Background

The 2000 CDC body mass index (BMI) growth charts were developed using national survey data from 1963 to 1994 and are routinely used to monitor growth and screen for obesity among youth aged 2-19 years. Parameters derived from the current BMI growth charts allow calculation of percentiles and their corresponding z-scores between the 3rd and 97th percentiles.Extrapolation of z-scores beyond the 97th percentile is not recommended because a wide range of high BMIs compress to a very narrow range of high z-scores. The objective of this study is to develop a data-driven method for overcoming this limitation.

Methods

BMI values from youth at or above the 95th percentile (the threshold for obesity) from combined national survey data between 1963 and 2016 were modeled as half-normal distributions at 6-month age intervals for boys and girls separately.Sex- and age-specific scale parameters were estimated using a direct survey-weighted population mean estimator and then smoothed using regression procedures.

Results

Parameters derived using BMI values from 1813 youth with obesity included in development of the original growth charts combined with 6964 youth with obesity from more recent surveys permitted calculation of BMI z-scores beyond the 97th percentile that do not result in the compressed values observed with the current method.

Conclusions

This new method preserves use of the current CDC growth charts for the usual calculations of percentiles and z-scores up to the 95th percentile.Supplementation with more recent national data allows calculation of percentiles and z-scores beyond what was previously recommended, and may be useful for monitoring extreme BMI throughout childhood development in clinical and research settings.