The association of annual and daily volume with in-hospital mortality in Dutch and Danish Emergency Departments

Emergency department (ED) outcomes may be influenced by two volume-related mechanisms: structural volume (large vs. smaller EDs), which may confer scale-related advantages, and relative daily workload (busy vs. quiet days). We examined associations between annual ED volume and relative daily volume with in-hospital mortality across the six most common presenting complaints (PCs). We conducted an international retrospective multicenter cohort study using the Netherlands Emergency department Evaluation Database (NEED; seven EDs) and the Danish Multicenter Cohort (DMC; five EDs), including the six most common PCs. We used mixed-effects logistic regression to examine associations of annual PC-specific volume and relative daily ED volume with in-hospital mortality, adjusted for demographics, urgency, vital signs, Glasgow Coma Scale, consultations, laboratory tests, imaging, and medication. Relative daily volume was defined as total ED visits on a given day compared with that ED’s usual daily volume and categorized as low (<–1 SD), usual-volume days (−1 to + 1 SD), or high (>+1 SD). We included 486,558 Dutch and 237,698 Danish ED visits. Annual PC-specific volume showed no association with mortality. Higher relative daily volume (busy days) was associated with lower mortality in the Netherlands (e.g. dyspnea AOR 0.89, 95% CI 0.86–0.92), with similar non-significant trends in Denmark. Categorical analyses showed that excess mortality on low-volume days largely drove these associations. Annual ED volume was not associated with mortality. Mortality per PC was higher on low-volume days, suggesting that (relative) daily volumes may be more relevant than structural scale (annual volumes).