Physiotherapy and occupational therapy staffing in day services was associated with slower functional decline over two years, Japanese cohort finds

A retrospective longitudinal cohort study used propensity-score matching to compare 24-month trajectories of activities of daily living (ADL) among 1,908 community-dwelling older adults in Japan who continuously used adult day services for at least 24 months, divided between services staffed with physiotherapists and occupational therapists and services without rehabilitation professionals Primary Study. After matching, 954 participants in each group were compared using a validated functional independence measure, assessed at baseline and at 6, 12, 18 and 24 months.

Both groups showed significant functional decline over 24 months, but deterioration occurred earlier in the unstaffed group, and by 24 months the physiotherapy/occupational-therapy-staffed group showed significantly smaller declines in both overall functional independence and transfer ability specifically — differences that exceeded the minimal detectable change threshold, indicating the difference is clinically as well as statistically meaningful.

This is a matched observational cohort, not a randomised trial, conducted within Japan’s long-term care day-service system; the authors themselves call for further research to clarify the specific mechanisms driving the effect and its longer-term outcomes.

The finding is consistent with allied health professional presence, and not only the content of what is delivered, mattering for functional preservation in day-service settings, though an observational design cannot separate the staffing from whatever else differs between the services that employ therapists and those that do not. For services designing or funding day-service provision, it is a reason to treat physiotherapy and occupational therapy staffing as a plausible lever against functional decline, worth testing rather than assuming. Falls and hospitalisation were not measured: the authors raise a reduced risk of both as a possibility, not a finding.