Financial Conflicts of Interest

The inherent tension built into remuneration for the healing arts has been recognized as far back as Plato, who devoted a small part of The Republic to the issue.³ In a classic fee-for-service arrangement, physicians benefit financially from the provision of more interventions, with patients and the market poorly positioned to make judgments regarding thenecessity of these services.⁵ Historically, a physician’s service largely equated with the physician’s presence, but systemic conflicts of interest could still arise, such as agreements for fee splitting from referrals or commissions from pharmacies. As the number and kinds of medical services have exploded over the last half-century, so has the potential for clinicians to profit from the profligate use of these services. The practice of medicine now provides the entrepreneurial physician with ample opportunities to develop ancillary business interests, such as owning radiology and other diagnostic or therapeutic centers or equipment, even entire hospitals.⁶ Clearly, the attendant financial gain in “self-referring” a patient for testing or intervention under such circumstances creates a conflict of interest as, logically, only a subset of patients will be likely to benefit from the additional procedures, whereas all patients (at least all those with the ability to pay) sent for testing or intervention would financially benefit the physician-owner. Empiric evidence amply demonstrates that such circumstances lead not simply to a perceived conflict of interest, but to a marked increase in utilization of services when compared to financially disinterested clinicians.⁷⁸⁹

The combination of rising health-care costs and the appearance of impropriety on the part of entrepreneurial clinicians has led to multiple attempts to remedy such conflicts of interest, from guidelines developed by professional organizations¹⁰ to federal statutory interventions, primarily the “Stark” laws.¹¹ Regulatory intervention, which initially was a blunt instrument, has continued to evolve over the last decade, recognizing that not all instances of physician ownership raise the same ethical concerns and that the absolute prohibition of physician ownership may actually disadvantage patients under some circumstances.¹² Currently, the Stark laws explicitly prohibit some forms of physician ownership while allowing exemptions for others. While the law remains subject to interpretation, and some physician-entrepreneurs will continue to take advantage of loopholes and uncertainties, clinicians should recognize that specific structures and systems providing financial benefit to them remain ethically problematic if such structures appear to promote decisions that are not clearly in the best interest of individual patients. Certainly, arrangements that both increase physician remuneration and improve the care of patients are theoretically possible and, if patient care can be demonstrated to improve, would be ethically preferred.¹³ But such arrangements demand a high burden of proof demonstrating that patients, and not simply clinicians, are benefited.

As expenditures on health care have grown and new managed care systems have been developed with an explicit goal of limiting costs, clinician behavior has become the primary target for those desiring to limit services to patients. Structured reimbursement plans have been designed to control physician-directed spending, creating a relatively new kind of conflict of interest for physicians, one that biases them toward withholding potentially beneficial care.¹⁴ However formally structured, these systems share the feature that clinician remuneration is inversely related to expenditures per patient. (More restrictive structures that prohibit physicians from recommending or prescribing specific treatments or interventions go beyond creating a conflict of interest to directly controlling physician behavior and have attracted attention due to the legal and ethical concerns related to not allowing these physicians to disclose these limitations of options to patients or the public [so-called gag rules].¹⁵) Traditional fee-for-service physicians may also find themselves under pressure from hospitals to limit care or inpatient stays, as payers reimburse through a fixed-payment system. Likewise, the current rise in in-house health care provided by large employers¹⁶ comes with an inherent conflict of interest for practitioners, whose salary and job security requires satisfying the demands of those employers. Such structures may have a coercive effect by threatening physicians with loss of income, either directly or by limiting the scope of their practice.¹⁷

As patients may be disadvantaged or harmed by either overutilization or underutilization of diagnostic and therapeutic interventions, the ethics of conflict of interest in managed care settings are no different than those in a fee-for-service environment. In either situation, the more likely a given system of clinician remuneration is to result in reasoning and action that does not coincide with the best interests of individual patients the more ethically problematic it becomes. Financial incentives for clinicians that are designed to improve the quality of care for individual patients that coincidentally decrease health-care costs (eg, increasing influenza vaccination rates in target populations) can avoid creating inherent conflicts. These “pay-for-performance” models raise a large number of practical and philosophical concerns¹⁸ and often result in unintended consequences, including the avoidance of the sickest patients and the improvement of documentation but not practice.¹⁹ Still, structuring systems to correlate clinician interest with patient interest is not only ethically defensible, but preferable.

