Is Sunlight Always the Best Disinfectant?
Are Financial Conflicts of Interest in Research Inherently Unhealthy?
SOURCE: iStockphoto
If the end goal is to encourage high quality science, we need to better understand the impact of financial conflicts of interest and get more information about whether existing policies to manage them are effective.
The past few years have seen an intensified focus on financial conflicts of interest in medical research. In efforts to remedy the problem, Congress and state legislatures have introduced or enacted laws addressing these issues,[1] organizations such as the American Association of Medical Colleges, or AAMC, have issued reports, and many individual institutions have developed and implemented their own policies.[2, 3]
The concern about possible conflicts of interest in medical research is not new. As far back as 1984, a study of biotechnology companies revealed that nearly one half of the existing firms were funding research in universities.[4] Other articles on the subject appeared in the medical literature in the early 1990s.[5] The National Bioethics Advisory Commission’s 2001 report on human subjects research also addressed the relevant issues.[6] Moreover, since federal regulations regarding conflicts of interest have been in place for many years, commercial sponsors have long been developing strategies to handle conflicts of interest.[7]
It is important to remember that many of the conflicts or potential conflicts that trouble industry watchdogs have nonetheless proven enormously valuable.
But despite all this research into the nature of potential conflicts and the development of policies to mitigate them, the discussion remains a dissonant array of voices. There is no consensus about what constitutes a financial conflict. There is also little empirical information about whether financial conflicts of interest hurt research quality or patient care. While some scholars have expressed grave concerns about industry-financed research, many benefits stem from the close relationship between industry and academia. In addition, we do not know whether mechanisms often used to address these conflicts effectively protect patients and research integrity or inadvertently hurt research efforts. More data are needed before further policies designed to manage conflicts are enacted.
The primary reason for the recent emphasis on conflicts of interest is the increased role of industry in financing clinical research. By the year 2000, 70 percent of the money for clinical drug trials in the United States was coming from industry.[8] Industry critics say policies to facilitate technology transfer of new drugs from the lab to the market as well as the financial pressure confronting academic institutions have led to inappropriately close relationships between academia and industry.[9]
A survey of the scientific literature, conducted in 2003, found that financial relationships among industry, investigators, and research institutions are “pervasive.” At the time of its publication, 23 percent to 28 percent of academic investigators were receiving research funding from industry.[10]
Many observers have argued that the dependency of academic research on industry dollars causes conflicts (or the appearance of conflicts), which diminishes patient care and research quality, compromises the integrity of the research infrastructure and affects the public trust in research objectivity.[11] They say financial relationships create conflicts between researchers’ obligations to follow scientific principles and their desire for financial gain. Studies have found correlations between financial relationships with industry and problematic trends such as publication of “pro-industry” results, and restraints on publication and data sharing.[12]
At the same time, it is important to remember that many of the conflicts or potential conflicts that trouble industry watchdogs have nonetheless proven enormously valuable. Financial relationships between academia and industry help bring new drugs to the market for patients and fuel economic development in states or regions. They also increase research budgets, supplementing funds obtained elsewhere.[13]
The federal government recognized these potential benefits almost 30 years ago, with the enactment of the Bayh-Dole Act.[14] The Act permitted universities and small businesses to own inventions made as a result of federal funding through patenting and authorized licensing to industry. The policy objective was to encourage the licensing of inventions developed in universities to industry, which would commercialize them into products for patients.
The Act has been enormously successful. The incentives it enabled have spurred research, led to the creation of new jobs, and facilitated the development of important public health products, including a Hepatitis B vaccine, human growth hormones, and synthetic penicillin.[15]
Congress recognized these successes through a resolution that passed the House of Representatives on December 6, 2006, which states:
The Bayh-Dole Act (Public Law 96-517) has made substantial contributions to the advancement of scientific and technological knowledge, fostered dramatic improvements in public health and safety, strengthened the higher education system in the United States, served as a catalyst for the development of new domestic industries that have created tens of thousands of new jobs for American citizens, strengthened States and local communities across the country, and benefited the economic and trade policies of the United States.[16]
Thus, we must consider the benefits, along with the dangers, of allowing industry funding for academic research. When addressing financial conflicts of interest, private and government policymakers must balance two competing objectives—maintaining research integrity and promoting important and beneficial collaborations.
