Background The clinical assessment of motor symptoms in Huntington's disease is usually performed with the Unified Huntington's Disease Rating Scale-Total Motor Score (UHDRS-TMS). A high interrater reliability is desirable to monitor symptom progression. Therefore, a teaching video and a system for annual online certification has been developed and implemented. Objectives The aim of this study is to investigate the interrater reliability of the UHDRS-TMS and of its subitems, and to examine the performance of raters in consecutive years. Methods Data from the online UHDRS-TMS certification were used. The interrater reliability was assessed for all first-time participants (n = 944) between 2009 and 2016. Intraclass correlation coefficients (ICC) were calculated for each year separately and the mean was taken as the total ICC. Results The UHDRS-TMS (ICC = 0.847), tandem walking (0.824), pronate/supinate hands left (0.713), and retropulsion pull test (0.706) showed good interrater reliability. Poor interrater reliability was found for maximal dystonia of the left and right upper extremity (0.187 and 0.322, respectively), maximal dystonia of the left and right lower extremity (0.200 and 0.256, respectively), and maximal dystonia of the trunk (0.389), tongue protrusion (0.266), and rigidity arms left (0.390). Raters performed significantly worse on follow-up certification compared to their first certification. Conclusions Our results suggest that the rating of dystonia (absent, slight, mild, moderate, or marked) is subjective and difficult to interpret, especially on video. Therefore, changing the dystonia items of the UHDRS-TMS should be explored. We also recommend that raters should watch the UHDRS-TMS teaching video before each certification.
