ISSN 1662-4009 (online)

ESPE Yearbook of Paediatric Endocrinology (2021) 18 11.11 | DOI: 10.1530/ey.18.11.11

ESPEYB18 11. Obesity and Weight Regulation Criticism of the previous BMI metrics for use in severely obese children (2 abstracts)

11.11. Evaluation of BMI metrics to assess change in adiposity in children with overweight and moderate and severe obesity

Barlow SE , Salahuddin M , Durand C , Pont SJ , Hoelscher DM & Butte NF


Department of Pediatrics, University of Texas Southwestern Medical Center, Dallas, Texas, USA; Children’s Health, Dallas, Texas, USA. sarah.barlow@utsouthwestern.edu.


Obesity (Silver Spring) 2020 Aug;28(8):1512–1518. https://pubmed.ncbi.nlm.nih.gov/32935936/

These authors analysed a cohort of n=399 overweight and obese children (age 2–12 years), who underwent a 1-year lifestyle intervention, to test whether possible alternatives to the BMI z-score (%BMIp95, %BMIp50, BMImz) might correlate better with changes in % total body fat (%BF) as assessed by bioelectrical impendence (BIA).

The findings showed that BMI z-score (calculated according to CDC growth charts) was the least suitable metric to indicate changes in %BF (R2=0.38). The association between delta BMI z-score and detla %BF was also limited by significant variation across age and weight categories. In contrast, other parameters: detla BMI, delta %BMIp95 and detla %BMIp50 performed better (R2=0.53, R2=0.53, R2=0.54, respectively).

This work adds to previous evidence that shows the BMI z-score is a poor indicator to evaluate obesity interventions in children and adolescents (1,2). The calculation of BMI z-scores (BMIz) includes a power transformation step to convert the strong positive skew (right sided tail) into a normal distribution – this causes a close bunching of high BMI values and hence under-emphasises even relatively large changes in BMI in overweight and obese individuals. Further work is also needed to quantify changes in body composition using more accurate methods than BIA, and to correlates those changes with alternative weight and height-dependent metrics. Furthermore, the correlations between these parameters and changes in metabolic health parameters needs to be investigated. Regardless of the many remaining open-ended questions, this work shows that the use of %BMIp95 and of %BMIp50 are more appropriate than BMI z-score in the assessment of an intervention in obese children and adolescents.

References: 1. Freedman D S, Butte N F, Taveras E M, Lundeen E A, Blanck H M, Goodman A B, Ogden C L. BMI z-Scores are a poor indicator of adiposity among 2- to 19-year-olds with very high BMIs, NHANES 1999–2000 to 2013–2014. Obesity (Silver Spring). 2017 Apr;25(4):739–746.2. Freedman D S, Goodwin Davies A J, Kompaniyets L, Lange S J, Goodman A B, Tam Phan T, Cole F S, Dempsey A, Pajor N, Eneli I, Christakis D A, Forrest C B. A Longitudinal Comparison of Alternatives to Body Mass Index Z-Scores for Children with Very High Body Mass Indexes. J Pediatr. 2021 Mar 5;S0022-3476(21)00211-0.

Article tools

My recent searches

No recent searches.