KMAN Publication Inc. Quality of Life and Health Sciences 2 Serial Number 4 2026 04 01 Explainable XGBoost Modeling of Quality of Life Among Patients with Fibromyalgia Using Pain, Sleep, and Psychological Flexibility Variables Explainable XGBoost Modeling of Quality of Life Among Patients with Fibromyalgia Using Pain, Sleep, and Psychological Flexibility Variables 1 14 10.61838/9khfxt92 EN Journal Article 2026 01 10 <table> <tbody> <tr> <td> <p><strong>Objective:</strong> This study aimed to develop and interpret an explainable XGBoost model for predicting quality of life among patients with fibromyalgia based on pain intensity, sleep quality, psychological flexibility, and relevant demographic and clinical variables.</p> <p><strong>Methods and Materials:</strong> This cross-sectional predictive modeling study was conducted among 318 patients with fibromyalgia receiving services from rheumatology and pain management clinics in Tehran, Iran. Data were collected using a demographic and clinical information form, the Short Form Health Survey for quality of life, the Visual Analog Scale for pain intensity, the Pittsburgh Sleep Quality Index for sleep quality, and the Acceptance and Action Questionnaire-II for psychological flexibility. After data screening and preprocessing, quality of life was considered the continuous outcome variable. Predictive modeling was performed using XGBoost regression, and model performance was compared with linear regression, support vector regression, and random forest regression. Model accuracy was evaluated using mean absolute error, root mean square error, coefficient of determination, and cross-validation. SHAP analysis was used to explain the final XGBoost model.</p> <p><strong>Findings:</strong> Correlation analysis showed that quality of life was significantly and negatively associated with pain intensity (r = -0.58, p &lt; 0.01) and poor sleep quality (r = -0.52, p &lt; 0.01), while it was significantly and positively associated with psychological flexibility (r = 0.49, p &lt; 0.01). The XGBoost model demonstrated the strongest predictive performance, with a test MAE of 7.84, test RMSE of 10.18, test R² of 0.66, and cross-validated R² of 0.63. SHAP analysis identified pain intensity as the most influential predictor, followed by poor sleep quality and psychological flexibility.</p> <p><strong>Conclusion:</strong> The findings indicated that quality of life among patients with fibromyalgia can be predicted with acceptable accuracy using an explainable XGBoost model, with pain intensity, sleep quality, and psychological flexibility emerging as the most important predictors.</p> </td> </tr> </tbody> </table> <p> </p> https://www.journals.kmanpub.com/index.php/qlhs/article/download/5759/10907