BackgroundA dementia diagnosis can lead to a decline in cognitive, social, and physical health, but people with dementia can live meaningful lives and participate actively in society with psychosocial support. This single-arm, non-randomised feasibility study explored the feasibility and acceptability of a Comprehensive REsilience-building psychoSocial intervenTion (CREST) for people with dementia, their caregivers, General Practitioners (GPs), and the public.MethodsNine people with dementia and their primary caregivers living in the community (n = 9 dyads) completed the CREST intervention which had three components (cognitive stimulation therapy [CST], physical exercise, and dementia education). Quantitative secondary outcomes were assessed at baseline and following the 15-week intervention; qualitative interviews were conducted during and post-intervention. All study components were assessed against pre-defined criteria, to determine the feasibility of conducting a future definitive trial.ResultsRecruitment of people with dementia and their caregiver was a significant challenge and led to considerable delays to the onset and conduct of the intervention. Only 13% of eligible GP practices agreed to assist in recruitment and achieved a 6% enrolment rate; a community-based recruitment strategy proved more effective, yielding a 29% enrolment rate. However, once recruited, participants maintained high attendance and adherence to the content of each component with average adherence rates of 98% for CST, exercise sessions and caregiver education. Adherence to secondary exercise measures was lower, with home exercise diary completion at 37% and Fitbit wear adherence at 80% during the day and 67% at night. The people with dementia felt their concentration and fitness had improved over the 15-week intervention and particularly enjoyed the social aspects (e.g. group classes, exercising with partners from the community). Caregivers felt they had better knowledge and understanding following their education component and reported that the social aspects (interacting and sharing experiences with each other) were important. Overall, participants reported that the three components of the intervention were feasible and acceptable. In addition, the quantitative measures and health economic tools employed were feasible. However, the secondary elements of the exercise component (recording home exercise diaries and Fitbit use) were not considered feasible. Overall, pre-defined criteria for progression to a definitive intervention were fulfilled in terms of acceptability, retention and fidelity but not recruitment.ConclusionWhile overall, the CREST intervention was feasible and acceptable to participants, significant difficulties with recruitment of people with dementia and their caregiver through GP practices impacted the viability of delivering the intervention. Recruitment through community-based groups proved a more feasible option and further work is needed to overcome barriers to recruiting this cohort before a larger-scale trial can be conducted.Trial registrationISRCTN25294519.