The prognostic significance of delirium in older hospital patients

OBJECTIVES: To determine whether delirium is an independent predictor of adverse outcomes of hospitalization in older patients. DESIGN: Cohort study PATIENTS: A total of 225 people admitted as an emergency to an acute geriatric unit in a university teaching hospital METHODS: Subjects were screened for delirium, defined by Diagnostic and Statistical Manual, 3rd Edition criteria, every 48 hours. Outcome measures included mortality, duration of hospital stay, hospital-acquired complications, and institutional placement. The influence of delirium on these outcomes was calculated after adjusting for age, illness severity on admission, burden of comorbidity, prior cognitive impairment, and level of disability. RESULTS: Delirium was present on admission in 41 patients (18%) and developed after admission in a further 53 patients (24%). Patients with delirium were more likely than nondelirious patients to have chronic cognitive impairment, severe acute illness, multiple comorbid conditions, and functional disability. Nevertheless, in multivariate analyses adjusting for these factors, delirium was independently associated with prolonged hospital stay, functional decline during hospitalization, increased risk of developing a hospital-acquired complication, and with increased admission to longterm care. CONCLUSION: Delirium is an independent predictor of adverse outcomes in older hospital patients.