Can waist circumference provide a new “third” dimension to BMI when predicting percentage body fat in children? Insights using allometric modelling

Alan M. Nevill, Elizabeth Bryant, Kate Wilkinson, Thayse Natacha Gomes, Raquel Chaves, Sara Pereira, Peter T. Katzmarzyk, José Maia, Michael J. Duncan

Research output: Contribution to journalJournal Article

Abstract

Introduction: Body mass index (BMI) is often criticized for not being able to distinguish between lean and fat tissue. Waist circumference (WC), adjusted for stature, is proposed as an alternative weight status index, as it is more sensitive to changes in central adiposity. Purpose: The purpose of the study is to combine the three dimensions of height, mass, and WC to provide a simple, meaningful, and more accurate index associated with percentage body fat (BF%). Methods: We employed a four independent sample design. Sample 1 consisted of 551 children (320 boys) (mean ± SD of age = 7.2 ± 2.0 years), recruited from London, UK. Samples 2, 3, and 4 consisted of 5387 children (2649 boys) aged 7 to 17 years recruited from schools in Portugal. Allometric modelling was used to identify the most effective anthropometric index associated with BF%. The data from samples 2, 3, and 4 were used to confirm and cross-validate the model derived in sample 1. Results: The allometric models from all four samples identified a positive mass exponent and a negative height exponent that was approximately twice that of the mass exponent and a waist circumference exponent that was approximately half the mass exponent. Consequently, the body shape index most strongly associated with BF% was BMI√WC. The √WC component of the new index can simply be interpreted as a WC “weighting” of the traditional BMI. Conclusions: Compared with using BMI and WC in isolation, BMI√WC could provide a more effective and equally noninvasive proxy for BF% in children that can be used in public and community health settings.
Original languageEnglish
JournalPediatric obesity
Volume14
Issue number4
DOIs
Publication statusPublished - 1 Apr 2019
Externally publishedYes

Fingerprint

Waist Circumference
Adipose Tissue
Body Mass Index
Portugal
Adiposity
Proxy
Public Health
Fats
Weights and Measures

Keywords

  • adiposity
  • bioelectrical impedance
  • multiplicative model
  • obesity

Cite this

Nevill, Alan M. ; Bryant, Elizabeth ; Wilkinson, Kate ; Gomes, Thayse Natacha ; Chaves, Raquel ; Pereira, Sara ; Katzmarzyk, Peter T. ; Maia, José ; Duncan, Michael J. / Can waist circumference provide a new “third” dimension to BMI when predicting percentage body fat in children? Insights using allometric modelling. In: Pediatric obesity. 2019 ; Vol. 14, No. 4.
@article{702e1f6d2650463981b723f27784ec8e,
title = "Can waist circumference provide a new “third” dimension to BMI when predicting percentage body fat in children? Insights using allometric modelling",
abstract = "Introduction: Body mass index (BMI) is often criticized for not being able to distinguish between lean and fat tissue. Waist circumference (WC), adjusted for stature, is proposed as an alternative weight status index, as it is more sensitive to changes in central adiposity. Purpose: The purpose of the study is to combine the three dimensions of height, mass, and WC to provide a simple, meaningful, and more accurate index associated with percentage body fat (BF{\%}). Methods: We employed a four independent sample design. Sample 1 consisted of 551 children (320 boys) (mean ± SD of age = 7.2 ± 2.0 years), recruited from London, UK. Samples 2, 3, and 4 consisted of 5387 children (2649 boys) aged 7 to 17 years recruited from schools in Portugal. Allometric modelling was used to identify the most effective anthropometric index associated with BF{\%}. The data from samples 2, 3, and 4 were used to confirm and cross-validate the model derived in sample 1. Results: The allometric models from all four samples identified a positive mass exponent and a negative height exponent that was approximately twice that of the mass exponent and a waist circumference exponent that was approximately half the mass exponent. Consequently, the body shape index most strongly associated with BF{\%} was BMI√WC. The √WC component of the new index can simply be interpreted as a WC “weighting” of the traditional BMI. Conclusions: Compared with using BMI and WC in isolation, BMI√WC could provide a more effective and equally noninvasive proxy for BF{\%} in children that can be used in public and community health settings.",
keywords = "adiposity, bioelectrical impedance, multiplicative model, obesity",
author = "Nevill, {Alan M.} and Elizabeth Bryant and Kate Wilkinson and Gomes, {Thayse Natacha} and Raquel Chaves and Sara Pereira and Katzmarzyk, {Peter T.} and Jos{\'e} Maia and Duncan, {Michael J.}",
year = "2019",
month = "4",
day = "1",
doi = "10.1111/ijpo.12491",
language = "English",
volume = "14",
journal = "Pediatric obesity",
issn = "2047-6302",
publisher = "Wiley-Blackwell for the International Association for the Study of Obesity",
number = "4",

