Developing a Nomogram Prediction Model for Dietary Behavior Adherence in Hemodialysis Patients

Area of Science:

• Nephrology
• Public Health
• Medical Informatics

Background:

• Dietary adherence is crucial for hemodialysis (HD) patients.
• Predictive models for dietary behavior adherence in HD patients are lacking.
• Individualized interventions require accurate risk assessment tools.

Purpose of the Study:

• Develop a nomogram prediction model for dietary behavior adherence in HD patients.
• Provide a simple and accurate tool for individualized risk prediction.
• Support effective intervention strategies for dietary management in HD.

Main Methods:

• Retrospective study including 236 patients (modeling group) and 121 patients (validation group).
• Lasso regression and multivariate logistic analysis identified independent predictors.
• Nomogram construction and validation using ROC curves, H-L test, calibration curves, and DCA.

Main Results:

• Age, years on dialysis, and poor self-care ability were risk factors for non-adherence.
• Higher education, dietary knowledge, social support, and family support were protective factors.
• Internal validation AUC: 0.884; External validation AUC: 0.902; both showed high clinical utility.

Conclusions:

• Key factors influencing dietary adherence in HD patients include age, dialysis duration, education, knowledge, social support, family support, and self-care ability.
• The developed nomogram demonstrates good discrimination and clinical benefit.
• The nomogram aids in early identification of high-risk patients for individualized dietary management.