Disclosure and prohibitions
The first challenge in addressing conflicts of interest is to analyze and define the sort of relationships that may cause a problem. Harvard University professor Dennis Thompson argued in the New England Journal of Medicine that the primary concerns in regulating conflicts of interest are maintaining the integrity of professional judgment and minimizing the influence of factors potentially leading to impropriety, such as financial gain.[17]
According to Thompson’s model, institutions must assess whether a given scenario presents a significant threat to research independence through consideration of a number of factors in sequence. These include: the value of the service that could be compromised, the accountability of the physician/researcher, the duration and closeness of the relationship between the research and the compromising interest, and the independence enjoyed by the researcher. Effective analysis of a potential conflict therefore depends on a thorough disclosure process.
Federal law does not prohibit—or even define—a conflict of interest.
In practice, there is near-universal agreement that disclosure processes are necessary for financial conflicts. Many organizations, professional societies, and governmental bodies have argued for the use of disclosure of financial interests as a mechanism to address conflicts.[18] Federal guidelines and regulations call for similar transparency and many institutions follow this approach and use specific, required disclosures as the key component of their policies.[19, 20]
Unfortunately, the institutional processes for disclosure are far from settled. Researchers have found wide variation in the content and requirements of institutions’ conflict of interest rules.[21] In addition, 40 percent of medical schools that responded to a 2004 survey by the AAMC do not require researchers to disclose significant financial interests in oral presentations of research results.[22] To make matters more complicated, a 2002 NIH survey found that 86 percent of responding institutions did not define the word “research” in their conflict-of-interest policies.[23]
However, nearly all policies identify a specific dollar amount as the threshold for a potential financial conflict of interest, which must therefore be reported. The AAMC found that 95 percent of institutions’ policies based their reportable compensation requirements on federal guidelines, which require disclosure of ties of $10,000 or more.[24]
Food and Drug Administration regulations add to the complexity and variety of approaches. Generally speaking, federal law only applies to those potential conflicts that could affect the reliability of the data in a marketing application submitted to the FDA. They focus on the bias that could arise from an investigator’s financial interest in the outcome of a study because of the way payment is arranged, because the investigator has a proprietary interest in the product, or because the researcher has an equity interest in the company sponsor of the study.
Federal law does not prohibit—or even define—a conflict of interest. Rather, FDA regulations address issues surrounding financial disclosure by investigators.[25] Sponsors must disclose financial arrangements with clinical investigators and certain interests of the clinical investigators in the product under study or in the sponsor. These interests are defined differently than those identified by many institutions. Sponsors must disclose investigator equity interests of $50,000 or more or if they pay $25,000 or more to the investigator or institution beyond the cost of the trial or other clinical studies. The FDA then uses this information as part of its assessment of the reliability of the clinical data represented.[26]
Some observers have argued that disclosure is an insufficient tool for management of financial conflicts of interest. For example, Dennis Thompson, mindful of the near-impossibility of discovering researchers’ motives, contends that it is safer and more ethically responsible to simply eliminate as many factors as possible that could serve as distractions to researchers and institutions from their intended medical and scholarly goals. Marcia Angell, MD, former editor-in-chief of The New England Journal of Medicine, has specifically suggested bans on equity interest and certain writing and speaking arrangements.[27] Moreover, some groups are encouraging federal and state governments to implement laws that completely ban financial relationships between industry and researchers.[28]
The FDA recently adopted a policy that prohibits certain financial relationships as part of its implementation of the Food and Drug Administration Amendments Act of 2007. In August 2008, the agency issued guidelines that, except in rare circumstances, prohibit a researcher’s participation on an Advisory Committee if that individual has a financial interest of $50,000 or more in the product being examined.[29]
Are these techniques effective?
Although disclosure is widely used as a technique to address conflicts, there is little empirical data about its effectiveness and impact. In fact, some published analyses of university financial disclosure policies cast doubt on whether a disclosure-based approach will effectively protect research participants and research integrity. These studies raise questions about whether such policies would truly help potential research participants make better-informed decisions as to whether to participate in a research project, or effectively deter researchers from accepting problematic financial conflicts.[30]
Subsequent studies further indicate that the disclosure of a researcher’s financial interests in a clinical trial may not affect a patient’s willingness to participate in the study, unless the researcher could earn money contingent upon the trial results.[31]
In an additional warning as to the possible dangers of overemphasized disclosure, a recent government report found that the FDA’s financial reporting requirements impeded the agency from recruiting scientific experts to its Advisory Committees.[32]
With regard to more distinct prohibitions of financial relationships, observers have noted that completely eradicating conflicts of interest is unlikely and could be counterproductive to the current realities of today’s market-driven environment. They argue that conflicts must be detected, analyzed, and managed in a sensible and effective manner, and that university and private organizations’ internal policies as well as state and federal statutes and regulations should be congruent with these objectives.[33]
What’s next?
