Abstract 4366703: Development of a Machine Learning-Based Prediction Model for Health-Related Quality of Life in Patients with Coronary Artery Disease

Background: Coronary artery disease (CAD) is the leading cause of mortality globally. As a chronic condition, CAD requires long-term management, and evaluating patients’ health-related quality of life (HRQoL) has become crucial from a patient-centered care perspective. HRQoL is a subjective assessment of health status that reflects an individual’s overall well-being. CAD patients tend to have lower HRQoL compared to the general population due to physical, psychological, and social factors. Objective: This study aimed to develop a machine learning-based model to predict HRQoL in patients with CAD using Wilson and Cleary’s conceptual model. Methods: This study extracted data from the 6th to 8th (2013–2020) Korea National Health and Nutrition Examination Survey (KNHANES). Adult patients diagnosed with angina or myocardial infarction were included. HRQoL was measured using EQ-5D; participants were classified as high-risk (EQ-5D < 0.678, n = 147) or non-risk (n = 1,163). SPSS 29.0 was used for complex sample analyses. Python 3.0 was used for data preprocessing, model development, evaluation, and feature importance analysis. Results: Six machine learning models were tested: logistic regression, decision tree, naive Bayes, random forest, support vector machine (SVM), and extreme gradient boosting (XGBoost). XGBoost showed the best performance (accuracy: 93%, AUC: 0.98). Key predictors included perceived health status, physical activity, discomfort, education, income, occupation, and activity limitation. Conclusions: Machine learning models, particularly XGBoost, demonstrated strong predictive performance for HRQoL in CAD patients. These findings may support personalized care strategies and the development of interventions to enhance quality of life in this population.