Cohen, Jordan J. MD
Headlines across the country electrify the nation almost daily with promises of a healthier tomorrow. Medical research is enjoying a golden age, making earth-shattering discoveries at a remarkable clip, and the public and policy-makers are responding enthusiastically. Support from private donors and public coffers alike is at an all-time high. Who could doubt that researchers at America's medical schools and teaching hospitals are on the side of the angels?
But what about those other headlines? The ones suggesting that those angelic investigators sometimes have devilish motives—motives stemming from a personal financial stake in their research; motives that are in direct conflict with their duties as disinterested scientists and responsible physicians? These latter headlines are raising some worrisome questions for all of us. Do financial conflicts of interest pose a threat to the objectivity of clinical research? Do they pose a threat to the safety of human subjects who participate in clinical research? Do they pose a threat to public trust in the integrity of clinical research? If the answer to any one of these questions is “yes,” we should indeed worry, and worry a lot.
I plan to address each of these concerns, but before I begin, I want to give you readers a homework assignment. Knowing that this topic—financial conflicts of interest in clinical research—is much too complicated for me to do justice to here, I urge you to go back and reread the AAMC's Guidelines for Dealing with Faculty Conflicts of Commitment and Conflicts of Interest in Research. Although published ten years ago, this monograph remains highly relevant and offers an excellent primer on the subject. The full text is available on our Web site at 〈http://www.aamc.org/〉, and copies are still in print.1
UNDERSTANDING CONFLICTS OF INTEREST
For now, though, let's begin with some definitions, starting with the term conflict of interest itself. All students of the subject agree that the term denotes a state of affairs, not a kind of behavior. This distinction is extremely important because it means that having a conflict of interest is not, ipso facto, a manifestation of improper behavior, much less scientific misconduct. Never mind that media reports on the subject often leave the impression that any investigator who has a conflict of interest of any kind is thereby guilty of, at the very least, bias, if not outright fraud. Quite the contrary, conflicts of interest are commonplace—ubiquitous, in fact, in complex settings such as those of academic medicine. We all have them, as do a growing number of our institutions.
A conflict of interest exists whenever an individual or an institution has a primary allegiance that requires certain actions and, simultaneously, has a secondary interest that (1) could abrogate that primary allegiance and (2) is sufficiently tempting to raise a reasonable possibility that it might actually do so. Hence, a conflict of interest is like potential energy; it has the capacity to cause something to happen, but until unleashed, is simply a lurking presence.
For an investigator, a conflict of interest exists when the prospect of some personal advantage is strong enough to pose a realistic possibility that he or she might compromise the researcher's primary obligation to adhere to sound scientific procedures in an unbiased search for the truth.
Financial Conflicts of Interest
Do conflicts of interest involving money pose any more of a threat to scientific objectivity than do those involving non-financial interests? What about one's self interest in career advancement, or in being recognized by peers, or in garnering the next National Institutes of Health (NIH) grant, or in publishing one's work in leading journals? Each of these non-financial interests poses its own temptation and, thus, can come into conflict with a researcher's primary commitment to maintain scientific objectivity, no matter what.
Like money, non-financial self-interests are double-edged. They have great potential to motivate individuals to do their very best work. But we know that they also have the potential to compromise judgment and undermine objectivity. Scientific objectivity is at perpetual risk from any investigator who has a strong desire for affirmative research results—whether it's to validate one's preconceptions, or to make one's data more publishable, or to garner praise from one's colleagues. The challenge to investigators has always been to recognize that keeping one's own interests in check is the very essence of academic integrity.
But I think we need to go further here. I would argue that conflicts of interest involving money, especially money from external sources, are distinctly different from all other potential conflicts, and thus warrant at least a consideration of the need for special, targeted safeguards.
▪ First is the fact that external financial interests are discretionary. As Kassirer and Angell2 have noted, no faculty member is obliged to enter into an external relationship that promises material gain. In each case, a choice is made whether to do so or not. By contrast, pursuing academic advancement is not optional for any faculty member who wishes to remain in good standing.
