Risk of in-hospital mortality following emergency department admission: results from the geriatric EDEN cohort study.

OBJECTIVE: To determine whether being admitted to emergency department (ED) for social disorders may predict a higher risk of in-hospital mortality among older inpatients. DESIGN: Prospective cohort study (mean follow-up: 9.1±10.0 days). SETTING: Angers University Hospital, France. PARTICIPANTS: Four hundred twenty-two inpatients (mean age 84.9±5.6years, 64.2% women). METHODS: At their admission to ED, inpatients aged 75 years and over received an assessment composed of 6 items: age, gender, number of drugs daily taken, history of falls during the past 6 months, usual place of life, and use of formal and/or informal home and social services. The reasons for admission to ED as well the diagnosis at the time of hospital discharge were separated into social and health disorders. The length of hospital stay was calculated in number of days using the hospital registry. Inpatients were separated into 2 groups based on the occurrence or not of death during the hospital stay. RESULTS: Older inpatients who died at hospital were more frequently institutionalized (P=0.034) and admitted to ED for social disorders (P=0.002) than those who did not. Multiple Cox regression model revealed that living in institution and social disorders as a reason for admission to ED were significantly associated with the occurrence of death at hospital (P=0.008 and P=0.036). Kaplan-Meier distributions of in-hospital mortality showed that home-living inpatients admitted to ED for social disorders died more and faster during hospitalization than those admitted for health disorders (P=0.016). CONCLUSION: Being admitted to ED for social disorders and living in institution predicted a higher risk of in-hospital mortality.