Temporal stability of the Unified Dyskinesia Rating Scale

Abstract

The objective of this study was to establish temporal stability characteristics for objective components of the Unified Dyskinesia Rating Scale (UDysRS). The UDysRS has strong internal consistency and a reliable factor structure, but the important issue of temporal stability has not been established. Using intraclass correlation coefficient (ICC) analyses, we examined UDysRS temporal stability for the objective scale components (Part III and IV) over an 8-hour observation period. We assessed ICCs for the single centralized rater, the on-site raters, and the agreement between the single centralized rater and the on-site raters. Kappa statistic assessed agreement between the single centralized and on-site raters for clinical state (ON vs OFF). For both the single centralized rater and the on-site raters, there was high temporal stability of the UDysRS Part III, Part IV, and Total Objective UDysRS in both ON and OFF states, with ICCs ranging from 0.822 (P < .0005) to 0.513 (P < .013). The agreement between the 2 rating techniques (centralized vs on-site) was significant for ON and OFF ratings of Part III, Part IV, and Total Objective UDysRS, ranging from 0.821 (P < .0005) to 0.703 (P < .0005). The UDysRS is highly stable for ON and OFF. Our data suggest that a single UDysRS evaluation for ON and for OFF states is highly representative of that state regardless of time. Likewise, if appropriate protocols need to assess dyskinesia in a field or community setting, the UDysRS can be filmed without an on-site rater and rated centrally with retained validity.