No single system of physician reimbursement can hope to avoid all potential conflicts of interest, but it is imperative that each clinician be cognizant of the conditions of the system in which one practices that may compel one to reason, recommend, or act in ways that do not benefit the patient-at-hand. Recognizing conflicts of interest remains the first step in ameliorating their effects.

The Clinician and the Pharmaceutical and Medical Device Industry

Over the last several years, particular professional and public attention has been focused on conflicts of interest that arise from various relationships between the pharmaceutical and medical device industries and clinicians.²⁰²¹²² Businesses that happen to make drugs or medical devices have interests that are no different than those of other businesses: maximizing shareholder value, generally by increasing sales.²² The profession of medicine, in general, and individual practitioners, in particular, continue to embrace a primary goal of improving and maintaining the health and well-being of individual patients. A conflict of interest develops when interactions with industry create circumstances in which the individual physician’s interest coincides with that of business, not patients. Beyond appearances, such conflicts have demonstrated the potential to alter physician practice in a manner that favors the pharmaceutical industry at the expense of the patient. Multiple specific interactions between industry and clinicians, including gift giving, consultancy arrangements, support of CME and guideline development, create conflicts of interest that vary in terms of effect, but each acts to bias the clinician away from patients and toward the interests of industry.²³

Gift giving is perhaps the most familiar of these interactions and may take the form of small trinkets (eg, pens, notepads, or candy) or more substantial premiums (eg, expensive meals, books, or travel). Other practices of industry, including consultancy fees or payment for participation in focus groups with little or no expectation of work on the part of the “consultant,” can also be viewed as gift giving. While such gift giving clearly concerns patients,²⁴²⁵ who might be expected to be negatively affected not only through the delivery of inappropriate care but also due to the higher cost of medications and health insurance, clinicians have been less concerned. Physicians, in particular, seem confident in denying that receiving these handouts alters their individual practice in any substantive way,²⁶ although they are concerned that their colleagues may not be quite so stalwart.²⁷²⁸ A rather large body of empiric literature from the social sciences²⁹ and specific studies involving physicians,³⁰³¹ however, has demonstrated consistently and convincingly that receiving gifts produces a measurable change in both attitudes and behavior that is favorable to the giver. This should not come as a surprise, as gifts given to clinicians by representatives of industry are not gifts at all, but rather a marketing of wares.²⁹ Physicians, it appears, are no more immune to the psychology of persuasion and influence than any other person.³²

Recognizing the negative public effects, actual and attitudinal,²⁴²⁵ of industry largesse toward medical professionals and fearing increasing federal regulation, a variety of professional organizations and the pharmaceutical industry itself have promulgated guidelines attempting to limit and define acceptable practices for transactions between industry and clinicians.³³ As rules and recommendations seek to limit the more egregious transactions between industry and clinicians, alternatives arise. For instance, generously reimbursed clinician participation in industry “focus groups” or “advisory panels,” where input and expertise is nominally but not actually sought, avoids prohibitions on gift giving. Many of the new guidelines provide for specific monetary limits on the value of gifts and vague restrictions on how often they can be received, with only the guidelines of the American Medical Student Association recommending the refusal of all gifts, no matter how small, provided by industry. Although compelling on its face, limiting the receipt of gifts based on monetary value assumes that “low-value” gifts are less ethically troublesome, presumably because they are less likely to influence behavior. (Every doctor may indeed have a price, but it is presumably over $100.) Unfortunately, no empiric evidence demonstrates a threshold effect of gift giving, a practice that most often induces change through the development of generally positive sympathies toward the giver rather than an explicit quid pro quo.²⁹ The receipt of gifts of any size, it appears, can effectively align a clinician’s interests with those of industry, creating a conflict of interest that may go unrecognized.

