Abstract
Context
No universal waist circumference (WC) percentile cutoffs used have been proposed for screening central obesity in children and adolescents.
Objective
To develop international WC percentile cutoffs for children and adolescents with normal weight based on data from 8 countries in different global regions and to examine the relation with cardiovascular risk.
Design and Setting
We used pooled data on WC in 113,453 children and adolescents (males 50.2%) aged 4 to 20 years from 8 countries in different regions (Bulgaria, China, Iran, Korea, Malaysia, Poland, Seychelles, and Switzerland). We calculated WC percentile cutoffs in samples including or excluding children with obesity, overweight, or underweight. WC percentiles were generated using the general additive model for location, scale, and shape (GAMLSS). We also estimated the predictive power of the WC 90th percentile cutoffs to predict cardiovascular risk using receiver operator characteristics curve analysis based on data from 3 countries that had available data (China, Iran, and Korea). We also examined which WC percentiles linked with WC cutoffs for central obesity in adults (at age of 18 years).
Main Outcome Measure
WC measured based on recommendation by the World Health Organization.
Results
We validated the performance of the age- and sex-specific 90th percentile WC cutoffs calculated in children and adolescents (6-18 years of age) with normal weight (excluding youth with obesity, overweight, or underweight) by linking the percentile with cardiovascular risk (area under the curve [AUC]: 0.69 for boys; 0.63 for girls). In addition, WC percentile among normal weight children linked relatively well with established WC cutoffs for central obesity in adults (eg, AUC in US adolescents: 0.71 for boys; 0.68 for girls).
Conclusion
The international WC cutoffs developed in this study could be useful to screen central obesity in children and adolescents aged 6 to 18 years and allow direct comparison of WC distributions between populations and over time.
Keywords: waist circumference, central obesity, cutoff points, children and adolescents
Obesity in children and adolescents has become a major global health challenge because the prevalence is increasing (1,2) and excess weight is associated with increased risk of morbidity and mortality, including hypertension, dislipidemia, type 2 diabetes, cardiovascular disease (CVD), and overall mortality (3). Body mass index (BMI), calculated as weight divided by height squared (kg/m2), is frequently used to classify overweight and obesity in epidemiological studies and in clinical practice. International age- and sex-specific BMI percentile cutoffs to define child overweight and obesity have been developed by Cole et al on behalf of the International Obesity Task Force (IOTF) (4) and are intended to correspond, at the age of 18 years, to the criteria in adults for overweight (BMI ≥ 25 kg/m2) and obesity (BMI ≥ 30 kg/m2). However, BMI cannot distinguish accurately between fat and fat-free mass (5). Waist circumference (WC), another adiposity measure indicating abdominal fat accumulation, may better predict risk of CVD and mortality (6, 7). Furthermore, WC is the essential component of metabolic syndrome definition endorsed by the International Diabetes Federation (IDF) (8, 9), the National Heart, Lung, and Blood Institute (10, 11) and other medical organizations (12).
Cutoffs of WC to define central obesity in adults have been proposed by the IDF in different populations (≥ 94/80 cm in male/female European, African, and Middle Eastern adults; ≥ 90/80 cm in male/female Asian adults) (13, 14). However, WC cutoffs to define central obesity in children and adolescents based on multiple populations in different regions have not yet been proposed (15), although several country-specific pediatric WC percentile cutoffs have been developed (16–25). Thus, it is important to develop international WC cutoffs that could be used as a common yardstick to enable direct comparison between populations and for monitoring central adiposity over time.
This study aims to develop international WC cutoffs for defining central obesity in children and adolescents aged 6 to 18 years, based on pooled data from 8 countries from several regions: Bulgaria (16), China (26), Iran (27), Korea (28), Malaysia (17), Poland (29, 30), Seychelles (31), and Switzerland (18).
Methods
Subjects
Data on WC for children and adolescents aged 4 to 20 years were available from 8 large population-based cross-sectional surveys from Bulgaria, China, Iran, Korea, Malaysia, Poland, Seychelles, and Switzerland (Table 1). While all surveys were representative of their underlying populations, the choice of the countries included in this study was based on the availability of adequate data and country selection can therefore be considered as convenient. In addition, although we aimed to develop international WC references for youth aged 6 to 18 years, we also included data on individuals aged 4 and 5 years, and 19 and 20 years in order to avoid left-edge and right-edge effects when modeling WC curves according to age (32). Details on sampling and methods of these studies have been described elsewhere (16-18, 26-31). Data from China, Korea, and Seychelles included pooled samples collected over several years; data from Malaysia were pooled from 2 national surveys conducted in 2008 and 2009; and other data (from Iran, Poland, and Switzerland) were based on single cross-sectional surveys. A total of 113,453 subjects aged 4 to 20 years from 8 countries were included in this study. Obesity, overweight, and underweight status were defined based on age- and sex-specific BMI percentile values of the IOTF (4). The prevalence of obesity, overweight, and underweight in the whole sample of each survey is shown in Table 2. Written informed consent was obtained from parents and children and adolescents. Each survey was approved by the respective Institutional Ethics Review Board in each country.
Table 1.
Description of Surveys Assessing Waist Circumference in Children and Adolescents Aged 4–20 Years From 8 Countries
| Country | Surveyed Year(s) | Description | Total | No. of Males | No. of Females | Age Range (y) | Ethnicity | Reference |
|---|---|---|---|---|---|---|---|---|
| Bulgaria | 2006–2007 | Varna representative growth and obesity survey | 3786 | 2040 | 1746 | 6–18 | European | (16) |
| China | 2000–2011 | Data pooled from 5 cycles of the China Health and Nutrition Survey | 8811 | 4693 | 4118 | 4–20 | Asian | (26) |
| Iran | 2011–2012 | National survey: Childhood and Adolescence Surveillance and Prevention of Adult Non- communicable Diseases | 18557 | 9373 | 9184 | 6–18 | Middle East | (27) |
| Korea | 2001–2013 | Data pooled from 5 cycles of the Korea National Health and Nutrition Examination Survey | 16547 | 8573 | 7974 | 4–20 | Asian | (28) |
| Malaysia | 2008–2009 | National surveys of the Nutritional Status of Primary and Secondary School Children in Malaysia | 16026 | 7972 | 8054 | 6–16 | Asian | (17) |
| Poland | 2007–2010 | National survey: Elaboration of Reference Blood Pressure Ranges for Children and Adolescents in Poland | 20495 | 9822 | 10673 | 4–18 | European | (29,30) |
| Seychelles | 2000–2014 | School-based National Surveillance Programs | 26940 | 13340 | 13600 | 4–18 | Black | |
| Switzerland | 2007 | National survey: Prevalence of Overweight and Obesity in 6–13 Year Old Children | 2291 | 1123 | 1168 | 6–13 | European | (18) |
| Total | 113453 | 56936 | 56517 | 4–20 |
Table 2.
Prevalence of Obesity, Overweight, and Underweight Based on the IOTF BMI Criteria in Children and Adolescents From 8 Countries
| Country | Survey Year(s) | Age Range | N | Obesity, n (%) | Overweight, n (%) | Underweight, n (%) |
|---|---|---|---|---|---|---|
| Bulgaria | 2006–2007 | 6–18 | 3786 | 306 (8.1) | 753 (19.9) | 172 (4.5) |
| China | 2000–2011 | 4–20 | 8811 | 156 (1.8) | 643 (7.3) | 1905 (21.6) |
| Iran | 2011–2012 | 6–18 | 18557 | 739 (4.0) | 2520 (13.6) | 4060 (21.9) |
| Korea | 2001–2013 | 4–20 | 16547 | 664 (4.0) | 2808 (17.0) | 1600 (9.7) |
| Malaysia | 2008–2009 | 6–16 | 16026 | 1227 (7.7) | 2163 (13.5) | 3985 (24.9) |
| Poland | 2007–2010 | 4–18 | 20495 | 643 (3.1) | 2570 (12.5) | 2556 (12.5) |
| Seychelles | 2000–2014 | 4–18 | 26940 | 1646 (6.1) | 3021 (11.2) | 5989 (22.2) |
| Switzerland | 2007 | 6–13 | 2291 | 41 (1.8) | 258 (11.3) | 187 (8.2) |
| Total | – | 4–20 | 113453 | 5422 (4.8) | 14736 (13.0) | 20454 (18.0) |
WC measurement
In all countries, WC was measured midway between the lowest rib and the superior border of the iliac crest at the end of a normal expiration with a flexible nonelastic anthropometric tape, to the nearest 0.1 cm, as recommended by the World Health Organization (33). The mean value of 2 measurements was used for data analysis. Quality control was performed in all the surveys including adequate training for the survey officers.
Design outline of this study
We pooled WC data from the 8 countries, consistent with the recommendation to include 5 to 10 sites when attempting to develop an international standard (34). Because different prevalence of obesity, overweight, or underweight in different populations may shift upward or downward the values of the country specific WC cutoffs (35) at the population level, we calculated age- and sex-specific 90th percentiles of WC in 4 different samples, including or excluding children with obesity, overweight, or underweight (4, 36) (Fig. 1). Sample 1 (N = 113,453) included the whole sample; Sample 2 (N = 108,031) excluded obese individuals based on the IOTF BMI cutoffs; Sample 3 (N = 93,295) excluded obese and overweight individuals; and Sample 4 (N = 72,841) excluded obese, overweight, and underweight individuals. We then compared the performance of the age- and sex-specific 90th percentile values of WC (ie, a sex- and age-specific binary WC variable defined as “high WC or nonhigh WC”) to predict cardiovascular (CV) risk in the 4 samples using receiver operator characteristics (ROC) curve analysis. This analysis was done in pooled data from the 3 populations (China, Iran, and Korea) in which data on CV risk factors were available. Based on the strength of the association between WC and CV risk, we determined which sample (ie, including or excluding children with obesity, overweight, or underweight) was the best base population to calculate WC 90th percentile cutoffs.
