Objectives: The aim of this study was to identify early indicators of prolonged hospital stays by elderly patients.
Methods: This prospective pilot study, conducted at Strasbourg University Hospital, included patients aged 75 years or older who were hospitalized via the emergency department (SAFES cohort: Sujet Agé Fragile: Evaluation et suivi, that is, Frail Elderly Subjects: Evaluation and Follow-up). A gerontologic evaluation of these patients during the first week of their hospitalization furnished the data for an exact logistic regression. Two definitions were used for prolonged hospitalization: 30 days and a composite number adjusted for diagnosis-related group according to the French classification (f-DRG).
Results: The analysis examined 137 hospitalizations. More than two thirds of the patients were women (73%), with a mean age of 84 years. Twenty-four hospitalizations (17%) lasted more than 30 days, but only 6 (4%) lasted beyond the DRG-adjusted limit. No social or demographic variables appeared to affect the length of stay, regardless of the definition of prolonged stay. No indicator was associated with the 30-day limit, but clinical markers were linked to prolongation assessed by f-DRG adjustment. A "risk of malnutrition" (OR=14.07) and "mood disorders" (OR=2,5) were both early markers for prolonged hospitalization. Although not statistically significant, "walking difficulties" (OR=2.72) and "cognitive impairment" (OR=5.03) appeared to be associated with prolonged stays. No association was seen with either the variables measured by Katz's Activities of Daily Living Index or its course during hospitalization.
Conclusion: Our study shows that when generally recognized indicators of frailty are taken into account, a set of simple items enables a predictive approach to the prolongation of emergency hospitalizations of the elderly.