Diagnostic performance of Body Mass Index, Waist Circumference and the Waist-to-Height Ratio for identifying cardiometabolic risk in Scottish pre-adolescents

Duncan Buchan, Gillian McLellan, Samantha Donnelly, Rosemary Arthur

Research output: Contribution to journalArticle

2 Citations (Scopus)
116 Downloads (Pure)


Background: Limited studies have examined the diagnostic performance of body mass index (BMI), waist circumference (WC) or waist to height ratio (WHtR) for identifying cardiometabolic risk (increased clustered glucose, triglycerides, mean arterial pressure and inv-HDL-cholesterol) in pre-adolescent youth. Aim: To compare the utility of BMI, WC and WHtR as predictors of cardiometabolic risk (CMR) in Scottish pre-adolescent children. Subjects and Methods: A cross-sectional analysis of 223 Scottish children (55.2% boys, mean age 8.4 years) was undertaken. BMI, WC and WHtR were used as exposure variables within multivariate logistic regression analysis and ROC analysis to examine the utility of these anthropometrical indices in identifying those at cardiometabolic risk. Results: Individuals with an elevated WHtR, WC and BMI were 3.51 (95% CI: 1.71-7.23; P < 0.001); 2.34 (95% CI: 1.35-4.06; P= 0.002) and 2.59 (95% CI: 1.42-4.73; P = 0.002) times more likely to be at cardiometabolic risk. The areas under the curves [AUC] to identify children with cardiometabolic risk were significant and similar among anthropometric indices (AUC’s = 0.60 - 0.65). When stratified by BMI, both WC and WHtR demonstrated a fair to good ability for identifying those at cardiometabolic risk (AUC = 0.75 - 0.81). Conclusions: Findings suggest that the combination of BMI with either WC or WHtR may provide an added benefit in the assessment of cardiometabolic risk amongst pre-adolescents.
Original languageEnglish
Pages (from-to)297-302
JournalAnnals of Human Biology
Issue number4
Early online date6 Nov 2016
Publication statusE-pub ahead of print - 6 Nov 2016



  • Weight status
  • Youth
  • Obesity
  • Cardiometabolic risk

Cite this