▪ Second, external financial interests are, in general, quite easy to quantify. Unlike the qualitative incentives at play within the academic advancement game, dollar values can be readily assigned to calibrate the magnitudes of most external financial incentives.
▪ Third, external financial interests are very easy for the public to understand. Everyone understands money and its ability to motivate—as a stimulus to do good works, as well as a lure to wander from the straight and narrow. By contrast, the non-financial sources of self-interest under which faculty labor are largely invisible to the general public and, even when on display, are well accepted as being intrinsic to the academic culture.
▪ Fourth, the economic interests of many universities, which create both opportunities and explicit incentives for faculty to pursue financial ends, have fostered an atmosphere quite distinct from that of yesteryear, when universities and their faculties were expected to pursue knowledge only for knowledge's sake—as an end in itself. But passage of the Bayh-Dole Act in 1980 and the prospects of substantial revenues from patentable discoveries have induced many ivory tower institutions to eagerly embrace the commercial world, especially when other sources of funds fall short of what is needed to support many promising research opportunities.
▪ And finally, for many if not most people, the prospect of making a lot of money legally from one's own intellectual effort is simply irresistible. “Threadbare and genteel” may have a noble ring to it. But, we are all brought up—certainly in this country—to believe that wealth is a legitimate reward for anyone who can devise a product or service that people deem valuable. In this respect financial self-interest operates in quite a different realm of the human psyche than does the more cultivated desire for academic distinction, even though that distinction typically brings some financial rewards along with it.
Well, you say, what's the worry? So, conflicts of interest are an unavoidable part of faculty life. So, financial interests carry a special warning label. So what? You're right, none of this means that we necessarily have anything to worry about. And you're right if you say that we already have safeguards in place to minimize inadvertent error as well as any willful abuse that might stem from self-interest of any kind.
The first line of defense as far as scientific objectivity is concerned is, of course, the scientific method itself. By calling for such precautions as proper controls, double-blind designs, and independent replication of key research findings, the architects of our time-honored scientific methodology meant expressly to minimize the influence of investigator bias on research results.
But even more important, in my view, is the intrinsic goodness of most people, certainly of those who select medical research as a career. Most investigators I know are highly ethical individuals—individuals who are strongly motivated by altruistic ideals. They want primarily to do good, and only secondarily to do well. Their instinctive bias toward integrity has, over the years, given rise to a set of powerful ethical norms to which our community manifestly aspires. Many of those norms have been institutionalized and constitute still another set of safeguards that are evident in the peer-review processes employed by journal and granting agencies.
Additionally, in the case of commercially sponsored research, we have the daunting series of regulatory hurdles that must be cleared before a new drug or device can be brought to market. The financial consequences—to say nothing of the ruined reputations—from running afoul of the oversights provided by the approval processes of the Food and Drug Administration, for example, exert enormous discipline on all investigators involved in clinical trials, especially the large number that require multi-center cooperation.
Nevertheless, as Bodenheimer3 and several others4–6 have documented, published reports of industry-supported drug trials are much more likely to favor the study drug over conventional therapy than is the case for non-industry-supported trials. In addition, some industry sponsors have clearly sought to inhibit free exchange and prompt and uncensored publication of research findings, thereby undermining a fundamental tenet of sound science. Thus, despite the safeguards currently in place, I think we do need to worry about the threat posed by external financial interests to objectivity in the conduct of clinical research.
And what about the second worry I raised at the outset? Should we worry that financial conflicts of interest are threatening the safety of human subjects who participate in clinical research? This is the worry, of course, underlying those headlines that burst on the scene in the aftermath of the unexpected deaths last year of patients enrolled in gene therapy trials. Media reports of those events are still reverberating throughout the country, and certainly in Washington. Whether journalists intended it or not, their reports left the strong impression in many people's minds that investigators with a financial stake in the outcome of their research had exercised faulty judgment and had placed their human subjects at unnecessary risk.
Reasonable questions were raised about whether the investigators had adhered fully to established inclusion criteria in selecting patients for their trials. Other questions were raised about whether the investigators had divulged all of the relevant facts to prospective subjects during the informed consent process.
