OP48Increased Mortality Associated with Weekend Admission; A Result of Higher Severity of Illness at the Time of Admission?

BackgroundA number of studies have shown that weekend admissions to hospital are associated with significantly higher mortality compared with weekday admissions. Previous studies have suggested that the observed variability in weekend-weekday mortality is a consequence of the quality of care provided in hospital. However, little information exists on whether the increased mortality associated with weekend admission is explained by the relative severity of illness at the time of admission.MethodsWe performed a retrospective analysis of all emergency acute care admissions in England between April 2009 and March 2010 using Hospital Episode Statistics data. In-hospital mortality among patients who were admitted as emergency admission on weekends as compared to weekdays was analysed using the following measures: (1) overall deaths in hospital; (2) deaths in hospital within and after 3 days of admission (3) deaths in hospital among patients with 'higher risk' conditions and 'lower risk' conditions. The 'higher risk' conditions consisted of 103 Primary Diagnosis codes (ICD-10), which accounted for the 74% of in-hospital deaths in the UK. The remaining Primary Diagnosis codes were categorised as 'lower risk' conditions.ResultsThere were 10,221,500 emergency hospital admissions during the two-year study period; 23.8% of the patients were admitted on weekends. A total of 391,938 patients died in hospital; a quarter of these died within 3 days of admission. Admission on weekends was associated with significantly higher in-hospital deaths as compared to weekday admission (Odds Ratio 1.11, 95% CI 1.10-1.12). Mortality within 3 days of admission (early deaths) was also higher for patients admitted on weekends (OR 1.16, 95%CI (1.14-1.17) and this difference persisted for both 'higher risk' and 'lower risk' conditions. However, when early deaths were excluded from the analysis, mortality was not affected by weekend admission.ConclusionHigher mortality was seen among patients admitted on weekends compared to those admitted on weekdays. However, as the odds ratio for deaths after 3 days of admission was not affected by day of admission, it is probable that the weekend admission effect on mortality is at least in part due to higher severity of illness at the time of admission rather than purely being a consequence of poorer quality of in hospital care at the weekend, and this warrants further investigation.