Figure 1.
Flow chart of the study design and analysis.
Finally, we generated age- and sex-specific trajectories of percentile values throughout childhood to the age of 18 years and examined how well these calculated trajectories linked, at the age of 18 years, with established adult WC cutoffs for central obesity as recommended by the IDF (ie, ≥ 94/80 cm for the male/female European, African, and Middle Eastern populations; ≥ 90/80 cm in the male/female Asian population). Previous studies had suggested differences in WC levels across different racial/ethnic populations in adults (13, 14).
Statistical analysis
Percentile curves
The data were pooled using weights according to the population size of each survey from the 8 countries (37). The generalized additive model for location, scale, and shape (GAMLSS) model for Box-Cox power exponential distribution with cubic spline smoothing (38) was used to construct smoothed age- and sex-specific 90th percentile curves of WC for each of the 4 samples (ie, including or excluding children with obesity, overweight, or underweight). In addition, we also calculated the age- and sex-specific WC percentile values in children and adolescents that linked, at the age of 18 years, with accepted adult WC cutoffs similarly to the method used by Cole et al to establish age- and sex-specific BMI percentile values for children and adolescents (4). The Box-Cox power exponential distribution has 4 parameters including μ, σ, ν, and τ which represent location (median), scale (approximate coefficient of variation), skewness (power transformation to symmetry), and kurtosis (degrees of freedom or power exponential parameter), respectively. The GAMLSS models were adjusted for skewness and kurtosis of WC from mixed data sources. Calculation of sex- and age-specific WC curves were based on data in children and adolescents aged 4 to 20 years to avoid an edge effect at low and high age values, but data in this paper are presented only for age of 6 to 18 years (32). Analyses were performed using the GAMLSS 4.3-1 library running under R 3.1.2 (39). Goodness-of-fit of the models was assessed by the Bayesian Information Criterion and by Q-Q plots (40).
WC and CV risk
Three populations (China: 750 children aged 7–17, 2009; Iran: 8393 children aged 6–17, 2003 and 2009; and Korea: 8688 children aged 10–17, 1998–2013) had data on CV risk and they were used to assess the performance of age- and sex-specific 90th percentile cutoffs of WC (ie, a sex- and age-specific binary WC variable) for predicting CV risk, and analysis was done in the 4 samples (ie, samples including or excluding obesity, overweight, or underweight).
In these 3 countries, measurements were available for systolic blood pressure (SBP), diastolic blood pressure (DBP), total cholesterol (TC), triglycerides (TG), high density lipoprotein cholesterol (HDL-C), low density lipoprotein cholesterol (LDL-C), and fasting glucose, as described elsewhere (41–44). High blood pressure (BP) was defined as SBP and/or DBP equal to or above age-, sex-, and height-specific 90th percentile of an international child BP reference (45). High TG (≥ 150mg/dL), low HDL-C (< 40mg/dL), and high fasting glucose (≥ 100mg/dL) were defined using cutoffs recommended by the IDF (46) and high TC (≥ 200mg/dL) and high LDL-C (≥ 130mg/dL) were defined based on the National Cholesterol Education Program (NCEP) ATP III (11). The prevalence of these 6 CV risk factors in the 3 populations is presented in Table 3.
Table 3.
Prevalence of Cardiovascular Risk Factors and Their Clustering in Children and Adolescents Aged 6–17 Years in China, Iran, and Korea
| Country | N | Age Range, Years | High BP | High Glucose | High TC | High TG | Low HDL | High LDL | ≥ 3 Risk Factors |
|---|---|---|---|---|---|---|---|---|---|
| China | |||||||||
| Males | 411 | 7–17 | 79 (19.2) | 41 (10.0) | 13 (3.2) | 45 (10.9) | 40 (9.7) | 15 (3.6) | 11 (2.7) |
| Females | 339 | 7–17 | 61 (18.0) | 20 (5.9) | 14 (4.1) | 30 (8.8) | 27 (8.0) | 17 (5.0) | 6 (1.8) |
| Total | 750 | 7–17 | 140 (18.7) | 61 (8.1) | 27 (3.6) | 75 (10.0) | 67 (8.9) | 32 (4.3) | 17 (2.3) |
| Iran | |||||||||
| Males | 4151 | 6–17 | 957 (23.3) | 328 (8.1) | 237 (5.7) | 370 (9.0) | 1399 (36.3) | 225 (6.3) | 193 (4.6) |
| Females | 4242 | 6–17 | 836 (20.0) | 408 (9.9) | 257 (6.1) | 400 (9.5) | 1456 (36.6) | 272 (7.5) | 210 (5.0) |
| Total | 8393 | 6–17 | 1793 (21.6) | 736 (9.0) | 494 (5.9) | 770 (9.3) | 2855 (36.5) | 497 (6.9) | 403 (4.8) |
| Korea | |||||||||
| Males | 4589 | 10–17 | 1057 (23.0) | 637 (14.0) | 269 (5.9) | 443 (9.7) | 899 (19.6) | 194 (4.5) | 253 (5.5) |
| Females | 4099 | 10–17 | 618 (15.1) | 497 (12.2) | 357 (8.7) | 413 (10.1) | 539 (13.2) | 250 (6.5) | 185 (4.5) |
| Total | 8688 | 10–17 | 1675 (19.3) | 1134 (13.1) | 626 (7.2) | 856 (9.9) | 1438 (16.6) | 444 (5.4) | 438 (5.0) |
Data are presented as n (%). Abbreviations: BP, blood pressure; HDL, high density lipoprotein; LDL, low density lipoprotein; TC, total cholesterol; TG, triglycerides.
In this study, we defined elevated CVD risk as having ≥ 3 out of 6 CV risk factors (ie, high BP, high TG, low HDL, high TC, high LDL, and high glucose) when testing sensitivity, specificity, and area under the curve (AUC) of the age- and sex-specific WC 90th percentile cutoffs to predict CV risk, consistent with previous studies (20, 47). Analysis was done in the 4 samples (ie, including or excluding children with obesity, overweight, or underweight).
In addition, the performance of several WC percentiles linking with adult WC cutoffs for specific racial/ethnic populations were also assessed for predicting CV risk (ie, the presence of ≥ 3 risk factors) in the pooled data (age 6-17 years) in the 3 populations (China, Iran, and Korea) and in 1 separate US adolescent population (age 12-17 years) with different racial/ethnic groups (data from the National Health and Nutrition Examination Survey, 1999-2014) (48). ROC curve analysis was performed using SAS v9.4 (SAS Institute, Cary, North Carolina).
Comparison of our international WC percentiles with European WC percentiles
Finally, the age- and sex-specific 75th and 90th percentiles of WC from the sample of children and adolescents with normal weight (ie, after exclusion of children with obesity, overweight, or underweight) were compared with the corresponding percentiles in the IDEFICS study (35), which presented the first WC percentiles based on normal-weight children aged 2 to 10 years from a sample size of 12,381 children from 8 European countries.
Results
Fig. 2 displays the age- and sex-specific 90th percentile values of WC from the 4 samples of children and adolescents aged 6 to 18 years based on pooled data from the 8 countries. In both males and females, the 90th percentile values of WC in Sample 1 (the whole sample) were, as expected, higher than those in Sample 2 (excluding individuals with obesity), and substantially higher than those in Sample 3 (excluding individuals with obesity or overweight) and Sample 4 (excluding individuals with obesity, overweight or underweight). However, the 90th percentile values in Sample 3 and Sample 4 were similar.
Figure 2.
Comparisons of the 90th percentile (P90) curves of waist circumference (WC) by age and sex among samples excluding children and adolescents with obesity, overweight, or underweight, based on pooled data from 8 countries. Notes: Sample 1 indicates the P90 of WC based on the whole population; Sample 2 indicates the P90 of WC based on the population excluding obese children; Sample 3 indicates the P90 of WC based on the population excluding obese and overweight children; Sample 4 indicates the P90 of WC based on the population excluding obese, overweight, and underweight children.
Tables 4 –7 present the smoothed 50th, 75th, 80th, 85th, 90th, 95th, 97th, and 99th percentiles of WC by age and sex for children and adolescents aged 6 to 18 years from the 4 samples (including or excluding individuals with obesity, overweight, or underweight) based on pooled data from the 8 countries.
Table 4.