People differ, of course, as to whether the evidence supporting these speculations rises to the level of certainty. But is that the point? Are we supposed to wait for irrefutable evidence that some patient died because an investigator with a financial conflict of interest was so blinded by greed that he or she failed to do all that could have been done to prevent the death? Or does the understandable public perception that such a nightmare is possible necessitate action now?
Which brings me to my third worry. Should we worry that financial conflicts of interest are posing a threat to public trust in the integrity of clinical research? Not just in the all-important matter of human-subjects protection, but as well in the more mundane matters of trustworthy research results, of valid conclusions, of reliable advice about prevention and treatment of disease? I confess that I do not know whether it's happened yet—whether all of the allegations to date relating to financial entanglements of clinical investigators have, in aggregate, eroded public confidence in the enterprise.
Perhaps we can take some comfort from the dozens of patient advocacy groups that willingly signed on to our proclamation to Capitol Hill last year reaffirming the shared commitment of patients and investigators to the ethical conduct of research.7 And from recent polls indicating that 95% of Americans still think that clinical research is critical for finding cures. Although we may be heartened by these reflections of today's level of public trust, I'm not persuaded to sit tight. The issue for me is the future.
I needn't belabor what's at stake here. Public trust is what fuels public support for medical research. Whether that research is funded by the the National Science Foundation, the Department of Defense, the Centers for Disease Control and Prevention, the Agency for Healthcare Research and Quality, or, for that matter, by some commercial enterprise, the ultimate source of the funds is the public. Imagine how willing the public and their representatives would be—let's say—to double the NIH budget if serious concern were widespread that financial interests on the part of university investigators biased their research, warping the results in their favor at the expense of objectivity. Imagine also how eager most patients would be to enroll in clinical trials if they became convinced that they were merely a means to their doctor's financial ends. And then imagine how proud you would be to be part of an enterprise viewed by the public with skepticism and mistrust rather than with the admiration and esteem such enterprises now enjoy.
So, whether or not external financial interests have resulted in an actual degrading of sound scientific practices; and whether or not financial interests have actually jeopardized the safety of human research subjects; and whether or not financial interests have actually caused some slippage in the public's confidence in what we do, we risk great peril if we fail to respond to the growing perception that financial conflicts of interest have gotten out of control. A perception, I would remind you, that is shared by Congress and by the Department of Health and Human Services.
FOSTERING PUBLIC TRUST
Educating the Public about Research Benefit
If we are to maintain that all-important trust, it seems to me that two kinds of responses are called for. The first is to better educate the public about how, with proper safeguards, limited financial incentives in the conduct of clinical research can work to the benefit of everyone. The other is to visibly strengthen those safeguards so that we can, with confidence, assure everyone that our financial conflicts of interest are being managed effectively.
First, the benefits. The public's interest is clearly served by ensuring that research discoveries are translated efficiently into beneficial products and services. The explosive growth in the availability of such new products and services since enactment of Bayh-Dole evidences the powerful catalytic role that financial incentives can play, at least at the institutional level. However much one might lament the commercial entanglements that inevitably accompany the close partnerships between universities and industry, one must concede that technology transfer has been greatly accelerated in the process. I believe people would have an easier time putting the matter of financial conflicts of interest in perspective if there were broader public understanding of this benefit.
Strengthening Safeguards: Eight Suggestions
But, as important as a better-informed public might be, our more urgent task in bolstering public confidence is to visibly strengthen our safeguards against the downsides of external financial interests.
I have eight suggestions for what we might do. I offer them not as a definitive set of initiatives, but as a starting point to provoke continued discussion:
1. Comply with existing full-disclosure requirements. Virtually everyone agrees that the first essential step in managing financial conflicts is full disclosure of any material interests that might bias one's judgment in the conduct of research. Current federal requirements call for filing financial conflict-of-interest statements at least annually with one's institution. Evidence suggests, however, that compliance with these requirements in some institutions, is falling short. Job one, therefore, is to ensure uniform adherence to this most basic principle so that all potential conflicts are subject to review and so that appropriate action can be taken when needed.
