Responsiveness and minimal important change of Turkish pain-related psychological measures in chronic non-specific low back pain

Abstract

Abstract Background The longitudinal interpretability of Turkish pain-related psychological measures in chronic non-specific low back pain (NSLBP) remains uncertain. This study evaluated the responsiveness and minimal important change (MIC) of the Turkish Fear-Avoidance Beliefs Questionnaire (FABQ), Tampa Scale of Kinesiophobia (TSK) and Pain Catastrophizing Scale (PCS). Methods In this prospective, single-arm measurement-property study, 110 adults with chronic NSLBP completed baseline and 4-week assessments during a standardized rehabilitation program. Distribution-based responsiveness was assessed using effect size and standardized response mean. Associations with Global Rating of Change (GRC)-defined improvement were examined using correlations and receiver operating characteristic analysis. MIC uncertainty was estimated by stratified bootstrap resampling with 20,000 iterations. Results Fifty-two participants were classified as improved and 58 as not improved. FABQ change scores were strongly correlated with GRC and showed excellent discrimination (area under the curve (AUC) = 0.943–0.977). GRC-defined MIC estimates (95% bootstrap confidence interval (CI)) were 5 (5–5) for FABQ-Physical Activity and 7 (6–8) for FABQ-Work. PCS correlations and discrimination were more modest (AUC = 0.647–0.700); MIC estimates ranged from 2 to 6 points, with the widest interval for PCS-Total [6 (3–11)]. TSK change was unrelated to GRC (rho = 0.014; AUC = 0.508), and no MIC was estimated. Conclusions In this rehabilitation context, FABQ change scores corresponded most strongly to GRC-defined improvement. PCS MIC estimates were less precise, whereas TSK change did not distinguish global improvement. These estimates are context-specific and require external validation.