Smoothed Percentile Values of WC (cm) by Age and Sex From Sample 1 (based on the whole population)
| Males | Females | |||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Age (years) | mu | sigma | nu | tau | P50 | P75 | P80 | P85 | P90 | P95 | P97 | P99 | mu | sigma | nu | tau | P50 | P75 | P80 | P85 | P90 | P95 | P97 | P99 |
| 6 | 54.048 | 0.094 | -1.549 | 1.523 | 54.0 | 57.4 | 58.4 | 59.7 | 61.5 | 64.5 | 66.9 | 72.2 | 53.058 | 0.099 | -1.057 | 1.521 | 53.1 | 56.5 | 57.5 | 58.7 | 60.5 | 63.5 | 65.7 | 70.5 |
| 7 | 55.748 | 0.104 | -1.652 | 1.637 | 55.7 | 59.8 | 61.0 | 62.5 | 64.6 | 68.3 | 71.1 | 77.7 | 54.578 | 0.109 | -1.123 | 1.666 | 54.6 | 58.6 | 59.8 | 61.3 | 63.3 | 66.7 | 69.2 | 74.7 |
| 8 | 57.669 | 0.115 | -1.723 | 1.768 | 57.7 | 62.4 | 63.9 | 65.7 | 68.2 | 72.5 | 75.9 | 83.8 | 56.423 | 0.118 | -1.204 | 1.800 | 56.4 | 61.2 | 62.5 | 64.2 | 66.5 | 70.4 | 73.3 | 79.6 |
| 9 | 59.753 | 0.125 | -1.685 | 1.888 | 59.8 | 65.3 | 66.9 | 69.0 | 71.9 | 76.9 | 80.8 | 89.7 | 58.476 | 0.126 | -1.240 | 1.911 | 58.5 | 63.9 | 65.4 | 67.3 | 69.9 | 74.3 | 77.5 | 84.5 |
| 10 | 61.890 | 0.134 | -1.534 | 1.962 | 61.9 | 68.1 | 70.0 | 72.3 | 75.4 | 80.9 | 85.1 | 94.6 | 60.588 | 0.130 | -1.265 | 2.001 | 60.6 | 66.5 | 68.2 | 70.3 | 73.1 | 77.8 | 81.2 | 88.8 |
| 11 | 64.045 | 0.139 | -1.388 | 1.950 | 64.0 | 70.7 | 72.7 | 75.1 | 78.5 | 84.3 | 88.6 | 98.5 | 62.718 | 0.131 | -1.374 | 2.047 | 62.7 | 69.0 | 70.7 | 72.9 | 75.9 | 80.9 | 84.6 | 92.8 |
| 12 | 66.159 | 0.139 | -1.310 | 1.861 | 66.2 | 72.9 | 74.9 | 77.4 | 80.9 | 86.8 | 91.3 | 101.7 | 64.706 | 0.128 | -1.528 | 2.042 | 64.7 | 71.0 | 72.9 | 75.1 | 78.2 | 83.4 | 87.3 | 96.1 |
| 13 | 68.170 | 0.135 | -1.271 | 1.729 | 68.2 | 74.7 | 76.7 | 79.2 | 82.6 | 88.7 | 93.2 | 103.9 | 66.370 | 0.124 | -1.645 | 2.000 | 66.4 | 72.6 | 74.4 | 76.6 | 79.7 | 85.0 | 88.9 | 97.9 |
| 14 | 70.040 | 0.130 | -1.259 | 1.611 | 70.0 | 76.3 | 78.2 | 80.6 | 84.0 | 90.0 | 94.6 | 105.3 | 67.695 | 0.119 | -1.748 | 1.941 | 67.7 | 73.7 | 75.5 | 77.7 | 80.7 | 85.9 | 89.9 | 99.0 |
| 15 | 71.682 | 0.125 | -1.166 | 1.513 | 71.7 | 77.6 | 79.4 | 81.7 | 85.0 | 90.8 | 95.2 | 105.5 | 68.641 | 0.114 | -1.814 | 1.879 | 68.6 | 74.4 | 76.1 | 78.2 | 81.2 | 86.3 | 90.2 | 99.2 |
| 16 | 73.051 | 0.121 | -0.975 | 1.462 | 73.1 | 78.8 | 80.5 | 82.7 | 85.9 | 91.3 | 95.4 | 104.9 | 69.238 | 0.109 | -1.823 | 1.824 | 69.2 | 74.8 | 76.4 | 78.4 | 81.2 | 86.1 | 89.9 | 98.5 |
| 17 | 74.193 | 0.119 | -0.802 | 1.482 | 74.2 | 79.9 | 81.6 | 83.7 | 86.7 | 91.9 | 95.7 | 104.3 | 69.638 | 0.106 | -1.793 | 1.800 | 69.6 | 75.0 | 76.5 | 78.5 | 81.2 | 85.9 | 89.4 | 97.5 |
| 18 | 75.176 | 0.118 | -0.725 | 1.568 | 75.2 | 81.0 | 82.7 | 84.9 | 87.8 | 92.7 | 96.4 | 104.2 | 69.925 | 0.104 | -1.715 | 1.812 | 69.9 | 75.2 | 76.7 | 78.6 | 81.2 | 85.7 | 89.0 | 96.5 |
Table 7.
Smoothed Percentile Values of WC (cm) by Age and Sex From the Sample 4 (based on the population excluding obese, overweight, and underweight children)
| Males | Females | |||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Age (years) | mu | sigma | nu | tau | P50 | P75 | P80 | P85 | P90 | P95 | P97 | P99 | mu | sigma | nu | tau | P50 | P75 | P80 | P85 | P90 | P95 | P97 | P99 |
| 6 | 53.875 | 0.069 | 0.106 | 1.561 | 53.9 | 56.2 | 56.9 | 57.7 | 58.7 | 60.4 | 61.5 | 63.8 | 52.921 | 0.073 | 0.182 | 1.565 | 52.9 | 55.4 | 56.0 | 56.9 | 57.9 | 59.7 | 60.9 | 63.3 |
| 7 | 55.349 | 0.073 | -0.155 | 1.613 | 55.3 | 58.0 | 58.7 | 59.6 | 60.7 | 62.6 | 63.8 | 66.5 | 54.419 | 0.078 | 0.096 | 1.648 | 54.4 | 57.2 | 57.9 | 58.8 | 60.0 | 61.9 | 63.2 | 65.8 |
| 8 | 56.992 | 0.078 | -0.376 | 1.684 | 57.0 | 59.9 | 60.7 | 61.7 | 62.9 | 65.0 | 66.4 | 69.3 | 56.162 | 0.083 | -0.085 | 1.709 | 56.2 | 59.2 | 60.0 | 61.0 | 62.3 | 64.4 | 65.8 | 68.7 |
| 9 | 58.797 | 0.081 | -0.499 | 1.743 | 58.8 | 62.0 | 62.9 | 63.9 | 65.3 | 67.6 | 69.1 | 72.3 | 58.102 | 0.086 | -0.321 | 1.735 | 58.1 | 61.4 | 62.3 | 63.4 | 64.9 | 67.2 | 68.8 | 72.0 |
| 10 | 60.730 | 0.085 | -0.526 | 1.769 | 60.7 | 64.2 | 65.2 | 66.3 | 67.8 | 70.3 | 71.9 | 75.3 | 60.131 | 0.088 | -0.573 | 1.727 | 60.1 | 63.7 | 64.7 | 65.9 | 67.5 | 70.0 | 71.8 | 75.5 |
| 11 | 62.787 | 0.087 | -0.530 | 1.755 | 62.8 | 66.5 | 67.5 | 68.7 | 70.4 | 73.0 | 74.8 | 78.4 | 62.176 | 0.089 | -0.883 | 1.718 | 62.2 | 65.9 | 67.0 | 68.3 | 70.0 | 72.8 | 74.8 | 78.9 |
| 12 | 64.909 | 0.088 | -0.520 | 1.707 | 64.9 | 68.7 | 69.8 | 71.1 | 72.8 | 75.5 | 77.5 | 81.4 | 64.098 | 0.088 | -1.174 | 1.712 | 64.1 | 68.0 | 69.1 | 70.4 | 72.2 | 75.2 | 77.4 | 82.0 |
| 13 | 67.005 | 0.088 | -0.495 | 1.643 | 67.0 | 70.9 | 71.9 | 73.3 | 75.0 | 77.9 | 79.9 | 84.0 | 65.762 | 0.087 | -1.378 | 1.719 | 65.8 | 69.7 | 70.8 | 72.2 | 74.1 | 77.2 | 79.4 | 84.2 |
| 14 | 69.013 | 0.086 | -0.429 | 1.574 | 69.0 | 72.8 | 73.9 | 75.3 | 77.0 | 79.9 | 82.0 | 86.3 | 67.150 | 0.085 | -1.533 | 1.745 | 67.1 | 71.1 | 72.2 | 73.6 | 75.5 | 78.6 | 80.9 | 85.8 |
| 15 | 70.836 | 0.085 | -0.251 | 1.507 | 70.8 | 74.6 | 75.7 | 77.0 | 78.8 | 81.7 | 83.8 | 88.1 | 68.237 | 0.083 | -1.568 | 1.761 | 68.2 | 72.2 | 73.3 | 74.7 | 76.5 | 79.6 | 81.9 | 86.7 |
| 16 | 72.415 | 0.084 | 0.019 | 1.481 | 72.4 | 76.2 | 77.2 | 78.6 | 80.3 | 83.2 | 85.2 | 89.4 | 69.052 | 0.081 | -1.465 | 1.759 | 69.1 | 72.9 | 74.0 | 75.4 | 77.2 | 80.2 | 82.4 | 86.9 |
| 17 | 73.795 | 0.084 | 0.239 | 1.521 | 73.8 | 77.6 | 78.7 | 80.0 | 81.8 | 84.6 | 86.5 | 90.5 | 69.708 | 0.081 | -1.309 | 1.760 | 69.7 | 73.6 | 74.7 | 76.0 | 77.8 | 80.7 | 82.8 | 87.2 |
| 18 | 75.038 | 0.084 | 0.343 | 1.625 | 75.0 | 79.1 | 80.2 | 81.5 | 83.2 | 86.0 | 87.8 | 91.6 | 70.281 | 0.081 | -1.147 | 1.778 | 70.3 | 74.2 | 75.3 | 76.6 | 78.4 | 81.2 | 83.2 | 87.4 |
Table 5.
