Financial Conflicts of Interest in Research Involving Human Subjects

 

Content Author

 

Robin N. Fiore, Ph.D.

Florida Atlantic University

 

1. Introduction

There is currently no comprehensive regulatory approach to reporting and managing conflicts of interest. Although the Public Health Service (PHS) has had regulations regarding financial conflicts of interest since 1995 and the Food and Drug Administration (FDA) has had regulations requiring financial disclosure for investigators since 1998, by and large, conflict of interest regulations, guidelines and policies have been oriented around the need to ensure good science. However, recent, high profile, adverse events involving researchers with significant financial interests in the outcomes of their studies have increased concerns about risks that financial conflicts of interest pose to the wellbeing of individuals who agree to participate in human subjects research. There is a growing consensus among investigators, regulators, and the public that existing guidelines have failed to deal adequately with problems associated with conflicts of interest and that additional measures are required to ensure the validity of research and protect human subjects.

 

The following two views represent a continuing, fundamental disagreement about the nature of the conflicts of interest:

 

•"Conflicts of interest are ubiquitous in academic life, indeed all professional life; conflicts of interest can never be eliminated. Moreover, the existence of conflicts of interest has to be accepted and not equated with scientific misconduct." (Korn 2000)

 

•"Financial conflicts of interest are not inherent to the research enterprise. They're entirely optional - unlike intellectual or personal conflicts of interest to which they're often compared." (Angell 2000)

 

On the first view, conflicts of interest are inescapable; what matters is whether professional misconduct actually ensues. On the second view, failing to forgo financial interests that pose conflicts is itself a form of research misconduct. Thus, proposed remedies vary, depending on which of these views is accepted. Nevertheless, the need to manage conflicts of interest cannot await consensus on this underlying dispute. Individual investigators and their institutions must cooperatively address the ethical challenges presented by financial conflicts of interest.

 

This module has the task of describing an issue that is under debate and in transition. It begins with a general discussion of conflicts of interest and the related ethical concerns that arise in the context of research involving human subjects. Subsequent sections summarize current reporting and disclosure requirements for investigators, and discuss the latest (May 2004) Federal guidance on the management of conflicts of interest.

 

2. What is a Conflict of Interest?

Professionals have a conflict of interest when their interests or commitments compromise their judgments, compromise their research reports, or compromise their communications to research subjects, participants, patients, and/or clients. (NHRPAC 2001). Conflicts of interest are of two major types (Rodwin 1993):

 

1.Conflicts between the professional's personal or financial interests and the interests of a subject/participant, patient or client.

2.Conflicts that involve competing loyalties, to two or more subjects, patients or clients. Alternatively, the conflict may be between a subject/participant, client or patient and a third party to whom the professional owes contractual duties, for example, sponsors of research, insurance companies, employers, etc.

 

The term competing interest rather than conflict of interest is preferred by some as a way of lessening any implicit sense of misconduct. Nevertheless, it is non-controversial that significant conflicts of interest increase the likelihood that professionals will abuse their trust. Law and professional ethics routinely require the avoidance of conflict of interest situations as a way of guarding against wrongful acts. Scientific misconduct is not always associated with conflicts of interest; but conflicts of interest increase the chance of scientific misconduct (Bodenheimer 2000). The approach taken in this module is to follow legal usage and distinguish conflicts of interest from actual breaches of obligation (Rodwin 1993).

 

As used here, conflict of interest describes a relationship, commitment or interest that constitutes a moral hazard i.e., something that has the likelihood of producing a moral error. In contrast, a breach of obligation describes an actual violation of a moral or legal duty. These include: violations of norms of professional scientific conduct, violations of human rights, neglect of professional role-related fiduciary responsibilities, and failure to acknowledge the existence of conflict of interest situations.

 

Adopting the distinction between conflicts of interest and breaches of obligation can ease defensiveness in disclosing and discussing the existence of conflicts of interest since their acknowledgment would not be an admission of wrongdoing. Clearly, distinguishing between circumstances that may influence conduct and actual misconduct avoids the error of referring to potential conflicts of interest. It also avoids what most commentators regard as an unhelpful differentiation between having a conflict of interest and (merely) having the appearance of a conflict of interest (Erde 1996).

