Qualitative Analysis of Medical Student Descriptions of Public Humiliation in the Graduation Questionnaire

Background

The Association of American Medical Colleges’ Graduation Questionnaire (GQ), administered annually to all graduating medical students in the U.S., indicates that public humiliation is the most frequently reported form of mistreatment, with one in five (19. 5%) students reporting at least one incident. To explore the type and frequency of “public humiliation” experiences students said were committed by health professionals other than physicians, we conducted a content analysis of written responses from the only year in which this question was asked: “Please briefly describe the behavior you personally experienced during medical school as being ‘publicly humiliated.’”

Methods

From the 2015 GQ survey, a random sample of 1,000 from the 5,438 responses to the public humiliation free-text question were selected. Using grounded theory methodology, three raters independently coded an initial test set of 150 items. Coding schemata were refined through virtual meetings and final coding was determined by agreement of two of three raters; where agreement was not achieved, raters discussed individual items. Descriptive statistics and inter-rater reliability were performed along with Chi-square analysis, where appropriate.

Results

Among the 1000 student responses selected, 1103 unique reports of humiliation were identified. Inter-rater reliability ranged from 0. 399-1 and achieved at least moderate agreement (>0. 41) on 82% of variables. Most responses attributed mistreatment to a specific perpetrator (n=731). Of these, physicians were implicated 83. 7% of the time (attendings 420 [57. 5%] and trainees 192 [26. 3%]), followed by nursing staff (29, 4. 0%), non-clinical faculty (27, 3. 7%) and operating room (OR) staff (25, 3. 4%). Of non-physician related events describing the setting (n=86), most occurred in the OR (n=29, 33. 7%) or during a clinical rotation (n=24, 27. 9%). Of those listing a specific clerkship (n=23), 34. 8% (n=8) implicated obstetrics/gynecology, 19. 2% (n=62) and 30. 4% (n=7) surgery.

Discussion

These findings demonstrate that many types of interprofessional mistreatment are not adequately captured by current survey methodology, and this mistreatment remains a pressing issue given its relationship with markers of wellbeing.

Significance

Mistreatment occurs throughout the learning environment and is not limited to physician-to-physician behaviors; thus, the entire learning environment must be addressed holistically through both education and assessment. Given the varied perspectives of different professions and historical training silos, surveying all health profession learners along the continuum of their education may better capture the nature of mistreatment in the clinical learning environment.