In-hospital mortality and length-of-stay in a cohort of ED patients admitted to internistic units

Abstract

Background: Several life-threatening acute diseases exhibit precise preferred times of onset, during the day (morning), the month or season of the year (winter), and the day of the week (Monday). Moreover, admission to the hospital during nighttime and weekend (WE) is associated with increased risk of in-hospital mortality (IHM) and length-of-hospital stay (LOS). Since most of these data have been obtained in North American settings, the aim of this study was to verify also in an Italian setting whether some factors related to ED evaluation, including time of arrival, and disease’s severity evaluated by a colour code, could be associated with IHM and LOS in ED patients admitted to internal medicine units.

Methods: This study was conducted between January 1, 2010 and July 31, 2012, at the St Anna Hospital of Ferrara. We analyzed age, sex, time of arrival to the ED, ED boarding time, and triage colour code of all patients admitted to the Department of Medicine. Friday to midnight Sunday was considered as WE, and the nine main national festive days were always considered as Sunday-WE. Thus, the time period was classified into holidays (WE + festive days) and weekdays (WD), and daytime (8:00 am - 8:00 pm) and nighttime (8:00 pm - 8:00 am). IHM and LOS were assumed as primary and secondary objectives, respectively. Patients were classified into deceased and survivors, and LOS was classified depending on length lower or higher than the median (seven days). T-test, chi-squared test, U Mann-Whitney test, and logistic regression analysis were performed.

Results: During the 18-month period, 13,237 consecutive patients visiting the ED (43.4% males, mean age 74.8 ± 13.6 years, 56.6% females, mean age 78.0 ± 13.3 years), were considered. IHM was related to: time of admission (daytime vs. nighttime, p = .015; WE vs. WD, p = .026, holidays vs. no holidays, p = .021), colour code (p < .001), and age (p < .001). Multivariate analysis showed an independent association with IHM for colour code (p < .001), and age (p < .001). LOS >7 days was related to: time of admission (daytime vs. nighttime, p < .001; WE or WD, p = .001; holidays vs. no holidays, p = .002). Furthermore, LOS >7 days depended on both colour code of admission (p < .001), and age (p < .001). Multivariate analysis showed a significant correlation with LOS >7 days for yellow colour code (p < .001), and age (p < .001).

Conclusions: Although several factors, such as admission colour code and age, influence some parameters of clinical outcome, i.e., IHM and LOS, a striking weekend effect is not evident in a cohort of consecutive patients admitted to an Italian Medical Department, characterized by intermediate-high level of urgency.