 

In sum, a conflict of interest is best understood as an objective fact situation in which there is increased potential for harm or wrongdoing as a result of compromised independence.

 

2.1 Financial Conflicts of Interest

The paradigm conflict of interest is financial interest. To be sure, non-financial (or only indirectly financial) forms of bias can pose serious risks to research and to human safety and dignity, but regulations and institutional oversight are primarily oriented to financial conflicts of interest. Significant financial interests must be disclosed to institutional officials and be appropriately managed [Title 42 CFR, Section 50, Parts 604 and 605]. A "significant financial interest," according to the PHS, is one that that could directly and significantly affect the design, conduct, or publication of research and thus bear on issues of human subjects protection. While the PHS defines "significant" in terms of a monetary threshold of a $10,000 interest or 5% ownership in an entity that would reasonably be affected by research. Neither PHS nor FDA spells out types of financial interests that may be held; each grantee institution must develop guidance, policies and mechanisms for reporting and managing conflicts of interest.

 

Financial interests include, but are not limited to:

 

•Compensation from employment (by other than grantee institution).

•Paid consultancy, advisory board service, etc.

•Stock ownership or options.

•Intellectual property rights (patents, copyrights, trademarks, licensing agreements, and royalty arrangements).

•Paid expert testimony.

•Honoraria, speakers' fees.

•Gifts.

•Trips.

 

Proprietary interests (intellectual property rights and equity) are of particular concern, not only because they offer potentially enormous financial rewards, but also because financial gain depends on the study outcome. (Note that investigators may have interests in competitor products not under study and have a conflict of interest despite having no direct financial association in a particular study.) The worry is that if researchers have a stake in the success of commercial ventures, their judgment and objectivity in the conduct of research may be affected - consciously or unconsciously - to the detriment of research subjects' wellbeing.

 

The goal of protecting human subjects is the best guide to determining whether a financial interest is ethically significant or not.

 

Basic questions that help reveal significant financial conflicts of interest are:

 

•Is compensation affected by the outcome of the research?

•Does a financial relationship suggest, to a reasonable person, that the investigator or institution would prefer one study outcome rather than another?

 

For additional questions to be considered in determining the significance of specific financial relationships see the 2004 Final Guidance [http://www.hhs.gov/ohrp/humansubjects/finreltn/fguid.pdf].

 

3. What are the ethical concerns relating to conflicts of interest in research?

There are two important ethical concerns relating to conflicts of interest: the preservation of sound science and the protection of human subjects. Each is discussed separately below, however, they are integrally related. Robust protection of human subjects - and those who use the products of science - depends on ensuring the soundness of research.

 

3.1 Preservation of Sound Science

Conflicts of interest pose a threat to scientific integrity by introducing forms of bias that affect the enterprise of science itself. First, financial relationships among investigators, academic research centers and private industry create incentives to serve commercial interests rather than the advancement of scientific knowledge. For example, sponsors may seek to restrict publication, citing protection of proprietary information, in order to avoid advancing the work of competitors. They may conceal negative study findings by maintaining control of publication, or avoid disclosing adverse events and side effects to the public (though these are disclosed to the FDA). Restricted or partial publication increases the cost of clinical progress and can jeopardize the health of future study subjects and future patients. It also impedes or disrupts the work of other scientists whose work would otherwise improve, build on or impeach prior investigations.

 

Secondly, the soundness of study results can be profoundly influenced by study design decisions: treatments to be tested, e.g., placebo control or active control, favorable and adverse endpoints, the characteristics of eligible and ineligible participants, stopping or modifying a trial, and so on (Brody 1996) . Without an investigator's being aware of it, conflicts of interest may influence the design and conduct of research in ways that render study results unsound, with the potential to misinform the practice of many physicians and to affect the health of patients. Thirdly, the availability of capitation payments - fees paid by study sponsors to physicians for each patient enrolled in a study - may increase the likelihood that basic science research will be less attractive to researchers and institutions than projects with more immediate application preferred by commercial sponsors. To the extent that capitation payments exceed actual overhead costs, the excess or "profit" benefits the investigator and/or the institution by providing a source of discretionary funds. Excessive capitation payments may serve as an inducement for researchers and research institutions to choose projects that are of interest to generous sponsors rather than alternatives that might be of more benefit to patients or society.

