Validation and cross-cultural adaptation of the COMPASS-31 in Croatian and Serbian patients with multiple sclerosis

AIM To validate and cross-culturally adapt Croatian and Serbian versions of composite autonomic symptom score-31 (COMPASS-31) for the detection of dysautonomia in patients with multiple sclerosis (MS). METHODS A total of 179 patients, 67 with clinically isolated syndrome (CIS) and 112 with MS, completed the COMPASS-31 at two MS centers in Zagreb and Belgrade between April 1 and October 31, 2016. Demographic and clinical data including age, gender, MS phenotypes, and the Expanded Disability Status Scale (EDSS) score were collected. RESULTS The Cronbach's alpha coefficient of COMPASS-31 total score was 0.844 for the Croatian MS sample and 0.779 for the Serbian MS sample. A joint analysis yielded Cronbach's alpha coefficients ranging from 0.394 to 0.796, with values in four domains higher than 0.700. In Croatian and Serbian samples and the total study sample, the Cronbach's alpha coefficient of COMPASS-31 was 0.785. Reproducibility measured by intra-class correlation coefficient (ICC) was acceptable (ICC=0.795). With regard to the clinical validity, significant correlation was found between EDSS and the COMPASS-31 total score (P<0.001). Furthermore, significant differences between MS phenotypes were detected for bladder and gastrointestinal domains and for the COMPASS-31 total score (PP<0.001, P=0.005, and P=0.027, respectively). Finally, significant differences between MS phenotypes in patients with score >0, which implies the existence of at least one of the symptoms investigated in each domain, were detected for secretomotor and bladder domains (P=0.015 and PP<0.001, respectively). CONCLUSION COMPASS-31 represents a valid and acceptable self-assessment instrument for the detection of dysautonomia in MS patients.