Metabolic syndrome predicts mobility decline in a community-based sample of older adults.

Published

Journal Article

OBJECTIVES: To determine whether metabolic syndrome is an independent predictor of decline in mobility in an elderly community sample. DESIGN: Biracial community-based prospective cohort study. SETTING: Urban and rural areas of central North Carolina. PARTICIPANTS: One thousand two hundred twenty-nine older African Americans and whites, mean age 77.0, who were participants in the Duke Established Populations for Epidemiologic Studies of the Elderly. MEASUREMENTS: Sociodemographic data and data on mobility (a subset of items from the Rosow-Breslau scale), depression (Center for Epidemiological Studies Depression Scale), self-report of medical conditions, body mass index, cognitive function (Short Portable Mental Status Questionnaire), blood pressure, height, and waist circumference. High-density lipoprotein cholesterol, triglycerides, and blood glucose were available from blood samples. Metabolic syndrome was calculated for subjects with complete data. RESULTS: Twenty-nine percent of the sample met criteria for metabolic syndrome. In bivariate analyses, age, sex, race, education, cognitive impairment, depression, impairment in mobility, history of stroke and heart disease, and metabolic syndrome at baseline were associated with a decline in mobility 4 years later. Regression analysis controlling for the above variables demonstrated that metabolic syndrome persisted as an independent and highly significant predictor of decline in mobility. CONCLUSION: These findings suggest that metabolic syndrome may be a distinct risk factor for mobility decline in community-dwelling older people.

Full Text

Duke Authors

Cited Authors

  • Blazer, DG; Hybels, CF; Fillenbaum, GG

Published Date

  • March 2006

Published In

Volume / Issue

  • 54 / 3

Start / End Page

  • 502 - 506

PubMed ID

  • 16551320

Pubmed Central ID

  • 16551320

International Standard Serial Number (ISSN)

  • 0002-8614

Digital Object Identifier (DOI)

  • 10.1111/j.1532-5415.2005.00607.x

Language

  • eng

Conference Location

  • United States