The influence of emergency physician gender on patient experience surveys.

backgroundPatient experience surveys are used to evaluate emergency physician (EP) performance, but they may be subject to gender bias. This study aimed to quantify the impact of EP gender on (1) the use of communal and agentic descriptors in free-text comments and (2) quantitative scores on patient experience surveys.

methodsThis retrospective study was conducted in a single urban, academic tertiary care emergency department (ED) that serves a diverse patient population. We included surveys for patients discharged from the ED during 1/1/19-12/31/21. Primary outcomes were the use of positive communal and agentic descriptors of the EP in the free-text portions of the patient experience surveys. Secondary outcome was mean quantitative physician score. Analyses included descriptive statistics and bivariate analyses, as well as a mixed effects model adjusted for patient and physician demographics and ED length of stay.

resultsWe studied 883 encounters (501 [57 %] female patients) with applicable free-text comments in the patient experience surveys. In a multivariable mixed effects model adjusted for patient, physician, and operational variables, all-women, all-men, and mixed gender physician teams were equally likely to be described by positive communal terms and positive agentic terms. We also studied 3707 encounters (2077 [56 %] female patients) with quantitative physician scores available in the patient experience surveys. In a similar multivariable mixed effects model, physician gender was not associated with differences in mean physician scores. The median age in the larger cohort was 50 (IQR 26-64), and older patient age was associated with higher quantitative scores (p = 0.004).

conclusionAmong ED patients, physician gender was not associated with the language used to describe physicians or the quantitative scores assigned to them.