitute for critical thinking." Rothman sees two critical problems in this trend. "The whole level of scientific discourse has been debased by this emphasis. It's not science anymore. It's who's pure. The second thing is that people are getting allegations thrown at them that they can't defend against." Rothman says that the upshot regarding disclosure should be: "Judge the work, not the author." Minsky disagrees. "To say that these articles stand on their own, that there's a kind of internal integrity to a study as it's reported, doesn't make sense to me as a procedure that you would want to claim validates the purity of the science." Margaret Dale, associate dean for faculty affairs at Harvard Medical School, says that researchers at her institution are required to disclose their financial interests. Harvard's policy states that a "faculty member publishing or formally presenting research results or providing expert commentary on a subject [must disclose] any financial interest in a business which owns or has a contractual relationship to the technology being reported or discussed or which sponsors the research being reported or discussed." The financial interest is usually described in a footnote in the published article. Dale says that she hasn't received any negative feedback from faculty members over this requirement. In her experience, "as more and more journals are requiring it, it's becoming an expected practice." While disclosure of financial interests is one issue, Blumenthal's findings address the very heart of the publish-or-perish world of science: total output of research papers and their influence on colleagues' work. Blumenthal says one possible explanation for his findings is that "people with mostly industry funding have been unable to attract federal funding, and that means that they probably are on average less productive, that their work hasn't been visible enough to give them credibility with study sections." MISSING THE BOAT: Edgar Haber notes that some contracts with industry can delay publication until information is passé Edgar Haber, the Elkan R. Blout Professor of Biological Sciences at the Harvard School of Public Health, explains the declining productivity as possibly being related to confidentiality agreements between academic scientists and industry. "The way I would interpret [the findings] is that industry-in their contracts with the investigators that they support-tends to delay the publication of information or even to prevent it all together until perhaps patents are filed and the information is pass^Î," he maintains. "So that could conceivably result in some investigators having lesser productivity or less interesting productivity. Or, they might not be able to publish it at all." Haber, a cardiovascular disease researcher with a diverse funding portfolio, including support from Bristol-Myers Squibb, gets around this problem by working closely with Joyce Brinton, director of the Office for Technology and Trademark Licensing at Harvard University. According to Brinton, "we have a general policy that applies across the board to all of our scientists. We cannot accept contractual terms that would delay publication or prevent publication. When and if we're willing to accept any kind of delay, it is a delay for the purposes of allowing the sponsor to ascertain whether something should become the subject of a patent application. This is a delay of 30 days prior to submission of an article for publication, with the option of the investigators agreeing to an additional 30 days if they need extra time to actually file." "We encourage researchers we sponsor to publish; otherwise, we're not going to attract top scientists," maintains Bristol-Myers Squibb's DeForrest. Regarding an article's clout within the scientific community, Rosenberg says that Blumenthal should have better clarified the difference between articles in clinical journals and a paper's degree of influence. "Anything in a clinical journal is going to be considered less influential," she comments. For Blumenthal, the most valuable take-home message to researchers is to maintain a diverse funding base. "I especially think having public funding is key so that there's a part of the work where the premium is placed on openness and dissemination." He doesn't know of any university that has a policy limiting the proportion of a researcher's funding to a certain type. "I think that most people would argue that you don't want to have rigid rules," he states. "I would say that any lab that is funded exclusively by industry for a long period of time at a university deserves very close scrutiny by the managers at that university for what it's contributing to the academic environment." Blumenthal sounds a special note of caution for clinical research. "The clinical side [of science] tends to be more heavily funded by industry than the non-clinical. Clinical scientists are more likely to fall into the category of getting a large proportion of their funding from industry. It's also that group that universities are depending on now to attract new industry funds. They're looking for money from drug trials. So there may come a time when universities have to ask what the purpose of their solicitation of funding is and whether they are providing a service or doing research. If they're providing a service, then I think they ought to separate it out rather than categorizing it as an academic endeavor." Journal TitleTotal Number of original articles on biology and genetics in 1992 Number of original articles with Ma. authors in 1992 Percentage of total number of articles Percentage of database (a) Disclosure of interests required by journal in 1992 American Journal of Human Genetics 27111 4.06 1.36 no Cell 357 51 14.29 6.29 no EMBO Journal 54628 5.13 3.45 no Genes & Development 21727 12.14 3.33 no Journal of Biological Chem. 3,8542265.86 27.87 no Journal of Cell Biology 51642 8.14 5.18 no Lancet 4612.17 0.12 no(b) Molecular and Cell Biology 61051 8.36 6.29 no Nature 53561 11.40 7.52 no Nature Genetics 12015 12.50 1.85 no New England Journal of Medicine 23521.74 0.62 yes(c) Plant Cell 1177 5.98 0.86 no Proceedings of the National Academy of Sciences 2,4542369.62 29.10 no(d) Science 482 50 10.37 6.17 yes(e) The database refers to the cellular and molecular biological and genetics articles published in 1992 and whose first or last author was affiliated with a Massachusetts nonprofit academic or research institution. Lancet's conflict of interest policy, introduced to authors on July 1993, expects authors to list all relevant sources of financial support that could potentially embarrass an author if the grant, business interest, or consultancy became known after publication. New England Journal of Medicine introduced its conflict of interest policy in July 1990. The journal expects that its authors will not have any financial interests in a company or competitor that makes a product discussed in the article. Proceedings of the National Academy of Sciences, as of May 1996, requires that all authors "disclose any commercial association that might be a conflict of interest in connection with the manuscript." Science instituted a policy of requiring authors "to reveal to us any relationship that they believe could be construed as causing a conflict of interest, whether or not the individual believes that is actually so" as of July 31, 1992. Source: S. Krimsky et al., Science and Engineering Ethics, 2:396-410, 1996.

Studies Call Attention To Ethics Of Industry Support

Interested in reading more?

Become a Member of