The effectiveness of community-based rehabilitation for providing services to people with stroke with functional limitations and participation restriction : a systematic review and implications

Abstract

We conducted a systematic review and meta-analysis of randomised and quasi-randomised trials to determine the effectiveness of community-based rehabilitation versus hospital/institution based rehabilitation in providing rehabilitative services to people with stroke with functional limitations and participation restriction. Data sources: Using a highly sensitive search strategy, duplicate searches were conducted for the following databases from January 1976 to May 2010: MEDLINE via PubMed, African Wide Information via EBSCO, Academic Search Premier via EBSCO, Cochrane CENTRAL, CINAHL, PsycInfo, PEDro. Review methods: Abstracts were scanned in duplicate for all randomised and quasi-randomised trials comparing the effectiveness of community-based rehabilitation with hospital/institution based rehabilitation in providing rehabiliative service to people with stroke with functional limitations and participation restriction. For this review, the primary outcome was functional independence while secondary outcomes included quality of life, physical, psychological and social functioning and, community participation of people with stroke and their caregivers. Results: Twelve randomised controlled trials with 2707 people with stroke were included. Percentage of males included in studies ranged from 42% to 75% and the participants ranged in age from a mean or median of 52 years to 78 years for those receiving the intervention and 55 years to 80 years for control participants. The treatment duration of community-based rehabilitation programmes ranged from three weeks to six months. Overall the meta-analysis found no evidence for the effectiveness of community-based rehabilitation as compared with hospital/institution based rehabilitation with respect to functional outcome (Standardised Mean Difference (SMD) 0.09; 95% Confidence Interval (CI) -0.08 to 0.26) or quality of life (Mean Difference (MD) 1.32; 95% CI -4.30 to 6.93) or carer strain (MD 0.76; 95% CI -0.19 to 1.77). Subgroup analyses at three months showed a significant effect for community-based rehabilitation over hospital/institution based rehabilitation on quality of life (MD 5.00; 95% CI 0.82 to 9.18); however, this effect was not maintained at six months. Cost-effectiveness tended towards a cost reduction associated with community-based rehabilitation.Conclusion: Use of community-based rehabilitation may be associated with positive and negative effects. However, there is currently insufficient supporting evidence to justify the implementation of community-based rehabilitation for stroke rehabilitation. A stronger evidence base is required to adequately inform health policy decisions and guide methods of service delivery to effectively improve stroke patient outcomes.