Psychometric properties of a Norwegian version of Multiple Sclerosis Impact Scale (MSIS-29)

Objectives - To examine psychometric properties of Multiple Sclerosis Impact Scale (MSIS-29) for use in Norwegian patients with multiple sclerosis. Methods - Translation was performed according to international guidelines. The questionnaire was answered by 64 patients prior to and at a screening session, and re-answered by 59 patients before and after 4 weeks of physiotherapy. Results - Internal consistency (alpha) was 0.92 for the physical and 0.85 for the psychological subscales. Reliability by intraclass correlation coefficients were 0.86 for the physical and 0.81 for the psychological subscales, smallest detectable change being 18.4 and 21.1, respectively. The physical but not the psychological subscale demonstrated mostly satisfactory associations with other physical measures. Responsiveness by area under the receiver operating characteristics curve was 0.83 and 0.76, respectively. As hypothesized, effect size was larger for the physical (1.01) than for the psychological (0.76) subscale after treatment. Conclusions - MSIS-29, Norwegian Version demonstrated satisfactory psychometric properties.