Local workings of acute admission avoidance: a critical realist examination of resource deployment in four hospital emergency departments

Background: Unplanned hospital admissions through Emergency Departments (EDs) in the United Kingdom have risen annually. National policy has suggested that many are avoidable with the right mix of community and hospital... [ view full abstract ]

Background: Unplanned hospital admissions through Emergency Departments (EDs) in the United Kingdom have risen annually. National policy has suggested that many are avoidable with the right mix of community and hospital service provision. With little empirical evidence to guide improvement initiatives, multiple local solutions have been developed. In the Avoidable Acute Admissions study we conducted a realist examination of these natural experiments in four hospitals in south west England. This enabled us to generate practical theory about how practitioner expertise and adaptations in hospital urgent care pathways can promote effective decision-making about emergency admission and discharge.

Design: This mixed-methods study incorporated value stream mapping (VSM) of events in patient journeys, organisational ethnography, and stakeholder participation.
We generated realist theory through analysis of two data sets:
1. VSM measurement of time spent by practitioners on key activities in 108 patient pathways.
2. Organisational ethnography using participant observation and semi-structured interviews with 65 patients, 30 carers and 282 practitioners, exploring their experiences in acute care systems on the four study sites. We prepared 24 qualitative case studies of patient journeys through emergency care. Qualitative data collected from clinical panels, learning sets and workshops were also analysed.
We used realist synthesis to develop provisional theories as to how the resources of hospital space, practitioner skills and new teams can contribute to reducing avoidable admissions. Initial analytic statements in the form of ‘if…then’ micro-hypotheses were synthesised into consolidated theories.

Results: On all sites, practitioners worked under considerable pressure to reduce admissions and avoid breaches in the four-hour target time for seeing patients in the ED. Innovative approaches to admission avoidance included: multi-disciplinary discharge/care of the elderly teams with rapid access to outpatient clinics and community-based care; a hospital-based acute GP service; diverse ways of deploying senior practitioners in front-line patient care and departmental oversight roles; and creation of new ‘holding spaces’ to observe and plan discharge. Despite these multiple efforts, alternatives to admission were not always known about and applied systematically. Accessibility of hospital and community resources was affected by variations in practitioners’ knowledge and experience, visibility of the resource, physical proximity to the ED and the institutional culture of communication between teams and organisations. The realist synthesis revealed a range of mechanisms used in specific contexts to deploy available resources more effectively.
Realist methodological learning points: Ethnographic observations and interviews provided rich data for generating the ‘if…then’ causal micro-theories. Original data sources often needed rechecking when multiple micro-hypotheses were combined into consolidated theories. Findings from quantitative and qualitative components of the study could be synthesised using realist methods. The quantitative VSM data often supplied data about outcomes, while qualitative data was more amenable to development of causal theories. Our final critical focus on the ‘how’ of resource deployment was supported by comparing findings across the four sites, where apparently similar resources were implemented differently.