Risk for short-term undesirable outcomes in older emergency department users: Results of the ER(2) observational cohort study

Launay, Cyrille P.; Galery, Kevin; Vilcocq, Christine; Afilalo, Marc; Beauchet, Olivier

BACKGROUND: The "Emergency Room Evaluation and Recommendations" (ER(2)) is a clinical tool designed to determine prognosis for the short-term Emergency Department (ED) undesirable outcomes including long length of stay (LOS) in ED and in hospital, as well as the likelihood of hospital admission during an index ED visit. It is also designed to guide appropriate and timely tailor-made geriatric interventions. This study aimed to examine whether ER(2) assessment part was: 1) usable by ED healthcare workers (e.g. nurses) and 2) scoring system associated with long LOS in ED and in hospital, as well as hospital admission in older ED users on stretchers. METHODS: Based on an observational, prospective and longitudinal cohort study 1,800 participants visiting the ED of the Jewish General Hospital (Montreal, Quebec, Canada) were recruited between September and December 2017. ER(2) assessment determined three risk-levels (i.e., low, medium and high) for short-term ED undesirable outcomes. The rate of ER(2) digital form completed, the time to fill ER(2) items and obtain ER(2) risk-levels, the LOS in ED and in hospital, and hospital admission were used as outcomes. RESULTS: ER(2) was usable by ED nurses in charge of older ED users. High-risk group was associated with both increased ED stay (coefficient of regression β = 3.81 with P≤0.001) and hospital stay (coefficient of regression β = 4.60 with P = 0.002) as well as with hospital admission (HR = 1.32 with P≤0.001) when low ER(2) risk level was used as referent level. Kaplan-Meier distributions showed that the three risk groups of participants differed significantly (P = 0.001). Those with high-risk level (P≤0.001) were discharged later from hospital to a non-hospital location compared to those with low risk. There was no significant difference between those classified in low-risk and in medium-risk groups (P = 0.985) and those in medium and high-risk groups (P = 0.096). CONCLUSION: The ER(2) assessment part is usable in daily practice of ED care and its risk stratifications may be used to predict adverse outcomes including prolonged LOS in ED and in hospital as well as hospital admission. TRIAL REGISTRATION: NCT03964311.