Essential requirements of quantitative measures for the evaluation of upper-extremity movement quality in post-stroke patients

Abstract

BackgroundAlthough kinematic assessment has potentials for better evaluation of the movement quality, they are not widely used in clinics. Besides the equipment cost, the contradictory results of their clinimetric analysis are presumed as a major reason. This raises concerns about whether the current kinematic indices are intrinsically sound and truly reflect the quality of the subjects’ movement. MethodsA list of essential requirements for the kinematic indices (KIs) was proposed. Monotonic response to the change in the movement quality, independence from the task geometry and the subjects’ upper limb dimensions and speed, and robustness against the measurement error was investigated. The compliance of the KIs was then evaluated. For this purpose, end-point speed profile (ESP) and end-point path geometry (EPG) were used in a simulation framework to generate hand movements with altered quality. ResultsAlmost all KIs showed monotonic behavior only in a limited range of movement quality. Many of the popular KIs (particularly, the geometry-based ones) were shown to be highly dependent on the task geometry and the subjects’ upper limb dimensions and speed. Velocity-based KIs generally showed more sensitivity to measurement error. Among the investigated KIs, only velocity deviation (VDEV) acceptably satisfied all requirements. ConclusionsFailure to meet the essential requirements was seen to be common among the popular KIs which may explain the mixed results of the clinimetric analysis in the pertinent literature. It is necessary to modify their formulations and/or seek alternative indices.