MedPage Today story.
Residents and medical students recalled clinical information with less accuracy after hearing a patient handoff rife with biased language, a survey study found.
Those who heard handoffs with blame-based bias had less accurate recall than those who heard neutral handoffs (77% vs 93%, P=0.005), according to Austin Wesevich, MD, MPH, MS, of the University of Chicago, and colleagues.
In addition, participants had less positive attitudes toward patients as measured by Provider Attitudes Toward Sickle Cell Patients Scale (PASS) scores after hearing biased handoffs (mean scores 22.9 vs 25.2, P<0.001), they reported in JAMA Network Open.
“People are getting it right more often when they hear the neutral [version],” Wesevich told MedPage Today. “It’s pretty compelling that there’s some sort of distraction or cognitive effect of having to neutralize out the biased phrases.”
In an invited commentary, Somnath Saha, MD, MPH, and Mary Catherine Beach, MD, MPH, of Johns Hopkins University in Baltimore, called the effect of biased language on information recall “intriguing.”
“At first glance, it seems that this effect might be a matter of simple distraction,” they wrote. “The human brain has finite capacity for retaining information, and the addition of clinically uninformative details might have crowded out information that was clinically important.”
The findings suggest that “by engendering more negative attitudes toward the patient,” they added, “stigmatizing language may lessen the clinician’s investment in the patient’s well-being and thereby reduce the clinician’s motivation to retain the information needed to provide optimal care for the patient.”
Wesevich and colleagues surveyed 169 residents and medical students (56% female, mean age 28.6), each of whom heard three recorded handoffs, which were a combination of neutral and biased. The biased handoffs were one of three types — stereotype, blame, or doubt — and they all came from a previous study by Wesevich’s group.
That study involved recording residents doing handoffs, and it found that biased phrases were more common in notes about patients who were Black or obese. While they changed small details to protect patient identity, the biased handoffs are largely true to the words used in the recorded handoffs in the earlier study, Wesevich told MedPage Today.
An example of a biased handoff with blame was about an adult patient with diabetes who “cut her dose in half 2 months ago because she was having some falls and thought she was getting hypoglycemic but never actually checked her blood glucose.” The neutral version of that handoff was rephrased to describe someone who “cut her dose in half after experiencing hypoglycemic symptoms.”
After residents listened to three handoffs, the researchers followed up with three items: a multiple-choice clinical information recall question; an adapted version of the PASS scale to assess attitudes towards patients; and three open-ended “key takeaway” responses.
Overall, Wesevich and colleagues found that those who heard biased handoffs had less accurate clinical information recall compared with those who heard neutral ones, but the difference wasn’t statistically significant.
They also found that those who reported bias as a key takeaway of the handoff had lower clinical information recall accuracy than those who did not (85% vs 93%; OR 0.47, 95% CI 0.26-0.85, P=0.01).
In addition, participants had lower empathy scores as measured by PASS item 2 when hearing biased handoffs compared with neutral ones (3.1 vs 3.3, P<0.001), and more positive attitudes toward patients were associated with higher clinical information recall accuracy (OR 1.12, 95% CI 1.02-1.22).
Wesevich and colleagues concluded that biased handoffs can “adversely impact the quality of care for Black patients” as they impeded the accurate transfer of key clinical information and diminished empathy, “potentially endangering patients and worsening health disparities.
“If you’re trying to add context to who someone is … and you choose negative phrases, you wield a lot of power about how that person is viewed,” Wesevich said, adding that it would be helpful for physicians to have trainings or monitoring or “accept the recommendations by multiple national organizations to standardize the handoff process.”
“It’s okay to recognize that we have biases and stereotypes,” he added. “If clinicians can accept that they’re human beings that have things to work on, and by taking some of these measures, we can try to reduce the impact of implicit biases on the care of patients — that level of honesty and introspection will go a long way.”
The study was limited because the handoffs were simulated, and thus could be different in clinical settings. Also, the findings may not be generalizable outside of residents and medical students.
