Prevalence and Factors Associated with Sarcopenia Among Brazilian Older Adults: An Exploratory Network Analysis

Objectives: This study aimed to explore the prevalence of sarcopenia and its intricate associations with sociodemographic factors, anthropometric and body composition data, cognitive levels, and depressive symptoms in community-dwelling older adults.Study Design: A randomized cross-sectional study was extracted from a probabilistic cluster conducted on elderly individuals aged 65 years or older residing in the community.Methods: Sarcopenia was defined according to the criteria of the European Working Group on Sarcopenia in Older People (EWGSOP2). Body composition was assessed using dual-energy X-ray absorptiometry (DXA). The Geriatric Depression Scale (GDS-15) was employed to screen for depressive symptoms. Associations were analyzed using networks based on mixed graphical models. Predictability indices of the estimated networks were assessed using the "proportion of explained variance" (R2) for numerical variables and the "proportion of correct classification" (CC*) for categorical variables.Results: The study included 278 participants, with a majority being female (61%). Among those with sarcopenia, 67% were women and 33% were men. In the network model, sex, education, race, income, waist circumference, weight, bone mass, muscle mass values, and depressive symptoms were associated with sarcopenia. The covariates collectively demonstrated a high accuracy (67.6%) in predicting sarcopenia categories.Conclusion: Network analysis of sarcopenia proves to be a valuable tool, enabling the exploration of complex relationships between covariates and this outcome. Furthermore, the results underscore the significance of early screening for the treatment of sarcopenia in elderly individuals.