Metabolic Risk Profiles: A Latent Class Analysis Involving Variables Related to Blood Glucose and Insulin Resistance

Background and Purpose: Older adults are more susceptible to the development of type 2 diabetes mellitus (T2DM) due to age-related changes in insulin secretion and signaling pathways. Given the multifactorial nature of metabolic disorders, the use of robust multivariate models is justified to explore associated risk factors. This study aimed to identify latent classes of metabolic profiles among older adults based on a cluster of variables associated with cardiovascular risk. Methods: The study included community-dwelling individuals aged 60 years or older residing in urban areas who participated in all three phases of data collection: questionnaires, clinical examinations, and blood sampling. Latent class analysis (LCA) was applied using dichotomized variables, with model selection based on criteria such as Akaike Information Criterion, Bayesian Information Criterion, G², log-likelihood value, and entropy estimation. Results: A total of 210 older adults were evaluated. Three latent classes were identified: low, moderate, and high metabolic risk. The high-risk class was characterized by a higher probability of altered HbA1c and triglyceride-glucose (TyG) index values (0.96), elevated fasting blood glucose, and a prior diagnosis of hypertension (0.87), as well as impaired homeostasis model assessment of insulin resistance (HOMA-IR) index and a prior diagnosis of T2DM (0.79). The moderate-risk class showed a greater likelihood of hypertension (0.87), altered TyG (0.87), and impaired HOMA-IR index (0.56). Conclusions: LCA proved to be a valuable tool in Public Health by enabling the identification of homogeneous subgroups within a heterogeneous population. These findings support the development of more targeted and effective preventive strategies based on specific metabolic risk profiles.