Weight change and lower body disability in older Mexican Americans

OBJECTIVES: To examine the association between 2-year weight change and onset of lower body disability over time in older Mexican Americans. DESIGN: Data were from the Hispanic Established Population for the Epidemiological Study of the Elderly (1993-2001). Weight change was examined by comparing baseline weight to weight at 2-year follow-up. Incidence of lower body disability was studied from the end of this period through an additional 5 years. SETTING: Five southwestern states: Texas, New Mexico, Colorado, Arizona, and California. PARTICIPANTS: One thousand seven hundred thirty-seven noninstitutionalized Mexican-American men and women aged 65 and older who reported no limitation in activities of daily living (ADLs) and were able to perform the walk test at 2-year follow-up. MEASUREMNTS: In-home interviews assessed sociodemographic factors, self-reported physician diagnoses of medical conditions (arthritis, diabetes mellitus, heart attack, stroke, hip fracture, and cancer), self-reported ADLs, depressive symptoms, and number of hospitalizations. Cognitive function, handgrip muscle strength, and body mass index (BMI) were obtained. The outcomes were any limitation of lower body ADL (walking across a small room, bathing, transferring from a bed to a chair, and using the toilet) and limitation on the walk test over subsequent 5-year follow-up period. General Estimation Equation (GEE) was used to estimate lower body disability over time. RESULTS: Weight change of 5% or more occurred in 42.3% of the participants; 21.7% lost weight, 20.6% gained weight, and 57.7% had stable weight. Using GEE analysis, with stable weight as the reference, weight loss of 5% or more was associated with greater risk of any lower body ADL limitation (odds ratio (OR)=1.43, 95% confidence interval (CI)=1.06-1.95) and walking limitation (OR=1.35, 95% CI=1.03-1.76) after controlling for sociodemographic variables and BMI at baseline. Weight gain of 5% or more was associated with greater risk of any lower body ADL limitation (OR=1.39, 95% CI=1.02-1.89), after controlling for sociodemographic variables and BMI at baseline. When medical conditions, handgrip muscle strength, high depressive symptomatology, cognitive function, and hospitalization were added to the equation, the relationship between 2-year weight change (> 5% loss or > 5% gain) and lower body disability decreased. CONCLUSION: Health conditions and muscle strength partially mediate the association between weight loss or gain and future loss of ability to walk and independently perform ADLs.