The impact of staffing model in a 6-day rehabilitation physiotherapy service

Background and Purpose: The impact of staffing model providing weekend physiotherapy rehabilitation has not been evaluated. This study aims to determine the impact of staffing a weekend rehabilitation service with physiotherapists currently working in rehabilitation compared to acute hospital physiotherapists, on length of stay (LOS), functional independence and gait and balance, and to determine the impact on neurological, orthopaedic, and reconditioning diagnostic groups. Methods: A prospective cohort study with historical control was completed in a private, metropolitan Australian rehabilitation unit. All participants admitted to the rehabilitation unit over two, 20‐week periods in 2011 and 2012 were included. Weekend physiotherapy was provided by physiotherapists working in rehabilitation in 2012 (intervention) and physiotherapists working in the acute wards in 2011 (control). Outcomes included LOS, Functional Independence Measure (FIM), and gait and balance measures. Results: Overall, there were 504 participants; 234 in 2012 and 270 in 2011. No difference was found in LOS between staffing models (mean difference−1.5 days, 95%CI −4.4 to 1.3). Greater FIM change (mean difference 3.5, 95%CI 0.3 to 6.7) and efficiency (FIM change/LOS: mean difference 0.3, 95%CI 0.1 to 0.5) were found with rehabilitation compared to acute staffing. There was no between‐group difference in gait or balance performance. When diagnostic groups were compared, no difference in LOS was found between staffing models. Participants with an orthopaedic diagnosis had a significantly greater FIM change (mean difference 3.8, 95%CI 0.4 to 7.1), whereas FIM efficiency was improved in neurological (mean difference 0.4, 95%CI 0.1 to 0.7) and orthopaedic populations (mean difference 0.3, 95%CI 0.03 to 0.5) with rehabilitation staffing. Discussion: Staffing a weekend rehabilitation service with physiotherapists currently working in rehabilitation influences functional independence. Different diagnostic groups appear to respond differently.