Comparison of WHO and CDC growth charts for defining weight status in the young population in Israel: a population-based cross-sectional study.

Background: In Israel, the absence of national growth charts leads to the use of CDC or WHO growth charts to assess pediatric growth indices. This population-based cross-sectional study compared BMI z-scores and weight classifications using CDC and WHO charts in children insured by Clalit Health Services (CHS).

Methods: The study analyzed the CHS electronic database for patients aged 2-18 years with at least one recorded BMI measurement from January 2017 to December 2023. Exclusion criteria included improbable BMI measurements (> 60 kg/m² or < 10 kg/m²). Demographic data, height, and weight were collected, and BMI z-scores were calculated using both CDC and WHO growth charts. Results were stratified by sex and age groups (2-5 years and > 5-18 years).

Results: The cohort included 1,475,543 children: 488,008 aged 2-5 years (52% male) and 987,535 aged > 5-18 years (51% male). In the younger group, the median CDC BMI z-scores were below 0 (< 50th percentile), while the median WHO BMI z-scores were above 0 for both sexes, with significant differences between methods (P < 0.001). BMI z-scores were lower in males than in females across both methods (P < 0.001). In the older group, for both methods, the median BMI z-scores were above 0. WHO z-scores had higher medians in males compared to CDC z-scores (P < 0.001), while in females, z-scores were similar between methods (P = 0.210). CDC showed lower median z-scores for males compared to females (P < 0.001), whereas WHO results were comparable between sexes (P = 0.337). There were significant discrepancies in weight classification, particularly in the 2-5 age group. Overweight rates were over 4% higher using CDC charts compared to WHO (P < 0.001), with minimal agreement (Kappa = 0.06 for males, 0.01 for females). In the older group, WHO classified 4% more children as overweight than CDC (P < 0.001), with moderate agreement in males (Kappa = 0.74) and strong agreement in females (Kappa = 0.81).

Conclusions: The study underscores the risk of misclassifying childhood overweight and obesity depending on the growth standard used, particularly in younger children. Policymakers should carefully choose appropriate standards and consider developing national growth charts tailored to the local pediatric population, while allocate resources for early interventions addressing both undernutrition and overnutrition.