Smoothed Percentile Values of WC (cm) by Age and Sex From Sample 2 (based on the population excluding obese children)
| Males | Females | |||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Age (years) | mu | sigma | nu | tau | P50 | P75 | P80 | P85 | P90 | P95 | P97 | P99 | mu | sigma | nu | tau | P50 | P75 | P80 | P85 | P90 | P95 | P97 | P99 |
| 6 | 53.745 | 0.084 | -0.719 | 1.654 | 53.7 | 56.7 | 57.6 | 58.6 | 60.0 | 62.2 | 63.8 | 67.1 | 52.773 | 0.089 | -0.373 | 1.666 | 52.8 | 55.9 | 56.7 | 57.8 | 59.2 | 61.4 | 63.0 | 66.2 |
| 7 | 55.322 | 0.093 | -0.970 | 1.747 | 55.3 | 58.8 | 59.8 | 61.0 | 62.7 | 65.3 | 67.2 | 71.1 | 54.218 | 0.098 | -0.498 | 1.764 | 54.2 | 57.8 | 58.8 | 60.0 | 61.7 | 64.2 | 66.0 | 69.7 |
| 8 | 57.109 | 0.102 | -1.152 | 1.845 | 57.1 | 61.2 | 62.4 | 63.8 | 65.7 | 68.8 | 71.0 | 75.7 | 55.984 | 0.107 | -0.647 | 1.861 | 56.0 | 60.2 | 61.3 | 62.7 | 64.6 | 67.5 | 69.6 | 73.9 |
| 9 | 59.073 | 0.110 | -1.220 | 1.949 | 59.1 | 63.8 | 65.1 | 66.8 | 69.0 | 72.5 | 75.1 | 80.5 | 57.982 | 0.115 | -0.768 | 1.953 | 58.0 | 62.8 | 64.1 | 65.6 | 67.7 | 71.1 | 73.4 | 78.3 |
| 10 | 61.132 | 0.118 | -1.181 | 2.043 | 61.1 | 66.5 | 68.0 | 69.8 | 72.2 | 76.2 | 79.0 | 85.0 | 60.064 | 0.119 | -0.876 | 2.037 | 60.1 | 65.3 | 66.8 | 68.5 | 70.8 | 74.5 | 77.1 | 82.4 |
| 11 | 63.261 | 0.123 | -1.107 | 2.066 | 63.3 | 69.1 | 70.7 | 72.6 | 75.2 | 79.5 | 82.5 | 88.9 | 62.174 | 0.120 | -1.033 | 2.081 | 62.2 | 67.7 | 69.3 | 71.1 | 73.6 | 77.6 | 80.4 | 86.2 |
| 12 | 65.402 | 0.123 | -0.994 | 1.983 | 65.4 | 71.3 | 73.0 | 75.0 | 77.7 | 82.0 | 85.2 | 91.8 | 64.157 | 0.118 | -1.187 | 2.072 | 64.2 | 69.8 | 71.4 | 73.3 | 75.8 | 80.0 | 82.9 | 89.1 |
| 13 | 67.483 | 0.121 | -0.885 | 1.848 | 67.5 | 73.3 | 74.9 | 76.9 | 79.6 | 84.0 | 87.2 | 93.9 | 65.844 | 0.114 | -1.286 | 2.027 | 65.8 | 71.4 | 73.0 | 74.9 | 77.4 | 81.6 | 84.6 | 90.9 |
| 14 | 69.444 | 0.117 | -0.782 | 1.719 | 69.4 | 75.0 | 76.6 | 78.5 | 81.2 | 85.6 | 88.7 | 95.5 | 67.210 | 0.110 | -1.363 | 1.970 | 67.2 | 72.6 | 74.2 | 76.0 | 78.5 | 82.7 | 85.7 | 92.1 |
| 15 | 71.188 | 0.113 | -0.610 | 1.614 | 71.2 | 76.5 | 78.0 | 79.9 | 82.5 | 86.8 | 89.9 | 96.4 | 68.220 | 0.106 | -1.405 | 1.916 | 68.2 | 73.5 | 74.9 | 76.8 | 79.2 | 83.3 | 86.2 | 92.6 |
| 16 | 72.666 | 0.111 | -0.367 | 1.566 | 72.7 | 77.9 | 79.4 | 81.2 | 83.7 | 87.8 | 90.7 | 96.9 | 68.903 | 0.102 | -1.411 | 1.871 | 68.9 | 74.0 | 75.4 | 77.2 | 79.6 | 83.5 | 86.4 | 92.6 |
| 17 | 73.919 | 0.109 | -0.168 | 1.597 | 73.9 | 79.2 | 80.6 | 82.4 | 84.9 | 88.8 | 91.6 | 97.3 | 69.395 | 0.100 | -1.401 | 1.857 | 69.4 | 74.4 | 75.8 | 77.5 | 79.8 | 83.7 | 86.5 | 92.6 |
| 18 | 75.013 | 0.109 | -0.069 | 1.702 | 75.0 | 80.4 | 81.9 | 83.7 | 86.1 | 89.9 | 92.5 | 97.9 | 69.773 | 0.099 | -1.356 | 1.876 | 69.8 | 74.7 | 76.1 | 77.8 | 80.1 | 83.9 | 86.7 | 92.5 |
Table 6.
Smoothed Percentile Values of WC (cm) by Age and Sex From Sample 3 (based on the population excluding obese and overweight children)
| Males | Females | |||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Age (years) | mu | sigma | nu | tau | P50 | P75 | P80 | P85 | P90 | P95 | P97 | P99 | mu | sigma | nu | tau | P50 | P75 | P80 | P85 | P90 | P95 | P97 | P99 |
| 6 | 53.138 | 0.074 | 0.018 | 1.639 | 53.1 | 55.7 | 56.3 | 57.2 | 58.3 | 60.0 | 61.2 | 63.6 | 52.098 | 0.079 | 0.177 | 1.644 | 52.1 | 54.7 | 55.4 | 56.3 | 57.4 | 59.2 | 60.5 | 62.9 |
| 7 | 54.486 | 0.079 | -0.203 | 1.699 | 54.5 | 57.3 | 58.1 | 59.0 | 60.2 | 62.2 | 63.5 | 66.2 | 53.365 | 0.085 | 0.089 | 1.720 | 53.4 | 56.3 | 57.1 | 58.1 | 59.4 | 61.3 | 62.7 | 65.4 |
| 8 | 55.999 | 0.084 | -0.375 | 1.759 | 56.0 | 59.1 | 60.0 | 61.0 | 62.4 | 64.5 | 66.0 | 69.0 | 54.908 | 0.091 | -0.052 | 1.774 | 54.9 | 58.2 | 59.1 | 60.2 | 61.6 | 63.8 | 65.3 | 68.3 |
| 9 | 57.666 | 0.089 | -0.460 | 1.809 | 57.7 | 61.1 | 62.1 | 63.2 | 64.7 | 67.1 | 68.7 | 72.0 | 56.685 | 0.096 | -0.212 | 1.808 | 56.7 | 60.4 | 61.3 | 62.5 | 64.1 | 66.5 | 68.2 | 71.6 |
| 10 | 59.462 | 0.093 | -0.448 | 1.834 | 59.5 | 63.3 | 64.3 | 65.5 | 67.2 | 69.8 | 71.5 | 75.1 | 58.604 | 0.099 | -0.370 | 1.827 | 58.6 | 62.6 | 63.7 | 64.9 | 66.7 | 69.4 | 71.2 | 75.0 |
| 11 | 61.436 | 0.097 | -0.408 | 1.823 | 61.4 | 65.5 | 66.6 | 67.9 | 69.7 | 72.5 | 74.4 | 78.3 | 60.623 | 0.100 | -0.565 | 1.833 | 60.6 | 64.8 | 66.0 | 67.3 | 69.2 | 72.1 | 74.1 | 78.3 |
| 12 | 63.554 | 0.098 | -0.330 | 1.773 | 63.6 | 67.8 | 68.9 | 70.3 | 72.1 | 75.1 | 77.1 | 81.2 | 62.592 | 0.099 | -0.764 | 1.823 | 62.6 | 66.9 | 68.1 | 69.5 | 71.4 | 74.5 | 76.6 | 81.1 |
| 13 | 65.711 | 0.097 | -0.230 | 1.705 | 65.7 | 69.9 | 71.1 | 72.5 | 74.4 | 77.4 | 79.4 | 83.7 | 64.338 | 0.096 | -0.900 | 1.805 | 64.3 | 68.6 | 69.8 | 71.3 | 73.2 | 76.4 | 78.6 | 83.2 |
| 14 | 67.811 | 0.095 | -0.122 | 1.636 | 67.8 | 72.0 | 73.2 | 74.6 | 76.5 | 79.5 | 81.6 | 85.9 | 65.814 | 0.094 | -0.993 | 1.786 | 65.8 | 70.1 | 71.3 | 72.7 | 74.7 | 77.8 | 80.1 | 84.8 |
| 15 | 69.703 | 0.093 | 0.032 | 1.579 | 69.7 | 73.9 | 75.0 | 76.4 | 78.3 | 81.3 | 83.4 | 87.7 | 66.974 | 0.091 | -1.009 | 1.768 | 67.0 | 71.2 | 72.3 | 73.8 | 75.7 | 78.8 | 81.1 | 85.7 |
| 16 | 71.290 | 0.093 | 0.245 | 1.566 | 71.3 | 75.5 | 76.6 | 78.0 | 79.9 | 82.8 | 84.9 | 89.1 | 67.830 | 0.089 | -0.950 | 1.762 | 67.8 | 72.0 | 73.1 | 74.5 | 76.4 | 79.5 | 81.6 | 86.2 |
| 17 | 72.613 | 0.093 | 0.426 | 1.614 | 72.6 | 76.9 | 78.1 | 79.5 | 81.3 | 84.2 | 86.2 | 90.2 | 68.492 | 0.088 | -0.866 | 1.782 | 68.5 | 72.6 | 73.8 | 75.2 | 77.0 | 80.0 | 82.1 | 86.5 |
| 18 | 73.753 | 0.094 | 0.491 | 1.715 | 73.8 | 78.2 | 79.4 | 80.9 | 82.7 | 85.6 | 87.5 | 91.4 | 69.032 | 0.088 | -0.752 | 1.828 | 69.0 | 73.2 | 74.4 | 75.8 | 77.6 | 80.6 | 82.6 | 86.8 |
Table 8 shows the performance of the age- and sex-specific 90th percentile WC cutoffs, from each of the 4 samples, for predicting CV risk in 3 countries (China, Iran, and Korea) with data on CV risk factors. The 90th percentile cutoffs for Sample 3 (excluding children with obesity or overweight) or Sample 4 (excluding children with obesity, overweight, or underweight) had higher AUC values for predicting CV risk (ie, presence of ≥ 3 CV risk factors) than Sample 1 (the whole sample) or Sample 2 (excluding children with obesity). There was no marked difference in predictive values between estimates from Sample 3 and Sample 4. Of note, findings were similar according to age categories (age 6-12 vs 13-17 yrs) (Table 8) and when stratified by country (Table 9). Based on these results, we chose to derive our final 90th WC percentile from Sample 4 (including children with normal weight and excluding those with obesity, overweight, or underweight).
