Heterogeneity in benefit finding among breast cancer patients: a latent profile analysis and influencing factors

This study investigates factors influencing benefit finding among breast cancer patients based on social cognitive theory and develops a nomogram to predict the probability of low benefit finding in breast cancer patients.

A study of 666 breast cancer patients in northern Anhui Province (January to December 2024) employed latent profile analysis to identify distinct benefit finding patterns. Potential predictors were identified through univariate analysis, least absolute shrinkage and selection operator regression, and multivariate analysis. Five machine learning algorithms were applied to predict low benefit finding, with performance evaluated via calibration and discriminative power metrics and internally validated using bootstrap resampling.

A two-classification model best fits the data, identifying the low benefit finding category (35%) and the high benefit finding category (65%). XGBoost outperformed other models and was selected as the final model. The model achieved an AUC of 0.945 on the validation set. SHAP analysis quantified each variable's contribution to predictions, revealing age, medication adherence, anxiety, social support, and depression as key determinants of benefit finding.

This study applied social cognitive theory to examine factors affecting benefit finding in breast cancer patients, focusing on environmental, individual, and behavioral domains. Results showed strong performance by the XGBoost classifier. The developed nomogram aids healthcare providers in swiftly identifying patients with low benefit finding, enabling personalized interventions to mitigate adverse psychological effects and improve long-term outcomes.