Mechanism evaluation of a digitally enabled rehabilitation intervention for people in aged care and neurological rehabilitation: mediation analysis of the AMOUNT trial.

Abstract

To investigate potential mechanisms of a digital rehabilitation intervention associated with improved mobility among adults undertaking rehabilitation. Causal mediation analysis of the AMOUNT trial (ACTRN12614000936628). Participants were randomised to digitally-enabled rehabilitation (virtual reality video games, activity monitors, and handheld computer devices prescribed by a physiotherapist) and usual care or usual care alone. Outcomes were mobility (Short Physical Performance Battery; continuous version; range 0-3), physical activity (average steps per day), and quality of life (EQ-5D-5L; utility score; range 0-1) measured at 6-months post-randomisation. Hypothesised mediators included mobility, physical activity, cognition, balance confidence, pain, activity and participation, and computer self-efficacy, assessed at 3-weeks post-randomisation. 216 participants with complete data were included. Three-week sit-to-stand ability mediated 6-month mobility (indirect effect 0.09 points, 95%CI 0.03-0.16), explaining 48% of the intervention's effect. Dynamic single-leg-stance balance mediated physical activity (indirect effect 345 steps per day, 95%CI 63-678) and composite mobility mediated quality of life (indirect effect 0.03 points, 95%CI 0.00-0.05). Digital rehabilitation outcomes appear to be mediated through sit-to-stand ability, dynamic single-leg-stance balance, and overall mobility. While future research is required to better understand these mediators, our findings recommend sit-to-stand training as a core element of digital rehabilitation interventions targeting mobility.