Table 8.
Performance of the 90th Percentile Cutoff of Waist Circumference (WC) Based on 4 Separate Participant Samples in 8 Countries (including or excluding children with obesity, overweight or underweight) for Predicting Cardiovascular Risk (≥3 risk factors) in Pooled Data From 3 Populations Aged 6 to 17 years (China, Iran, and Korea)
| Males | Females | |||||
|---|---|---|---|---|---|---|
| Reference Population | Sensitivity (%) | Specificity (%) | AUC (95% CI) | Sensitivity (%) | Specificity (%) | AUC (95% CI) |
| All (n = 17831) | ||||||
| Sample 1 | 36.3 | 91.1 | 0.64 (0.61, 0.67) | 28.4 | 91.9 | 0.60 (0.57, 0.63) |
| Sample 2 | 45.1 | 87.4 | 0.66 (0.63, 0.69) | 34.2 | 88.9 | 0.61 (0.58, 0.65) |
| Sample 3 | 60.0 | 78.6 | 0.69 (0.67, 0.72) | 45.1 | 81.7 | 0.63 (0.60, 0.66) |
| Sample 4 | 58.6 | 79.6 | 0.69 (0.66, 0.72) | 43.1 | 83.6 | 0.63 (0.60, 0.66) |
| 6–12 years (n = 7479) | ||||||
| Sample 1 | 37.0 | 91.0 | 0.64 (0.59, 0.69) | 27.0 | 93.0 | 0.60 (0.55, 0.65) |
| Sample 2 | 48.1 | 86.6 | 0.67 (0.63, 0.72) | 36.2 | 89.9 | 0.63 (0.58, 0.68) |
| Sample 3 | 64.1 | 76.7 | 0.70 (0.66, 0.75) | 48.6 | 81.9 | 0.65 (0.61, 0.70) |
| Sample 4 | 61.9 | 78.2 | 0.70 (0.66, 0.74) | 46.5 | 83.7 | 0.65 (0.61, 0.70) |
| 13–17 years (n = 10352) | ||||||
| Sample 1 | 35.9 | 91.2 | 0.64 (0.60, 0.67) | 29.6 | 91.0 | 0.60 (0.56, 0.65) |
| Sample 2 | 43.1 | 88.0 | 0.66 (0.62, 0.69) | 32.4 | 88.2 | 0.60 (0.56, 0.65) |
| Sample 3 | 57.2 | 79.9 | 0.69 (0.65, 0.72) | 42.1 | 81.6 | 0.62 (0.58, 0.66) |
| Sample 4 | 56.5 | 80.6 | 0.69 (0.65, 0.72) | 40.3 | 83.6 | 0.62 (0.58, 0.66) |
Sample 1: Whole sample.
Sample 2: Sample excluding obese children.
Sample 3: Sample excluding obese and overweight children.
Sample 4: Sample excluding obese, overweight, and underweight children.
Notes: Cardiovascular risk factors include high blood pressure, high total cholesterol, high triglycerides, low high-density lipoprotein (HDL) cholesterol, high low-density lipoprotein (LDL) cholesterol, and high fasting glucose.
Abbreviations: AUC: Area under the curve of ROC analysis.
Table 9.
Performance of the 90th Percentile Cutoffs of Waist Circumference (WC) Based on 4 Separate Samples From 8 Countries (including or excluding children with obesity, overweight or underweight) for Predicting Cardiovascular Risk (≥ 3 risk factors) in 3 Separate Populations Aged 6 to 17 years (China, Iran, and Korea)
| Males | Females | |||||
|---|---|---|---|---|---|---|
| Reference population | Sensitivity (%) | Specificity (%) | AUC (95% CI) | Sensitivity (%) | Specificity (%) | AUC (95% CI) |
| China (7–17 years, n = 750) | ||||||
| Sample 1 | 18.2 | 93.5 | 0.56 (0.37, 0.74) | 0 | 92.8 | 0.46 (0.25, 0.68) |
| Sample 2 | 27.3 | 89.8 | 0.58 (0.40, 0.77) | 16.7 | 91.0 | 0.54 (0.29, 0.78) |
| Sample 3 | 36.4 | 82.5 | 0.59 (0.41, 0.78) | 33.3 | 84.4 | 0.59 (0.34, 0.84) |
| Sample 4 | 36.4 | 83.8 | 0.60 (0.42, 0.78) | 33.3 | 85.6 | 0.60 (0.34, 0.85) |
| Iran (6–17 years, n = 8393) | ||||||
| Sample 1 | 21.8 | 93.7 | 0.58 (0.53, 0.62) | 31.9 | 90.4 | 0.61 (0.57, 0.65) |
| Sample 2 | 28.5 | 90.8 | 0.60 (0.55, 0.64) | 36.7 | 87.3 | 0.62 (0.58, 0.66) |
| Sample 3 | 43.0 | 83.7 | 0.63 (0.59, 0.68) | 46.7 | 81.2 | 0.64 (0.60, 0.68) |
| Sample 4 | 43.0 | 84.1 | 0.63 (0.59, 0.68) | 45.2 | 82.7 | 0.64 (0.60, 0.68) |
| Korea (10–17 years, n = 8688) | ||||||
| Sample 1 | 48.2 | 88.5 | 0.68 (0.65, 0.72) | 25.4 | 93.3 | 0.59 (0.55, 0.64) |
| Sample 2 | 58.5 | 84.1 | 0.71 (0.68, 0.75) | 31.9 | 90.4 | 0.61 (0.56, 0.66) |
| Sample 3 | 73.9 | 73.5 | 0.73 (0.70, 0.77) | 43.8 | 82.1 | 0.63 (0.58, 0.67) |
| Sample 4 | 71.5 | 75.1 | 0.73 (0.70, 0.77) | 41.1 | 84.4 | 0.63 (0.58, 0.67) |
Sample 1: whole sample
Sample 2: sample excluding obese children
Sample 3: sample excluding obese and overweight children
Sample 4: sample excluding obese, overweight, and underweight children
Notes: Cardiovascular risk factors include high blood pressure, high total cholesterol, high triglycerides, low high-density lipoprotein (HDL) cholesterol, high low-density lipoprotein (LDL) cholesterol, and high fasting glucose.
Abbreviations: AUC, area under the curve of receiver operator characteristics analysis
Table 10 presents the age- and sex-specific WC 90th percentile and other percentiles at the age of 18 years which linked with adult WC cutoffs for central obesity (85 cm, 90 cm, and 94 cm in males and 80 cm in females) based on the sample of individuals with normal weight (Sample 4).
Table 10.
Age- and Sex-Specific Waist Circumference (WC) for the 90th Percentile, and Virtual Trajectories of WC During Childhood Estimated With GAMLSS, Which Link With WC Cutoffs for Central Obesity in Adults From the IDF or the Obesity in Asia Collaboration
| Males | Females | |||||
|---|---|---|---|---|---|---|
| WC Cutoffs for Adult Central Obesity | WC Cutoff for Adult Central Obesity | |||||
| Age (years) | P90, cm | 85 cm† | 90 cm‡ | 94 cm§ | P90, cm | 80 cm†‡§ |
| 6 | 58.7 | 59.8 | 62.9 | 65.4 | 57.9 | 58.9 |
| 7 | 60.7 | 61.9 | 65.4 | 68.2 | 60.0 | 61.1 |
| 8 | 62.9 | 64.3 | 68.1 | 71.2 | 62.3 | 63.6 |
| 9 | 65.3 | 66.8 | 70.9 | 74.3 | 64.9 | 66.2 |
| 10 | 67.8 | 69.4 | 73.9 | 77.6 | 67.5 | 69.0 |
| 11 | 70.4 | 72.0 | 76.9 | 80.9 | 70.0 | 71.6 |
| 12 | 72.8 | 74.6 | 79.7 | 84.1 | 72.2 | 74.0 |
| 13 | 75.0 | 76.9 | 82.3 | 86.8 | 74.1 | 75.8 |
| 14 | 77.0 | 78.9 | 84.5 | 89.2 | 75.5 | 77.3 |
| 15 | 78.8 | 80.7 | 86.3 | 91.0 | 76.5 | 78.3 |
| 16 | 80.3 | 82.2 | 87.6 | 92.2 | 77.2 | 78.9 |
| 17 | 81.8 | 83.6 | 88.8 | 93.1 | 77.8 | 79.5 |
| 18 | 83.2 | 85.0 | 90.0 | 94.0 | 78.4 | 80.0 |
†WC ≥ 85 cm for men and ≥ 80 cm for women for Asian adults recommended by the Obesity in Asia Collaboration.
‡WC ≥ 90 cm for men and ≥ 80 for women for South Asian, Chinese, and Japanese adults recommended by the IDF.
§WC ≥ 94 cm for men and ≥ 80 for women for European, African, and Eastern Mediterranean and Middle East adults recommended by the IDF.
Notes: For males, WC = 85 cm at the age of 18 years (ie, adult) corresponds to P93.6 of WC for children and adolescents aged 6 to 17 years; WC = 90 cm at the age of 18 years corresponds to P98.4 of WC; WC = 94 cm to P99.5 of WC. For females, WC = 80 cm at age of 18 corresponds to P93.3 of WC.
Abbreviations: BMI, body mass index; GAMLSS, generalized additive model for location, scale and shape, (which uses Box-Cox power exponential distribution with cubic spline smoothing); IDF, International Diabetes Federation; IOTF, International Obesity Task Force; WC, waist circumference.
All estimates are calculated based on data excluding children with obesity, overweight or underweight based on the IOTF BMI pediatric criteria.
