Abstract Introduction There is limited understanding of long-term delirium care after discharge from hospital for older people. A realist approach was used to investigate the contextual factors and mechanisms of care that influence recovery from delirium. Realist evaluation is fundamentally theory-driven. A preliminary programme theory was used as the foundation for theory testing and refinement to develop the RecoverED intervention. Method Realist interviewing techniques were used to obtain real-world and lived experiences of delirium recovery and service use in the community for theory-building and testing. Semi-structured interviews were conducted with a purposive sample of people with delirium (N = 7), informal carers (N = 14), and healthcare professionals (N = 24). Data from the interviews were analysed using a deductive codebook of Context-Mechanism-Outcome (CMO) configurations. Open coding was also performed to identify inductive themes, which were then aggregated to elicit explanatory statements. Results There was support for a multicomponent delirium intervention including cognitive and physical rehabilitation, and psychosocial support. The analysis revealed the need for an additional component which focused on improving awareness and understanding about delirium among those with lived experience. In the context of insufficient knowledge about delirium, people experienced increased fear and anxiety among other negative outcomes. Offering a focused educational component as part of the intervention is expected to contribute to recovery outcomes. This was associated with CMOs identifying the need for positive relationships with staff, improving communication with staff and sense-making through staff emotional support. Conclusion(s) The preliminary programme theory was refined based on the realist analysis data. Additional components were included, one of which was targeted education for people with delirium and carers. Following a consultation with an expert panel, the intervention is being tested in a feasibility trial and process evaluation, which will analyse data from multiple sources using realist methods to further refine the intervention.
Read full abstract