Understanding Low Breast Cancer Screening Rates by Resident Physicians.

Abstract

Background: While research has shown that medical residents often achieve poorer cancer screening rates than faculty in the same practice, residents' disadvantages in experience and continuity may be overcome in high-functioning primary care practices. To explore this observation, we compared breast cancer screening rates of resident and faculty physicians at an academic primary care practice with robust practice support.

Methods: We measured breast cancer screening compliance in CY2025 using tools embedded in EPIC. Patients were considered compliant if they were women age 40-74 with a mammogram within 2 years and no disqualifying characteristics. We abstracted patient data from the electronic medical record and collected physician data on gender, language facility, and years in practice. We compared breast cancer screening rates by physician type in bivariate and multivariable logistic regression models.

Results: Resident patients comprised 1024 (12%) of the 8825 patients eligible for breast cancer screening. Faculty panels included a higher percentage of patients who were white, English speakers, on commercial insurance, and with more medical comorbidities. The resident panel had more patients with serious mental illness and fewer primary care encounters. Faculty physician exceeded resident performance overall (83% vs. 66%, p < 0.001). The odds ratio of a resident patient having up-to-date breast cancer screening was 0.41 (95% confidence interval 0.35-0.47) in the unadjusted model. The adjusted odds ratios improved from 0.5 (0.40-0.62) in model 1 (patient sociodemographic only) to 0.59 (0.48-0.74) in model 2 (clinical and utilization variables) to 0.73 (0.54-0.98) in model 3 (physician attributes).

Conclusion: In a teaching practice with a robust primary care infrastructure, disparities in breast cancer screening between faculty and resident panels improved with rigorous adjustment. However, residency training programs may need to be redesigned to eliminate remaining disparities in care.