Patient-physician sex concordance and real-world disparities in cancer treatment practices and outcomes in medical oncology

Abstract

The impact of sex inequities in healthcare may be magnified in medical oncology, where patients and physicians often navigate life-limiting illnesses and intensive treatments. We examined the associations among patient-physician sex concordance, treatment practices, and cancer outcomes. This was a population-based, retrospective cohort study of adults diagnosed with stage II-IV lung or colon cancer in 2013–2020 in Alberta, Canada and referred to a medical oncologist. We classified patient-physician dyads as sex concordant (female-female, male-male) or discordant (female-male, male-female). We analyzed time-to-event data using Kaplan-Meier methods and associations with Cox and logistic regression models. We identified 12,047 patients treated by 209 medical oncologists. In multivariable analysis of the propensity score-matching (PSM) cohort, sex concordance was not independently associated with overall survival (OS) or cancer-specific survival (CSS) in the overall cohort and in female patients. But among male stage IV patients treated by female physicians, sex discordance was significantly associated with lower OS and CSS ( p = 0.02). Upon propensity score-matching, more oncologists prescribing systemic therapy for stage II-III disease were female (25.8% vs. 22.6%, p = 0.02), whereas in stage IV disease, more prescribing oncologists were male (64.7% vs. 57.9%, p < 0.001). Sex discordance was not associated with worse OS or CSS among the overall cohort. Cancer management and outcomes may be prone to the effects of sex bias in specific patient-physician relationships.