Abstract
OBJECTIVE: Prior research, mostly from North America, suggests improved clinical outcomes for female patients treated by female physicians. Whether these findings apply in European healthcare systems and how underlying processes of care vary by sex remains unclear. This study aimed to assess whether in-hospital outcomes and processes of care differ by patient sex, physician sex or their interaction, in a European setting. DESIGN: Retrospective cohort study. SETTING: General internal medicine division of a Swiss tertiary teaching hospital. PARTICIPANTS: Adult inpatients (≥18 years) hospitalised between 2014 and 2024 and their primarily responsible physicians, classified by administrative sex (male vs female). The cohort included 20 094 hospitalisations (44.6% female patients) and 216 physicians (48.1% female). OUTCOME MEASURES: Outcomes included in-hospital mortality, 30-day mortality and 30-day readmission, as well as processes of care (resource use, advance care planning and cardiovascular low-value care). Multilevel mixed-effects regression models adjusted for patient and physician characteristics. RESULTS: Female patients had lower in-hospital mortality (OR 0.72, 95% CI 0.59 to 0.89) and 30-day mortality (OR 0.75, 95% CI 0.65 to 0.87), lower hospitalisation costs (-4.26%, 95% CI -6.08% to -2.41%), fewer diagnostic and therapeutic procedures (-6.44%, 95% CI -9.76% to -3.01%), fewer blood tests (-8.95%, 95% CI -12.98% to -4.73%) and were less likely to have resuscitation orders (OR 0.64, 95% CI 0.58 to 0.71) or intensive care unit transfer orders (OR 0.64, 95% CI 0.55 to 0.73). They were, however, more likely to receive non-indicated antihypertensive treatment (OR 1.86, 95% CI 1.38 to 2.51). No significant differences were observed by physician sex or patient-physician sex interaction. CONCLUSIONS: In a European tertiary teaching hospital with universal healthcare coverage, in-hospital outcomes and processes of care did not differ by physician sex or patient-physician sex interaction. Nevertheless, disparities by patient sex persisted, underscoring the need for sex-disaggregated quality monitoring and sex-sensitive medical training.