In the 3 countries with data on CV risk factors, the WC percentile in childhood linking, at the age of 18 years, with the WC cutoff of 85 cm for adult men recommended by the Obesity in Asia Collaboration (49, 50) predicted CV risk better than the WC percentile in childhood linking with the WC cutoff of 90 cm for adult men recommended by the IDF (AUC of 0.68 vs 0.64) (Table 11). In addition, the WC percentiles in adolescents linking, at the age of 18 years, with the adult WC cutoffs of 94 cm for men and 80 cm for women (ie, the WC cutoffs recommended by the IDF for European and African adults) performed relatively well in predicting CV risk (AUC of 0.71 for males and 0.68 for females) (Table 12).
Table 11.
Performance of Those Percentiles of WC Linking With Adult WC Cutoffs for Predicting Cardiovascular Risk (≥ 3 risk factors) in the Pooled Data From 3 Test Populations Aged 6 to 17 Years From China, Iran, and Korea
| WC Cutoffs | Sensitivity (%) | Specificity (%) | AUC (95% CI) |
|---|---|---|---|
| Males (n = 9151) | |||
| 85 cm (at age 18 years)† | 47.7 | 83.3 | 0.68 (0.65, 0.71) |
| 90 cm (at age 18 years)‡ | 37.2 | 91.2 | 0.64 (0.61, 0.67) |
| Females (n = 8680) | |||
| 80 cm (at age 18 years)†‡ | 37.9 | 86.8 | 0.62 (0.59, 0.65) |
Notes: Cardiovascular risk factors include high BP, high TC, high TG, low HDL, high LDL, and high glucose
†WC ≥85 cm for men and ≥80 cm for women for Asian adults recommended by the Obesity in Asia Collaboration.
‡WC ≥90 cm for men and ≥80 for women for South Asian, Chinese, and Japanese adults recommended by the IDF.
Table 12.
Performance of Those Percentiles Of WC Linking With Adult WC Cutoffs for Predicting Cardiovascular Risk (≥ 3 risk factors) in US Adolescents Aged 12-17 Years Based on Data From the NHANES 1999-2014
| WC Cutoffs | Sensitivity (%) | Specificity (%) | AUC (95% CI) |
|---|---|---|---|
| Males (n = 1927) | |||
| 94 cm (at age 18 years)§ | 60.5 | 81.2 | 0.71 (0.66, 0.76) |
| Females (n = 1837) | |||
| 80 cm (at age 18 years)§ | 83.1 | 52.1 | 0.68 (0.61, 0.74) |
Notes: Cardiovascular risk factors include high BP, high TC, high TG, low HDL, high LDL, and high glucose
§WC ≥ 94 cm for men and ≥ 80 for women for European, African, and Eastern Mediterranean and Middle East adults recommended by the IDF.
Fig. 3 displays the comparisons of our age- and sex-specific 75th and 90th WC percentiles in normal weight children and adolescents (Sample 4) with the corresponding WC percentiles among normal weight European children in the IDEFICS study. Our WC percentile values were similar to those from the IDEFICS study (in male children aged 6-8 years and in female children aged 6-10 years), but slightly higher for males aged 9 and 10 years (eg, our 90th WC percentile is +1.1 cm at age 9 years and +1.6 cm at age 10 years).
Figure 3.
Comparisons of the 75th and 90th percentile values of waist circumference (WC) after exclusion of children and adolescents with underweight, overweight or obesity compared to corresponding percentiles of normal-weight European children in the IDEFICS study
Discussion
To our knowledge, this is the first study presenting age- and sex-specific WC cutoffs to define elevated central obesity in children and adolescents aged 6 to 18 years based on data from several countries in different regions. These international WC percentile cutoffs can be useful for identifying central obesity in different countries and for allowing direct comparison of the prevalence of central obesity between populations and trends over time.
Given the adverse effects of central obesity on several health outcomes observed among children and adolescents, it is important to establish standard WC percentile cutoffs to define elevated WC in children and adolescents, which could be used in different countries. However, it is challenging to choose a reference population for the establishment of universally valid WC cutoffs because overweight and obesity (defined by BMI criteria) largely differ between countries. In this study, we adopted several reasonable assumptions to develop pediatric percentile-based WC cutoffs that can be validly used in different countries. First, our data came from 8 countries from several regions. It is recommended by some experts that the number of countries needed to establish an international standard should range from 5 to 10 (34). Second, we used a weighted sampling design for pooling data from studies from different countries, to account for differences in sample sizes (4,51). Third, since different prevalence of unhealthy weight (ie, overweight/obesity or underweight) may shift upwards or downwards the distribution of weight-influenced parameters such as WC (52,53), we developed 90th percentile WC cutoffs in terms of best CV risk prediction, which was found to occur in the population sample (Sample 4) in which individuals with obesity, overweight, or underweight were excluded. We believe that these different assumptions and steps that we used to establish our reference WC percentiles allow for generalization of their validity and use to identify abdominal adiposity in different countries.
We chose the 90th WC percentile as the cutoff to identify central obesity in children and adolescents for 2 main reasons. First, the 90th percentile WC cutoff is also used by the IDF (46) and the modified ATP III (10). Second, the 90th percentile WC cutoff linked best, at the age of 18 years, with several criteria for adult central obesity in the present study. We assessed CV risk, based on CV risk factors (presence of ≥ 3 of 6 CV risk factors) consistent with an absolute CVD risk score approach. A similar approach was also used for validating country-specific WC percentile cutoffs in children and adolescents (20, 54). We found that the 90th WC percentile in children with normal weight performed well to predict CV risk. With regard to how WC percentiles in children and adolescents linked with adult criteria for abdominal obesity (13, 14), we used a method similar to that used to establish international BMI criteria to define child overweight and obesity (4); we found good linkage of the 90th WC percentile with several criteria of adult central adiposity (ie, 85 cm and 80 cm for Asian males and females, respectively).
The National Cholesterol Education Program ATP III (11) and the American Heart Association (55) have recommended WC cutoffs > 102 cm for men and > 88 cm for women to prevent CVD risk in the US adult population. However, several studies have shown that these high WC cutoffs performed poorly to predict CV risk in the US adults (47). Thus, when linking childhood WC percentiles to adult WC cutoffs, we preferred to use the adult WC values recommended by the IDF (94 cm for men and 80 cm for women), which have been shown to have fairly high sensitivity and specificity to identify CV risk (47).
Furthermore, 2 large prospective studies conducted by the Obesity in Asia Collaboration suggested that a WC cutoff of 85 cm, rather than 90 cm, is more suitable to predict hypertension and diabetes in Asian adult males, while the WC cutoff of 80 cm for adult females was shown to be equally valid in the Asian population vs other populations (49, 50). Hence, we also calculated the corresponding WC percentile values for Asian boys that link with a WC of 85 cm in male adults. Among males, our results show that the percentile WC values linking, at the age of 18 years, with an adult WC cutoff of 85 cm (Obesity in Asia Collaboration) performed better than the percentile WC values in boys linking with an adult WC of 90 cm (IDF). It should be noted that our 90th percentile WC cutoffs at the age of 18 years, based on normal-weight individuals, are close to the adult WC percentile recommended by the Obesity in Asia Collaboration (ie, 85 cm for men and 80 cm for women). In addition, the IDF recommended the 90th percentile WC cutoffs for defining central obesity for youth aged 6 to 15 years, but the adult WC cutoffs for adolescents aged 16 to 17 years (46). However, the modified ATP III recommends 90th percentile WC cutoffs for youth until 17 years of age (10). Thus, we also calculated the 90th percentile WC cutoffs for adolescents aged 16 to 17 years (Table 3), so that researchers or clinicians can use either the 90th percentile WC cutoffs or the adult WC cutoffs for adolescents aged 16 to 17 years.
Cole et al used a reference sample that largely preceded the obesity epidemic to derive the IOTF BMI cutoffs (4), namely, data that included few children with overweight or obesity. Unfortunately, it is not possible to derive WC cutoffs from these historical survey data because WC was not routinely collected in children until recently. To avoid the distorting effect of largely different prevalences of obesity among different populations in recent decades, it is reasonable to determine WC cutoffs for central obesity using virtually comparable populations, that is, in subsamples of current surveys that only include individuals with normal weight. This approach is further supported by our finding that the 90th WC percentile predicted CV risk better in sample of children and adolescents with normal weight rather than in samples including varying proportions of overweight or obese children. It should be further emphasized that WC reference cutoff values to identify elevated WC should, as much as possible, reflect the normal biological variation in a healthy population (35). While it is not useful to exclude individuals with abnormal weight when establishing normative data unrelated to adiposity (eg, height) (56), excluding individuals with extreme values is useful when developing normative data related to body weight status, which can be used universally (56).
Our study has two main strengths. First, we used data from 8 population-based samples involving more than 110,000 children and adolescents in several regions, which strengthens the generalizability of our findings to other populations. Second, we used a combination of several CV risk factors to assess the association of CV risk with the 90th percentile WC cutoffs. However, several limitations should also be noted. First, the number of countries in our study is not large and the data lack populations from some parts of the world, such as North and South America; however, the number of countries included in our study is still significant (n = 8) and represents 4 different global regions. This is consistent with recommendations for establishing international standards (where a minimal number of 5-10 countries is advised); hence, our WC norms have a good, albeit not perfect, potential for generalizability to other countries (34). Second, our analysis linking WC cutoffs with CV risk relied on data from only 3 countries. Future studies should further evaluate the performance of our proposed WC percentile cutoffs in other populations. Third, the predictive ability of WC to predict CV risk was limited, with AUC values of 0.69 for boys and 0.63 for girls. In addition, WC percentile (about 10 cm higher than the 90th percentile) linked relatively well with established WC cutoffs for central obesity in adults (eg, AUC in US adolescents: 0.71 for boys; 0.68 for girls), although lower than the more complex Framingham risk score (AUC of 0.75 for predicting 10-year CVD risk) (57). This is, however, still impressive given that it predicts CV risk on the basis of WC as a sole risk factor which is, furthermore, dichotomized (ie, elevated WC vs non-elevated WC). Fourth, we developed our final WC cutoff values after excluding children with obesity or overweight based on BMI. Sensitivity and AUC of WC to predict CV risk were lower in the total population (ie, also including individuals with overweight/obesity or underweight) than found after excluding children with obesity, overweight or underweight. However, our use of assessing overweight, obesity and underweight based on selected BMI cutoffs is somehow arbitrary. Fifth, although WC measurements in each country followed the recommendations by the World Health Organization, differences in accuracy and precision of WC measurements may have occurred between countries. In addition, the instruments used for analyses of blood samples in China, Iran, and Korea were different, which might have influenced the comparability of blood variables.