}

Can waist circumference provide a new “third” dimension to BMI when predicting percentage body fat in children? Insights using allometric modelling. / Nevill, Alan M.; Bryant, Elizabeth; Wilkinson, Kate; Gomes, Thayse Natacha; Chaves, Raquel; Pereira, Sara; Katzmarzyk, Peter T.; Maia, José; Duncan, Michael J.

In: Pediatric obesity, Vol. 14, No. 4, 01.04.2019.

Research output: Contribution to journalJournal Article

TY - JOUR

T1 - Can waist circumference provide a new “third” dimension to BMI when predicting percentage body fat in children? Insights using allometric modelling

AU - Nevill, Alan M.

AU - Bryant, Elizabeth

AU - Wilkinson, Kate

AU - Gomes, Thayse Natacha

AU - Chaves, Raquel

AU - Pereira, Sara

AU - Katzmarzyk, Peter T.

AU - Maia, José

AU - Duncan, Michael J.

PY - 2019/4/1

Y1 - 2019/4/1

N2 - Introduction: Body mass index (BMI) is often criticized for not being able to distinguish between lean and fat tissue. Waist circumference (WC), adjusted for stature, is proposed as an alternative weight status index, as it is more sensitive to changes in central adiposity. Purpose: The purpose of the study is to combine the three dimensions of height, mass, and WC to provide a simple, meaningful, and more accurate index associated with percentage body fat (BF%). Methods: We employed a four independent sample design. Sample 1 consisted of 551 children (320 boys) (mean ± SD of age = 7.2 ± 2.0 years), recruited from London, UK. Samples 2, 3, and 4 consisted of 5387 children (2649 boys) aged 7 to 17 years recruited from schools in Portugal. Allometric modelling was used to identify the most effective anthropometric index associated with BF%. The data from samples 2, 3, and 4 were used to confirm and cross-validate the model derived in sample 1. Results: The allometric models from all four samples identified a positive mass exponent and a negative height exponent that was approximately twice that of the mass exponent and a waist circumference exponent that was approximately half the mass exponent. Consequently, the body shape index most strongly associated with BF% was BMI√WC. The √WC component of the new index can simply be interpreted as a WC “weighting” of the traditional BMI. Conclusions: Compared with using BMI and WC in isolation, BMI√WC could provide a more effective and equally noninvasive proxy for BF% in children that can be used in public and community health settings.

AB - Introduction: Body mass index (BMI) is often criticized for not being able to distinguish between lean and fat tissue. Waist circumference (WC), adjusted for stature, is proposed as an alternative weight status index, as it is more sensitive to changes in central adiposity. Purpose: The purpose of the study is to combine the three dimensions of height, mass, and WC to provide a simple, meaningful, and more accurate index associated with percentage body fat (BF%). Methods: We employed a four independent sample design. Sample 1 consisted of 551 children (320 boys) (mean ± SD of age = 7.2 ± 2.0 years), recruited from London, UK. Samples 2, 3, and 4 consisted of 5387 children (2649 boys) aged 7 to 17 years recruited from schools in Portugal. Allometric modelling was used to identify the most effective anthropometric index associated with BF%. The data from samples 2, 3, and 4 were used to confirm and cross-validate the model derived in sample 1. Results: The allometric models from all four samples identified a positive mass exponent and a negative height exponent that was approximately twice that of the mass exponent and a waist circumference exponent that was approximately half the mass exponent. Consequently, the body shape index most strongly associated with BF% was BMI√WC. The √WC component of the new index can simply be interpreted as a WC “weighting” of the traditional BMI. Conclusions: Compared with using BMI and WC in isolation, BMI√WC could provide a more effective and equally noninvasive proxy for BF% in children that can be used in public and community health settings.

KW - adiposity

KW - bioelectrical impedance

KW - multiplicative model

KW - obesity

U2 - 10.1111/ijpo.12491

DO - 10.1111/ijpo.12491

M3 - Journal Article

VL - 14

JO - Pediatric obesity

JF - Pediatric obesity

SN - 2047-6302

IS - 4

ER -