Carnegie Mellon University
June 19, 2012

Press Release: Carnegie Mellon Researchers Find Physicians Are Biased When Evaluating Medical Conflict of Interest Policies

Same Patterns Found in Finance Industry; Researchers Suggest Need For Third Parties in Conflict of Interest Policy-Making Process

Contact: Shilo Rea / 412-268-6094 / shilo@cmu.edu

LoewensteinPITTSBURGH—Medical institutions have been under pressure to develop and implement policies to avoid conflicts of interest between physicians and pharmaceutical companies. In most cases, medical professionals who have a stake in the issues at hand craft the conflict of interest policies.

New research from Carnegie Mellon University's George Loewenstein and Zachariah Sharek and the University of Pittsburgh's Robert Schoen investigated whether medical professionals making conflict of interest policy decisions are able to separate their policy judgments from their personal, vested interests. The research, which will be published in an upcoming issue of the Journal of Law, Medicine and Ethics, shows that physicians are subject to motivated bias when it comes to assessing the policies intended to regulate their behavior. The research team also tested financial planners and found similar stronger patterns of motivated bias.

Loewenstein, the Herbert A. Simon University Professor of Psychology and Economics, noted that the genesis of the research was "when we became aware of a hospital implementing a conflict of interest policy, heard the objections that were being raised against the policy, and wondered whether they would seem as ridiculous if they were disguised as dealing with a different field.

"Generally, the medical and financial industries have made their own rules," continued Loewenstein, an expert on bias, conflicts of interest and decision-making. "This research suggests that, while medical and financial professionals should be involved in drafting and implementing their respective conflict of interest policies, they should also bring third parties into the process."

For the study, the researchers surveyed a mix of physicians, financial planners and a comparably educated control group who were randomly assigned to provide their reactions to a conflict of interest policy that was presented as applying either in a medical context, involving relationships between physicians and the pharmaceutical industry, or in a financial context, dealing with relationships between personal financial planners and companies that market investments. After reading the proposed policies, participants rated how reasonable they thought the policies were. Then, they were presented with objections to the policies and asked to evaluate those.  Finally, in light of the objections they had seen and evaluated, they were asked whether their views on the original policy had changed.

The results revealed a strikingly consistent pattern of motivated bias for both physicians and financial planners. Physicians evaluating a conflict of interest policy in a medical context evaluated it negatively, and perceived objections to the policy as largely reasonable.  However, when examining the same policies in a financial planning context, physicians were supportive of policies to limit conflict of interest. Financial planners displayed a similar pattern, reacting negatively to the policy that would affect them, but positively when examining the restrictions in a medical context.  Finally, the control group with no vested interest evaluated both policies positively, and dismissed the objections as being unreasonable.  

"Physicians and financial planners were supportive of a conflict of interest policy, as long as it did not apply to their own industry," emphasized Schoen, professor of medicine at Pitt.

Sharek, a doctoral candidate in CMU's Tepper School of Business and lead author of the study, added, "These results suggest that people with vested interests are not only biased, but that they are not aware that they are biased. Additionally, there is nothing to indicate that physicians and financial planners are unique in this regard, so we would expect that this pattern of bias would occur in other domains, such as politics, public policy, and business."

###

Pictured above is George Loewenstein, the Herbert A. Simon University Professor of Psychology and Economics.