BACKGROUND: Recovery Colleges offer peer-led, coproduced psychoeducational courses to support people to have meaningful lives. There is limited knowledge of their use in the context of dementia. This study used a realist programme theory approach to develop an in-depth understanding of how United Kingdom Recovery College dementia courses lead to outcomes for people with dementia, families and staff. OBJECTIVES: To build knowledge through mapping Recovery College dementia courses across the United Kingdom; conduct a realist review of evidence; identify key components of effective Recovery College dementia courses within a realist programme theory; identify outcome measures for evaluating Recovery College dementia courses; understand diverse ethnic and cultural needs of people with dementia in accessing Recovery College dementia courses; and coproduce resources for implementing Recovery College dementia courses. DESIGN AND METHODS: A mixed-methods design to examine what works for who and in what circumstances, coproduced with people with lived experience of dementia and staff from National Health Service Recovery Colleges and memory services. This involved: a United Kingdom-based staff survey of memory services and Recovery College dementia course provision, delivery and attendance; a realist review combining literature and stakeholder knowledge to build an initial programme theory; a realist evaluation using ethnographic observation within five case studies, realist interviews and documentary evidence, with analysis involving stakeholders using realist logic to explore causal processes operating in different contexts and intended and unintended outcomes (i.e. Contexts, Mechanisms and Outcomes); a scoping review of potential outcome measures for evaluation; and coproduced resources using three rounds of stakeholder workshops. SETTING AND PARTICIPANTS: Set in United Kingdom National Health Service mental health organisations, staff recruitment from Recovery Colleges and memory services and people with dementia/families from Recovery Colleges. RESULTS: Twelve (from 51) Recovery Colleges offered dementia-specific courses, and 210 memory service staff completed the survey. Thirty-five documents and discussions with 19 stakeholders (7 people with dementia, 2 family carers, 10 staff) informed the initial programme theory. A trusted person endorsing a course encouraged attendance. Shared coproduction values underpinned the success of setting up and running courses. Through co-facilitation of recovery-focused content by peer-tutors with well-developed facilitation skills, attendees appeared to mediate self-stigma, manage emotional uncertainty and make meaningful social connections in ways which engendered hope for the future. Course evaluation was challenging as people often left without completing a written survey. One attempt at pre and post course well-being measures failed to capture follow-up date. LIMITATIONS: Four case sites were recruited, limiting diversity in course delivery methods and participant backgrounds. Numbers of people with dementia attending courses was lower than expected, raising questions about awareness and accessibility. Course evaluation was limited to standard feedback sheets. Sufficient data on coproduction processes and staff experiences were lacking, thus restricting refined theories about coproduction and its impact on practice. CONCLUSIONS: Recovery-focused post-diagnostic courses can enable people with dementia to consider a hopeful future. Resources available may stimulate inclusivity and accessibility. The authenticity of peer tutors with dementia resonated with course attendees, and lived experience insights from people who were living positive lives with dementia is a key strength of this form of post-diagnostic support. FUTURE WORK: Longer-term outcomes (e.g. hopefulness) for people with dementia attending recovery-focused courses is possible, but further validation of identified measures to address responsiveness, interpretability, inclusion of personal-recovery domains and cultural sensitivity for diverse populations is a pre-requisite. FUNDING: This synopsis presents independent research funded by the National Institute for Health and Care Research (NIHR) Health and Social Care Delivery Research programme as award number NIHR131676.