Vivian La

Published in Science on August 5, 2024

By Vivian La

In hospital emergency rooms, female patients are less likely to receive pain medication than male patients who reported the same level of distress, a new study finds. Reported today in the Proceedings of the National Academy of Sciences, the result further documents that because of sex bias, women often receive less or different medical care than men.

The study is “really impressive,” says Diane Hoffmann, a health lawyer at the University of Maryland who studies pain management disparities and was not involved in the new work. Previous studies about sex bias in emergency rooms were either too small in scale or didn’t account for factors such as hospital crowds, pregnancy, or type of pain, she says, which could have explained any differences in pain management.

Because pain is so subjective, doctors instruct patients to rate it from a painless zero to an unbearable 10 on a scale that often includes cartoon faces mimicking the various levels of discomfort. Based on this measure, doctors and nurses in emergency rooms make quick decisions about whether to prescribe pain medication and, if so, to choose the type and strength of painkiller. The numerical scale is supposed to help standardize pain relief treatment.

Nevertheless, women often fear their pain won’t be taken as seriously as men’s pain despite the standardized scale, notes Vineet Arora, a doctor at the University of Chicago who was not involved in the new study. That’s because centuries-old stereotypes that cast women as overly emotional and dramatic have resulted in a well-documented bias that they’re similarly exaggerating their pain. “You know when your body is not working as well as it should, but the feeling that it’s being dismissed … I have definitely felt that,” Arora says. “I have also definitely felt patients feel that.”

Decisions about pain management get trickier when there isn’t an obvious physical source of trauma such as a broken leg to examine, so doctors rely on their perception of a patient’s pain. “That’s when bias creeps in,” Hoffmann says.

To probe the depth of such bias, Shoham Choshen-Hillel, a social psychologist at the Hebrew University Business School and the Federmann Center for the Study of Rationality, and colleagues examined discharge records for more than 17,000 emergency room patients at the Hadassah University Medical Center in Israel from 2015 to 2019. Records included patient demographic information, medications prescribed, a patient’s zero-to-10 pain rating, and other data related to the emergency room visit, such as the sexes and ages of the doctors who treated each patient. (The records examined noted a patient’s sex as either male or female and included no information on a patient’s gender identity.) The researchers focused on pain without obvious causes, such as headache or chest pain.

The team found a sex bias against female patients across all age groups and types of pain relief medication. Among female patients, 38% received some type of pain relief medication, compared with 47% of male patients with similar symptoms. Although female patients, on average, rated their pain slightly lower than male patients—6.64, compared with 6.81—the team controlled for this and found a similar disparity across all pain ratings, regardless of whether opioids or a less risky nonopioid such as ibuprofen was prescribed. 

To see whether these disparities transcend national borders, the researchers also analyzed more than 4000 discharge records of emergency room patients at University of Missouri Health Carein the United States. They found similar patterns across the same variables of age, pain levels, and type of medication.

Moving beyond medical records, the team also set up an experiment to demonstrate the bias. More than 100 nurses at the Missouri hospital were presented with a written scenario describing a patient’s case as well as a pain rating of nine out of 10. Each nurse was then asked to rate their perception of this pain from zero to 100, requiring them to make their own assessment of the case and not just repeat the given number. When the patient described was a man, nurses rated his pain closer to 90, averaging 80. When the patient was described as a woman, nurses rated her pain lower, averaging 72. In both the experiment and the data studies, disparities persisted regardless of the health care provider’s sex.

Between the two countries and the different variables, the results were “amazingly consistent,” Choshen-Hillel says. These disparities show a clear bias, she says, as health care providers are required by federal and hospital guidelines to prescribe pain medication based on reported pain scores, regardless of sex.

But many factors contribute to pain management decisions, and there might be more to the story than just a sex bias, Arora says. When she was training as a doctor, pain was considered a vital sign that had to be treated. It led to overprescription of pain medication and potentially contributed to the opioid epidemic, studies have shown. “There’s been a sea change in the way people have thought about treating pain, and maybe that brings the bias to the fore,” she says. This could mean doctors are being overly cautious toward certain patients about prescribing pain medication, resulting in a sex disparity. Women could be communicating their pain differently to doctors or just feeling less inclined to ask for medication.

In the future, Choshen-Hillel and her team want to further explore ways to address this bias in a clinical setting. That could look like a “nudge” in the form of a pop-up message when a nurse or doctor enters in a patient’s pain score, reminding them of hospital guidelines. “We can use a simple reminder, but let the doctors still decide because each case may be different.”

Although more research needs to be done about potential solutions, there could be immediate benefits to a study like this, says Jeff Linder, a primary care physician at Northwestern University who wasn’t involved in the study. Acknowledging to patients that there’s evidence of potential undertreatment could build trust, he says. “Doctoring, as much as we like to think of as a scientific endeavor, is a social, emotional activity.”