Objectives: Health services have an imperative to reduce prolonged patient length of stay (LOS) in ED. Our objective is to develop and validate an accurate prediction model for patient LOS in ED greater than 4 hours using a data mining technique. Methods: Data were collected from a regional Australian public hospital for all ED presentations between 1 January 2016 and 31 December 2017. A decision tree algorithm was built to predict patients with an ED LOS >4 hours. A total of 33 attributes were analysed. The performance of the final model was internally validated. Clinically relevant patterns from the model were analysed. Results: The accuracy of the model was 85%. We identified that patients at our site who were at high risk of ED LOS >4 hours were those who were waiting in ED for a medical consultation, or those who were waiting for a urology, surgical, orthopaedic or paediatric consultation if the request for consultation occurred more than 2 hours after the patient was first seen by an ED doctor. Conclusion:This model performed very well in predicting ED LOS >4 hours for each individual patient and demonstrated a number of clinically relevant patterns. Identifying patterns that influence ED LOS is important for health managers in order to develop and implement interventions targeted at those clinical scenarios. Future work should look at the utility of displaying individual patient risk of ED LOS >4 hours using this model in realtime at the point-of-care.

Time‐based targets (TBTs) for ED stays were introduced to improve quality of care but criticised as having harmful unintended consequences. The aim of the review was to determine whether implementation of TBTs influenced quality of care. Structured searches in medical databases were undertaken (2000–2019). Studies describing a state, regional or national TBTs that reported processes or outcomes of care related to the target were included. Harvest plots were used to summarise the evidence. Thirty‐three studies (n = 34 million) were included. In some settings, reductions in mortality were seen in ED, in hospital and at 30 days, while in other settings mortality was unchanged. Mortality reductions were seen in the face of increasing age and acuity of presentations, when short‐stay admissions were excluded, and when pre‐target temporal trends were accounted for. ED crowding, time to assessment and admission times reduced. Fewer patients left prior to completing their care and fewer patients re‐presented to EDs. Short‐stay admissions and re‐admissions to wards within 30 days increased. There was conflicting evidence regarding hospital occupancy and ward medical emergency calls, while times to treatment for individual conditions did not change. The evidence for associations was mostly low certainty and confidence in the findings is accordingly low. Quality of care generally improved after targets were introduced and when compliance with targets was high. This depended on how targets were implemented at individual sites or within jurisdictions, with important implications for policy makers, health managers and clinicians.

Time‐based targets for ED length of stay were introduced in England in 2000, followed by the rest of the UK, Canada, Ireland, New Zealand, and Australia after ED crowding was associated with poor quality of care and increased mortality. This systematic review evaluates qualitative literature to see if ED time‐based targets have influenced patient care quality. We included 13 studies from four countries, incorporating 617 interviews. We conclude that time‐based targets have impacted on the quality of emergency patient care, both positively and negatively. Successful implementation depends on whole hospital resourcing and engagement with targets.

Scenario 1Ms J is 81 years old, an active retiree and lives alone. She has been brought to the ED by her daughter feeling tired and flat. She felt well getting up early this morning, but an hour later felt very fatigued, mildly short of breath and nauseated. She returned to bed awaiting her daughter's visit. She has a past history of hypertension, diabetes and hypothyroidism. Her vital signs are normal. The junior ED registrar asks how she should work up this patient?

Objective This scoping review explores the structure and process‐level strategies that are associated with medical retrieval outcomes. A secondary aim is to identify the range of medical retrieval outcomes used to assess the performance of remote retrieval services. Design A scoping review of peer‐reviewed literature from PubMed, CINAHL and the Web of Science was undertaken following guidelines set by the Johanna Briggs Institute manual for scoping reviews. All articles were assessed by two reviewers. Themes were derived inductively from the data extracted. Setting Medical retrievals in sparsely populated remote locations in high‐income countries. Participants Staff and clients of remote medical retrieval services. Interventions Structures and processes (e.g. resource availability, retrieval staff structures and governance protocols) that aimed to improve medical retrieval outcomes. Outcomes Patient health outcomes and service efficiency. Results Twenty‐four articles were included. Three broad themes, related to the nature of the interventions, were included: optimising prehospital management of retrievals, staffing and resourcing of retrieval services and retrieval model evaluation. Mortality was the most frequently used outcome indicator in these studies, but was not measured consistently across studies. Conclusions This review highlights significant gaps in the literature that describes the structure and processes of retrieval models operating in remote areas and a dearth of literature evaluating specific operational strategies implemented within medical retrieval models. The available literature does not meaningfully assist with identifying key outcome indicators for developing a consistent monitoring and evaluation framework for retrieval services in geographically, culturally and demographically diverse remote contexts.