Cross-cultural adaptation and preliminary validation of the Japanese version of the Pain Assessment in Advanced Dementia (PAINAD-J): reliability and convergent validity

Objective To translate and culturally adapt the Pain Assessment in Advanced Dementia (PAINAD) into Japanese (PAINAD-J) and to evaluate its preliminary psychometric properties. Methods This study included 30 residents with dementia (Mini-Mental State Examination [MMSE] ≤20) in a long-term care facility. The PAINAD-J was translated and culturally adapted using forward–backward translation and expert committee review. Participants were categorized into pain and control groups (n = 15 each) based on clinical observations during physical therapy. Video recordings obtained before, during, and after therapy were independently evaluated by thirteen raters using the PAINAD-J and the Faces Pain Scale (FPS). Internal consistency (Cronbach’s α), intra- and inter-rater reliability (intraclass correlation coefficients [ICC]), and convergent validity (correlation with FPS) were assessed. Results The PAINAD-J showed a high Cronbach’s α (0.90), providing preliminary evidence of internal consistency. Intra- and inter-rater reliability were good (ICC(1,1) = 0.78, 95% CI: 0.72–0.83; ICC(2,1) = 0.78, 95% CI: 0.72–0.83). A strong positive correlation was observed between PAINAD-J and FPS scores (r = 0.89), supporting convergent validity. Conclusions The PAINAD-J showed preliminary evidence of internal consistency, good intra- and inter-rater reliability, and strong convergent validity. These findings suggest that it may serve as a reliable observational tool for pain assessment in people with dementia in Japan. However, structural validity, measurement error, and responsiveness were not established. Further multicenter studies are required to confirm its full psychometric properties.