In a new study published in the journal of the Proceedings of the National Academy of Sciences, a group of Yale researchers linked publicly available information on more than 42,000 practicing physicians to a list of party affiliations that’s regularly used by political campaigns to target communications. After winnowing down their data to correct for unaffiliated and unreachable physicians, they mailed out a survey to a sample of the doctors.
The 233 physicians who responded did not know that the survey was about their politics. Rather, they thought it was about how they administer the "social" survey, the interview that primary-care doctors give patients before they first treat them. These conversations can be lengthy and cover everything from employment to health risks, so researchers hoped they would be a good place to glean information about how doctors deal with politically charged issues with their patients.
Participants were asked to evaluate nine hypothetical scenarios they might face with patients. With each vignette, they were asked to evaluate how concerned they’d be about the issue and how likely they’d be to engage in a list of potential treatment options. The scenarios covered everything from a patient who had regular sexual contact with sex workers to one who commutes to work by motorcycle but doesn’t wear a helmet.
The responses showed clear differences between providers identified independently as Democrats or Republicans when it came to politically charged issues. For example, Republicans were most concerned about scenarios that included multiple prior abortions and marijuana use. Democrats were most concerned about firearm access and patients who had sexual relationships with sex workers.
The treatment plans doctors said they would probably pursue also differed by party affiliation. Democratic doctors were less likely to discuss the health risks of marijuana, highlight its legal risks or encourage the patient to cut down use. Republican doctors were twice as likely as Democrats to discourage patients from having more abortions in the future and 35 percent more likely to discuss mental health in connection with abortions.
Overall, Republican physicians were more likely than their Democratic counterparts to say that they would engage in “active treatment” options such as encouraging a patient to stop seeing sex workers or urge a patient to cut down on marijuana use. But when it came to less politicized issues such as consuming alcohol or smoking cigarettes, that difference diminished.
“We found that Democrats, not surprisingly, are more concerned about a scenario where a patient who had young kids is storing a firearm,” said Eitan Hersh, an assistant professor of political science at Yale University and the paper’s first author. Democratic respondents were 66 percent more likely to urge the patient not to store firearms at home, he says, “but were less likely than the Republican physicians to discuss storage practices.” Perhaps this demonstrates a knowledge gap, he says, that goes along with political sentiment.
Hersh says that strong partisan differences were immediately apparent despite the study’s small sample size. The researchers controlled for variables such as the demographic composition of a physician’s patient population and the gender, age and religious attendance of doctors. But since the study deals with hypothetical situations, it’s impossible to know whether party affiliation could affect outcomes for patients in real life.
Although the researchers acknowledge that the cause of the partisan differences could be some factor aside from political affiliation, the study offers no evidence of what that cause might be. But, Hersh says “Nothing predicts the outcome the way that partisanship does. I feel very confident that the results are real.”
It was hard for Hersh to get funding for what he calls “this controversial issue,” so he funded this project from his own research account. Now, he hopes it will serve as proof of concept for more work on how party affiliation affects patients in real life, from end-of-life decisions to LGBT health care.
Hersh hopes the study will open up a dialogue about how political bias affects medical decision-making and make practitioners more aware of ways in which their personal beliefs might affect the care they give patients. Depending on the study’s response, he says, he may even consider making his data on physician names and party affiliations public. Until political data can be linked to actual outcomes, though, it won’t be clear to what extent politics affect physicians’ decisions.
He is likely to encounter resistance among doctors who don’t think that their political views inform their practice decisions. “I never once treated a Republican or Democrat cancer in my life,” says Rep. Phil Roe (R-Tenn.), a retired obstetrician-gynecologist who co-chairs the GOP Doctors Caucus, an 18-member group of congressional medical providers devoted to health-care policy. “When a patient walked into my office, I didn’t know if they were a Republican or a Democrat, and I honestly didn’t care.”
Roe said in an emailed response to questions that he hopes patients will seek out the best care available regardless of their doctor’s political affiliation. “Party affiliation should have nothing to do with patient care.”
That may be true, but, Hersh says, there are good reasons to check out your doctor’s party affiliation before heading into the office. “Patients in a medical examination room are in a fairly vulnerable position,” he says. “They’ve put a lot of trust in their doctor. It’s important for all of us to understand how a doctor’s ideological biases might affect their judgment.”