 

3.2 Protection of Human Subjects

As relationships between industry and academic medicine become increasingly more complex, clinicians may assume multiple roles of physician, investigator, sponsor, and sometimes, institutional official, and institutional review board (IRB) member as well. The primary role, however, must be that of fiduciary. A fiduciary is a person trusted with acting on behalf of others for their benefit, or in the public interest. Compromised loyalties are inconsistent with acting as a fiduciary. Professionals with fiduciary responsibilities must avoid interests (or roles) that pose a hazard to the welfare of those for whom they are responsible.

 

Human subject recruitment in industry-sponsored trials conflicts with the fiduciary role. Investigators, sponsors, grantee institutions, and physicians in private practice all stand to benefit by a patient's participation in research. Since studies must enroll a sufficient number of subjects to obtain funding, there is an inevitable conflict between potential subject/participant interests and those of investigators. The concern is that investigators - under pressure to recruit - may undermine the consent process by misrepresenting the research or inappropriately influencing patients to participate. The most objectionable financial arrangements in clinical research involve "trading in patients" or their medical information. For example, the practice of paying finders' fees to physicians for enrolling their patients is rightly regarded as ethically equivalent to fee splitting (Lind 1990). According to The American Medical Association, "it is unethical to for physicians to accept payment solely for referring patients to research studies" [Opinion 6.03]. Other recruitment techniques, such as naming private-practice physicians as co-investigators, or retaining them as consultants in order to gain access to their patients, increase the likelihood that financial considerations may influence clinical judgment with respect to whether patients are benefited by being enrolled in a study.

 

4. What are the current requirements applicable to individual investigators?

Before beginning a study, individual investigators are required to disclose financial interests that may be affected by the outcome of research to designated institutional officials. Institutions are required to report the existence of conflicts of interest - but not substantive details - to PHS funding agencies and to take steps to reduce, eliminate or manage conflicts of interest [Title 42 CFR, Section 50: Parts 603 and 604].

 

The FDA requires sponsors and individual investigators to report certain financial arrangements as part of marketing applications for drugs, biologics and medical devices [Title 21CFR, Section 54; Part 4].

 

4.1 Institutional Assurance

The regulations state that information regarding financial conflicts of interest should be obtained from all investigators according to institutional policies and procedures. Thus, research institutions are formally responsible for developing and communicating a process for reviewing, authorizing and monitoring arrangements that present conflicts of interest. However, the specifics are left up to each institution, and vary widely. A national survey of academic medical centers reported that the responsibility for conflict of interest policies tends to be distributed among several administrative units, so investigators are advised to make sure they are in compliance with all of their institution's policies relating to conflicts of interest (GAO 2001). In the event that investigators fail to comply with conditions or restrictions imposed to manage conflicts of interest, the institution must report the delinquency and how it is being handled. Improperly managed conflicts of interest can result in the PHS suspending funding to the offending grantee. The PHS may also require the institution to ensure that investigators disclose the conflicting interest in each public presentation of the results of research [Title 42 CFR Section 50: Parts 604-606].

 

5. What is the impact of the 2004 Guidance on current requirements?

On May 12, 2004, the Department of Health and Human Services (DHHS) published final guidance for IRB's Investigators, research institutions and others entitled "Financial Relationships and Interests in Research Involving Human Subjects: Guidance for Human Subject Protection". The Final Guidance [http://www.hhs.gov/ohrp/humansubjects/finreltn/fguid.pdf] has now been published. The 2004 final document states that it "does not change any existing regulations or requirements, and does not impose any new requirements." It applies to research conducted or supported by HHS or regulated by the FDA and replaces earlier "draft interim guidance." Guidance documents are often used by Federal agencies to augment existing regulations without triggering the cumbersome statutory rule-making process. It is probably right to regard this draft guidance a statement of "best practices" and to incorporate its recommendations into current policies and procedures. The 2004 guidance defines a "conflicting financial interest" as "a financial interest related to a research study that will, or may reasonably be expected to create a bias." It continues to rely on institutional assurance rather than prescription. While clearly anticipating that IRB's and investigators will take more active roles in ensuring that financial interests do not compromise the protection of research subjects, it stops short of prescribing, the approach taken in earlier guidance documents.

