Breast Cancer Knowledge, Occupational Exposures, and Screening Practice Among Female Healthcare Workers in Ethiopia: A Machine Learning-Based Analysis

Methods: An institution-based cross-sectional study was conducted among 293 female healthcare workers from five hospitals in Addis Ababa. Data were collected using structured questionnaires and face-to-face interviews. Analyses included descriptive statistics, chi-square tests, binary logistic regression, and Extreme Gradient Boosting (XGBoost) models with SHAP interpretability.

Results: The mean age of participants was 29.0 ± 5.9 years, and the mean breast cancer knowledge score was 5.2 ± 1.5 (out of 8). Only 12.3% had ever undergone biopsy or mammography. Higher educational level, older age, and greater exposure scores were significantly associated with better knowledge (p < 0.0001). XGBoost models demonstrated excellent predictive performance (AUC = 0.951 for knowledge, 0.988 for exposure, and 0.920 for screening history). Key predictors included age, contraceptive use, reproductive history, and exposure–knowledge interactions. Occupational chemical exposure was widespread (88.8%), particularly to alcohol-based disinfectants.

Conclusion: Despite moderate knowledge, breast cancer screening uptake among female healthcare workers in Addis Ababa remains alarmingly low. Occupational exposures are common, underscoring the need for workplace safety measures, continuous professional development, and institutional screening programs to empower healthcare workers as agents of change in Ethiopia’s breast cancer control efforts.