Challenges in Quantifying Upper Extremity Dysmetria in the Scale for the Assessment and Rating of Ataxia (SARA)

Background Assessments of upper extremity control in ataxia are susceptible to measurement error and inherent variability. Objectives Quantify trial-to-trial variability in the Scale for the Assessment and Rating of Ataxia (SARA) nose-to-finger (NTF) and finger chase (FC) tasks, and test agreement between scores derived from live assessment and retrospective video review. Methods 17 patients with cerebellar ataxia were video-recorded performing tasks behind a transparent, calibrated 1-cm grid. NTF tremor amplitude and FC undershoot/overshoot distance derived from live scores were compared with measurements from video review. Results NTF tremor amplitude and FC under/overshoot varied between consecutive trials. Patients received 2.4 ± 0.7 (NTF) and 2.6 ± 0.8 (FC) different SARA scores for limb excursions within one session. Live and video score match was NTF: 64.7% left, 70.6% right; FC: 64.7% left, 41.2% right. Live scoring often overestimated impairment. Conclusions Trial-to-trial variability and live/video discordance limit the precision of SARA upper-limb tasks.