Conclusions
This study provides, for the first time, cutoffs for increased WC, based on data from several countries in different regions, which are demonstrably associated with an increased CV risk and closely correspond to adult criteria for abdominal adiposity at the age of 18 years. These WC cutoffs may be useful to assess abdominal adiposity in children and adolescents aged 6 to 18 years in different countries. While recognizing that country-specific norms for elevated WC may also be useful, our international WC cutoffs have the advantage of providing a standard metric, particularly for countries that have not developed their own national WC references, and allow direct comparison of the prevalence of central obesity in children and adolescents between countries and over time.
Acknowledgments
We thank Professor Tim J Cole (Population, Policy and Practice Programme, University College London, Great Ormond Street Institute of Child Health, London, United Kingdom) for useful advice. We also thank the National Center for Health Statistics of the U.S. Centers for Disease Control and Prevention, the National Institute of Nutrition and Food Safety of China Center for Disease Control and Prevention, and the Carolina Population Center of the University of North Carolina at Chapel Hill for sharing their valuable data.
Financial Support: This work was supported, in part, by National Natural Science Foundation of China (81673195); National Institutes of Health (NIH) (grants R01-HD30880, DK056350, R24-HD050924, and R01-HD38700); a Fellowship Grant of “Medical Science Fund” of Varna Medical University; a Fellowship Grant from Government of India; the Swiss Federal Office of Public Health; Isfahan University of Medical Sciences; a Ministry of Science, Ministry of Science, Technology and Innovation ScienceFund research grant (06-01-02-SF0314); Nestle Products (Malaysia) Sdn. Bhd. (NN-002-2007); a grant funded by European Economic Area (E015/P01/2007/01/85-PL 0080); a grant from the National Centre for Research and Development in Poland (NR13000206). The study funders had no role in the study design, data collection, analysis or interpretation, or writing of the paper or in the decision to submit the paper for publication.
Author Contributions: B.X., X.Z., R.K., P.B., M.L., Y.M.H., B.K.P., L.M.S., S.V.G., I.H-A., T.N., M.K-W., and A.K. designed the study. B.X., X.Z., and P.B. led the writing of the manuscript. Z.K., H.S.K., M.E.M., A.T.R., V.M.I., A.G., M.N.I., A.K., R.H., V.S., A.R-Ś., G.A., M.Q., A.Ś-L., L.O-N., V.E., V.K., M.Z., A.J.V., T.D., and M.Z. were involved in data collection. X.Z. analyzed the data. All authors contributed to the interpretation of the results and revision of the manuscript for important intellectual content and approved the final version of the manuscript. B.X. is guarantor.
The International Child Waist Circumference References Establishment Consortium consists of B.X., X.Z., R.K., M.L., Y.M.H., B.K.P., L.M.S., S.V.G., I.H-A., T.N., M.K-W., A.K., M.D.S., H.N., A.S., Z.K., H.S.K., B.S-W., M.E.M., A.T.R., V.M.I., A.G., M.N.I., A.K., R.H., V.S., A.R-Ś., G.A., M.Q., A.Ś-L., L.O-N., Y.Y., V.E., V.K., A.J.V., T.D., M.Z., C.G.M., and P.B.
Glossary
Abbreviations
- ATP III
Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III)
- AUC
area under the curve
- BMI
body mass index
- BP
blood pressure
- CV
cardiovascular
- CVD
cardiovascular disease
- HDL-C
high density lipoprotein cholesterol
- IDF
International Diabetes Federation
- IOTF
International Obesity Task Force
- GAMLSS
general additive model for location, scale, and shape
- LDL-C
low density lipoprotein cholesterol
- ROC
receiver operator characteristics
- TC
total cholesterol
- TG
triglycerides
- WC
waist circumference
Additional Information
Disclosure Summary: The authors have nothing to disclose.
Data availability: Restrictions apply to the availability of data generated or analyzed during this study to preserve patient confidentiality or because they were used under license. The corresponding author will on request detail the restrictions and any conditions under which access to some data may be provided.
References
- 1. Han JC, Lawlor DA, Kimm SY. Childhood obesity. Lancet. 2010;375(9727):1737–1748. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Xi B, Mi J, Zhao M, et al. ; Public Health Youth Collaborative and Innovative Study Group of Shandong University. Trends in abdominal obesity among U.S. children and adolescents. Pediatrics. 2014;134(2):e334–e339. [DOI] [PubMed] [Google Scholar]
- 3. Park MH, Falconer C, Viner RM, Kinra S. The impact of childhood obesity on morbidity and mortality in adulthood: a systematic review. Obes Rev. 2012;13(11):985–1000. [DOI] [PubMed] [Google Scholar]
- 4. Cole TJ, Bellizzi MC, Flegal KM, Dietz WH. Establishing a standard definition for child overweight and obesity worldwide: international survey. Bmj. 2000;320(7244):1240–1243. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Freedman DS, Wang J, Maynard LM, et al. Relation of BMI to fat and fat-free mass among children and adolescents. Int J Obes (Lond). 2005;29(1):1–8. [DOI] [PubMed] [Google Scholar]
- 6. van Dijk SB, Takken T, Prinsen EC, Wittink H. Different anthropometric adiposity measures and their association with cardiovascular disease risk factors: a meta-analysis. Neth Heart J. 2012;20(5):208–218. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Lee CM, Huxley RR, Wildman RP, Woodward M. Indices of abdominal obesity are better discriminators of cardiovascular risk factors than BMI: a meta-analysis. J Clin Epidemiol. 2008;61(7):646–653. [DOI] [PubMed] [Google Scholar]
- 8. Alberti KG, Zimmet P, Shaw J. Metabolic syndrome–a new world-wide definition. A Consensus Statement from the International Diabetes Federation. Diabet Med. 2006;23(5):469–480. [DOI] [PubMed] [Google Scholar]
- 9. Zimmet P, Alberti G, Kaufman F, et al. ; International Diabetes Federation Task Force on Epidemiology and Prevention of Diabetes. The metabolic syndrome in children and adolescents. Lancet. 2007;369(9579):2059–2061. [DOI] [PubMed] [Google Scholar]
- 10. Cook S, Weitzman M, Auinger P, Nguyen M, Dietz WH. Prevalence of a metabolic syndrome phenotype in adolescents: findings from the third National Health and Nutrition Examination Survey, 1988-1994. Arch Pediatr Adolesc Med. 2003;157(8):821–827. [DOI] [PubMed] [Google Scholar]
- 11. National Cholesterol Education Program (NCEP) Expert Panel on Detection E, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III),. Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III) final report. Circulation. 2002;106(25):3143–3421. [PubMed] [Google Scholar]
- 12. Alberti KG, Eckel RH, Grundy SM et al. ; International Diabetes Federation Task Force on Epidemiology and Prevention; Hational Heart, Lung, and Blood Institute; American Heart Association; World Heart Federation; International Atherosclerosis Society; International Association for the Study of Obesity. Harmonizing the metabolic syndrome: a joint interim statement of the International Diabetes Federation Task Force on Epidemiology and Prevention; National Heart, Lung, and Blood Institute; American Heart Association; World Heart Federation; International Atherosclerosis Society; and International Association for the Study of Obesity. Circulation. 2009;120(16):1640–1645. [DOI] [PubMed] [Google Scholar]
- 13. Rao G, Powell-Wiley TM, Ancheta I, et al. ; American Heart Association Obesity Committee of the Council on Lifestyle and Cardiometabolic Health Identification of Obesity and Cardiovascular Risk in Ethnically and Racially Diverse Populations: a Scientific Statement From the American Heart Association. Circulation. 2015;132(5):457–472. [DOI] [PubMed] [Google Scholar]
- 14. Lear SA, James PT, Ko GT, Kumanyika S. Appropriateness of waist circumference and waist-to-hip ratio cutoffs for different ethnic groups. Eur J Clin Nutr. 2010;64(1):42–61. [DOI] [PubMed] [Google Scholar]
- 15. de Moraes AC, Fadoni RP, Ricardi LM, et al. Prevalence of abdominal obesity in adolescents: a systematic review. Obes Rev. 2011;12(2):69–77. [DOI] [PubMed] [Google Scholar]
- 16. Galcheva SV, Iotova VM, Yotov YT, Grozdeva KP, Stratev VK, Tzaneva VI. Waist circumference percentile curves for Bulgarian children and adolescents aged 6-18 years. Int J Pediatr Obes. 2009;4(4):381–388. [DOI] [PubMed] [Google Scholar]
- 17. Poh BK, Jannah AN, Chong LK, Ruzita AT, Ismail MN, McCarthy D. Waist circumference percentile curves for Malaysian children and adolescents aged 6.0-16.9 years. Int J Pediatr Obes. 2011;6(3-4):229–235. [DOI] [PubMed] [Google Scholar]
- 18. Aeberli I, Gut-Knabenhans I, Kusche-Ammann RS, Molinari L, Zimmermann MB. Waist circumference and waist-to-height ratio percentiles in a nationally representative sample of 6-13 year old children in Switzerland. Swiss Med Wkly. 2011;141:w13227. [DOI] [PubMed] [Google Scholar]
- 19. Khadilkar A, Ekbote V, Chiplonkar S, et al. Waist circumference percentiles in 2-18 year old Indian children. J Pediatr. 2014;164(6):1358–62.e2. [DOI] [PubMed] [Google Scholar]
