Identifying High-Disability Pain in Patients with Temporomandibular Disorders Using Penalized Logistic Regression: A Graded Chronic Pain Scale-Based Study

Purpose: This study aimed to develop a clinically interpretable risk stratification model for identifying patients with the Graded Chronic Pain Scale (GCPS)-defined high-impact pain among patients with painful temporomandibular disorders (TMD) using L1-regularized logistic regression.Methods: This secondary cross-sectional analysis included 1,488 adult patients with painful TMD diagnosed according to Research Diagnostic Criteria for Temporomandibular Disorders Axis I. High-disability pain was defined as GCPS grades III-IV, and lowdisability pain as grades I-II.Candidate predictors included demographic factors, TMD diagnostic category, pain severity (PS) and interference, pain catastrophizing, subjective sleep quality, and psychological distress measures.Variables with multicollinearity or conceptual overlap with GCPS grading were excluded.An L1-regularized logistic regression model with class weighting was fitted, and performance was assessed using the area under the receiver operating characteristic curve (ROC-AUC).Results: Among 1,488 patients, 518 (34.8%) were classified as having high-impact pain.The final model retained PS, pain catastrophizing, poor subjective sleep quality, and somatization as key predictors.PS showed the strongest association with high-impact pain (odds ratio [OR]=2.99,95% confidence interval [CI] 2.54-3.52),followed by pain catastrophizing (OR=1.39,95% CI 1.20-1.61),poor sleep quality (OR=1.16,95% CI 1.01-1.32),and somatization (OR=1.17,95% CI 1.02-1.34).Sex and the TMD diagnostic category were not significant predictors.The model demonstrated acceptable discrimination, with an AUC of 0.777 on the held-out test set.Conclusions: GCPS-defined high-disability pain in painful TMD was characterized mainly by PS, catastrophizing, poor sleep quality, and somatic symptom burden.L1-regularized logistic regression provided a reduced-complexity and clinically interpretable model for biopsychosocial stratification of high-disability pain.