Evaluating Perspectives on Empathy in Anesthesiology Trainees: A Mixed-Methods Study of Pre-Anesthesia Encounters

Citation:

Tomobi O, Gouge C, Brashear A, Totzkay D, Chen F, Huffmyer J, Scalzo D, Nemergut EC, Guillow R. Evaluating Perspectives on Empathy in Anesthesiology Trainees: A Mixed-Methods Study of Pre-Anesthesia Encounters. Anesth Analg. 2026;XXX(00):1-8. doi:10.1213/ANE.0000000000007919

This prospective mixed-methods study from a single United States academic center followed twenty four anesthesiology trainees across one hundred fifty one pre-anesthesia encounters, measuring empathy from trainee, patient, and observer perspectives alongside trainee wellness and encounter time. Empathy scores were generally high, but statistical analyses suggested that more senior training level, trainee gender, wellness, encounter duration, and patient age were associated with small differences in empathy ratings. Because of serious risk of confounding, small sample size, measurement limitations, and cross-sectional design, certainty is very low and the findings are best seen as hypothesis-generating prompts for reflection on communication and education rather than as a basis for practice change.

Study at a glance

- Design and setting: Prospective observational mixed-methods study at a single United States academic anesthesiology department, enrolling anesthesiology trainees (postgraduate years 1 through 5 and fellows) and their adult surgical patients in pre-anesthesia holding areas between March 2024 and June 2024. Quantitative data included empathy and wellness scales plus timed encounters; qualitative data involved thematic analysis of perspectives on empathy (not detailed in this episode).

- Participants and measures: Twenty four anesthesiology trainees contributed one hundred fifty one observed pre-anesthesia encounters (mean patient age 56 years, approximately sixty two percent female, mostly American Society of Anesthesiologists physical status II–III). Trainees completed the Jefferson Scale of Empathy (self-rated empathy) and a ten item Resident Wellness Scale; patients completed the Jefferson Scale of Patient Perceptions of Physician Empathy after the encounter; a physician observer completed the Jefferson observer empathy scale and timed each encounter.

- Key quantitative findings: Mean self-rated empathy on the Jefferson physician scale was 118 (standard deviation 13.7) on a possible range of 20 to 140, indicating medium to high empathy. Mean patient-rated empathy was 32.9 (standard deviation 3.0) and mean observer-rated empathy was 33.9 (standard deviation 1.27) on scales ranging from 5 to 35, both in the high range. More senior postgraduate year was statistically correlated with shorter encounter time (correlation coefficient minus 0.52, P = 0.01) and lower scores in the “walking in the patient’s shoes” empathy subdomain (correlation coefficient minus 0.45, P = 0.03).

- Gender, wellness, and multivariable results: In gender-stratified analyses, male and female trainees had similar self-rated and observer-rated empathy, but patient-rated empathy was slightly higher for male trainees (mean 33.1 versus 32.2; P = 0.02). Multivariable linear regression suggested trainee gender remained associated with patient-rated empathy (confidence interval for the gender coefficient 0.321 to 1.80; P = 0.011) after adjustment for selected covariates. Higher trainee wellness scores were correlated with lower self-rated empathy (correlation coefficient minus 0.55, P = 0.01) and shorter encounter time (correlation coefficient minus 0.421, P = 0.041), and older patient age was correlated with lower patient-rated empathy (correlation coefficient minus 0.41, P = 0.047).

- Risk of bias and certainty of evidence: Overall risk of bias is judged serious, driven mainly by likely residual and unmeasured confounding, potential selection bias in trainee participation, and multiple exploratory analyses without prespecified primary hypotheses. Outcome measurement is based on validated scales, but observer-rated empathy shows poor internal consistency (Cronbach alpha 0.38), and clustering of encounters within trainees was not clearly handled. For all key outcomes (self-rated, patient-rated, and observer-rated empathy, encounter time, and wellness), certainty of evidence about any causal effects of trainee characteristics is rated very low by the GRADE approach.

- Practice implications: The study suggests that empathy in pre-anesthesia encounters is generally perceived as high, and that there may be subtle associations between training level, gender, wellness, encounter duration, and empathy scores. However, because the evidence is very low certainty and observational, it should not drive structural changes such as preferential assignment by gender or assumptions that shorter visits are inherently less empathic. Reasonable responses include using the findings to stimulate trainee reflection on maintaining empathic behaviors as efficiency increases, expanding explicit teaching and feedback around empathic communication in pre-anesthesia consultations, and designing more rigorous multicenter studies with adequate adjustment for confounding and clustering to clarify causal relationships.