Objective: To examine changes in leisure participation following stroke/transient ischaemic attack (TIA) and explore its relationship to modifiable and non-modifiable participant characteristics. Design: An observational study design with self-report questionnaires collected at two time points (baseline and 6-months). Setting: The study was conducted across 21 hospital sites in England, Wales, and Northern Ireland. Participants: Participants were aged 18+ and had experienced a first or recurrent stroke or TIA and had a post-stroke/TIA modified Rankin score (mRS) of ≤3. Procedure: Research practitioners at each site approached potential participants. Individuals who agreed to participate completed a baseline questionnaire whilst an inpatient or at a first post-stroke/TIA clinic appointment. A follow-up questionnaire was posted to participants with a freepost return envelope. Two questionnaires were developed that collected demographic information, pre-stroke/TIA mRS, social circumstances (e.g., employment situation) and incorporated the shortened Nottingham Leisure Questionnaire (sNLQ). Results: The study recruited eligible participants (N = 3295); 2000 participants returned questionnaires at follow-up. Data showed three participant variables were significant predictors of engagement in leisure activities post-stroke/TIA: age, sex, and deprivation decile. There was an overall decline in the number and variety of leisure activities, with an average loss of 2.2 activities following stroke/TIA. Only one activity, “exercise/fitness” saw an increase in engagement from baseline to follow-up; watching TV remained stable, whilst participation in all other activities reduced between 10% and 40% with an average activity engagement reduction of 22%. Conclusions: Some groups experienced a greater reduction in activities than others—notably older participants, female participants, and those living in a low socioeconomic area. Registration: researchregistry4607. Strengths and limitations of this study: 1. This is the largest-ever study to survey life and leisure activity engagement following stroke/TIA. 2. Survey responses were self-reported retrospectively and, therefore, may have been misreported, or misremembered. 3. Despite the large cohort, there were few participants, and so respondents, from ethnic minority groups.