Feasibility and Sensitivity of Wearable Sensors for Daily Activity Monitoring in Spinal Cord Injury Trials.

Abstract

BackgroundThe aim of clinical trials for spinal cord injury (SCI) is to improve everyday-life activity outcomes, which requires reliable methods for monitoring patient activity. This study evaluates sensor-derived activity metrics in comparison to established clinical assessment methods.MethodsWearable inertial sensors collected data from 69 individuals with acute, traumatic cervical SCI participating in the Nogo-A Inhibition in Spinal Cord Injury trial (NCT03935321), a phase 2b, multicenter, randomized, placebo-controlled trial. During inpatient rehabilitation, participants wore up to 5 inertial sensors for up to 3 consecutive days each week. An estimation of average daily energy expenditure (EE) was used as an indicator of physical activity and compared to the recovery of Upper Extremity Motor Scores (UEMS) and Spinal Cord Independence Measures (SCIM).ResultsParticipants in the verum (n = 41; 59.4%) and placebo (n = 28; 40.6%) groups showed similar initial activity levels, however, the verum group exhibited a significantly greater weekly increase in average daily EE (ΔEE = 11.6 kcal/day/week, 95% CI [1.5, 21.8], P = .025). In contrast, no significant group differences were observed in changes in UEMS (ΔUEMS = 0.1/week, 95% CI [-0.2, 0.3], P = .603) or SCIM (ΔSCIM = 0.2, per week 95% CI [-0.7, 1.1], P = .644).ConclusionContinuous sensor-based activity monitoring offers objective and sensitive insights into changes in physical capabilities, effectively complementing periodic clinical assessments. Thus, sensor-derived outcome measures offer potential for improving the evaluation of clinical studies in individuals with SCI.Clinical Trail Registration:https://clinicaltrials.gov; NCT03935321.