 

As a general guideline, the 2004 Guidance states that the Belmont principles − respect for persons, beneficence, justice − should not be undermined by financial interests. In particular:

 

•IRBs should consider whether management of conflicts of interest adequately protects subjects, or actions are required.

•IRBs should determine the level of detail to be provided to subjects regarding sources of funding, funding arrangements, financial interests of parties involved, financial interest management actions, etc.

 

6. What are the primary strategies for eliminating, reducing and managing conflicts of interest?

The primary strategies for managing conflicts of interest are disclosure and prohibition. A number of influential professional societies, researchers and institutions have advocated a total ban on paid consultancies and equity holdings in entities related to their research; some have recommended barring investigators from investments in fields in which they are conducting research. Less drastic approaches include:

 

•Peer review of the study design.

•Independent oversight of the research.

•Insulating investigator from knowledge about the impact of financial interests through blind-trust type devices.

•Insulating the subject/participant from the influence of financial considerations on professional judgment by having an investigator with a conflict abstain from problematic aspects of the study.

•Disclosure of the financial interest to subjects on the consent form.

 

6.1 Disclosure and Consent

Although nearly all research institutions have the requisite policies calling for investigators to disclose "significant financial interests" to designated institutional officials, fewer than 10% of institutions in a national survey required disclosure to funding agencies or to journals; only 1% required disclosure to IRB's and/or subjects (McCrary et al. 2000) .Current PHS and FDA regulations are silent on what if any information about financial conflicts of interest ought to be shared with prospective research subjects. Critics of the current regulations argue that the failure to disclose financial conflicts of interest violates the ethical obligation to provide subjects and potential participants with information that is relevant to the choice to participate. Further, some argue that failure to disclose is fundamentally deceptive and that without complete information about the risks of participation, including the possibility that financial arrangements of the investigator might influence their judgment, consent is invalid. It is worth noting that the World Medical Association's 2000 revision of the Declaration of Helsinki now lists "sources of funding" as information that should be provided to subjects. Others have argued that disclosure merely confuses or disturbs subjects and does nothing to actually eliminate or reduce conflicts of interest. On this view, subjects are unlikely to be able to make effective use of information about conflicts of interest because they are, variously, mistaken about the nature of clinical trials, confused about relative and objective risk, or ill-informed about financial interrelationships among investigators, institutions and industry.

 

Rather than addressing the question of how much transparency is enough, the U.S. National Human Research Protections Advisory Committee (NHRPAC) has argued that research should not be allowed to proceed until the risk from "troubling financial relationships" has been reduced to a level below "significant" through appropriate management strategies. In that event, the consent form need only state, for example, that the review committee believes that there are no conflicts of interest that, taken together with the management strategies agreed to by the investigator, will influence the way the study will be conducted (NHRPAC 2001). Investigators should not rely on disclosure alone to do the work of protecting the rights and welfare of human subjects, as Dr. Greg Koski - the former Director of the Federal Office for Human Research Protections (OHRP) - acknowledged when he declared, "disclosure is not enough in most instances and, yet, at the same time openness is essential."

 

6.2 Prohibiting Problematic Financial Interests

The evidence is persuasive that investigators with financial ties to companies whose products they are studying are much more likely to publish studies favorable to those products (Bodenheimer 2000). Ideally, investigators should be paid on the basis of time and effort, not the outcome of trials. There is a fairly broad consensus in favor of prohibiting academic researchers and their families from having financial interests in companies sponsoring their research, in companies that manufacture a product or device being tested, and in companies that manufacture competing products. The difficulty is that the current regulatory approach that relies on individual research institutions to assure that conflicts of interest are managed cannot ensure a level playing field, i.e. that policies are consistent across institutions. Without binding regulations or voluntary agreements, institutions fear that they risk losing investigators to institutions with more lenient policies.