Another industry practice that has led to concerns regarding conflict of interest is the increasing subsidization of CME, including professional publications³⁴³⁵ and guideline development.³⁶ Underwriting the cost of medical education programs, and thereby decreasing or eliminating the costs associated with participation for individual clinicians, might be viewed as simply another form of gift giving, particularly given the meals, liquor, and entertainment that often accompany industry-sponsored CME programs. Yet, the influence of industry on CME deserves specific mention and careful analysis, as the creation of conflict of interest through CME is more insidious than other more overt efforts of industry to change physician behavior. The investment of the pharmaceutical industry in CME in the United States was > $1.1 billion in 2005, representing almost two thirds of the total amount spent on CME that year.³⁷ Most physicians and many other clinicians must accrue a certain number of hours of accredited CME in order to maintain licensure, a powerful motivator that ensures a market for and attendance at CME activities.

Although providers of CME must be accredited in order to offer certification, the Accreditation Council for Continuing Medical Education (ACCME) has certified numerous private, for-profit companies, known as medical education and communication companies, to present accredited CME programs, almost all of which are entirely underwritten by pharmaceutical companies.³⁸ The inherent conflicts of interest involved would seem clear and concerning, with a blurring of the line between marketing and education, as clinicians receive a larger and larger percentage of their ongoing education viewed through the filter of an industry that is focused on increasing profits.³⁸³⁹ Counterargument centers on the overall value of CME for clinicians, with industry defending their support as an effort to establish goodwill and aid physicians in providing “the best, most appropriate, and most up-to-date health care to their patients.”⁴⁰ This defense of current practice recognizes that many new therapies and innovations offer real benefit and that clinicians need to be aware of these. Allowing an industry focused on maximizing profits to be not only the primary content provider of CME but the primary arbiter of what constitutes “the best, most appropriate” health care, however, is problematic.

The creation of clinical practice guidelines represents an attempt to define the best and most appropriate diagnostic and/or treatment strategies for particular diseases and conditions. With the potential to alter the behavior of a large number of practicing clinicians, guidelines influenced by the pharmaceutical industry have the potential to magnify and project that inherent conflict of interest into the entire medical community. Authors of clinical practice guidelines, in fact, often have significant financial relationships with the pharmaceutical industry, and those relationships are only occasionally disclosed.³⁶ In addition, pharmaceutical companies may directly underwrite the creation and promulgation of practice guidelines that are expected to directly benefit revenue, as Eli Lilly, the makers of recombinant activated protein C, did with the “Surviving Sepsis” campaign.⁴¹

Even when there has been significant industry influence in the creation of such guidelines, professional organizations may be used to provide an imprimatur to such documents, further obscuring the inherent conflict of interest. As members and officers of professional organizations, clinicians can insist that the responsible bodies within those organizations exercise due diligence to identify and neutralize conflicts of interest prior to endorsing clinical practice guidelines. Many professional organizations, including the American College of Chest Physicians,⁴² have recently developed and instituted strict policies governing conflicts of interest among members of guideline development committees that are aimed at ensuring objective and scientifically sound products. Still, the users of clinical practice guidelines should take the time to review the disclosure information of the guideline authors, and adjust their perceptions of the objectivity and authority of the recommendations accordingly. Supplanting pharmaceutical marketing material with practice guidelines as a method of informing and educating physicians offers no scientific or ethical advantage if industry manages to control the subject, form, and content of these guidelines.

Ethically, support from industry for CME is not, in itself, inherently unacceptable. The pharmaceutical industry benefits greatly from the clinical practice of medicine, and some mechanism for the industry to support that practice would be morally defensible. Such support becomes problematic only insofar as industry influence conspires to shape clinician’s reasoning in a manner that is more likely to benefit the industry than patients. Some observers believe that any participation of industry in CME must bring with it an element of marketing and, thus, should be absolutely prohibited.²⁰ Others remain open to at least the possibility of industry support without industry influence, provided that strict regulations designed to maintain this distinction are developed and honored.⁴³⁴⁴ As with gift giving, empirical evidence strongly suggests that industry support and influence in CME not only creates an apparent conflict of interest, but actually limits the scope of the topics presented to clinicians⁴⁴ and alters their subsequent behavior in a manner that is consistent with the sponsor’s interest.⁴⁵

