Experiences of US medical students

Purpose: To determine if medical students of different races/ethnicities or genders have different perceptions of bias in the United States (US).

Methods: An IRB-approved, anonymous survey was sent to US medical students from November 2022 through February 2024. Students responded to statements regarding perceptions of bias toward them from attendings, patients, and classmates. Chi-square tests, or Fisher’s exact tests, when appropriate, were used to calculate if significant differences exist among genders or races/ethnicities in response to these statements.

Results: 370 students responded to this survey. Most respondents were women (n=259, 70%), and nearly half were White (n=164, 44.3%). 8.5% of women agreed that they felt excluded by attendings due to their gender, compared to 2.9% of men (p=0.018). 87.5% and 73.3% of Hispanic and Black students agreed that bias due to race negatively impacted research opportunities compared to 37.2% of White students (p<0.001). 87% and 85.7% of White and Asian/Pacific Islander students, respectively, believed that an attending would treat multiple trainees the same regardless of religion compared to 50% of Black students (p<0.001). A greater proportion of Hispanic and Black students than White and Asian/Pacific Islander students believe that bias due to gender affects research opportunities. More than a quarter of students personally experienced some form of bias from attending with no significant difference based on gender or race/ethnicity. Most women (66%) reported that they were assumed to be a nurse compared to 1.9% of men (p<0.001). 30% of Black, 18.8% of Hispanic, and 12.2% of Asian/Pacific Islander students felt excluded by their classmates due to their race compared to 4.3% of White students (p<0.001).

Conclusion: Gender or race/ethnicity affects perceptions of bias from attendings, patients, and classmates. This study lends insight into the experiences and perceptions of medical students of different racial/ethnic and gender identities backgrounds in the US.

Methods
This data was collected through Google Forms, and respondents were asked to log in with their email addresses to make sure that they could only submit their responses once. Data was processed in R studio.