Since relationships between industry, institutions, and researchers are frequently beneficial, it seems imprudent to prohibit them. Bright line tests and dollar thresholds that automatically trigger prohibitions do not account for the complexity and value of relationships among industry and academia. Rather, mechanisms must be developed to ensure the safety of research participants and the integrity of research, while promoting—or at least not stifling—productive relationships. Balancing these competing interests is critical.
So how to do it? First, agencies must enforce existing rules. Reports continue to indicate that financial conflicts of interest are not sufficiently monitored. A recent study by the Office of the Inspector General at the Department of Health and Human Resources found that the FDA does a poor job of policing the financial conflicts of physicians who conduct clinical trials.[34]
Second, we need more consistency. There is no consensus about what constitutes a conflict or even how to decide what constitutes a conflict. Disclosure, as a tool, has value, but that utility may be limited in its practical consequence, since data suggests this glut of information doesn’t greatly influence the decisions of most research participants in many situations.[35]
Third, there is no consensus about what financial information should be disclosed and the best method for that disclosure. Nor is there agreement about whether defining certain financial conflicts in terms of a dollar limit is effective; and if so, what those amounts should be. Perhaps reflecting this lack of consensus, disclosure requirements vary widely among institutions, and between institutions and FDA, resulting in a confusing patchwork quilt of standards. This, in turn, weakens the impact of attempted enforcement, and harms the public trust.[36]
Thus, prior to broad statutory or regulatory changes, more research is necessary. There is little empirical data about financial relationships between commercial sponsors, investigators, and institutions—and decision makers need more information about techniques for addressing conflicts. Existing analyses suggesting specific methodologies are few, and they largely focus on disclosure of financial conflicts. It is also important to note that financial ties merely represent one possible type of conflict in the research environment. Other types of conflicts that have been identified include: “the desire for faculty advancement, to compete successfully and repetitively for sponsored research funding, to receive accolades from professional peers and win prestigious research prizes, and to alleviate pain and suffering.”[37] More tools are needed for identifying, evaluating, and addressing these other conflicts.
Recent empirical studies and calls for more consensus in standards are a good start towards effective policy, but more work is needed. In the meantime, efforts meant to limit relationships, if taken too far, could disrupt a research enterprise that has led to products that have cured and treated millions of patients. The success of the Bayh-Dole Act demonstrates convincingly the value of industry’s role in research. Meanwhile, the prospects of prohibiting conflicts are increasingly dim, and exhaustive disclosure of financial interests may have a surprisingly limited effect on patient decision making. It therefore seems that researchers have considerable work ahead in coming to an effective approach to conflicts of interest between industry and academic medicine. While finding a way to handle conflicts in a manner that preserves research integrity and maintains the pubic trust is critical, few universalizable standards have emerged from the ongoing debate. As such, policymakers should tread lightly until they have more data.
Michael Werner is President of The Werner Group, a Washington, DC-based consulting firm. He represents life sciences companies, patient advocacy groups, and research institutions and is an expert on regulatory and bioethics issues in research and can be reached at: mwerner [at] thewernergroup.net. Ari Stern is a Research Associate for The Werner Group. The views expressed are the authors’ own.
Endnotes
[1] See for example, The Physician Payments Sunshine Act of 2009, S. 301, 111th Congress, 1st Session.
[2] “Protecting Subjects, Preserving Trust, Promoting Progress: Policy and Guidelines for the Oversight of Individual Financial Interests in Human Subjects Research, Task Force on Financial Conflicts of Interest in Clinical Research,” AAMC (December, 2001).
[3] Ehringhaus, S, and Korn, D. “U.S. Medical School Policies on Individual Conflicts of Interest, Results of an AAMC Survey,” AAMC publication (September, 2004).
[4] Blumenthal, D, et al.. “Industrial support of university research in biotechnology,” Science 231, no. 4735 (January 17, 1986): 242-246.
[5] See for example, Brennan, T. A, “Buying editorials,” New England Journal of Medicine 331, no. 10 (1994):673-5, (1994).
[6] “Ethical and Policy Issues in Research Involving Human Participants,” Report and Recommendations of the National Bioethics Advisory Commission (August, 2001).
[7] Werner, MJ and Price EP, “Managing conflicts of interest: a survival guide for biotechs,” Nature Biotechnology 25, no. 2 (February 2007): 161-163 .
[8] Bodenheimer, Thomas, “Uneasy Alliance — Clinical Investigators and the Pharmaceutical Industry,” New England Journal of Medicine 342, no. 20 (May 18, 2000): 1539-1544 .
[9] Angell, M., “Is academic medicine for sale?” New England Journal of Medicine 342 (2000): 508-510.