2. Incorporate disclosure of financial interests in the informed consent process for human subjects research. I agree with those who argue that the cognizant institutional review board, or some other appropriate body, should determine when and to what extent such disclosure is warranted. The general principle here is that patients deserve to know about an investigator's relevant financial interests before agreeing to participate in human-subjects research.I am aware that strong arguments exist against this recommendation. Concerns about the privacy rights of investigators, about raising unnecessary fears in prospective volunteers, about confusing patients rather than truly informing them, and about how such disclosure might actually weaken rather than strengthen trust in one's physician. These are legitimate concerns. But not as compelling, in my view, as the consequences of not disclosing. Consider, for example, the damage to public confidence of having undisclosed financial interests come to light only after a tragic adverse event has occurred. Whatever the potential disadvantage of prospective disclosure, avoiding that nightmare, it seems to me, is more than worth it.But the clincher for me is that prospective disclosure of relevant financial interests is simply the right thing to do. Truth telling is, after all, a fundamental virtue of our profession.
3. Regularize the policies and strengthen the procedures we use to manage financial conflicts of interest. Institutions receiving federal research support are required to have written policies and procedures for determining whether a conflict of interest exists and, if so, for determining how such a conflict should be managed, reduced, or eliminated, as well as for sanctioning faculty who violate institutional rules. The precise nature of those policies and procedures, however, has been left, so far, largely to local discretion rather than to dictates handed down by the feds.As a result, not surprisingly, the ways in which these matters are handled across the country are highly variable. Compounding the variability among institutions is the inconsistency with which many individual institutions appear to apply their own policies. Processes as variable as those currently in place to manage financial conflicts of interest breed cynicism not only among members of the public but within our own community, as well.One clear way for us to signal our eagerness to address public concern about financial conflicts of interest is to agree voluntarily to abide by a common set of principles for managing those conflicts. We need to develop a consensus about how to determine when an individual investigator's significant financial interest is acceptable, when it needs monitoring, and when it is unallowable. We also need a consensus about what sanctions are appropriate for various kinds of infractions of conflict-of-interest policies.To help forge the needed consensus, and to address several related matters that I'll mention in a moment, I have just appointed a new AAMC task force on conflicts of interest, to be chaired by Dr. William Danforth, former chancellor of Washington University. Its first assignment will be to recast and extend the AAMC's existing and time-honored conflict-of-interest guidelines as a set of contemporary principles that can become a charter of accountability for our community in dealing with these matters.Critical to the success of this task force will be broad representation from inside and outside the academic medicine community, and especially from our patient partners in the research enterprise. Equally critical to its success will be close communication with university presidents, who bear responsibility for managing conflicts of interest across all academic sectors. You may know that the Association of American Universities (AAU), which represents 61 of our major research institutions, has formed its own task force to examine these issues in that larger context. We have been working and will continue to work closely with the AAU on these matters. But, given the special circumstances and exquisite sensitivities surrounding conflicts of interest in human-subjects research, we are the ones who must take responsibility for this high-stakes arena—take responsibility, in other words, for fulfilling the medical profession's obligation to regulate itself.
4. Gather the experts. One group that we have not engaged sufficiently to assist us in strengthening our self-regulatory safeguard is the legion of highly ethical clinical investigators on our faculties who are conducting impeccable science while maintaining significant financial ties with for-profit enterprises. Such individuals are arguably in the best position to help us all understand the nature of the temptations they and their colleagues experience, as well as the means they rely upon to hold those temptations in check. We would be remiss not to enlist their help in crafting meaningful and sensitive measures for optimizing the benefits of financial incentives while minimizing the threats. Our new task force also will include ample representation from this knowledgeable group.