 

Some have recommended that prohibitions on stock ownership and stock options be imposed only on investigators who are responsible for sensitive elements of the study process: the selection of subject-participants, obtaining informed consent, and clinical management. Critics have responded that bias also may occur in the design of the study itself, and in the interpretation of study results, and so have endorsed broader prohibitions on the ownership of stock/stock options (Lo et al. 2000).

 

There is general agreement that the most problematic financial interests are equity holdings and paid consultancies. However, internal compensation to investigators and institutions from grants can be just as significant a source of conflicts of interest (Brody 1996). Since grants typically cover more than the marginal costs of research, anything that reduces grant funding, such as decisions to stop a trial early or failure to enroll enough patients with the consequent elimination of a center from a trial, have both financial and career implications.

 

An alternative to outright prohibition of "profit" that exceeds true direct and indirect costs - whether from excess capitation payments/enrollment bonuses or overhead - is to insulate the investigator from the surplus. An alternative suggested by The Institute of Medicine is that the excess be placed in a pool administered by the institution for all researchers, rather than allowing the investigator or the investigator's unit to appropriate it.

 

7. Conclusion

The goal of managing conflicts of interest is to minimize the extent to which the design and conduct of research is influenced - consciously or unconsciously - by financial considerations. However, merely complying with the letter of current conflict of interest regulations and institutional policies is unlikely to achieve the twin objectives of maximizing protection of human subjects and ensuring sound science. Responsible conduct of research calls for investigators to exercise their best judgment when entering into financial relationships that constitute conflicts of interest, to acknowledge non-financial biases, and to disqualify themselves when appropriate.

 

Reference List

 

Angell, M. (2000). Remarks. HHS Conference on Human Subject Protection and Financial Conflicts of Interest: August 16-17, 2000, NIH Campus, Bethesda MD.

 

Bodenheimer, T. (2000). Remarks. HHS Conference on Human Subject Protection and Financial Conflicts of Interest: August 16-17, 2000, NIH Campus, Bethesda MD.

 

Bodenheimer, T. (2000). Uneasy Alliance--Clinical Investigators and the Pharmaceutical Industry. New England Journal of Medicine 342(20): 1539-1544.

 

Brody, B. A. (1996). Conflicts of Interest and the Validity of Clinical Trials. In Conflicts of Interest in Clinical Practice and Research. J. Roy G. Spence, D. S. Shimm and A. E. Buchanan. New York, Oxford University Press: 407-417.

 

Erde, E. L. (1996). Conflicts of Interest in Medicine: A Philosophical and Ethical Morphology. In Conflicts of Interest in Clinical Practice and Research. J. Roy G. Spence, D. S. Shimm and A. E. Buchanan. New York, Oxford University Press: 12-41.

 

GAO (2001). Financial Conflicts in Biomedical Research: Report to the Ranking Minority Member, U.S. Senate Subcommittee on Public Health, Committee on Health, Education, Labor, and Pensions. Washington, DC, U.S. General Accounting Office: 1-39.

 

Korn, D. (2000). Conflicts of Interest in Biomedical Research. Journal of the American Medical Association 284: 2234-2237.

 

Lind, S. (1990). Finder’s Fees for Research Subjects. New England Journal of Medicine 323: 192-195.

 

Lo, B., L. E. Wolf, Berkeley, A. (2000). Conflict of Interest Policies for Investigators in Clinical Trials. New England Journal of Medicine 343: 1616-1620.

 

McCrary, S.V., Anderson, C.B., Jakovljevic, J., Khan, T., McCullough, L.B., Wray, N.P., Brody, B.A.. (2000). A National Survey of Policies on Disclosure of Conflicts of Interest in Biomedical Research. New England Journal of Medicine 343: 1621-1626.

 

National Human Research Protections Advisory Committee (2001). NHRPAC Recommendations on HHS’s Draft Interim Guidance on Financial Relationships in Clinical Research.

 

Rodwin, M. A. (1993). Medicine, Money and Morals: Physicians Conflicts of Interest, Oxford University Press.