Specialized stroke services: A meta-analysis comparing three models of care

Background: Using previously published data, the purpose of this study was to identify and discriminate between three different forms of inpatient stroke care based on timing and duration of treatment and to compare the results of clinically important outcomes. Methods: Randomized controlled trials, including a recent review of inpatient stroke unit/rehabilitation care, were identified and grouped into three models of care as follows: ( a) acute stroke unit care ( patients admitted within 36 h of stroke onset and remaining for up to 2 weeks; n = 5), (b) units combining acute and rehabilitative care ( combined; n = 4), and ( c) rehabilitation units where patients were transferred onto the service approximately 2 weeks following stroke (postacute; n = 5). Pooled analyses for the outcomes of mortality, combined death and dependency and length of hospital stay were calculated for each model of care, compared to conventional care. Results: All three models of care were associated with significant reductions in the odds of combined death and dependency; however, acute stroke units were not associated with significant reductions in mortality when this outcome was analyzed separately (OR 0.80; 95% CI: 0.61-1.03). Postacute stroke units were associated with the greatest reduction in the odds of mortality (OR 0.60; 95% CI: 0.44-0.81). Significant reductions in length of hospital stay were associated with combined stroke units only ( weighted mean difference -14 days; 95% CI: -27 to -2). Conclusions: Overall, specialized stroke services were associated with significant reductions in mortality, death and dependency and length of hospital stay although not every model of care was associated with equal benefit. Copyright (c) 2007 S. Karger AG, Basel.