Mortality, admission rates and outpatient use among frequent users of emergency departments: a systematic review

OBJECTIVE: This systematic review examines whether frequent emergency department (ED) users experience higher mortality, hospital admissions and outpatient visits than non-frequent ED users. DESIGN: We published an a priori study protocol in PROSPERO. Our search strategy combined terms for 'frequent users' and 'emergency department'. At least two independent reviewers screened, selected, assessed quality and extracted data. Third-party adjudication resolved conflicts. Results were synthesised based on median effect sizes. DATA SOURCES: We searched seven electronic databases with no limits and performed an extensive grey literature search. ELIGIBILITY CRITERIA FOR SELECTING STUDIES: We included observational analytical studies that focused on adult patients, had a comparison group of non-frequent ED users and reported deaths, admissions and/or outpatient outcomes. RESULTS: The search strategy identified 4004 citations; 374 were screened by full text and 31 cohort and cross-sectional studies were included. Authors used many different definitions to describe frequent users; the overall quality of the included studies was moderate. Across seven studies examining mortality, frequent users had a median 2.2-fold increased odds of mortality compared with non-frequent users. Twenty-eight studies assessing hospital admissions found a median increased odds of admissions per visit at 1.16 and of admissions per patient at 2.58. Ten studies reported outpatient visits with a median 2.65-fold increased risk of having at least one outpatient encounter post-ED visit. CONCLUSIONS: Frequent ED users appear to experience higher mortality, hospital admissions and outpatient visits compared with non-frequent users, and may benefit from targeted interventions. Standardised definitions to facilitate comparable research are urgently needed. REVIEW REGISTRATION NUMBER: PROSPERO (CRD42013005855).