- 20. Ma GS, Ji CY, Ma J, et al. Waist circumference reference values for screening cardiovascular risk factors in Chinese children and adolescents. Biomed Environ Sci. 2010;23(1):21–31. [DOI] [PubMed] [Google Scholar]
- 21. Nawarycz LO, Krzyzaniak A, Stawińska-Witoszyńska B, et al. Percentile distributions of waist circumference for 7-19-year-old Polish children and adolescents. Obes Rev. 2010;11(4):281–288. [DOI] [PubMed] [Google Scholar]
- 22. Mukherjee S, Leong HF, Wong XX. Waist circumference percentiles for Singaporean children and adolescents aged 6-17 years. Obes Res Clin Pract. 2016;10(Suppl 1):S17–S25. [DOI] [PubMed] [Google Scholar]
- 23. Bravo J, Raimundo AM, Santos DA, Timon R, Sardinha LB. Abdominal obesity in adolescents: development of age-specific waist circumference cut-offs linked to adult IDF criteria. Am J Hum Biol. 2017;29(6):e23036. [DOI] [PubMed] [Google Scholar]
- 24. Rönnecke E, Vogel M, Bussler S, et al. Age- and Sex-Related Percentiles of Skinfold Thickness, Waist and Hip Circumference, Waist-to-Hip Ratio and Waist-to-Height Ratio: results from a Population-Based Pediatric Cohort in Germany (LIFE Child). Obes Facts. 2019;12(1):25–39. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Shah M, Radia D, McCarthy HD. Waist circumference centiles for UK South Asian children. Arch Dis Child. 2019;105(1):80–85. [DOI] [PubMed] [Google Scholar]
- 26. Liang YJ, Xi B, Song AQ, Liu JX, Mi J. Trends in general and abdominal obesity among Chinese children and adolescents 1993-2009. Pediatr Obes. 2012;7(5):355–364. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Ahadi Z, Qorbani M, Kelishadi R, et al. Association between breakfast intake with anthropometric measurements, blood pressure and food consumption behaviors among Iranian children and adolescents: the CASPIAN-IV study. Public Health. 2015;129(6):740–747. [DOI] [PubMed] [Google Scholar]
- 28. Kweon S, Kim Y, Jang MJ, et al. Data resource profile: the Korea National Health and Nutrition Examination Survey (KNHANES). Int J Epidemiol. 2014;43(1):69–77. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Gurzkowska B, Kułaga Z, Litwin M, et al. The relationship between selected socioeconomic factors and basic anthropometric parameters of school-aged children and adolescents in Poland. Eur J Pediatr. 2014;173(1):45–52. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Świąder-Leśniak Z, Kułaga Z, Grajda A, et al. References for waist and hip circumferences in Polish children and adolescents 3–18 year of age. Standardy Medyczne. 2015;12:137–150. [Google Scholar]
- 31. Chiolero A, Paradis G, Madeleine G, Hanley JA, Paccaud F, Bovet P. Discordant secular trends in elevated blood pressure and obesity in children and adolescents in a rapidly developing country. Circulation. 2009;119(4):558–565. [DOI] [PubMed] [Google Scholar]
- 32. Borghi E, de Onis M, Garza C, et al. ; WHO Multicentre Growth Reference Study Group. Construction of the World Health Organization child growth standards: selection of methods for attained growth curves. Stat Med. 2006;25(2):247–265. [DOI] [PubMed] [Google Scholar]
- 33. Physical status: the use and interpretation of anthropometry. Report of a WHO Expert Committee. World Health Organ Tech Rep Ser. 1995;854:1–452. [PubMed] [Google Scholar]
- 34. Cole TJ. The international growth standard for preadolescent and adolescent children: statistical considerations. Food Nutr Bull. 2006;27(4 Suppl Growth Standard):S237–S243. [DOI] [PubMed] [Google Scholar]
- 35. Nagy P, Kovacs E, Moreno LA, et al. ; IDEFICS consortium. Percentile reference values for anthropometric body composition indices in European children from the IDEFICS study. Int J Obes (Lond). 2014;38 (Suppl 2):S15–S25. [DOI] [PubMed] [Google Scholar]
- 36. Cole TJ, Flegal KM, Nicholls D, Jackson AA. Body mass index cut offs to define thinness in children and adolescents: international survey. BMJ. 2007;335(7612):194. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. van Buuren S, Hayes DJ, Stasinopoulos DM, Rigby RA, ter Kuile FO, Terlouw DJ. Estimating regional centile curves from mixed data sources and countries. Stat Med. 2009;28(23):2891–2911. [DOI] [PubMed] [Google Scholar]
- 38. Rigby RA, Stasinopoulos DM. Generalized additive models for location scale and shape. Appl Statist. 2005;54:507–554. [Google Scholar]
- 39. Stasinopoulos DM, Rigby RA. Generalized additive models for location scale and shape (GAMLSS) in R. J Stat Softw. 2007;23:1–46. [Google Scholar]
- 40. Royston P, Wright EM. Goodness-of-fit statistics for age-specific reference intervals. Stat Med. 2000;19(21):2943–2962. [DOI] [PubMed] [Google Scholar]
- 41. Yan S, Li J, Li S, et al. The expanding burden of cardiometabolic risk in China: the China Health and Nutrition Survey. Obes Rev. 2012;13(9):810–821. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42. Khashayar P, Heshmat R, Qorbani M, et al. Metabolic Syndrome and Cardiovascular Risk Factors in a National Sample of Adolescent Population in the Middle East and North Africa: the CASPIAN III Study. Int J Endocrinol. 2013;2013:702095. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43. Kelishadi R, Gouya MM, Adeli K, et al. ; CASPIAN Study Group. Factors associated with the metabolic syndrome in a national sample of youths: CASPIAN Study. Nutr Metab Cardiovasc Dis. 2008;18(7):461–470. [DOI] [PubMed] [Google Scholar]
- 44. Chung JY, Kang HT, Shin YH, Lee HR, Park BJ, Lee YJ. Prevalence of metabolic syndrome in children and adolescents - the recent trends in South Korea. J Pediatr Endocrinol Metab. 2013;26(1-2):105–110. [DOI] [PubMed] [Google Scholar]
- 45. Xi B, Zong X, Kelishadi R, et al. ; International Child Blood Pressure References Establishment Consortium. Establishing International Blood Pressure References Among Nonoverweight Children and Adolescents Aged 6 to 17 Years. Circulation. 2016;133(4):398–408. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46. Zimmet P, Alberti G, Kaufman F, et al. ; International Diabetes Federation Task Force on Epidemiology and Prevention of Diabetes. The metabolic syndrome in children and adolescents. Lancet. 2007;369(9579):2059–2061. [DOI] [PubMed] [Google Scholar]
- 47. Zhu S, Heymsfield SB, Toyoshima H, Wang Z, Pietrobelli A, Heshka S. Race-ethnicity-specific waist circumference cutoffs for identifying cardiovascular disease risk factors. Am J Clin Nutr. 2005;81(2):409–415. [DOI] [PubMed] [Google Scholar]
- 48. McQuillan GM, McLean JE, Chiappa M, Corporation H, Lukacs SL. National Health and nutrition examination survey biospecimen program: NHANES III (1988–1994) and NHANES 1999–2014. Vital Health Stat 2. 2015. (170):1–14. [PubMed] [Google Scholar]
- 49. Obesity in Asia Collaboration. Is central obesity a better discriminator of the risk of hypertension than body mass index in ethnically diverse populations? J Hypertens. 2008;26(2):169–177. [DOI] [PubMed] [Google Scholar]
- 50. Huxley R, Barzi F, Lee CM, et al. ; Obesity in Asia Collaboration. Waist circumference thresholds provide an accurate and widely applicable method for the discrimination of diabetes. Diabetes Care. 2007;30(12):3116–3118. [DOI] [PubMed] [Google Scholar]
- 51. Hayes DJ, van Buuren S, ter Kuile FO, Stasinopoulos DM, Rigby RA, Terlouw DJ. Developing regional weight-for-age growth references for malaria-endemic countries to optimize age-based dosing of antimalarials. Bull World Health Organ. 2015;93(2):74–83. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 52. de Onis M, Onyango AW, Borghi E, Siyam A, Nishida C, Siekmann J. Development of a WHO growth reference for school-aged children and adolescents. Bull World Health Organ. 2007;85(9):660–667. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 53. Rodd C, Metzger DL, Sharma A; Canadian Pediatric Endocrine Group (CPEG) Working Committee for National Growth Charts Extending World Health Organization weight-for-age reference curves to older children. BMC Pediatr. 2014;14:32. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 54. Gracia-Marco L, Moreno LA, Ruiz JR, et al. Body Composition Indices and Single and Clustered Cardiovascular Disease Risk Factors in Adolescents: providing Clinical-Based Cut-Points. Prog Cardiovasc Dis. 2016;58(5):555–564. [DOI] [PubMed] [Google Scholar]
- 55. Grundy SM, Cleeman JI, Daniels SR, et al. ; American Heart Association; National Heart, Lung, and Blood Institute. Diagnosis and management of the metabolic syndrome: an American Heart Association/National Heart, Lung, and Blood Institute Scientific Statement. Circulation. 2005;112(17):2735–2752. [DOI] [PubMed] [Google Scholar]
- 56. Júlíusson PB, Brannsether B, Kristiansen H, Hoppenbrouwers K, Bjerknes R, Roelants M. Should children with overweight or obesity be excluded from height references? Arch Dis Child. 2015;100(11):1044–1048. [DOI] [PubMed] [Google Scholar]
- 57. Cook NR, Paynter NP, Eaton CB, et al. Comparison of the Framingham and Reynolds Risk scores for global cardiovascular risk prediction in the multiethnic Women’s Health Initiative. Circulation. 2012;125(14):1748–56, S1. [DOI] [PMC free article] [PubMed] [Google Scholar]