Cueing-assisted gamified augmented-reality home rehabilitation for gait and balance in people with Parkinson disease: feasibility and effectiveness in the clinical pathway.

There was no control group and no randomisation: everyone had six weeks of usual care before starting, and several measures improved during that period, so some of the change credited to the platform may have come from usual therapy, time or repeated testing. The changes on the clinical tests were smaller than the threshold generally taken as clinically important, so the authors themselves say no meaningful effect was shown. Participants were relatively high functioning to start with (baseline Timed Up and Go averaged around 8.5 seconds) and only Hoehn and Yahr stages 1 to 3 were included, with people who could not walk independently for 30 minutes excluded, so the results may not apply to more disabled patients or those with more severe cognitive, visual or hearing problems. Assessments were carried out by 28 different therapists across 15 sites and the paper says they could not always be scheduled at the same time of day, so medication fluctuations may have influenced the results. Whether participants actually did their usual prescribed home exercises during the comparison period was not recorded. Therapists had no previous experience with augmented reality, which the authors suggest may have limited how well the exercises were tailored. About 15% of those who took the platform home dropped out and only 73 of the 100 enrolled had the final assessment.

The paper states that the funder played no role in the design, conduct or reporting of the study, but does not name the funder in the text supplied. The Strolll platform was developed in collaboration with Strolll Limited, the company that makes it, and the authors say their findings were shared with the company for future optimisation of the platform.

Do not conclude that this platform improves gait and balance. There was no control group, participants also improved on several tests during the six weeks of usual care before training began, and the change on the primary outcome was 0.34 seconds on the Timed Up and Go, below the smallest difference considered clinically important.