We included 19 studies involving 619 participants. The trials included participants who had some residual motor power of the paretic arm, the potential for further motor recovery and with limited pain or spasticity, but tended to use the limb little if at all. Only five studies had adequate allocation concealment. The majority of studies were underpowered (median number of included patients was 15) and we cannot rule out small-trial bias. Six trials (184 patients) assessed disability immediately after the intervention, indicating a significant standard mean difference (SMD) of 0.36, 95% confidence interval (CI) 0.06 to 0.65. For the most frequently reported outcome, arm motor function (11 studies involving 373 patients), the SMD was 0.72 (95% CI 0.32 to 1.12). There were only two studies that explored disability improvement after a few months of follow up and found no significant difference, SMD -0.07 (95% CI -0.53 to 0.40).