Conflicts of interest in medical education: Disclosure may not help

Yesterday’s New York Times had an interesting op-ed, “Stumbling into Bad Behavior,” about corruption and unethical conduct in corporate and financial settings.  The authors, Max H. Bazerman and Ann E. Tenbrunsel, are academics who wrote a book about ethical blind spots.  They note that regulators, prosecutors, and journalists tend to focus on corruption caused by willful actions or ignorance, but this overlooks unintentional lapses:  “Our legal system often focuses on whether unethical behavior represents ‘willful misconduct’ or ‘gross negligence.’ Typically people are only held accountable if their unethical decisions appear to have been intentional — and of course, if they consciously make such decisions, they should be. But unintentional influences on unethical behavior can have equally damaging outcomes.”

This caught my attention as it relates to conflicts of interest in medicine.  For example, I have long expressed ethical concerns regarding the willful participation of physicians in pharmaceutical promotion.  It is a clear conflict of interest to purport to be an unbiased advisor to patients, while at the same time choosing to attend (or deliver) overtly slanted marketing presentations.  However, defenders of such participation say they deserve more credit: They cannot be corrupted, and would never willingly deliver biased medical advice no matter how drug or device manufacturers reward them.

These positions are reconcilable given that bias is often unconscious and unwilled.  Bazerman and Tenbrunsel note:  “[M]uch unethical conduct that goes on, whether in social life or work life, happens because people are unconsciously fooling themselves. They overlook transgressions … because it is in their interest to do so.”

Psychiatry, of course, has a lot to say about how people fool themselves.  We discount our own lapses to maintain our self-esteem.  We may employ psychological denial to make our troubling inconsistencies disappear, or utilize projection to attribute our faults to others.  We may reframe liabilities to look like assets — and we may do all of these outside of our own awareness.  Thus, it is entirely consistent sincerely to consider oneself principled, ethical, and unbiased, and yet to be undermined by one’s own unconscious mind.

To me, the most striking paragraph of the op-ed questioned the value of disclosing conflicts of interest, an issue I’ve raised in the past.  As chair of Continuing Medical Education (CME) at my medical center, I am forever badgering CME speakers to provide a “disclosure slide” at the start of their talks.  (CME is required to maintain medical licensure and stay up to date, so all physicians must attend many hours of CME annually.)  This disclosure of financial ties to industry, and other potential sources of bias, is required by state and national CME oversight bodies; the medical center risks its CME accreditation if this rule is not followed.  Yet the value of disclosure has always felt tenuous to me.  Maybe it’s better than nothing, I thought, but simply disclosing potential conflicts of interest hardly guarantees that the talk will not be biased anyway, nor that the physician audience will know how to evaluate the imparted information given the disclosure.

The New York Times op-ed cites a 2005 study that clarifies this matter of disclosure in a very useful if sobering way.  “The Dirt on Coming Clean: Perverse Effects of Disclosing Conflicts of Interest” by Daylian M. Cain, George Loewenstein, and Don A. Moore discusses what disclosure intends to remedy, versus what it may actually do given unconscious as well as conscious factors.  The authors point out that people generally do not discount advice from biased advisors as much as they should, even when advisors’ conflicts of interest are disclosed.  Moreover, disclosure can actually increase the bias in the delivered information because it leads the disclosers to feel morally licensed and strategically encouraged to exaggerate their advice even further.  The paper reports an empirical study conducted with Carnegie Mellon University undergraduates that supports these concerns.  The authors conclude:

[D]isclosure cannot be assumed to protect recipients of advice from the dangers posed by conflicts of interest. Disclosure can fail because it (1) gives advisors strategic reason and moral license to further exaggerate their advice and (2) it may not lead to sufficient discounting to counteract this effect. The evidence presented here casts doubt on the effectiveness of disclosure as a solution to the problems created by conflicts of interest. When possible, the more lasting solution to these problems is to eliminate the conflicts of interest.

I couldn’t have said it better myself.  The op-ed likewise concludes that, “Good people unknowingly contribute to unethical actions, so reforms need to address the often hidden influences on our behavior.”  When it comes to unbiased medical education, neither good intentions nor disclosure of potential bias is sufficient.  The solution is to admit we are fallible humans, and to avoid sources of bias, conscious and unconscious.  Psychiatry could help — if it weren’t so complicit itself.