Background: Environmental enrichment involves organisation of the environment and provision of equipment to facilitate engagement in physical, cognitive and social activity. In animals with stroke, it promotes brain plasticity and recovery. Aims: To assess the feasibility and safety of a patient-driven model of environmental enrichment incorporating access to communal and individual environmental enrichment. Methods: A non-randomised cluster trial with blinded measurement involving people with stroke (n=193) in 4 rehabilitation units was carried out. Feasibility was operationalised as activity 10 days after admission to rehabilitation and availability of environmental enrichment. Safety was measured as falls and serious adverse events. Benefit was measured as clinical outcomes at 3 months, by an assessor blinded to group. Results: The experimental group (n=91) spent 7% (95% CI -14 to 0) less time inactive, 9% (95% CI 0 to 19) more time physically, and 6% (95% CI 2 to 10) more time socially active than the control group (n=102). Communal environmental enrichment was available 100% of the time, but individual environmental enrichment was rarely within reach (24%) or sight (39%). There were no between-group differences in serious adverse events or falls at discharge or 3 months nor in clinical outcomes at 3 months. Conclusions: This patient-driven model of environmental enrichment was feasible and safe. However, the very modest increase in activity by people with stroke, and the lack of benefit in clinical outcomes 3 months after stroke do not provide justification for an efficacy trial. Clinical Trial Registration: ANZCTR 12613000796785 Words: 245
