Comparison of two frailty measures in the Conselice Study of Brain Ageing.

OBJECTIVES: Uncertainty about the definition of frailty is reflected by the development of many ways to identify frail people. We aimed to compare the validity of two frailty measures in participants of the Conselice Study of Brain Aging. DESIGN: Prospective population-based study with 4 year follow up. PARTICIPANTS/SETTING: 1,016 subjects aged 65 and over in a rural Italian population. METHODS: For each participant, a Frailty Index (FI) and a Conselice Study of Brain Aging Score (CSBAS) were determined. The FI was created from 43 deficits according to a standardized methodology; 7 variables derived from a previously validated Easy Prognostic Score comprised the CSBAS. RESULTS: The FI had characteristic properties described in other population samples, with a gamma distribution, a 99% limit of about 0.64 and higher values in women than men. CSBAS and FI were strongly correlated with each other (r = 0.72) and both correlated with age (r = 0.32, r = 0.27, respectively). Each was independently predictive of death in a multivariate model, with greater specificity and sensitivity than age alone. CONCLUSIONS: Frailty can be measured by different tools and facilitates a more direct quantification of individual vulnerability than chronological age alone. Though the Frailty Index and the Conselice Study of Brain Aging Score are underpinned by different rationales, clinical utility will continue to motivate their development.

Physical activity and dementia risk in the elderly: findings from a prospective Italian study.

OBJECTIVE: To examine the effect of physical activity on risk of developing Alzheimer disease (AD) and vascular dementia (VaD) in the elderly. METHODS: Data are from a prospective population-based cohort of 749 Italian subjects aged 65 and older who, in 1999/2000, were cognitively normal at an extensive assessment for clinically overt and preclinical dementia and, in 2003/2004, underwent follow-up for incident dementia. Baseline physical activity was measured as energy expenditure on activities of different intensity (walking, stair climbing, moderate activities, vigorous activities, and total physical activity). RESULTS: Over 3.9 +/- 0.7 years of follow-up there were 86 incident dementia cases (54 AD, 27 VaD). After adjustment for sociodemographic and genetic confounders, VaD risk was significantly lower for the upper tertiles of walking (hazard ratio [HR] 0.27, 95% CI 0.12 to 0.63), moderate (HR 0.29, 95% CI 0.12 to 0.66), and total physical activity (HR 0.24, 95% 0.11 to 0.56) compared to the corresponding lowest tertile. The association persisted after accounting for vascular risk factors and overall health status. After adjustment for sociodemographic and genetic confounders, AD risk was not associated with measures of physical activity and results did not change after further adjustment for vascular risk factors and overall health and functional status. CONCLUSIONS: In this cohort, physical activity is associated with a lower risk of vascular dementia but not of Alzheimer disease. Further research is needed about the biologic mechanisms operating between physical activity and cognition.