ESPE Abstracts

Background: Although Centers for Disease Control and Prevention (CDC) and World Health Organization growth charts, dichotomizing “girls versus boys,” are commonly used, scenarios exist where this binary approach may not be ideal. These scenarios include care for transgender youth undergoing transitions, non-binary youth, and rare diseases where sex-specific growth chart creation is impractical. There is a need for growth charts and z-score calculators that age smooth differences in pubertal timing between sexes to determine how youth are growing as “children” versus “girls or boys.”

Objective: To develop age-adjusted, sex non-specific growth charts for height, weight, and body mass index (BMI), and z-score calculators for these parameters using similar statistical techniques and datasets used to create CDC 2000 growth charts.

Methods: We analyzed data from five US cross-sectional nationally representative surveys including National Health Examination Surveys II-III and National Health and Nutrition Examination Surveys I-III. We used the Lambda, Mu, Sigma semi-parametric approach in a Generalized Additive Models for Location, Scale, and Shape (GAMLSS) technique to model growth. Box-Cox Power distribution families in GAMLSS with additive age splines were used to calculate estimates of our sex non-specific height, weight, and BMI reference data tables.

Results: Data from 39,119 participants (49.5% female; 66.7% non-Hispanic White; 21.7% non-Hispanic Black were included in development of our growth charts, reference ranges, and z-score calculators. Respective growth curves were largely superimposable through age 10 after which, coinciding with pubertal onset timing, differences became apparent. Age-adjusted sex non-specific z-score calculators for height, weight, and BMI are available on this website (http://tsaheight2020.shinyapps.io/gender0growthcharts).

Conclusions: Given the increasing prevalence of youth seeking transgender care and recognized limitations of current approaches, a need has arisen in terms of accurately and appropriately tracking growth parameters in these individuals. For transgender youth, our growth charts could help in terms of properly assessing a transgender youth’s near adult height prediction and expectations surrounding this, weight classification, and decisions regarding further therapy related to growth and weight status that is more robust than the common practice of comparing male and female charts side-by-side in clinical decision-making. These tools provide an intermediate reference between male-specific and female-specific data points, and will help assess growth outcomes in a systematic fashion until transgender-specific, longitudinal data are available. Moreover, these growth charts may have utility as they relate to rare diseases where it can be challenging to create separate growth charts given overall low prevalence.