Implementation processes in a cognitive rehabilitation intervention for people with dementia: a complexity-informed qualitative analysis

Objectives: Healthcare is often delivered through complex interventions. Understanding how to implement these successfully is important for optimising services. This article demonstrates how the complexity theory concept of ‘self-organisation’ can inform implementation, drawing on a process evaluation within a randomised controlled trial of the GREAT (Goal-oriented cognitive Rehabilitation in Early-stage Alzheimer’s and related dementias: a multi-centre single-blind randomised controlled Trial) intervention which compared a cognitive rehabilitation intervention for people with dementia with usual treatment. Design: A process evaluation examined experiences of GREAT therapists and participants receiving the intervention, through thematic analysis of a focus group with therapists and interviews with participants and their carers. Therapy records of participants receiving the intervention were also analysed using adapted framework analysis. Analysis adopted a critical realist perspective and a deductive-inductive approach to identify patterns in how the intervention operated. Setting: The GREAT intervention was delivered through home visits by therapists, in eight regions in the UK. Participants: Six therapists took part in a focus group, interviews were conducted with 25 participants and 26 carers, and therapy logs for 50 participants were analysed. Intervention: A 16-week cognitive rehabilitation programme for people with mild-to-moderate dementia. Results: ‘Self-organisation’ of the intervention occurred through adaptations made by therapists. Adaptations included simplifying the intervention for people with greater cognitive impairment, and extending it to meet additional needs. Relational work by therapists produced an emergent outcome of ‘social support’. Self-organised aspects of the intervention were less visible than formal components, but were important aspects of how it operated during the trial. This understanding can help to inform future implementation. Conclusions: Researchers are increasingly adopting complexity theory to understand interventions. This study extends the application of complexity theory by demonstrating how ‘self-organisation’ was a useful concept for understanding aspects of the intervention that would have been missed by focusing on formal intervention components. Analysis of self-organisation could enhance future process evaluations and implementation studies. Trial registration number: ISRCTN21027481.