Abstract Background Comprehensive Geriatric Assessment (CGA) provides detailed information on clinical, functional and cognitive aspects of older patients and is especially useful for assessing frail individuals. Although a large proportion of hospitalized older adults demonstrate a high level of complexity, CGA was not developed specifically for this setting. Our aim was to evaluate the application of a CGA model for the clinical characterization and prognostic prediction of hospitalized older adults. Methods This was a prospective observational study including 746 patients aged 60 years and over who were admitted to a geriatric ward of a university hospital between January 2009 and December 2011, in Sao Paulo, Brazil. The proposed CGA was applied to evaluate all patients at admission. The primary outcome was in-hospital death, and the secondary outcomes were delirium, nosocomial infections, functional decline and length of stay. Multivariate binary logistic regression was performed to assess independent factors associated with these outcomes, including socio-demographic, clinical, functional, cognitive, and laboratory variables. Impairment in ten CGA components was particularly investigated: polypharmacy, activities of daily living (ADL) dependency, instrumental activities of daily living (IADL) dependency, depression, dementia, delirium, urinary incontinence, falls, malnutrition, and poor social support. Results The studied patients were mostly women (67.4%), and the mean age was 80.5±7.9 years. Multivariate logistic regression analysis revealed the following independent factors associated with in-hospital death: IADL dependency (OR=4.02; CI=1.52-10.58; p = . 005); ADL dependency (OR=2.39; CI=1.25-4.56; p =.008); malnutrition (OR=2.80; CI=1.63-4.83; p