Mode of presentation rather than the 'weekend effect' is a major determinant of in-hospital mortality

Abstract

BackgroundThe influence of patient demographics and mode of admission on the ‘weekend effect’ remains unclear. This study examins the relationship between day of admission, patient demographics, mode of presentation and survival. MethodsHospital admissions over a three-year period were studied. Patients with an inpatient stay less than 24 h and those who were discharged from the emergency department were excluded. In-hospital mortality was correlated with day of admission, age, gender and mode of presentation in a binary logistical regression analysis. ResultsThere were 448,827 admissions, of which 350,648 (85.7%) occurred during a weekday. 256,777 (62.7%) were emergency presentations, which was closely related to a weekend admission (92.3% vs 57.8%, p < 0.001). There were 8099 deaths of which 6336 (78.2%) related to a weekday admission and 1736 (21.4%) related a weekend admission. Mortality for elective admissions was 78 (0.05%) compared to 8021 (3.12%), p < 0.001 in emergency admissions. Univariable regression analysis revealed a weekend admission (Odds Ratio (OR) 1.68 (95% confidence interval (CI) 1.60–1.78, p < 0.001) and emergency presentation (OR 63.02 (95%CI 50.42–78.77), p < 0.011) were associated with weekend mortality. On multivariable analysis the OR for weekend admission reduced to 1.07 (95%CI 1.01–1.13), p = 0.013 and the OR for emergency presentation increased to 76.68 (95%CI 61.40–96.00), p < 0.001. ConclusionThis study highlights that higher weekend mortality rates are a consequence of a lower proportion of elective admissions. Extending the working week to seven days might reduce weekend mortality without reducing the total number of deaths.