Identifying Overweight and Obesity among Finnish School-aged Children: Comparing Waist Circumference and Waist-to-Height Ratio with Body Mass Index
Slayter, Chantelle M. H. (2020)
2020
Master's Degree Programme in Public and Global Health
Yhteiskuntatieteiden tiedekunta - Faculty of Social Sciences
This publication is copyrighted. Only for Your own personal use. Commercial use is prohibited.
Hyväksymispäivämäärä
2020-05-20
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:tuni-202004294518
https://urn.fi/URN:NBN:fi:tuni-202004294518
Tiivistelmä
Background The rising prevalence of childhood overweight and obesity, both in Finland and abroad, calls for continued research to determine an optimal method for detecting it. Current methods, such as body mass index (BMI), only measures general obesity, yet central obesity is more attributable to obesity-health risks. Waist circumference (WC) and waist-to-height ratio (WHtR) measure central obesity and can be better measures of childhood overweight and obesity.
Aim of Study The aim of the study is to look at the WC, WHtR, and BMI distributions among Finnish school-aged children, exploring the agreement between WC and WHtR against BMI in terms of detecting childhood overweight and obesity using different internal and external criteria.
Methodology The sample consisted of 896 children 7, 11, and 14 years old, with data used from the LATE study, a cross-sectional study in Finland. Descriptive statistics and bivariate correlation were used to explore distributions. Selected cut-offs, using the Youden Index, from receiver operating characteristic (ROC) curve analysis and cross-tabulation was utilized for comparing the agreement between BMI, WC, and WHtR.
Results Means and percentile tables for BMI were comparable to international obesity task force (IOTF) BMI and Finnish BMI references. In the sample, WC percentiles differ by sex and age. WC, WHtR, and BMI were all strongly correlated between each other across all ages and both sexes (p>0.001). The optimal cut-offs calculated from the ROC analysis yielded sensitivity and specificity values of 85% to 100%. Cross-tabulation of internal WC 85th and 95th percentiles yielded sensitivity 50% to 89% and specificity 94% to 100%, external WC 85th and 95th percentile thresholds yielded sensitivity 83% to 100% and specificity 80% to 92%. A 0.5 WHtR cut-off had sensitivity values from 49% to 80%, and specificity values from 96% to 100%. Of the children who were identified as having overweight or obesity by WC and WHtR, up to 70% of them have normal BMI.
Conclusion Overall, the sensitivity and specificity of the recommended WC optimal cut-offs from the ROC curve analysis were highest as compared to all other internal and external thresholds in the study. Further, BMI missed a high proportion of children who would be at risk for obesity-related health risks by central obesity measures. WC and WHtR, measures of central obesity, are viable alternatives to BMI to accurately identify childhood overweight and obesity. However, the recommended cut-offs from the ROC analysis still are not optimal thresholds, despite having high sensitivity and specificity. More research is needed to determine the optimal age- and sex- specific cut-off values to accurately identify childhood overweight and obesity based on central obesity.
Aim of Study The aim of the study is to look at the WC, WHtR, and BMI distributions among Finnish school-aged children, exploring the agreement between WC and WHtR against BMI in terms of detecting childhood overweight and obesity using different internal and external criteria.
Methodology The sample consisted of 896 children 7, 11, and 14 years old, with data used from the LATE study, a cross-sectional study in Finland. Descriptive statistics and bivariate correlation were used to explore distributions. Selected cut-offs, using the Youden Index, from receiver operating characteristic (ROC) curve analysis and cross-tabulation was utilized for comparing the agreement between BMI, WC, and WHtR.
Results Means and percentile tables for BMI were comparable to international obesity task force (IOTF) BMI and Finnish BMI references. In the sample, WC percentiles differ by sex and age. WC, WHtR, and BMI were all strongly correlated between each other across all ages and both sexes (p>0.001). The optimal cut-offs calculated from the ROC analysis yielded sensitivity and specificity values of 85% to 100%. Cross-tabulation of internal WC 85th and 95th percentiles yielded sensitivity 50% to 89% and specificity 94% to 100%, external WC 85th and 95th percentile thresholds yielded sensitivity 83% to 100% and specificity 80% to 92%. A 0.5 WHtR cut-off had sensitivity values from 49% to 80%, and specificity values from 96% to 100%. Of the children who were identified as having overweight or obesity by WC and WHtR, up to 70% of them have normal BMI.
Conclusion Overall, the sensitivity and specificity of the recommended WC optimal cut-offs from the ROC curve analysis were highest as compared to all other internal and external thresholds in the study. Further, BMI missed a high proportion of children who would be at risk for obesity-related health risks by central obesity measures. WC and WHtR, measures of central obesity, are viable alternatives to BMI to accurately identify childhood overweight and obesity. However, the recommended cut-offs from the ROC analysis still are not optimal thresholds, despite having high sensitivity and specificity. More research is needed to determine the optimal age- and sex- specific cut-off values to accurately identify childhood overweight and obesity based on central obesity.