Optimum Hospice at Home Services for End-of-Life care: realist evaluation and co-production

Overview: Hospice at home (H@H) services aim to enable people to have a “good death” at home. While this accords with UK policy, statutory services are ill-equipped to meet this demand and there is limited evidence from the perspective of service users, as participants or co-producers, of what aspects are most helpful. Our evaluation asked what features of H@H models work, for whom, how and under what circumstances? We combined a realist evaluation with co-production embedded throughout: phase 1 - a national survey to map service provision and models of H@H care; phase 2 - qualitative and economic data collection with commissioners, service providers and patient-carer dyads; phase 3 - national consensus workshops with stakeholders. The funding application was developed in partnership with ten public members. Two of these became co-applicants, both former carers with experience of hospice services. Utilising a co-production approach, the lay co-applicants contributed to all stages of the research process and were an integral part of the project team. This paper focuses on their role in qualitative data analysis (phase 2); developing and refining programme theories and testing context-mechanism-outcome (CMO) configurations. Key points: The co-applicants initially participated in clearly delineated tasks such as commenting on patient/carer resources. At the end of phase 1, they requested greater involvement, which translated into participating in data analysis. Additional training in realist philosophy and data analysis was provided. The co-applicants attended team coding meetings. Preparation included listening to and reading carer interviews, and coded according to provisional CMOs. Co-applicants fed back at coding meetings as part of the iterative process of refining CMOs. They provided a fresh perspective on the data, reflecting on the underlying narrative that proved invaluable in helping researchers to step back from the minutiae of coding and consider the carer’s perspective. Implications: Public members as an integral part of a realist research team add a different lens to interrogating CMO configurations, enhancing rigour in developing programme theories. While the process evolved overtime, we would recommend early discussions around what activities co-applicants would like to participate in, training requirements, time commitment and emotional demands.