When Conflicts of Interest Threaten Scientific Integrity

The growing commercialization of scientific research has increasingly forced biomedical publishers to grapple with real and perceived conflicts of interest. Every scientist has a competing interest when it comes to getting published in a high-profile journal—a prestigious publication record brings status, research funds, job security, and other personal benefits—but more and more scientific research is funded by companies with a vested interest in the outcome of that research.

A 2003 systematic review of 37 high-impact biomedical journals published in JAMA concluded that “Financial relationships among industry, scientific investigators, and academic institutions are widespread. Conflicts of interest arising from these ties can influence biomedical research in important ways.” Yet, the authors found, in 2001 just 43% of the journals surveyed had policies requiring disclosure of conflicts of interest.

PLoS defines a competing interest as "anything that interferes with, or could reasonably be perceived as interfering with, the full and objective presentation, review, or publication of research findings, or of articles that comment on or review research findings. Competing interests can be financial, professional, or personal; hidden or declared; actual or perceived." PLoS editors may decide not to publish original research or other articles if the editors believe the competing interests may have compromised the work, analyses, or interpretations presented in the paper.

Though the PLoS competing interest policy applies to all PLoS journals, the potential for industry corruption of science raises the most concern among the medical editors. The main difference between medical and basic biology journals is that the research results in medical journals typically have greater practical application. Medical results often have implications for health policy and clinical practice, and might affect the viability of a prospective commercial product when, for example, a study suggests that a prospective drug or chemical in mass circulation poses a risk to human health. For the most part, results published in a more basic biology journal like PLoS Biology do not have such direct implications and potential conflicts. There are exceptions, of course – for example, a set of results might doom a planned development project because it threatens an endangered species.

Industries with a substantial financial interest in the outcome of a study often do everything in their power to protect their interests by funding their own studies and publishing only those results they like, for example, or by manufacturing uncertainty around studies reporting results they don't like, or by attempting to influence the promulgation or implementation of policies and rules at the appropriate regulatory agencies: FDA (drugs), FWS (endangered species), or EPA (toxics). Biases arising from industry-sponsored studies are well documented in clinical medicine. The lastest evidence comes from a recent PLoS Medicine study of potential biases in drug-drug comparisons, which found that randomized controlled trials comparing statins with other drugs are more likely to report results and conclusions favoring the sponsor's product over the other drug in the study.

But knowing a study is industry funded isn't enough. More important is having independent reviewers evaluate the validity of the study design, methods, protocols, and determine whether the conclusions are consistent with the data. This is how rigorous peer review is supposed to work. But peer review can miss fatal flaws in studies—even when authors provide disclosure statements. That's why many researchers working at the intersection of science and policy want to see transparent review processes and independent scientific review panels oversee research results destined for use in policymaking (or drug approval) decisions.

In "Why Not the Best: How Science Failed the Florida Panther" (which I wrote), former Conservation Biology editor Reed Noss argued that endangered species management policies should be modeled after California law, which requires independent scientific review at several stages of the conservation planning process, “so consultants working for counties or developers are not able to get away with using flawed scientific methodologies.” The published reports of such independent reviews can help the public navigate an otherwise impenetrable, seemingly arcane debate.

A new feature article published this week in PLoS Biology called “The Toxic Origins of Disease” (which I also wrote) describes how the chemical industry hired scientists to "replicate" the work of a researcher who reported that in utero exposures to a mass-produced chemical (called bisphenol A) caused developmental and reproductive defects in mice. Industry scientists found no such effects, but when an independent review panel assessed their studies, they found fatal flaws in both study design and interpretation of the results.

Toxicology journals, like medical journals, have a long history of publishing studies biased toward industry interests. Just as a 2003 study published in the British Medical Journal found that “studies sponsored by pharmaceutical companies were more likely to have outcomes favoring the sponsor than were studies with other sponsors”, a 2005 commentary published in Environmental Health Perspectives found that of 115 published studies concerning effects of low doses of bisphenol A in experimental animals, 94% of publicly funded studies found evidence of harm while 100% of chemical industry studies found no evidence of harm.

When researchers have a vested interest in the outcome of a research question, readers have a right to know that information. The 2003 JAMA study reported that industry investment in US biomedical research and development almost doubled between 1980 and 2000, as federal support dropped. As the line between science and business becomes increasingly blurred, transparency about conflicts of interest is the minimum requirement to give readers all the information they need to evaluate scientific results. The Center for Science in the Public Interest has been leading efforts to manage conflicts of interest that allow corporations to exploit the credibility of peer-reviewed journals to advance their own interests. CSPI urges all journals to adopt a strong policy regarding disclosing conflicts of interest and publishing those disclosures “to allow scientists, the public, and policy makers to make more informed judgments about research reports, letters, commentaries, editorials, book and literature reviews, and news articles, and to safeguard the credibility of scientific peer review.” Although such dislosures can't guarantee that policymakers will base their decisions on the best scientific evidence available, they at least flag studies that may require more careful scrutiny.