Health Service Management Study for Stroke: A Randomized Controlled Trial to Evaluate Two Models of Stroke Care

Abstract

Background The most effective and efficient model for providing organized stroke care remains uncertain. This study aimed to compare the effect of two models in a randomized controlled trial. Methods Patients with acute stroke were randomized on day one of admission to combined, co-located acute/rehabilitation stroke care or traditionally separated acute/rehabilitation stroke care. Outcomes measured at baseline and 90 days post-discharge included functional independence measure, length of hospital stay, and functional independence measure efficiency (change in functional independence measure score ÷ total length of hospital stay). Results Among 41 patients randomized, 20 were allocated co-located acute/rehabilitation stroke care and 21 traditionally separated acute/rehabilitation stroke care. Baseline measurements showed no significant difference. There was no significant difference in functional independence measure scores between the two groups at discharge and again at 90 days postdischarge (co-located acute/rehabilitation stroke care: 103·6 ± 22·2 vs. traditionally separated acute/rehabilitation stroke care: 99·5 ± 27·7; P = 0·77 at discharge; co-located acute/rehabilitation stroke care: 109·5 ± 21·7 vs. traditionally separated acute/rehabilitation stroke care: 104·4 ± 27·9; P= 0·8875 at 90 days post-discharge). Total length of hospital stay was 5·28 days less in co-located acute/rehabilitation stroke care compared with traditionally separated acute/rehabilitation stroke care (24·15 ± 3·18 vs. 29·42 ± 4·5, P = 0·35). There was significant improvement in functional independence measure efficiency score among participants assigned to co-located acute/rehabilitation stroke care compared with traditionally separated acute/rehabilitation stroke care (co-located acute/rehabilitation stroke care: median 1·60, interquartile range: 0·87–2·81; traditionally separated acute/rehabilitation stroke care: median 0·82, interquartile range: 0·27–1·57, P = 0·0393). Linear regression analysis revealed a high inverse correlation ( R2 = 0·89) between functional independence measure efficiency and time spent in the acute stroke unit. Conclusion This proof-of-concept study has shown that co-located acute/rehabilitation stroke care was just as effective as traditionally separated acute/rehabilitation stroke care as reflected in functional independence measure scores, but significantly more efficient as shown in greater functional independence measure efficiency. Co-located acute/rehabilitation stroke care has potential for significantly improved hospital bed utilization with no patient disadvantage.