RESUMEN
OBJECTIVE: To assess the longitudinal association between cognitive impairment and sarcopenia in a sample of Brazilian community-dwelling older adults. DESIGN: Nine-year observational prospective study. SETTING AND PARTICIPANTS: A total of 521 community-dwelling older adults from 2 Brazilian sites of the Frailty in Brazilian Older Adults (FIBRA in Portuguese) study. METHODS: Sarcopenia was defined as low hand-grip strength and low muscle mass. Cognitive impairment was determined at baseline using the Mini-Mental State Examination, with education-adjusted cutoff scores. The logistic regression model was used to assess the association between cognitive impairment and incident sarcopenia after adjusting for gender, age, education, morbidities, physical activity, and body mass index. Inverse probability weighting was applied to correct for sample loss at follow-up. RESULTS: The mean age of the study population was 72.7 (±5.6) years, and 365 were women (70.1%). Being 80 years and older [odds ratio (OR), 4.62; 95% CI, 1.38-15.48; P = .013], being under- and overweight (OR, 0.29; 95% CI, 0.11-0.76; P = .012, and OR, 5.12; 95% CI, 2.18-12.01; P < .001, respectively) and having cognitive impairment (OR, 2.44; 95% CI, 1.18-5.04; P = .016) at baseline predicted sarcopenia after 9 years. CONCLUSION AND IMPLICATIONS: Cognitive impairment may predict sarcopenia in Brazilian older adults. More studies are necessary to identify the main mechanisms shared by sarcopenia and cognitive decline, which could support the development of prevention interventions.