Accepted for/Published in: JMIR mHealth and uHealth
Date Submitted: May 29, 2023
Date Accepted: Aug 26, 2024
An evaluation of the effect of app-based exercise prescription using RL on satisfaction and exercise intensity: a randomised crossover trial.
ABSTRACT
Background:
Mobile technologies (e.g., smartphone apps and wearable activity trackers) are a cost-effective, scalable way of remotely delivering exercise programs to users. This double-blind randomised crossover study sought to evaluate the effect of personalising app-based exercise sessions via a reinforcement learning (RL) model on user satisfaction and exercise intensity.
Objective:
The primary aim was to investigate the impact of the i80 BPM app, implementing machine learning for exercise prescription, on user satisfaction and exercise intensity among a general population. A secondary objective was to assess the effectiveness of machine learning-generated exercise programmes for remote prescription of exercise to members of the public.
Methods:
Participants were randomised to complete three exercise sessions per week for 12 weeks using the i80 BPM mobile application, crossing over weekly between intervention and control conditions. The 'intervention' condition involved individualising exercise sessions using RL, based on user preferences such as exercise difficulty, selection, and intensity, whereas under the 'control' condition, exercise sessions were not individualised. Exercise intensity (measured by the 10-item Borg scale) and user satisfaction (measured by the Physical Activity Enjoyment Scale, PACES-8) were recorded post-workout.
Results:
All 69 participants (27 males, 42 females; mean age 42 years) completed 559 exercise sessions over 12 weeks. The General Estimating Equations analysis showed that participants were more likely to exercise at a higher intensity (mean intensity: intervention, 5.82, 95% CI 5.59-6.05; control, 5.19, 95% CI 4.97-5.41) and report higher satisfaction (mean satisfaction: RL, 4; baseline, 3.73) in the RL model condition. User satisfaction remained consistent in the RL condition while decreasing in the baseline condition (p = 0.015).
Conclusions:
These findings provide preliminary evidence that machine learning-generated exercise programmes can effectively increase exercise intensity and user satisfaction in a remote setting. Despite the promising results, further research is needed to determine the long-term impact and sustainability of this approach. Clinical Trial: N/A
Citation
Request queued. Please wait while the file is being generated. It may take some time.
Copyright
© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.