Classification of childhood obesity using longitudinal clinical body mass index and its validation

Area of Science:

• Pediatrics
• Cardiology
• Public Health

Background:

• Childhood adiposity is linked to long-term cardiometabolic risks.
• Current obesity classification relies on cross-sectional Body Mass Index (BMI), potentially underestimating risks.
• A need exists for improved methods to classify childhood obesity and its associated health outcomes.

Purpose of the Study:

• To develop and validate a childhood obesity classification system using longitudinal clinical data.
• To assess the association between longitudinal BMI classification and cardiometabolic risk.
• To compare the predictive power of longitudinal versus cross-sectional BMI classification for adverse health outcomes.

Main Methods:

• An observational study utilizing electronic health record data from a tertiary care hospital, with replication in an independent cohort.
• Longitudinal classification based on median BMI percentile and persistence in obesity class for individuals with ≥ 3 BMI measurements.
• Analysis of associations between longitudinal BMI class, early-onset obesity, socioeconomic status (SES), and cardiometabolic risk.

Main Results:

• Obesity (BMI ≥ 95th percentile) affected 24.1% and severe obesity 10.6% of children.
• Early-onset obesity (≤ 6 years) was associated with persistent or higher obesity classes.
• Longitudinal BMI classification demonstrated a higher Area Under the Curve (AUC) for predicting cardiometabolic risk (80.0%) compared to cross-sectional BMI (75.8%).

Conclusions:

• Longitudinal BMI classification offers a more accurate reflection of long-term cardiometabolic risk in children.
• This refined classification system can guide focused intervention strategies.
• It may also assist in identifying children who could benefit from genetic testing for obesity-related conditions.