Background: Stroke patients conventionally undergo a substantial part of their rehabilitation in hospital. Services have been developed that offer patients ea rly discharge from hospital with rehabilitation at home (early supported discharge [ESD]). W e have assessed the effects and costs of such services. Methods: We did a meta- analysis of data from individual patients who took part in randomised trials tha t recruited patients with stroke in hospital to receive either conventional care or any ESD service intervention that provided rehabilitation and support in a c ommunity setting with the aim of shortening the duration of hospital care. The p rimary outcome was death or dependency at the end of scheduled follow-up. Findi ngs:Outcome data were available for 11 trials (1597 patients). ESD services were mostly provided by specialist multidisciplinary teams to a selected group (medi an 41%) of stroke patients admitted to hospital. There was a reduced risk of de ath or dependency equivalent to six (95%CI one to ten) fewer adverse outcomes f or every 100 patients receiving an ESD service (p=0.02). The hospital stay was 8 days shorter for patients assigned ESD services than for those assigned convent ional care (p < 0.0001). There were also significant improvements in scores on t he extended activities of daily living scale and in the odds of living at home a nd reporting satisfaction with services. The greatest benefits were seen in the trials evaluating a coordinated multidisciplinary ESD team and in stroke patient s with mild to moderate disability.Interpretation: Appropriately resourced ESD s ervices provided for a selected group of stroke patients can reduce long-term d ependency and admission to institutional care as well as shortening hospital sta ys. Background: Stroke patients conventionally undergo a substantial part of their rehabilitation in hospital. Services have been evaluated that offer patients ea rly discharge from hospital with rehabilitation at home (early supported discharge [ESD]). services: We did a meta-analysis of data from individual patients who took part in randomized trials tha tha recruited patients with stroke in hospital to receive either conventional care or any ESD service intervention that provided rehabilitation and support in ac ommunity settings with the aim of shortening the duration of hospital care. The p rimary outcome was death or dependency at the end of scheduled follow-up. Findi ngs: Outcome data were available for 11 trials (1597 patients). There was a reduced risk of de ath or dependency equiva The hospital stay was 8 days shorter for patients assigned ESD services than for those assigned conventional care (p = 0.02) <0.0001). There were also significant improvements in scores on t he extended activities of daily living scale and in the odds of living at home a nd reporting with services. The greatest benefits were seen in the trials evaluating a coordinated multidisciplinary ESD team and in stroke patient s with mild to moderate disability.Interpretation: Appropriately resourced ESD s ervices provided for a selected group of stroke patients can reduce long-term d ependency and admission to institutional care as well as shortening hospital sta ys.