After stroke, up to 81% of individuals develop shoulder subluxation, a condition frequently associated with poor upper limb function. Recently, electrical stimulation has been applied to shoulder muscles to treat shoulder subluxation. The purpose of this meta-analysis was to examine the efficacy of surface electrical stimulation for the prevention or reduction of shoulder subluxation after stroke. A meta-analysis of all eligible randomised or quasi-randomised trials of electrical stimulation for the treatment of shoulder subluxation identified by computerised and hand searches of the literature was carried out. The primary outcome measure of interest was subluxation. Seven (four early and three late) trials met the inclusion criteria. The mean PEDro score out of 10 for quality of the methods was 5.8 for the four early trials and 4.3 for the three late trials. Data were pooled when subluxation was measured in millimetres. Analysis found that, when added to conventional therapy, electrical stimulation prevented on average 6.5mm of shoulder subluxation (weighted mean difference, 95% CI 4.4 to 8.6) but only reduced it by 1.9mm (weighted mean difference, 95% CI -2.3 to 6.1) compared with conventional therapy alone. Therefore, evidence supports the use of electrical stimulation early after stroke for the prevention of, but not late after stroke for the reduction of, shoulder subluxation.