Cross-diagnostic validity of the SF-36 physical functioning scale in patients with stroke, multiple sclerosis and amyotrophic lateral sclerosis: a study using Rasch analysis.

OBJECTIVE The aim of this study was to investigate unidimensionality and differential item functioning of the SF-36 physical functioning scale (PF10) in patients with various neurological disorders. PATIENTS PATIENTS post-stroke (n = 198), with multiple sclerosis (n = 151) and amyotrophic lateral sclerosis (n = 193) participated. METHODS Unidimensionality of the PF10 within the patient groups was investigated by performing a separate Rasch analysis for each group. Differential item functioning was investigated in a pooled Rasch analysis of the 3 groups. RESULTS Within each group, all items fitted the Rasch model, except the "bathing/dressing" item in the amyotrophic lateral sclerosis group. The pooled analysis showed inadequate fit to the Rasch model for one item ("walking several hundred metres"). Of the other 9 fitting items, 5 showed differential item functioning for stroke vs multiple sclerosis and amyotrophic lateral sclerosis, while no differential item functioning was found between multiple sclerosis and amyotrophic lateral sclerosis. CONCLUSION All items of the PF10, except one for the amyotrophic lateral sclerosis group, form a unidimensional scale, supporting the use of a sum score as a measure of physical functioning within these diagnostic groups. When comparing the data of patients after stroke, with that of patients with multiple sclerosis and/or amyotrophic lateral sclerosis patients, adjustments for differential item functioning are required.