Clinician versus machine: Reliability and responsiveness of motor endpoints in Parkinson's disease

Abstract Background Enhancing the reliability and responsiveness of motor assessments required to demonstrate therapeutic efficacy is a priority for Parkinson's disease (PD) clinical trials. The objective of this study is to determine the reliability and responsiveness of a portable kinematic system for quantifying PD motor deficits as compared to clinical ratings. Methods Eighteen PD patients with subthalamic nucleus deep-brain stimulation (DBS) performed three tasks for evaluating resting tremor, postural tremor, and finger-tapping speed, amplitude, and rhythm while wearing a wireless motion-sensor unit (Kinesia) on the more-affected index finger. These tasks were repeated three times with DBS turned off and at each of 10 different stimulation amplitudes chosen to yield small changes in treatment response. Each task performance was video-recorded for subsequent clinician rating in blinded, randomized order. Test–retest reliability was calculated as intraclass correlation (ICC) and sensitivity was calculated as minimal detectable change (MDC) for each DBS amplitude. Results ICCs for Kinesia were significantly higher than those for clinician ratings of finger-tapping speed ( p