[10] Bekelman, JE, et al, “Scope and Impact of Financial Conflicts of Interest in Biomedical Research,” JAMA 289 (2003): 454-465.
[11] Tereskerz PM, Moreno J., “Ten steps to developing a national agenda to address financial conflicts of interest in industry sponsored clinical research,” Accountability in Research 12 (2005):1-17.
[12] Bekelman, JE, “Scope and Impact of Financial Conflicts of Interest in Biomedical Research.”
[13] Johnston, J, “Conflict of Interest in Biomedical Research,” Bioethics Briefing Book, The Hastings Center (October 2008).
[14] P.L. 96-517, codified at 35 U.S.C. §§200-12.
[15] “The Bayh-Dole Act: Important to our Past, Vital to our Future,” available at http://www.autm.net/Content/NavigationMenu/About/PublicPolicy/BDTalkPts031407.pdf (accessed on January 16, 2009).
[16] H. Con. Res. 319, 109th Congress, 2nd Session.
[17] Thompson, D, “Understanding Financial Conflicts of Interest”, New England Journal of Medicine Volume 329, no. 8 (August 19, 1993): 573-576.
[18] Weinfurt, KP, Dinan, MA, et al, “Policies of Academic Medical Centers for Disclosing Financial Conflicts of Interest to Potential Research Participants,” Academic Medicine 81, no. 2 (February 2006): 113-118.
[19] 40 CFR Part 50, Subpart F and 21 CFR 54.
[20] Ehringhaus, S, and Korn, D. “U.S. Medical School Policies on Individual Conflicts of Interest, Results of an AAMC Survey.”
[21] Weinfurt, KP, Dinan, MA, et al, “Policies of Academic Medical Centers for Disclosing Financial Conflicts of Interest to Potential Research Participants.”
[22] Ehringhaus, S, and Korn, D. “U.S. Medical School Policies on Individual Conflicts of Interest, Results of an AAMC Survey.”
[23] National Institutes of Health, Office of Extramural Research “Conflict of Interest Objectivity in Research–NIH Review of Financial Conflict of Interest Policies of Grantee Institution,” available at http://grants.nih.gov/grants/policy/coi/nih_review.htm (last updated July 18, 2002, last accessed February 2, 2008).
[24] Ehringhaus, S, and Korn, D. “U.S. Medical School Policies on Individual Conflicts of Interest, Results of an AAMC Survey.”
[25] 21 CFR 54.1.
[26] 21 CFR 54.
[27] Angell, Marcia, “Is Academic Medicine for Sale?” New England Journal of Medicine 342, no. 20 (May 18, 2000):1516-1518,
[28] Prescription Project Model Legislation, “Prescription Drug and Medical Device Marketing Restrictions and Disclosure Act,” NLARx (January 18, 2008).
[29] Guidance for the Public, FDA Advisory Committee Members, and FDA Staff on Procedures for Determining Conflict of Interest and Eligibility for Participation in FDA Advisory Committees, US Food and Drug Administration, available at http://www/fda/gov/ohrms/dockets (August 2008).
[30] Weinfurt, KP, Dinan, MA, et al, “Policies of Academic Medical Centers for Disclosing Financial Conflicts of Interest to Potential Research Participants.”
[31] “Knowing Doctor’s Financial Interests Doesn’t Deter Clinical Trial Participants,” Science Daily, April 2, 2008.
[32] FDA Advisory Committees, “Process for Recruiting Members and Evaluating Potential Conflicts of Interest, US General Accounting Office Report,” GAO-08-640 (September, 2008).
[33] Korn, D, “Conflicts of Interest in Biomedical Research,” JAMA 284 (Nov 2000): 2234 – 2237.
[34] Harris, G, “FDA is Lax on Oversight During Trials, Inquiry Finds,” The New York Times, January 12, 2009, page A10.
[35] Weinfurt, et al found that researcher equity interests in the project sponsor may affect patient decision-making and trust. KP Weinfurt, MA Hall, MA Dinan, V Depuy, JY Friedman, JS Allsbrook, J Sugarman, “Effects of Disclosing Financial Interests on Attitudes Toward Clinical Research,” Journal of general internal medicine: official journal of the Society for Research and Education in Primary Care Internal Medicine, 23, no. 6 (2008): 860-6.
[36] Tereskerz PM, Moreno J., “Ten steps to developing a national agenda to address financial conflicts of interest in industry sponsored clinical research.”
[37] Korn, D, “Conflicts of Interest in Biomedical Research.”
Comments on this article
By clicking and submitting a comment I acknowledge the Science Progress Privacy Policy and agree to the Science Progress Terms of Use. I understand that my comments are also being governed by Facebook's Terms of Use and Privacy Policy.
