The Effect of Wearable-Based Real-Time Feedback on Running Injuries and Running Performance: A Randomized Controlled Trial

Background: Running technique and running speed are considered important risk factors for running injuries. Real-time feedback on running technique and running speed by wearables may help reduce injury risk. Purpose: To investigate whether real-time feedback on spatiotemporal metrics and relative speed by commercially available pressure-sensitive insoles would reduce running injuries and improve running performance compared with no real-time feedback. Study Design: Randomized controlled trial; Level of evidence, 1. Methods: A total of 220 recreational runners were randomly assigned into the intervention and control groups. Both groups received pressure-sensitive insoles, but only the intervention group received real-time feedback on spatiotemporal metrics and relative speed. The feedback aimed to reduce loading on the joint/segment estimated to exhibit the highest load. Injury rates were compared between the groups using Cox regressions. Secondary outcomes compared included injury severity, the proportion of runners with multiple injuries, changes in self-reported personal best times and motivation (Behavioral Regulation in Exercise Questionnaire–2), and interest in continuing wearable use after study completion. Results: A total of 160 participants (73%) were included in analyses of the primary outcome. Intention-to-treat analysis showed no significant difference in injury rate between the groups (Hazard ratio [HR], 1.11; P = .70). This was expected, as 53 of 160 (33%) participants ended up in the unassigned group because they used incorrect wearable settings, nullifying any interventional effects. As-treated analysis showed a significantly lower injury rate among participants receiving real-time feedback (HR, 0.53; P = .03). Similarly, the first-time injury severity was significantly lower (–0.43; P = .042). Per-protocol analysis showed no significant differences in injury rates, but the direction favored the intervention group (HR, 0.67; P = .30). There were no significant differences in the proportion of patients with multiple injuries (HR, 0.82; P = .40) or changes in running performance (3.07%; P = .26) and motivation. Also, ~60% of the participants who completed the study showed interest in continuing wearable use. Conclusion: Real-time feedback on spatiotemporal metrics and relative speed provided by commercially available instrumented insoles may reduce the rate and severity of injuries in recreational runners. Feedback did not influence running performance and exercise motivation. Registration: NL8472 (Dutch Trial Register).