Background: The preservation of autonomy and the ability of elderly to carry out the basic activities of daily living, beyond the therapeutic care of any pathologies, appears as one of the main objectives of care during hospitalization.
Objectives: To identify early clinical markers associated with the loss of independence in elderly people in short stay hospitals.
Methods: Among the 1,306 subjects making up the prospective and multicenter SAFEs cohort study (Sujet Agé Fragile: Evolution et suivi-Frail elderly subjects, evaluation and follow-up), 619 medical inpatients, not disabled at baseline and hospitalized through an emergency department were considered. Data used in a multinomial logistic regression were obtained through a comprehensive geriatric assessment (CGA) conducted in the first week of hospitalization. Dependency levels were assessed at baseline, at inclusion and at 30 days using Katz's ADL index. Baseline was defined as the dependence level before occurrence of the event motivating hospitalization. To limit the influence of rehabilitation on the level of dependence, only stays shorter than 30 days were considered.
Results: About 514 patients were eligible, 15 died and 90 were still hospitalized at end point (n = 619). Two-thirds of subjects were women, with a mean age of 83. At day 30 162 patients (31%) were not disabled; 61 (12%) were moderately disabled and 291 severely disabled (57%). No socio-demographic variables seemed to influence the day 30 dependence level. Lack of autonomy (odds ratio (OR) = 1.9, 95% confidence interval (CI) = 1.2-3.6), walking difficulties (OR = 2.7, 95% CI = 1.3-5.6), fall risk (OR = 2.1, 95% CI = 1.3-6.8) and malnutrition risk (OR = 2.2, 95% CI = 1.5-7.6) were found in multifactorial analysis to be clinical markers for loss of independence.
Conclusions: Beyond considerations on the designing of preventive policies targeting the populations at risk that have been identified here, the identification of functional factors (lack of autonomy, walking difficulties, risk of falling) suggests above all that consideration needs to be given to the organization per se of the French geriatric hospital care system, and in particular to the relevance of maintaining sector-type segregation between wards for care of acute care and those involved in rehabilitation.