News Release

ER patients may care less about a doctor's race and gender than previously thought

Black or white, man or woman, providers got similar satisfaction and confidence scores from simulated patients in new research with implications for quality reporting

Peer-Reviewed Publication

Michigan Medicine - University of Michigan

When a patient arrives at an American emergency room today, they have a higher chance than ever before of seeing a doctor who's a woman or a person of color. And they're also more likely than ever to get a survey after they go home, asking how satisfied they were with their ER care.

But some have worried that a combination of increased physician diversity, underlying patient biases, and more emphasis on ER satisfaction ratings could create issues - especially if hospitals with low ER patient ratings face public reporting or even financial penalties in future.

A new study by a team from Yale University and the University of Michigan may ease some of those fears.

Published in JAMA Network Open, the study finds that patients rated their satisfaction with a simulated ER visit just as highly if the doctor randomly assigned to them was white or African-American, and man or woman. Confidence in their doctor also didn't vary based on physician gender or race.

This held true even after the researchers took into account variation in respondents' demographic characteristics and attitudes like prejudice and sexism.

Surprise findings

Lead author Rachel Solnick, M.D., M.Sc., notes that the study doesn't mean that emergency physicians won't ever face individual situations where a patient expresses bias against them because of race, gender, age or other factors.

In fact, it was her own experience with such biases as an emergency medicine resident, as well as previous research by others, that led Solnick to do the study.

The fact that ER patients don't get to choose which doctor they see, unlike with most other fields of medicine, adds to the potential for bias to creep into the doctor-patient interaction.

So, she says, "We were really surprised that even after looking at these data in many different ways, we did not see evidence of racial or gender bias affecting patient satisfaction or confidence. This is not to say that people are bias-free, but it did not appear to enter into their ratings of care in surveys. This is good news, with a grain of salt."

More about the study

Solnick, now a National Clinician Scholar at the U-M Institute for Healthcare Policy and Innovation, sees patients as an emergency medicine physician at Michigan Medicine, U-M's academic medical center. She's one of the approximately 25% of American ER physicians who are women; 5% of all ER physicians identify as African-American.

Working with Yale colleagues including Kyle Peyton, Ph.D., Gordon Kraft-Todd, Ph.D. (now at Boston College) and Basmah Safdar, M.D., M.Sc., the team designed a study that drew its "patients" from two online platforms commonly used in academic research. More than 1,600 of them were recruited from the Amazon Mechanical Turk interface, and another 1,600 were recruited through Lucid using a quota sample matched to Census demographics.

Each "patient" received the same scenario: they had gone to the ER for stomach symptoms, and received tests, a diagnosis of viral gastroenteritis and supportive treatment plan from the simulated physician randomly assigned to them.

Alongside the physician diagnosis was a contradictory diagnosis from an online symptom checker, which suggested their symptoms might indicate a much more serious issue: an infected appendix. This contrast was meant to instill doubt in the doctor's competence in providing appropriate care.

Assessing patient bias toward or against certain types of ER physicians is especially important at this time, Solnick adds.

The federal Centers for Medicare and Medicaid Services has been testing the feasibility of a new survey tool called the Emergency Department Patient Experiences with Care (EDPEC) for hospitals to use with ER patients who did not get admitted to the hospital.

Such a tool could be used in future to impose financial penalties on hospitals with low ER satisfaction ratings, just as patient ratings for other types of care already do. The new CMS survey tool is in addition to private ones that hospitals already use in-house.

"Patient satisfaction surveys have really altered the fabric of the emergency doctor/patient relationship, because in addition to thinking about the medical treatment we also are considering how we will be 'graded' at the end of the encounter," says Solnick.

This isn't necessarily a bad thing, she emphasizes. In fact, it has made clear communication throughout an ER visit even more important -- even as ERs cope with crowding and have to accept every patient who comes to them.

Next steps

The current research didn't investigate patient assessments of doctors from other racial and ethnic backgrounds, immigration statuses, ages or national origins. Also, the scenario the researchers studied was based in an ER setting, and other areas of the healthcare system where patients have a longer term relationship with their doctor or exercise greater discretion in physician choice may reveal different results. That could be another avenue for follow-up studies.

Although this study did not find that physician race or gender negatively affected patient satisfaction, on average, there is compelling evidence that increasing the diversity of the physician workforce has positive benefits for public health. For example, a recent field experiment in Oakland, CA found that black men treated by black male physicians reported greater satisfaction and took more preventative health measures than black men treated by white male physicians.

But at the same time, minority physicians and women physicians often experience higher rates of burn out from their job and some report higher rates of job turnover due to discrimination.

Solnick and colleagues feel that this new data points the research spotlight towards other sources of potential workplace discrimination- colleagues, superiors, and policies- and the need for institutions to identify and evolve in order to sustain and encourage a diverse workforce.


Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.