Consistency in Administration and Response for the Backward Push and Release Test: A Clinical Assessment of Postural Responses

Background and Purpose: The backward push and release test (PRT) is a standardized clinical test of postural responses elicited by perturbations. Our goal was to determine reliability of administration and response. This will inform clinical administration and determine whether to develop an instrumented version. Methods: One examiner administered 10 backward PRT trials to adults with Parkinson disease (12), multiple sclerosis (14) and controls (12). We used three-dimensional motion analysis, force plates and instrumented gloves to measure administration and response. Administration variables were angle of posterior trunk lean and the distance of the centre of mass (CoM) behind the ankle. Postural response variables were latency of postural response from release to step initiation and first compensatory step length. Reliability was measured using the range of variables across trials, comparison of first and later trials, intraclass correlations (ICCs) to measure consistency and correlations between administration and response. Results: There was inherent variability in administration, which affected postural response characteristics. Larger trunk angle and greater CoM-ankle distance were correlated with shorter postural response latencies and larger step lengths. Participant height also had an effect; taller participants had larger trunk angles prior to release resulting in longer latencies and larger step lengths. Using ICCs, consistency of trunk angle was likely acceptable and CoM-ankle distance was high. Consistency of latency was low, while step length was likely acceptable. Discussion: Despite variability in administration and inconsistency in response, different postural response characteristics were detected between patients with different disease states. Based on these results, we will create algorithms to instrument the PRT using inertial movement sensors to collect more sensitive measures of postural responses than observational clinical rating scales. Feedback for appropriate lean angle and calibration for participant height will improve consistency and usefulness of the instrumented PRT.