Hormone-defined metabolic phenotypes in institutionalized older adults: a cluster analysis

Introduction
Ageing is associated with progressive declines in metabolic and functional integrity, increasing susceptibility to insulin resistance, type 2 diabetes, dyslipidemia, and cardiovascular disease. Hormonal regulation, including leptin, ghrelin, and insulin, plays a central role in coordinating energy ba­lance, appetite, and nutrient metabolism. However, conventional anthropometric and metabolic markers may not fully capture inter-individual variability in metabolic health, particularly in older adults with excess adiposity. This study aimed to identify distinct metabolic phenotypes in institutionalized psychogeriatric older adults using a data-driven, hormone-centered approach.

Material and methods
Ninety-five participants underwent assessment of anthropometry, blood pressure, nutritional markers, and fasting serum concentrations of leptin, ghrelin, insulin, glucose, high-density lipoprotein cholesterol, triglycerides, non-esterified fatty acids, and albumin. Standardized variables were subjected to k-means cluster analysis to define homogeneous phenotypes, and between-cluster differences were evaluated using ANOVA with Welch correction.

Results
Two clusters were identified. Cluster 1 (n = 69) exhibited relatively stable endocrine and metabo­lic profiles despite elevated body mass index (BMI) and central adiposity, consistent with a metabolically healthy obesity pattern. Cluster 2 (n = 26) displayed higher standardized levels of leptin, ghrelin, and insulin, while glucose, triglycerides, non-esterified fatty acids, albumin, and BMI did not differ significantly. This hormonal profile suggests early alterations in endocrine regulation and a potential predisposition to insulin resistance and metabolic instability, even in the absence of overt changes in classical metabolic markers.

Conclusions
Our findings indicate substantial metabolic heterogeneity in institutionalized older adults and highlight the value of hormone-centered phenotyping for detecting subtle endocrine dysregulation. Early identification of at-risk individuals may inform precision interventions aimed at promoting metabolic stability and healthy ageing.