People might prefer to think of doctors as rational experts, assessing the patient’s health without bias or judgement, but the truth can be very different. Depending on their political affiliations, doctors responded very differently to the same patient stories in a new study.

Yale University's Dr Eitan Hersh and Matthew Goldenberg note in the Proceedings of the National Academy of Sciences that “political beliefs have been shown to spill over into nonpolitical domains, such as consumer spending, choice of romantic partner, and job hiring.”

To test if this effect influences the health care patients get, Hersh and Goldenberg compared a list of 20,000 primary care physicians with voter registration data in American states that record party affiliation. They were able to select more than a thousand doctors registered as either Democrats or Republicans, 233 of whom responded to a survey on how they would react to patients presenting with certain histories and symptoms.

Political affiliation didn't affect how doctors reported they would respond to patients presenting with histories of depression or heavy drinking, but the same did not apply to more politically charged questions. Democrat doctors expressed more alarm at the idea of parents with young children having firearms in the house. Republicans, on the other hand, were likely to treat marijuana smokers' health as being at risk, even without physical symptoms.

The largest partisan difference was in how the doctors said they would respond to a patient who reported having had two abortions in the previous five years, but described no health effects and was not pregnant at the time. Male Republican doctors in particular were very likely to describe the patient's condition as serious, blame any health issues on the abortions, and lecture the patient on alleged consequences.

To avoid the possibility that the physicians' approach might be shaped by the demographics of their regular clients, the group was weighted towards examples where doctors of opposing political affiliations worked together. Participants in the study were not alerted to the purpose of the study or that their affiliations were known.

The project was inspired by a Florida state law (upheld last year by a Federal Court of Appeal) that bans doctors from discussing gun safety with their patients unless it is “directly relevant to patient care”. As the authors note, conversations between doctors and patients can touch on many politically sensitive topics, of which guns are not necessarily the most likely to prove problematic.

For people susceptible to medical prejudice, the paper's findings may not be news. Lists of lesbian, gay, bisexual and transgender-friendly doctors exist precisely because so many people in those communities have had bad experiences with doctors who jump to the conclusion that their sexuality must be responsible for unrelated health problems. Hersh and Goldenberg's work confirms this experience empirically, and shows how many issues it can extend to.