Who is less likely to die in association with improved National Emergency Access Target (NEAT) compliance for emergency admissions in a tertiary referral hospital?

Abstract

Objective The aim of the present study was to identify patient and non-patient factors associated with reduced mortality among patients admitted from the emergency department (ED) to in-patient wards in a major tertiary hospital that had previously reported a near halving in mortality in association with a doubling in National Emergency Access Target (NEAT) compliance over a 2-year period from 2012 to 2014. Methods We retrospectively analysed routinely collected data from the Emergency Department Information System (EDIS) and hospital discharge abstracts on all emergency admissions during calendar years 2011 (pre-NEAT interventions) and 2013 (post-NEAT interventions). Patients admitted to short-stay wards and then discharged home, as well as patients dying in the ED, were excluded. Patients included in the study were categorised according to age, time and day of arrival to the ED, mode of transport to the ED, emergency triage category, type of clinical presentation and major diagnostic codes. Results The in-patient mortality rate for emergency admissions decreased from 1.9% (320/17022) in 2011 to 1.2% (202/17162) in 2013 (P<0.001). There was no change from 2011 to 2013 in the percentage of deaths in the ED (0.19% vs 0.17%) or those coded as in-patient palliative care (17.9% vs 22.2%). Although deaths were not associated with age by itself, the mortality rate of older patients admitted to medical wards decreased significantly from 3.5% to 1.7% (P=0.011). A higher mortality rate was seen among patients presenting to ED triage between midnight and 12 noon than at other times in 2011 (2.5% vs 1.5%; P<0.001), but this difference disappeared by 2013 (1.3% vs 1.1%; P=0.150). A similar pattern was seen among patients presenting on weekends versus weekdays: 2.2% versus 1.7% (P=0.038) in 2011 and 1.3% versus 1.1% (P=0.150) in 2013. Fewer deaths were noted among patients with acute cardiovascular or respiratory disease in 2013 than in 2011 (1.7% vs 3.6% and 1.5% vs 3.4%, respectively; P<0.001 for both comparisons). Mode of transport to the ED or triage category was not associated with changes in mortality. These analyses took account of any possible confounding resulting from differences over time in emergency admission rates. Conclusions Improved NEAT compliance as a result of clinical redesign is associated with improved in-patient mortality among particular subgroups of emergency admissions, namely older patients with complex medical conditions, those presenting after hours and on weekends and those presenting with time-sensitive acute cardiorespiratory conditions. What is known about the topic? Clinical redesign aimed at improving compliance with NEAT and reducing time spent within the ED of acutely admitted patients has been associated with reduced mortality. To date, no study has attempted to identify subgroups of patients who potentially derive the greatest benefit from improved NEAT compliance in terms of reduced risk of in-patient death. It also remains unclear as to what extent non-patient factors (e.g. admission practices and differences in coding of palliative care patients) affect or confound this reduced risk. What does this paper add? The present study is the first to reveal that enhanced NEAT compliance is associated with lower mortality among particular subgroups of emergency patients admitted to in-patient wards. These include older patients with complex medical conditions, those presenting after hours or on weekends or those with time-sensitive acute cardiorespiratory conditions. These results took account of any possible confounding resulting from differences over time in emergency admission rates, deaths in the ED, numbers of short-stay ward admissions and coding of palliative care deaths. What are the implications for practitioners? Efforts aimed at improving NEAT compliance and efficiencies at the ED-in-patient interface appear to be worthwhile in reducing in-patient mortality among particular subgroups of emergency admissions at high risk. More research is urgently needed in identifying patient- and system-level factors that predispose to higher mortality rates in such populations, but are potentially amenable to focused interventions aimed at optimising transitions of care at the ED-in-patient interface and increasing NEAT compliance for patients admitted to in-patient wards from the ED.