Abstract
OBJECTIVE: Adult spinal deformity (ASD) have lower health-related quality of life (HRQoL) compared to the general population. Applying Rasch measurement theory (RMT), this study tested the revised Scoliosis Research Society-22 (SRS-22r) HRQoL instrument among symptomatic adult patients with degenerative spinal disorders and varying degrees of ASD. METHODS: SRS-22r data from 637 outpatient spine clinic patients with degenerative spine conditions were investigated for unidimensionality, item/scale fit, differential item functioning (DIF), scale coverage/targeting, and person separation index (PSI) using RMT. RESULTS: Unidimensionality of the SRS-22r was not supported for either the total score or for 3 of its 5 domains. Item fit was acceptable for 11/22 items. The individual domains showed good coverage despite the degree of structural disorders. Ordered thresholds were achieved by merging response categories in some of the items. DIF towards age or sex was found in 11/22 items and in some domain items. The PSI exceeded 0.7 for the SRS-22r total score. CONCLUSION: The individual domain scores of the SRS-22r perform better than the total score providing good coverage and targeting among patients with ASD. Refinements of items and domains may improve the structural validity of the instrument to meet the criteria for measuring ASD patients, even when multidimensionality persists.