Background: In England, 70% of people who were homeless on admission to hospital were discharged back to the street without having their care and support needs addressed (Department of Health, 2013). Furthermore they experienced high hospital readmission rates and often resorted to inappropriate use of the Emergency Department. Following these data, Government funding was provided to enable hospitals to work with local partners to develop specialist integrated homeless hospital discharge schemes, including intermediate or step-up/step-down care. In this paper we report preliminary findings from a realist evaluation which explored the effectiveness of the different schemes established. Schemes took many different formats, and tended to be either uniprofessional, comprising housing workers, or multi-professional, comprising: GPs; occupational therapists; nurses; social workers; housing workers; and peer navigators.
Methods: The study was carried out in England between September 2015 and February 2018. It was funded by the National Institute of Health Research (NIHR). It employs a realist evaluation methodology, comparing sites with specialist integrated care (n=4) to those with standard care (n=2). Realist evaluation is designed to explore the relationship between the local context, the mechanisms and resources deployed and the outcomes. The overall aim is to generate hypotheses about ‘what works for whom, in what circumstances and why’. Across the six study sites in-depth observational fieldwork was undertaken. This encompassed interviews with 60 practitioners and stakeholders, and 60 people who were homeless on admission to hospital. People who were homeless were interviewed shortly after discharge and again three months later. An economic evaluation and ‘data linkage’ (across a total of 20 sites) were also undertaken to interrogate the emerging realist hypotheses.
Preliminary Results: Our emerging hypothesis is that there is no ‘magic bullet’ or single most effective solution. Findings so far indicate that improving hospital discharge arrangements for people who are homeless requires action on many different fronts to weave together a range of potentially effective mechanisms and resources. To illustrate this, we will present examples of different local configurations evolving over time, in rural and urban contexts, mapping the advantages and disadvantages of each. We will pinpoint a number of key attributes of effective systems (e.g. clear protocols for the discharge of people who are homeless, discharge co-ordination and intermediate care) and the factors that may explain why these are necessary and workable in some but not all locations. Throughout, we will tease out the concept of ‘specialist integrated care’ and whether this is indicative of improved quality or further fragmentation.
Limitations and future research: The limitations of the research are that it is focused on England. We hope that during the conference we can engage with international colleagues with a view to scoping ideas for future comparative research in the field of integrated care transitions.