In a response to concerns regarding undue industry influence in the form, content, and presentation of accredited CME programs, the ACCME revised its rules on industry sponsorship of CME in 2004.⁴⁶ The new standards demand the explicit disclosure of financial conflicts of interest among participants in CME planning and for presentations to be devoid of commercial content. Whether the new standards can or will have an actual effect in practice remains to be seen.⁴³⁴⁷ The federal government appears no longer willing to wait on the sideline, as evidenced by the US Senate Finance Committee recently requesting information from the ACCME regarding the influence of industry in CME.⁴⁸ Without waiting for the Senate hearings, a thoughtful clinician will recognize the permeability of the “firewall”⁴⁴ being constructed between industry support and CME simply by reflecting on the content of any program supported by an “unrestricted educational grant,” for there is scarcely an example in which the topic being taught does not bear directly on a major product of the underwriter. With renewed federal interest in the state of CME, it certainly behooves the medical profession to reexamine its relationships with industry in this regard. As academic medical centers occupy a prominent place in CME, both through direct CME offerings and by providing a large proportion of the speakers even for industry-sponsored events, advocates for reform ahead of regulation have called on academia to be the agent of change.³⁹⁴⁹⁵⁰ Yet, practicing clinicians, as the consumers of CME, hold the power to eliminate industry’s influence on CME directly, simply by refusing to attend programs underwritten or otherwise supported by industry. Certainly, clinicians voice a preference for CME devoid of industry influence,⁵¹ though it is not yet clear that we are willing to pay the increased price for such a product. Foregoing industry support of CME would require clinicians to bear the real cost of that education personally. But, failing to assume those costs may ultimately prove more costly to the public respect and deference still paid to the medical profession.²²³⁹

Ameliorating Conflicts of Interest: Avoidance and Disclosure

Conflicts of interest will remain an ever-changing and unavoidable part of clinical medicine. That fact, however, should not lead to resignation and an unqualified acceptance of all conflicts of interest, for certainly not all conflicts are unavoidable, nor do all present the same ethical peril. Persistent and systemic conflicts of interest, which may compromise the care of many patients, raise serious ethical concerns and demand attention. Prior to any possibility of ameliorating conflicts of interest, conflicts must first be recognized and elucidated, which is no small task in the multifaceted and complex health-care system of the United States. Once identified, the common responses of denial and self-delusion must be avoided. These defensive reactions generally take the form of refusing to acknowledge that the conflict has negatively influenced behavior and denying a provable causal link,³⁹ often attended by a sense of outrage that personal and professional integrity could be questioned. Such a response not only fails to recognize that the appearance of a conflict of interest is problematic, whether or not the conflict can be shown to change behavior, but also stands in the face of significant empirical evidence that medical management does change when systemic conflicts of interest exist. Recognizing the importance of ameliorating conflicts of interest does not demand a sense of personal moral culpability, however, only that clinicians recognize that any practice system is improved on when the interests of clinician are more closely aligned with the interests of patients.

Once identified, the two primary methods for dealing with conflicts of interest are disclosure and avoidance. Disclosure has lately been promulgated as a remedy for virtually all conflicts of interest affecting physicians.⁵² The act of disclosure will look different for different conflicts of interest. In a managed care organization, the requirement would entail informing patients of circumstances creating conflicts of interest for clinicians and the prohibition of gag rules, while in CME, disclosure of the industry sponsor of the program and financial conflicts of interest for speakers is mandated.⁴⁶ The value of disclosure rests on its explicitness, presumably offering parties a chance to opt out or, at a minimum, adjust their expectations of one another. But disclosure alone is likely to do little to protect patient interests, for most patients cannot choose the system that provides them care, and the penetration of industry into CME has left few speakers, programs, and professional organizations untainted. Disclosure without alternatives offers little and, by itself, fails to protect patients.⁵³

Clearly, the most ethically sound, and most practically difficult, approach to conflicts of interest in clinical medicine is avoidance. Acknowledging the inevitability of conflicts of interest does not provide an excuse to simply tolerate every one. Rather, each specific circumstance creating a conflict demands an answer to the question “Must it be so?” The costs of conflicts of interest in clinical medicine are high, both in terms of the effects on individual patients and on the credibility and integrity of the profession as a whole. As such, clinicians need to look for conflicts of interest in their practice, acknowledge and disclose them when discovered, and move beyond disclosure to avoid or eliminate them. Failing to do so is tantamount to rejecting our personal and professional obligation to provide a right and good healing action for our patients.