Effects of an Exercise-Assisting Mobile App (Osteoarthritis-Rehabilitation Assistant [O-RA]) on Rehabilitation Outcomes in Older Adults: Randomized Controlled Parallel Clinical Trial.

Abstract

Background: Mobile apps and biofeedback using motion analysis have both been used separately to increase compliance with exercise programs. We developed a mobile app, Osteoarthritis-Rehabilitation Assistant (O-RA), that uses motion analysis technology in the mobile app to assist older adults with performing a knee exercise program.

Objective: This study aimed to evaluate the effects of the O-RA app on the compliance and correctness of the exercise program by older adults.

Methods: We conducted an assessor-blind, parallel-design, randomized controlled trial with 40 older adults (aged 60 years or older) who had no symptoms and no diagnosis of knee osteoarthritis. Participants were divided into 2 groups: O-RA app (intervention) group and standard treatment (control) group. Both groups were taught 4 types of exercise programs by a physical therapist for 15 minutes and were instructed to do exercises at home every day for 1 week. The number of exercises, the percentage between observed and prescribed exercises, the correctness of exercises, and overall pain during the program were assessed in both groups.

Results: The control group had significantly higher compliance with the exercise program than the intervention group (t38=3.5044, P=.001). There was no statistically significant difference in the correctness of the exercise program between the intervention and control groups. The difficulty of use and satisfaction were 47 and 59, respectively, out of the full score of 100. The main problems were the instability and the difficulty using the app.

Conclusions: In older adults without knee osteoarthritis symptoms or diagnosis, the O-RA app was not a facilitator but a barrier to the lower extremity exercise program. An updated version, aiming to increase the stability and make it more user-friendly, should be developed; however, more comprehensive data, including qualitative user feedback and standardized usability metrics, will be needed to effectively guide its design.