88 - Assessment of Biased Language in Nightly Resident Handover

Poster Number: 88
Publication Number: 88.211

Austin Wesevich, University of Chicago Division of the Biological Sciences The Pritzker School of Medicine, Chicago, IL, United States; Sonya V. Patel-Nguyen, Duke University School of Medicine, Durham, NC, United States; Ilona Fridman, University of North Carolina, Carboro, NC, United States; Erica Langan, Duke University, Durham, NC, United States; Victoria Parente, Duke University School of Medicine, Durham, NC, United States

Background: Black and Latinx patients disproportionately experience negative hospital outcomes compared to White non-Latinx patients. Although many factors likely contribute, clinician implicit bias may exacerbate these disparities. Evidence suggests bias can be transmitted from one clinician to another in written documentation and negatively impact management decisions. Despite the ubiquity of patient handovers in the hospital, few have investigated presence of bias in these communications.

Objective: In this study, we aimed to measure stigmatizing language in resident handovers and describe types of bias in handover of White patients and families versus patients and families of color.

Design/Methods: This cross-sectional observational study used audio recordings of resident handover of general medicine and pediatric inpatient teams to objectively measure stigmatizing language. Audio recordings were performed on 55 intern patient handovers. Recordings were transcribed then independently coded by two physicians for occurrences of stereotypes, blame, and doubt. Discrepancies were reconciled. Patient race, ethnicity and gender were obtained through the electronic health record. Occurrences of bias were compared by race, ethnicity, and gender using Chi-squared and Fisher exact tests.

Results: A total of 302 patient handovers were transcribed and coded, 183 from general medicine (61%) and 119 from general pediatrics (39%). Patients in handovers were 46% Black, 35% White, 10% Hispanic, and 9% another race/ethnicity (Asian, American Native, Multiracial, or Other).  Approximately half (52%) were female. Bias was identified in 70 handovers (23%), with 34 showing stereotypes (49%), 39 showing blame (56%), and 15 showing doubt (21%). Handovers of Black patients were more likely to have stereotype bias than White patients (18% vs. 5%, p=0.001). Handovers of Hispanic patients had similar levels of any bias as those of non-Hispanic patients (17% vs. 24%, p=0.5). Handovers of patients identifying as another race/ethnicity (not Black, White, or Hispanic) were more likely to have any bias than White patients (43% vs 14%, p=0.005). Gender was not associated with the presence of bias.Conclusion(s): Stigmatizing language was present in almost a quarter of resident handovers. Biased language was more common in handovers of patients of color, which may negatively impact subsequent providers clinical care of that patient.  Future studies are needed to evaluate the clinical impact of the presence of bias in handover and the effectiveness of interventions to reduce biased language in discussions of patients and their families.