KMAN Publication Inc. Quality of Life and Health Sciences 2 Serial Number 4 2026 04 01 Elastic Net Regression and Gradient Boosting for Predicting Health-Related Quality of Life Among Patients With Chronic Obstructive Pulmonary Disease Elastic Net Regression and Gradient Boosting for Predicting Health-Related Quality of Life Among Patients With Chronic Obstructive Pulmonary Disease 1 16 10.61838/nmxhg438 EN Journal Article 2026 01 06 <table> <tbody> <tr> <td> <p><strong>Objective:</strong> This study aimed to develop and compare Elastic Net Regression and Gradient Boosting Regression models for predicting health-related quality of life among patients with chronic obstructive pulmonary disease in Tehran.</p> <p><strong>Methods and Materials: </strong>This cross-sectional predictive modeling study was conducted on 286 patients with chronic obstructive pulmonary disease who were receiving outpatient pulmonary care in Tehran, Iran. Data were collected using a demographic and clinical information form, spirometric indices, the St. George’s Respiratory Questionnaire, the COPD Assessment Test, the modified Medical Research Council dyspnea scale, the Hospital Anxiety and Depression Scale, and physical activity assessment. The total score of the St. George’s Respiratory Questionnaire was considered the primary outcome variable. After data screening, preprocessing, standardization of continuous predictors, and coding of categorical variables, two predictive models were developed: Elastic Net Regression and Gradient Boosting Regression. The dataset was divided into training and testing subsets, and model performance was evaluated using root mean squared error, mean absolute error, and coefficient of determination.</p> <p><strong>Findings: </strong>Bivariate analyses showed that poorer health-related quality of life was significantly associated with higher COPD Assessment Test scores, greater dyspnea severity, lower FEV1 percentage predicted, lower FEV1/FVC ratio, more frequent exacerbations, more hospitalizations, higher anxiety and depression scores, lower physical activity, longer disease duration, higher body mass index, and older age. In the testing dataset, Elastic Net Regression explained 61.9% of the variance in health-related quality of life, with RMSE = 10.12 and MAE = 7.81. Gradient Boosting Regression demonstrated superior predictive performance, explaining 70.4% of the variance, with RMSE = 8.91 and MAE = 6.83. The most important predictors across models were symptom burden, dyspnea, lung function, depression, exacerbation frequency, anxiety, and physical activity.</p> <p><strong>Conclusion: </strong>Gradient Boosting Regression provided more accurate prediction of health-related quality of life than Elastic Net Regression, although both models identified clinically meaningful predictors. These findings support the use of multidimensional predictive modeling to identify chronic obstructive pulmonary disease patients at greater risk of poor quality of life.</p> </td> </tr> </tbody> </table> https://journals.kmanpub.com/index.php/qlhs/article/download/5785/10974