What Are the "Top 3" Changes Needed to Improve US Healthcare?

Robert M. Centor, MD; Pennie Marchetti, MD; R.W. Donnell, MD; Roy M. Poses, MD


November 13, 2006

In This Article

Roy M. Poses, MD

There are so many problems afflicting contemporary healthcare that one hardly knows what solutions to propose first. I will stick to one area with which I am familiar.

Conflicts of interest in healthcare have become a hot topic. A recent commentary by Brennan and colleagues in JAMA suggested that physicians at academic medical centers should be banned from receiving gifts from pharmaceutical and device companies, even small gifts such as pens and coffee mugs with the companies' logos; from getting funding from such companies to attend conferences; from receiving pharmaceutical samples; and from eating meals supplied by these companies.[18] The rationale was that even small gifts can have important effects. "Social science research demonstrates that the impulse to reciprocate for even small gifts is a powerful influence on people's behavior," the authors wrote. "People who give or accept gifts with no explicit 'strings attached' still carry an expectation of some kind of reciprocity."[18]

Several prominent academic medical centers have subscribed to these policies.[19] While I applaud this trend, what is most striking about the current movement to ban "free lunches" is that it ignores the much bigger conflicts of interest that pervade healthcare. If even a small gift, like a pen or a coffee mug, may affect physicians' behavior, it stands to reason that a more substantial financial relationship could affect behavior even more. The Brennan article addressed some such relationships, such as faculty serving on speakers' bureaus for pharmaceutical and device companies, but it did not recommend banning them, only insisting that such work be governed by contracts with "specific deliverables."

Common sense suggests, however, that being paid $10,000 or $100,000 a year to consult for a company might make someone more hesitant to criticize that company than getting a coffee mug.

And yet, quite a few faculty and leaders of academic medical centers consult or otherwise work for pharmaceutical and device companies. A 1999 study found that 7.6% of faculty at one school disclosed such financial relationships with industry.[20] Almost half of the controlled trials reported in major psychology journals from 2001 to 2003 included authors with a disclosed conflict of interest.[21] Even worse, sometimes potential conflicts are not disclosed. For example, 7 of 13 authors in a study on depression in pregnancy[22] failed to disclose financial relationships with companies that make antidepressants.[23] All 6 authors of another study on the relationship between migraine headaches and cardiovascular disease[24] failed to disclose financial relationships with companies that make drugs for these diseases.[25]

In fact, there is evidence that some pharmaceutical companies employ physicians as consultants or speakers not just for their technical expertise, but also to induce them to become key opinion leaders who will help market the companies' products. For example, legal documents from the gabapentin investigation revealed how Parke-Davis used physicians in its part-time employ as "local champions of the drug" and "key influencers."[26] An article in a marketing magazine even explained "creating a pharma buzz" for pharmaceutical companies: "...while buzz should always appear to be spontaneous, it should, in fact, be scientifically crafted and controlled...."[27] It identified prime "buzz drivers" as individuals to whom people listen -- "they may be physician opinion leaders in a given specialty."[27]

Furthermore, the JAMA article did not even address relationships that could represent even larger conflicts of interest, such as hospital leaders serving on the boards of for-profit healthcare companies. As directors, they have a fiduciary duty to protect the interests of the company and its stockholders, which could have an even stronger impact on their thinking and behavior than just being paid as a consultant. Ironically, in such a situation, a lowly intern could be punished for accepting a free coffee mug while reporting to a director of the company whose logo was on the mug.

Thus, I suggest the following 3 steps to deal more comprehensively with the conflicts of interest that now pervade healthcare:

Disclosure: Anyone with decision-making power over individual patients or healthcare organizations should disclose any potential conflict of interest. Such disclosures should occur any time they speak, write, or take an action that conceivably could be related to the conflict. Such disclosures should be extensive enough to reveal the intensity and nature of the potential conflict. For example, a professor giving a talk about a particular disease should disclose any relationship with a manufacturer of a drug used to treat that disease; in fact, he should also reveal about how much he gets for that consulting, what he consults about, and how the drug company's interests might be related to his talk.

Regulation: Clear regulations should be put in place to guide disclosures. The regulations should be conceptually similar for practicing physicians and for leaders of academic medical centers or other decision makers. Once disclosures become wide-spread, the scope of the problem may be more apparent. At that point, certain conflicts may need to be banned outright. If there is a rationale to ban small gifts to doctors, there is an even better rationale to ban consulting contracts, speaker honoraria, and board memberships.

Enforcement: Incentives must be established to encourage disclosure. Failure to disclose important relationships should result in stronger negative consequences than public reprimands.

Unfortunately, since many high-level healthcare decision makers have conflicts of interests, and many of these relationships are very lucrative, I do not expect a rush to follow my recommendations. In the meantime, though, it would be best to be highly skeptical about whose interests are served by healthcare decision makers who work for 2 or more masters.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.