Conflicts of interest and your physician: psychological processes that cause unexpected changes in behavior

The ubiquitous nature of medical conflicts of interest is attracting increased attention from physicians, policymakers, and patients. However, little work has examined the psychological processes at play in the presence of such conflicts. I investigate the subtle influences arising from conflicts of interest that change behavior in both physicians and patients. First, I explore why physicians accept gifts from pharmaceutical companies and medical device manufacturers that appear, to many critics, to be unethical. I review evidence from my published and ongoing research that demonstrates two psychological processes that enable physicians to accept industry gifts: (a) a sense of entitlement and (b) a sense of invulnerability to the biasing effects of conflicts of interest. Second, I investigate the situations that may increase or decrease bias. I find that people, subject to a financial conflict of interest, show greater bias in their advice when they feel less able to identify with the advice-recipient(s). This, perversely, leads to advisors giving more biased advice to groups of people than to one identified individual. Finally, I examine the impact of the conflicted advice on the patient and the success of policies intended to manage such conflicts. Mandatory second opinions and disclosure are often advocated as potential solutions to deal with conflicts of interest. However, both policies have limitations and can sometimes make matters worse. A primary advisor who knows about a second advisor may give even more biased advice since the presence of a second advisor undermines the relationship with the primary advisor. Also, although disclosure of a conflict of interest does have the intended effect of causing patients to trust the advice they receive less, I find that it also has an additional unintended consequence: it creates increased pressure to comply with the (distrusted) advice. This increased pressure occurs because patients want to avoid appearing as though they believe that the (now disclosed) conflict of interest has corrupted their doctors' advice. Thus, instead of being merely a warning, disclosure can become a burdensome request to comply with advice the patient trusts less. These results highlight potential pitfalls that new policies intended to address conflicts of interest should seek to avoid.