Development and validation of a nomogram for predicting the probability of medication adherence to adjuvant endocrine therapy in breast cancer patients: a predictive modeling study.

[PURPOSE] Adherence to adjuvant endocrine therapy (AET) is critical for breast cancer prognosis, yet there is a current lack of convenient predictive tools that integrate multidimensional factors. This study aimed to develop a nomogram prediction model for forecasting AET adherence in breast cancer patients.

[METHODS] Clinical data from 403 breast cancer patients were collected and analyzed. Patients were randomly divided into training (n = 281) and validation (n = 122) cohorts at a 7:3 ratio. Risk factors influencing treatment adherence were screened using univariate and multivariate logistic regression. The nomogram was constructed and validated using R software, with its predictive performance and clinical utility evaluated through receiver operating characteristic (ROC) curves, calibration curves, and decision curve analysis (DCA).

[RESULTS] Multivariate analysis identified medical insurance type (OR = 3.435, 95% CI: 1.230-9.592, P = 0.019), psychological assessment (OR = 0.779, 95% CI: 0.712-0.853, P < 0.001), and perceived social support (OR = 1.131, 95% CI: 1.088-1.177, P < 0.001) as independent predictors of AET adherence. The resulting nomogram achieved AUC values for the training cohort and validation cohort of 0.933 (95% CI: 0.905-0.961) and 0.891 (95% CI: 0.826-0.957), respectively. Calibration curves and DCA demonstrated excellent consistency and clinical applicability.

[CONCLUSIONS] The study identified medical insurance type, psychological assessment, and perceived social support as key factors influencing adherence to AET. The developed nomogram on this basis provides a visual tool for identifying high-risk populations with poor adherence to AET, which helps to carry out personalized interventions for different patients in the future.