Exploring Gender Bias in Nursing Evaluations of Emergency Medicine Residents

OBJECTIVES Nursing evaluations are an important component of residents' professional development as nurses are present for interactions with patients and nonphysician providers. Despite this, there has been few prior studies on the benefits, harms, or effectiveness of using nursing evaluations to help guide emergency medicine residents' development. We hypothesized that gender bias exists in nursing evaluations and that female residents, compared to their male counterparts, would receive more negative feedback on the perception of their interpersonal communication skills. METHODS Data were drawn from nursing evaluations of residents between March 2013 and April 2016. All comments were coded if they contained words falling into four main categories: standout, ability, grindstone, and interpersonal. This methodology and the list of words that guided coding were based on the work of prior scholars. Names and gendered pronouns were obscured and each comment was manually reviewed and coded for valence (positive, neutral, negative) and strength (certain or tentative) by at least two members of the research team. Following the qualitative coding, quantitative analysis was performed to test for differences. To evaluate whether any measurable differences in ability between male and female residents existed, we compiled and compared American Board of Emergency Medicine in-training examination scores and relevant milestone evaluations between female and male residents from the same period in which the residents were evaluated by nursing staff. RESULTS Of 1,112 nursing evaluations, 30% contained comments. Chi-square tests on the distribution of valence (positive, neutral, or negative) indicated statistically significant differences in ability and grindstone categories based on the gender of the resident. A total of 51% of ability comments about female residents were negative compared to 20% of those about male residents (χ2 = 11.83, p