Does disclosure really mitigate conflicts of interest?

The first posting in this series  on behavioral ethics provided an overview of that emerging and important area of knowledge and what it might mean for the C&E field.    In this post we examine what behavioral ethics teaches about COI-related compliance.

Conflict-of-interest compliance regimes are, for the most part, based on disclosure requirements.  That is, while some types of conflicting interests are prohibited in  all circumstances, a more common approach is to require appropriate disclosure of the interests in question (and, in some instances, approval from specified parties before the COI condition is permitted to continue).

But how effective is this way of addressing conflicts?  In “Disclosing Conflicts of Interest – Does Experience and Reputation Matter?, Christopher W. Koch and Carsten Schmidt (replicating the results of an earlier study – “The Dirt on Coming Clean: Perverse Effects of Disclosing Conflicts of Interest,”  by Daylian  M. Cain, George Loewenstein and Don A. Moore)  found that in some circumstances  disclosure could be a cause of, rather than cure for, unethical behavior: “information providers whose conflicts are not disclosed will feel morally bound to report accurately to information users, because they would consider it unfair to lie to someone who is unaware of the misalignment of incentives.  Disclosing conflicts of interest would have the effect of removing the moral bound and providing a moral license to misreport.”

The particulars of the experiments undertaken for these studies (Koch and Schmidt used an audit-related setting) are, in my view, less important for C&E professionals than is the larger message, which is that disclosure should not be considered a panacea for COIs.  We have already seen in the Blog how disclosure can be ineffective (when patients don’t use information from publicly accessible data bases of  pharma company payments to health care providers) or even lead to “reverse conflicts of interest.” The findings reported in these two papers suggest that even when disclosed and approved COIs may need to be actively managed (to the extent that can be done in a given set of circumstances).

Part of that process should be educating both the individuals with the COIs and those managing the COIs on the nature of the challenge facing them.  As described by Cain,  ,  people often fail to “understand how big a problem conflicts of interest” are and he further notes that COIs can affect the judgment of even well-meaning individuals, i.e., someone “need not be intentionally corrupt to have difficulty in objectively navigating a conflict of interest.” 

Finally, one unsurprising but still important aspect of this research is showing that disclosure is likely to be more effective as the level of relevant experience/expertise of the party to whom the disclosure grows.  This, in turn, suggests that COI approvals generally should not be made by line management alone, but should involve a C&E officer, who is likely to be a more sophisticated estimator of the impact of a COI in an organization.

Coming up: what  behavioral ethics teaches us about C&E risk assessment and training.