Factor Analysis and Clustering of the Movement Disorder Society-Non-Motor Rating Scale

Background The primary validation of the Movement Disorder Society Non-Motor Rating Scale was recently published, but 2 important structural analyses were not included. The objective of this study was to examine the structural characteristics of the Movement Disorder Society Non-Motor Rating Scale by factor and cluster analyses. Methods Data came from the validation study, an international multicenter cross-sectional study of 402 Parkinson's disease patients. Demographic and clinical data, the Movement Disorder Society Non-Motor Rating Scale, and Hoehn and Yahr staging were used. Exploratory and confirmatory factor analysis and nonhierarchical cluster analysis were performed. Results The exploratory factor analysis showed that all 13 domains of the Movement Disorder Society Non-Motor Rating Scale, except 1, and the Non-Motor Fluctuations subscale performed as unidimensional (variance explained: 0.36, sleep and wakefulness; -0.82, orthostatic hypotension). The confirmatory factor analysis could be carried out in 9 domains and showed that 6 of them and the Non-Motor Fluctuations subscale adjusted to the model satisfactorily according to the root mean square error of approximation. Furthermore, all domains had comparative fit index values >0.95, except depression and pain (both, 0.94) and sleep and wakefulness (0.90). The Non-Motor Fluctuations subscale showed satisfactory root mean square error of approximation (0.07), but a low comparative fit index value (0.91). A 5-cluster solution, correctly classifying 96% of the cases, was found. Conclusions Overall, most subscales of the Movement Disorder Society Non-Motor Rating Scale are unidimensional, and although each subscale is distinct in terms of content covered, factors and clusters that are of clinical relevance are discernible and contribute to our understanding of possible nonmotor subtypes in Parkinson's disease. (c) 2020 International Parkinson and Movement Disorder Society