Process evaluation of APPLE-Tree (active prevention in people at risk of dementia through lifestyle behaviour change and technology to build resilience): dementia prevention study focused on health and lifestyle changes

BACKGROUND: This concurrent, exploratory, mixed-methods process evaluation, embedded within a randomised controlled trial, investigates how the 'active prevention in people at risk of dementia through lifestyle behaviour change and technology to build resilience' (APPLE-Tree) secondary dementia prevention intervention might support behavioural and lifestyle goal attainment, through determining the contexts influencing engagement and testing intervention theoretical assumptions. AIMS: We aimed to investigate (a) intervention reach, dose and fidelity, (b) contexts influencing engagement and (c) alignment of findings with theoretical assumptions about how the intervention might have supported participants to meet personalised behavioural and lifestyle goals. METHOD: We measured intervention reach and dose. We selected interviewees for setting, gender and ethnic diversity from the 374 APPLE-Tree trial participants randomised to the intervention arm. We interviewed 25 intervention participants, 12 facilitators and 3 study partners. Additionally, we analysed 11 interviews previously conducted during or after intervention delivery for an ethnography, and 233 facilitator-completed participant goal records. We thematically analysed data, combining inductive/deductive approaches informed by the 'capability, opportunity and motivation-behaviour' (COM-B) behaviour change model. We video-recorded a randomly selected tenth of sessions and rated fidelity. RESULTS: A total of 346 of 374 (92.5%) intervention arm participants received some intervention (reach), and 305 of 374 (81.6%) attended ≥5 main sessions (predefined as adhering: dose). According to facilitator records, participants met a mean of 5.1 of 7.5 (68.3%) goals set. We generated three themes around (a) building capability and motivation, (b) connecting with other participants and facilitators and (c) flexibility and a tailored approach. CONCLUSIONS: The intervention supported behaviour change, through increasing knowledge and providing space to plan, implement and evaluate new strategies and make social connections. Feedback indicated that the intervention was flexible and inclusive of diverse preferences and needs.