Beyond BMI: a comprehensive approach to pediatric obesity assessment

Abstract

Background: Traditional Body Mass Index based obesity classification presents limitations in pediatric populations, particularly among physically active children. The 2025 Obesity Classification Framework proposed by The Lancet Diabetes & Endocrinology Commission integrates body fat distribution and metabolic biomarkers, aiming to enhance diagnostic accuracy in pediatric obesity.

Methods: We evaluated 111 physically active children (aged 5-11 years) from the Monterrey Football League in Mexico using both the traditional BMI-based classification and the new 2025 Obesity Classification Framework, which incorporates body composition (measured by bioelectrical impedance analysis), waist-to-height ratio, and metabolic biomarkers. Each participant was classified with both frameworks, and outcomes were compared against metabolic risk markers. Normality was assessed using the Shapiro-Wilk test. Non-normally distributed variables (fat mass, visceral fat, triglycerides, creatinine, and pCr) were analyzed using non-parametric tests, while parametric tests were applied for normally distributed data. Agreement between classifications was determined using Cohen's kappa coefficient.

Results: Agreement between classifications was moderate (κ = 0.532, P < 0.001). Using the new framework, 20 children previously classified as overweight by BMI were reclassified as having preclinical obesity, reflecting excess adiposity previously unrecognized. Conversely, four participants initially categorized as obese by BMI were reclassified as non-obese, reflecting elevated lean mass rather than adiposity. Participants categorized as having preclinical obesity exhibited significantly higher levels of LDL cholesterol and apolipoprotein B compared to non-obese peers.

Conclusions: The 2025 Obesity Classification Framework provides greater precision than traditional BMI-based assessments by effectively differentiating between excess adiposity and increased lean mass in physically active children. Although bioelectrical impedance analysis was selected due to its practicality, cost-effectiveness, and non-invasiveness, it has inherent measurement variability compared to dual-energy x-ray absorptiometry. Future research validating these results against DXA or other reference standards is recommended. Adopting this comprehensive assessment strategy may facilitate earlier and more targeted interventions for children at risk of obesity-related complications.

Clinical trial registration: https://doi.org/10.1186/ISRCTN12172320, identifier ISRCTN12172320.

Keywords: body composition; body mass index; childhood nutrition; metabolic health; pediatric obesity; physical activity.

Figure 1

Patient-recruitment flow chart.

Figure 2

Distribution of obesity prevalence according to different classification methods: non-obesity, overweight, preclinical obesity, and clinical obesity, represented for each method.

Figure 3

Correspondence of obesity classifications in comparison with (a) LDL-cholesterol and (b) ApoB. Mann–Whitney U Test, Student's t-test for independent samples, One-way ANOVA.

Figure 4

Classification of pediatric athletes: comparison between traditional BMI classification and the 2025-OCF. A total of 48 participants remained classified as normal weight. Four participants with a BMI ≥85th percentile were reclassified as non-obese because they did not meet criteria for excess body fat or increased waist-to-hip ratio (WHR). Thirty-six participants with a BMI ≥85th percentile had both excess body fat and increased WHR, placing them in the preclinical obesity category. Four participants with a BMI ≥95th percentile were reclassified as non-obese, as their high BMI resulted from increased skeletal muscle mass (above the 90th or 95th percentile) rather than excess adiposity or WHR elevation. Fifteen participants with a BMI ≥95th percentile, excess body fat, and increased WHR but no metabolic abnormalities were classified as preclinical obesity, bringing the total to 19 preclinical obesity cases when combined with those classified based on body fat percentage alone. Only four participants met the criteria for clinical obesity, having a BMI ≥95th percentile, excess body fat, increased WHR, and abnormal metabolic markers, including elevated fasting glucose, cholesterol, or ApoB.