BACKGROUND: Rehabilitation based upon research evidence gives stroke survivors the best chance of recovery. There is substantial research to guide practice in stroke rehabilitation, yet uptake of evidence by healthcare professionals is typically slow and patients often do not receive evidence-based care. Implementation interventions are an important means to translate knowledge from research to practice and thus optimise the care and outcomes for stroke survivors. A synthesis of research evidence is required to guide the selection and use of implementation interventions in stroke rehabilitation. OBJECTIVES: To assess the effects of implementation interventions to promote the uptake of evidence-based practices (including clinical assessments and treatments recommended in evidence-based guidelines) in stroke rehabilitation and to assess the effects of implementation interventions tailored to address identified barriers to change compared to non-tailored interventions in stroke rehabilitation. SEARCH METHODS: We searched CENTRAL, MEDLINE, Embase, and eight other databases to 17 October 2019. We searched OpenGrey, performed citation tracking and reference checking for included studies and contacted authors of included studies to obtain further information and identify potentially relevant studies. SELECTION CRITERIA: We included individual and cluster randomised trials, non-randomised trials, interrupted time series studies and controlled before-after studies comparing an implementation intervention to no intervention or to another implementation approach in stroke rehabilitation. Participants were qualified healthcare professionals working in stroke rehabilitation and the patients they cared for. Studies were considered for inclusion regardless of date, language or publication status. Main outcomes were healthcare professional adherence to recommended treatment, patient adherence to recommended treatment, patient health status and well-being, healthcare professional intention and satisfaction, resource use outcomes and adverse effects. DATA COLLECTION AND ANALYSIS: Two review authors independently selected studies for inclusion, extracted data, and assessed risk of bias and certainty of evidence using GRADE. The primary comparison was any implementation intervention compared to no intervention. MAIN RESULTS: Nine cluster randomised trials (12,428 patient participants) and three ongoing trials met our selection criteria. Five trials (8865 participants) compared an implementation intervention to no intervention, three trials (3150 participants) compared one implementation intervention to another implementation intervention, and one three-arm trial (413 participants) compared two different implementation interventions to no intervention. Eight trials investigated multifaceted interventions; educational meetings and educational materials were the most common components. Six trials described tailoring the intervention content to identified barriers to change. Two trials focused on evidence-based stroke rehabilitation in the acute setting, four focused on the subacute inpatient setting and three trials focused on stroke rehabilitation in the community setting. We are uncertain if implementation interventions improve healthcare professional adherence to evidence-based practice in stroke rehabilitation compared with no intervention as the certainty of the evidence was very low (risk ratio (RR) 1.19, 95% confidence interval (CI) 0.53 to 2.64; 2 trials, 39 clusters, 1455 patient participants; I2 = 0%). Low-certainty evidence indicates implementation interventions in stroke rehabilitation may lead to little or no difference in patient adherence to recommended treatment (number of recommended performed outdoor journeys adjusted mean difference (MD) 0.5, 95% CI -1.8 to 2.8; 1 trial, 21 clusters, 100 participants) and patient psychological well-being (standardised mean difference (SMD) -0.02, 95% CI -0.54 to 0.50; 2 trials, 65 clusters, 1273 participants; I2 = 0%) compared with no intervention. Moderate-certainty evidence indicates implementation interventions in stroke rehabilitation probably lead to little or no difference in patient health-related quality of life (MD 0.01, 95% CI -0.02 to 0.05; 2 trials, 65 clusters, 1242 participants; I2 = 0%) and activities of daily living (MD 0.29, 95% CI -0.16 to 0.73; 2 trials, 65 clusters, 1272 participants; I2 = 0%) compared with no intervention. No studies reported the effects of implementation interventions in stroke rehabilitation on healthcare professional intention to change behaviour or satisfaction. Five studies reported economic outcomes, with one study reporting cost-effectiveness of the implementation intervention. However, this was assessed at high risk of bias. The other four studies did not demonstrate the cost-effectiveness of interventions. Tailoring interventions to identified barriers did not alter results. We are uncertain of the effect of one implementation intervention versus another given the limited very low-certainty evidence. AUTHORS' CONCLUSIONS: We are uncertain if implementation interventions improve healthcare professional adherence to evidence-based practice in stroke rehabilitation compared with no intervention as the certainty of the evidence is very low.