Investigating Association Between Sex and Faculty Teaching Evaluation in General Surgery Residency Programs: A Multi-Institutional Study

Video-Based Assessment of Microsurgical Trainees: An Evaluation of Gender Bias

BACKGROUND

 Gender bias in graduate medical evaluations remains a challenging issue. This study evaluates implicit gender bias in video-based evaluations of microsurgical technique, which has not previously been described in the literature.

METHODS

 Two videos were recorded of microsurgical anastomosis; the first was performed by a hand/microsurgery fellow and the second by an expert microsurgeon. A total of 150 surgeons with microsurgical experience were recruited to evaluate the videos; they were told these videos depicted a surgical trainee 1 month into fellowship followed by the same trainee 10 months later. The only variable was the name ("Rachel" or "David") that each participant was randomly assigned to evaluate. Participants were asked to score each video for quality, technique, efficiency, as well as overall progression and development after the second video compared with the initial video. To focus on bias, these outcome measures were selected to be purposefully subjective and all ratings were based on a subjective 1to 10 scale (10 = excellent).

RESULTS

 The analysis included 150 participants (75% male). There were no statistically significant differences in scores between the "female" and "male" trainee. The trainees received the same median initial (1-month video) and final (11th-month video) scores for all criteria except initial technique, in which the female trainee received a 7 and the male trainee received an 8. Notably, 11-month scores were consistently the same or lower than 1-month scores for both study groups (p < 0.001). There were also no differences within either study group based on participant sex. Microsurgery practitioners overall rated both groups lower than those who do not currently practice microsurgery.

CONCLUSION

 Our study did not identify a gender bias in this evaluation method. Further investigation into how we assess and grade trainees as well as the presence and impact of implicit biases on varying surgical assessment methods is warranted.

Bias in the trauma bay: A multicenter qualitative study on team communication

BACKGROUND

Team communication and bias in and out of the operating room have been shown to impact patient outcomes. Limited data exist regarding the impact of communication bias during trauma resuscitation and multidisciplinary team performance on patient outcomes. We sought to characterize bias in communication among health care clinicians during trauma resuscitations.

METHODS

Participation from multidisciplinary trauma team members (emergency medicine and surgery faculty, residents, nurses, medical students, emergency medical services personnel) was solicited from verified level 1 trauma centers. Comprehensive semistructured interviews were conducted and recorded for analysis; sample size was determined by saturation. Interviews were led by a team of doctorate communications experts. Central themes regarding bias were identified using Leximancer analytic software (Leximancer Pty Ltd., Brisbane, Australia).

RESULTS

Interviews with 40 team members (54% female, 82% White) from 5 geographically diverse Level 1 trauma centers were conducted. More than 14,000 words were analyzed. Statements regarding bias were analyzed and revealed a consensus that multiple forms of communication bias are present in the trauma bay. The presence of bias is primarily related to sex but was also influenced by race, experience, and occasionally the leader's age, weight, and height. The most commonly described targets of bias were females and non-White providers unfamiliar to the rest of the trauma team. Most common sources of bias were White male surgeons, female nurses, and nonhospital staff. Participants perceived bias being unconscious but affecting patient care.

CONCLUSION

Bias in the trauma bay is a barrier to effective team communication. Identification of common targets and sources of biases may lead to more effective communication and workflow in the trauma bay.

LEVEL OF EVIDENCE

Prognostic and Epidemiological; Level IV.

Beyond the ratings: gender effects in written comments from clinical teaching assessments

Assessment of clinical teachers by learners is problematic. Construct-irrelevant factors influence ratings, and women teachers often receive lower ratings than men. However, most studies focus only on numeric scores. Therefore, the authors analyzed written comments on 4032 teacher assessments, representing 282 women and 448 men teachers in one Department of Medicine, to explore for gender differences. NVivo was used to search for 61 evidence- and theoretically-based terms purported to reflect teaching excellence, which were analyzed using 2 × 2 chi-squared tests. The Linguistic Index and Word Count (LIWC) was used to categorize comment data, which were analyzed using linear regressions. The only significant difference in NVivo was that men were more likely than women to have the word "available" in a comment (OR 1.4, p < .05). A subset of LIWC variables showed significant gender differences, but all effects were modest. Men teachers had more positive emotion words written about them, while negative emotion words appeared equally. Significant differences occurred more often between the men and women residents who wrote the comments, rather than those attributed to the gender of the teachers. For example, women residents used more social and gender-related words (β 1.87, p < 0.001) and fewer words related to power or achievement (β -3.78, p < 0.001) than men residents. Profound gender differences were not found in teacher assessment comments in this large, diverse academic department of medicine, which differs from other studies. The authors explore possible reasons including differences in departmental culture and issues related to the methods used.