ISQUA17-2693MULTIDISCIPLINARY EMBEDDED RESEARCH TO IDENTIFY SOLUTIONS TO EMERGENCY DEPARTMENT OVERCROWDING

Abstract

Objectives: Emergency department (ED) overcrowding is a common and persistent problem in many countries, impacting on the quality of care, patient safety, staff morale and cost. There is existing literature on the causes and effects of overcrowding, and evidence that it can be alleviated by a range of different quality improvement interventions. However, interventions don't always have the desired effect and many health care staff and systems are suffering “change fatigue” and lack the means to prioritise and focus their improvement effort strategically. Our objective was to determine a prioritised set of evidence-based interventions for reducing ED overcrowding that were attuned to the context-specific problems and organisational potential for change in an acute hospital in the UK, and to develop generic guidance for use by other organisations to identify compatible interventions. Methods: Mathematical modelling based on queuing theory was used to explore the intrinsic limits and drivers of ED performance in an acute hospital in the UK under different operating conditions, given exogenous factors such as arrival rates and bed capacity elsewhere in the hospital. ED performance was defined as the percentage of patients admitted or discharged within 4 hours in a given day. Ethnographic research (observations (20 hours), staff shadowing (15 hours) and interviews (n = 35)) explored the organisational practices and perceptions of staff in the hospital in relation to overcrowding. Soft systems methodology was used to integrate modelling and ethnographic findings within a structured multi-stakeholder process to identify and prioritise interventions. Results: Staff reported various system, process and cultural factors impacting on overcrowding, such as lack of communication between ED staff and staff in other departments which lead to delays in clinical decision-making, difficulties maintaining good team dynamics in the ED, irregularity of some services such as imaging and ambulatory care, and exit blocks produced by problems with flow across the hospital. A mathematical queueing model was developed and parametrised using data routinely collected in the ED, emerging findings from the ethnography and expert clinical input. It was used to assess quantitatively the relative scale of reductions in ED overcrowding achievable through improving different aspects of the process. For example, we determined the extent to which delays caused by the availability of specialist clinicians and/or hospital beds are currently limiting ED performance and quantified achievable gains through requesting specialists and/or beds earlier in the process and/or shifting the current pattern of hospital discharges to earlier in the day. Existing evidence of effective interventions from the literature were characterised with respect to their impact on different aspects of crowding and the context of implementation. These interventions were filtered and prioritised accounting for how desirable, acceptable and feasible to implement staff considered them to be. In parallel, generic guidance was developed for identifying evidence-based interventions that might be compatible with a given organisation based on their characteristics. Conclusion: Quantitative modelling coupled with qualitative research enabled an acute hospital to assess the potential impact of interventions aimed at reducing ED overcrowding and so where strategically to focus their efforts to affect change. Generic guidance for other organisations would need to be tested to understand the potential for spread.