This cohort study analyzed data from 229,695 female patients with early breast cancer to identify factors associated with nonpersistence and nonadherence to prescribed endocrine therapy. Using a deep learning model, the study found that age, treatment type, past adherence, and income were significant predictors of nonpersistence and nonadherence. The study highlights the importance of addressing socioeconomic and behavioral factors in improving treatment adherence for breast cancer patients.
