Abstract
Traditional reliance on Body Mass Index (BMI) as a diagnostic tool for obesity is increasingly challenged due to its inability to differentiate between fat and lean mass and to capture fat distribution. Emerging evidence-including findings from our longitudinal study in Latino patients with obesity and insights from the 2025 Lancet Commission on Obesity-suggests that a comprehensive evaluation of body composition is essential for accurate risk stratification. This review synthesizes historical perspectives and recent developments in obesity phenotyping, detailing how the field has evolved from simple BMI-based assessments to multifaceted approaches incorporating bioelectrical impedance analysis (BIA) and supplementary anthropometric measures such as waist circumference and waist-to-hip ratio. We also examine the metabolic, genetic, and hormonal mechanisms underlying phenotypic variability, which help explain why individuals with similar BMIs may exhibit markedly different health risks. By integrating our original data with an extensive review of current literature, we demonstrate that refined obesity phenotyping can serve as an early indicator of progression from preclinical to clinical obesity. Such nuanced classifications offer the potential for more personalized therapeutic interventions aimed at optimizing weight loss outcomes and reducing cardiometabolic risk. Overall, our findings advocate for a multidimensional approach to obesity assessment that promises to improve clinical outcomes through tailored, phenotype-based strategies.