Cognition has already been considered as a component of frailty, and it has been demonstrated that it is associated with adverse health outcomes. We estimated the prevalence of frailty syndrome in an Italian older population and its predictive role on all-cause mortality and disability in nondemented subjects and in demented patients. We evaluated 2,581 individuals recruited from the Italian Longitudinal Study on Aging, a population-based sample of 5,632 subjects, aged 65-84 years old. Participants received identical baseline evaluation at the 1st survey (1992-1993) and were followed at 2nd (1995-1996) and 3rd survey (2000-2001). A phenotype of frailty according to partially modified measurement of Cardiovascular Health Study criteria was operationalized. The overall prevalence of frailty syndrome in this population-based study was 7.6% (95% confidence interval (CI) 6.55-8.57). Frail individuals noncomorbid or nondisable were 9.1% and 39.3%, respectively, confirming an overlap but not concordance in the co-occurrence among these conditions. Frailty was associated with a significantly increased risk of all-cause mortality over a 3-year follow-up (hazard ratio (HR) 1.98, 95% confidence interval (CI) 1.52-2.60) and over a 7-year follow-up (HR 1.74, 95% CI 1.44-2.16), but with significant increased risk of disability only over a 3-year follow-up (HR 1.32, 95% CI 1.06-1.86 over a 3-year follow-up and HR 1.16, 95% CI 0.88-1.56 over a 7-year follow-up). Frail demented patients were at higher risk of all-cause mortality over 3- (HR 3.33, 95% CI 1.28-8.29) and 7-year follow-up periods (HR 1.89, 95% CI 1.10-3.44), but not of disability. Frailty syndrome was a short-term predictor of disability in nondemented older subjects and short- and long-term predictor of all-cause mortality in nondemented and demented patients.