The Dyskinesia Impairment Scale, Second Edition: Development, construct validity, and reliability.

Abstract

To create a shortened, more user-friendly Second Edition of the Dyskinesia Impairment Scale (DIS-II) to assess dystonia and choreoathetosis, and evaluate its construct validity and reliability. Scale development included an online expert meeting (n=21) and iterative discussions within the research group (n=6). A Rasch measurement model analysis on DIS scores from individuals with dyskinetic cerebral palsy or inherited/idiopathic dystonia (n=123, 74 males, mean age14 years, SD5 years) was performed to evaluate the construct validity and reliability of the DIS-II. The DIS-II evaluates dystonia and choreoathetosis in action and rest in 11 body regions, with action items scored from 0 to 3 and rest items 0 to 2. The number of videos to record are reduced from 26 to 14 and the items to score are reduced from 144 to 88. Rating scale functioning, goodness-of-fit evaluation, principal component analysis, and targeting met the predefined quality criteria of the study and construct validity was therefore considered good. Furthermore, person reliability indicated that the DIS-II can separate individuals into eight distinct ability levels. The DIS-II provides valid and reliable measures for dystonia and choreoathetosis, and reduces the administration and scoring time compared with the DIS. The DIS-II logit scores (interval level data) enhance comparison over time and between individuals in clinical practice and research. Compared with the Dyskinesia Impairment Scale (DIS), the shortened edition (DIS-II) requires half of the number of videos to be scored. The DIS-II has a simplified rating scale, requiring scoring of 88 instead of 144 items. The DIS-II has shown excellent reliability and good construct validity. The interval properties of the DIS-II are superior to the ordinal level outcome measures of the DIS.