OBJECTIVES: To identify, appraise and synthesise existing design evidence for inpatient stroke rehabilitation facilities; to identify impacts of these built environments on the outcomes and experiences of people recovering from stroke, their family/caregivers and staff. DESIGN: A convergent segregated review design was used to conduct a systematic review. DATA SOURCES: Ovid MEDLINE, Scopus, Web of Science and Cumulative Index to Nursing and Allied Health Literature were searched for articles published between January 2000 and November 2020. ELIGIBILITY CRITERIA FOR SELECTING STUDIES: Qualitative, quantitative and mixed-methods studies investigating the impact of the built environment of inpatient rehabilitation facilities on stroke survivors, their family/caregivers and/or staff. DATA EXTRACTION AND SYNTHESIS: Two authors separately completed the title, abstract, full-text screening, data extraction and quality assessment. Extracted data were categorised according to the aspect of the built environment explored and the outcomes reported. These categories were used to structure a narrative synthesis of the results from all included studies. RESULTS: Twenty-four articles were included, most qualitative and exploratory. Half of the included articles investigated a particular aspect of the built environment, including environmental enrichment and communal areas (n=8), bedroom design (n=3) and therapy spaces (n=1), while the other half considered the environment in general. Findings related to one or more of the following outcome categories: (1) clinical outcomes, (2) patient activity, (3) patient well-being, (4) patient and/or staff safety and (5) clinical practice. Heterogeneous designs and variables of interest meant results could not be compared, but some repeated findings suggest that attractive and accessible communal areas are important for patient activity and well-being. CONCLUSIONS: Stroke rehabilitation is a unique healthcare context where patient activity, practice and motivation are paramount. We found many evidence gaps that with more targeted research could better inform the design of rehabilitation spaces to optimise care. PROSPERO REGISTRATION NUMBER: CRD42020158006.