The subject of industry influence on medicine in general and psychiatry in particular appears to have turned into a toxic one for many physicians, like the clinician who stood up after a lecture on the subject by past APA President Paul Appelbaum, M.D., at the APA Institute on Psychiatric Services last month. He protested that it was all a grand "overreaction" to a nonproblem that had caused physicians, and especially psychiatrists, to become scared and defensive.

That the subject had become a polarizing one was a point that Appelbaum had already acknowledged. He opened his address by citing the words of Harvard physician Thomas Stossel, M.D., who called the persistent demand for regulation of financial conflicts of interest in medicine "a damaging solution in search of a problem." (Stossel's words came from a 2007 article in the journal Perspectives in Biology and Medicine, "Regulation of Financial Conflicts of Interest in Medical Practice and Medical Research: A Damaging Solution in Search of a Problem.")

Yet however angry and beleaguered physicians may feel, Appelbaum said the high tide of oversight and regulation rising around industry-physician relations is unlikely to turn. "Psychiatrists and other physicians are caught in a quickly changing regulatory and ethical landscape with simultaneous flux in professional standards, institutional policies, and statutory requirements," he emphasized at the institute.

For this reason, he suggested that physicians, as well as policymakers and medical organizations, might do well to reconceptualize the issue not as an ethical one, with its implications of bad faith and bad behavior, but as a public-policy problem. And as with any public-policy dilemma, working out how physicians will relate to industry in the future will require weighing costs and benefits of a variety of possible solutions and a realization that there is likely to be no perfect solution.

"It may detoxify the issue to view it less as a matter of ethics and more an issue of policy, allowing us to shun reflexive judgment and engage instead in a systematic analysis of options," Appelbaum said. "We need to weigh options against each other and against the option of doing nothing at all. Sometimes we will have data [to back up our decisions] and sometimes not. In all cases, there will be costs associated with our choices, so we need to conceptualize our goal as finding the best approach, not the perfect approach."

As a model of how physician conflicts of interest associated with pharmaceutical-industry relationships might be more dispassionately arbitrated, Appelbaum offered the "principal-agent" problem—a ubiquitous scenario in economics in which any hired "agent" may have interests that diverge from those of the principal, or the hiring party.

The principal-agent problem arises in the most prosaic situations: a department-store clerk is hired to induce as many customers as possible to make purchases, but the clerk may have any number of divergent interests—from taking breaks, to answering text messages, to managing a side business on eBay—that conflict with what he or she was hired to do.

So, too, can conflicts of interest arising from physician relationships with industry be viewed as a variation of the principal-agent problem. Appelbaum summarized evidence showing how financial relationships with the pharmaceutical industry can create potentially conflicting interests for doctors in continuing medical education, clinical practice, and research.

He cited, for example, industry-supported presentations that can influence physicians' prescribing in the direction desired by the companies, sometimes contrary to the interests of patients. Or, he noted, a review of studies on meetings with drug-company representatives found increased requests from physicians for formulary additions for promoted drugs and altered prescribing practices by both residents and practicing physicians. These actions favored newly promoted drugs despite their additional cost and sometimes despite evidence suggesting that they may have been inappropriate.

Meta-analyses and reviews consistently indicate that research supported by industry is more likely to report positive findings than studies with other sources of funding. Papers in which at least one author had financial ties to sponsors of psychiatric clinical trials also show a higher rate of positive findings, Appelbaum said.

Possible solutions for resolving principal-agent problems that can be applied to medicine include education, disclosure, management, and alignment of interests.

Education has an intrinsic appeal, but its effectiveness as a tool in resolving or preventing conflicts may be limited, according to Appelbaum.

Reviews of education with house staff about the impact of meetings with drug reps showed some short-term effects on perceptions and behavior. What may be the only study in psychiatry found that a one-hour educational program led to no change in residents' attitudes toward drug reps, but a reduced acceptance of noneducational gifts, he said.

Disclosure, a response to the popular demand for "transparency," operates essentially as a proxy for monitoring whose interests the doctor is serving and is the most ubiquitous strategy for managing potential conflicts. Today, presenters at sessions sponsored by the Accreditation Council for Continuing Medical Education (ACCME) must reveal companies from which they have received funding; journals routinely require authors to divulge financial ties with industry; the National Institutes of Health now asks institutions to report whether researchers' income from a company exceeds $10,000; and state and federal "sunshine" bills seek Web-based disclosure of industry payments.

Similarly, "management" is a strategy that is growing in popularity. ACCME accreditation standards mandate that program content be determined independently of the program's funder, and speakers can be required to balance their presentations. In addition, hospitals require drug reps to make appointments to see physicians, restrict them from distributing literature to residents, and bar them from patient-care areas. Meanwhile by 2004 more than three-fourths of U.S. medical schools had established committees to review and manage researchers' relationships with industry, Appelbaum reported.

"Alignment of interests" tends to be the strategy most commonly favored by economists in other principal-agent scenarios, and it consists essentially of eliminating the divergent or conflicting interests of the agent. "In the context of medicine this means abstaining from certain relationships so we are not confused by multiple sources of influence on our behavior," Appelbaum said.

For instance, the Association of American Medical Colleges (AAMC) and Institute of Medicine of the National Academy of Sciences have recommended that physicians not accept personal gifts or food from industry, and the AAMC recommends that researchers with financial interests related to a study not be allowed to participate unless they give up those interests. Academic medical centers have banned presentations by pharma employees and restricted faculty from joining industry speaking bureaus.

Appelbaum emphasized that the principal-agent model is useful not so much as a means of arriving at the perfect solution to conflicts of interest as for understanding that there is no perfect solution: all strategies will have limitations and costs, a fact that may dampen the ardor of those who would eliminate all industry involvement with medicine.

"Cutting off relationships with industry inhibits flow of clinical input to the development of therapeutics and eliminates industry support for potentially beneficial programs," Appelbaum stated.

"Viewing the issue as a principal-agent problem can focus everyone on the questions that need to be asked: What is the undesirable behavior we are trying to prevent? How serious are the consequences? How effective are solutions, and what are the costs? And are they really going to work?"