5. Seek agreement on upper limits for allowable financial interests. Variation among institutions with respect to the allowable upper level of external financial interest is enormous, ranging from stringent ceilings on investments and on income from consulting, at one end of the spectrum, to the complete absence of limits, at the other end.Limitless financial entanglements in commercial entities interested in one's own research, no matter how fully and widely disclosed, constitute a brazen assault on public confidence in university science. The public's willingness to believe in the objectivity of academic research has understandable limits. So, too, should a faculty member's financial involvement with those who stand to profit from the outcomes of his or her research.For this reason, one of the charges I am giving to our new AAMC task force on conflicts of interest is to recommend limits that would strike a reasonable balance between the need, on the one hand, to motivate investigators to move their discoveries as quickly as possible from bench to bedside, and the danger, on the other hand, of fostering the view that our research enterprise operates more like a bazaar than like a public service.
6. Inaugurate a certification process for research faculty. The ethical conduct of medical research has become much more demanding in recent times than many of us remember from the days when giants strode the earth. I believe the time has come to expect research faculty to demonstrate formally that they understand the myriad regulations governing the conduct of modern medical research and the numerous ethical dilemmas with which they must grapple. Understanding the intricacies of financial conflicts of interest is but one of the competencies that all research faculty should be expected to have.To fulfill this expectation, I am also asking our new task force to consider the feasibility of inaugurating a rigorous, voluntary, institution-based certification process based on national standards. By way of analogy, consider how well the board certification process for individual physicians has not only fostered highquality health care but also served to assure the public that the profession is serious about its commitment to excellence in the practice of medicine. A comparable process for certifying individual clinical investigators could have a similarly salutary effect on adherence to high standards of scientific conduct and would go a long way toward bolstering the public's confidence in the integrity of academic medicine's research mission.
7. Invest more resources in the research infrastructure. More money and more people are clearly needed if we are to identify all existing conflicts, fairly evaluate their potential for bias, monitor those that are worrisome, hear appeals from those that are judged unacceptable, and scrutinize everyone's compliance with institutional policies. Both federal and non-federal sponsors of research should certainly be pressed for more help in strengthening our conflict-of-interest safeguards, and I assure you that the AAMC will continue to push hard for this kind of help. But, in the final analysis, we cannot plead poverty as grounds for shirking public accountability. Whatever help we do or don't get from external sponsors, we must find a way to shore up our procedures. Money may be tight, but public trust is priceless.
8. Establish principles governing institutional conflicts of interest. Although the primary focus of public concern in these matters, at present, is on the financial interests of individual faculty members, an equally pressing issue is the increasing extent to which our institutions themselves are engulfed in financial conflicts of interest. Many of our leading medical schools and teaching hospitals now have significant equity positions in spin-off companies and enjoy sizable revenues from licensing fees and royalties on patented inventions.
Such institutional interests raise serious questions about how credible and effective medical schools and teaching hospitals can be in dealing with faculty conflicts. Can an institution that has its own financial stake in a faculty member's research be trusted to judge whether that research is susceptible to bias? Should institutions be prohibited from having a financial stake at all in a faculty member's research? Still more important, to what extent will the ever-expanding financial conflicts of interest of institutions contribute to the erosion of public trust in the objectivity of academic research?
Thus, the final and perhaps most challenging request I am making of our new task force is to consider what additional safeguards are needed to address the potential downsides of financial conflicts at the institutional level.
In closing, let me return to the central issue of trust. Trust us, we say, and we'll make a difference. Support our science, and we'll make your lives better.
In making these claims, we would do well to remembe that trust is not a birthright of individuals or institutions who do research; trust is a distinction that is earned by merit. Our ability to continue to pursue the venerated mission of clinical research dependes entirely on our willingness to merit public trust. Maintaining public confidence in the integrity of what we do requires more than assuring ourselves that external financial interests have not tainted our scientific and ethical standards. We must also reassure the public that we have done everything in our power to ensure that their interests are not sub-ordinate to ours.
Let me end with a quote from the proclamation I referenced earlier.
Whereas, the future of medicine depends upon an enduring trust between researchers and patients, we pledge ourselves to uphold the principles of beneficence, justice, and respect for individuals, and to ensure the ethical conduct of research.7
Please join me in reaffirming that pledge.