Abstract
Aims/Introduction
The aim of the present study was to compare the cut‐off values and prediction effect of different obesity indices by different definitions of metabolic syndrome (MetS) in Zhejiang Province of China.
Materials and Methods
We carried out a cross‐sectional survey of 10,100 individuals (age 40 years and older) in Jiashan, Zhejiang Province. Receiver operating characteristic analysis was used to examine discrimination and find optimal cut off values of waist circumference (WC), body mass index (BMI), waist‐to‐hip ratio (WHR) and waist‐to‐height ratio (WHtR) to predict two or more non‐adipose components of MetS by The National Cholesterol Education Program Expert Panel on Detection, Evaluation, And Treatment of High Blood Cholesterol In Adults (Adult Treatment Panel III) definition modified by the Asia–Pacific region criteria, International Diabetes Federation definition for the Chinese population and Chinese Diabetes Society definitions of MetS.
Results
The age‐standardized prevalence of MetS was 23.78% vs 28.76% vs 19.37% by The National Cholesterol Education Program Expert Panel on Detection, Evaluation, And Treatment of High Blood Cholesterol In Adults (Adult Treatment Panel III), International Diabetes Federation and Chinese Diabetes Society definitions, respectively. Cut‐off values of BMI were approximately 24 kg/m2 both in men and women by three definitions; the average cut‐off values of WC, WHR and WHtR were 83 cm in men vs 81 cm in women, 0.89 in men vs 0.86 in women and 0.50 in men vs 0.51 in women, respectively. The area under receiver operating characteristic curve of BMI was larger than WC both in men and women (P < 0.05); in women, the area under receiver operating characteristic curve of WHtR was larger than WC, and WHR was smaller.
Conclusions
MetS is prevalent in Zhejiang Province of China, especially in the female population. BMI and WHtR might be more useful than WC and WHR for predicting two or more non‐adipose components of MetS.
Keywords: Body mass index, Metabolic syndrome, Waist‐to‐hip ratio
Introduction
Metabolic syndrome (MetS) is the name given to a clustering of metabolic and cardiovascular risk factors that have been widely discussed for at least 20 years. Since the first official definition of MetS put forward by a working group of the World Health Organization (WHO) in 19991, a number of different definitions have been proposed. The most widely accepted of these definitions is the original WHO definition and alternatives proposed by the European Group for the Study of Insulin Resistance (EGIR)2, and the US National Cholesterol Education Program Adult Treatment Panel III (NCEA‐ATPIII)3. In 1999, the WHO definition included a measure of obesity and defined obesity in terms of either body mass index (BMI) or waist‐to‐hip ratio (WHR). The EGIR (1999) and ATPIII (2001) definitions also introduced waist circumference (WC) as a measurement of adiposity. The Chinese Diabetes Society (CDS) also introduced a MetS definition in 2004, and it defined BMI as a measurement of obesity4. The latest definition is the one of the International Diabetes Federation (IDF), which takes into account evidence that abdominal obesity is an important component of MetS5, and proposed WC as an indication of abdominal obesity6, 7.
Various alternative methods of the indices of obesity, such as WC, BMI, WHR and waist‐to‐height ratio (WHtR), have been proposed, but their role in defining MetS has not been fully evaluated. The aim of the present study was to carry out a comparative validation of WC, BMI, WHR and WHtR for defining MetS under different definitions in a Chinese population aged 40 years and older in the southeast of China. We used the ATPIII, IDF and CDS definitions of MetS, excluding the measure of obesity, to evaluate which of the WC, BMI, WHR and WHtR obesity measures, and an appropriate cut‐off, is most closely predictive of the non‐adipose components of each MetS definition.
Methods
Study population
The present study was a community‐based, cross‐sectional survey that was funded by the Chinese Society of Endocrinology (CSE), and carried out during March to September of 2011 in Jiashan of Zhejiang, China. A total of 10,100 participants aged 40 years and older were selected by the cluster random sample method. The procedure of the study was designed by RuiJin Hospital, Shanghai Jiaotong University School of Medicine, and approved by the ethics committee. Written informed consent was obtained from all participants.
We excluded 21 observations, because data were missing for fasting glycemia, 2‐h glycemia or glycated hemoglobin. The final sample size was 10,079.
Data collection
Trained investigators completed questionnaires, measured anthropometric variables including height, weight, WC and hip circumference (HC), and carried out the oral glucose tolerance test (OGTT). The questionnaires included questions related to the diagnosis and treatment of diabetes. If participants had a history of diabetes or were taking drugs to treat diabetes, 75‐g oral dextrose tests were carried out. If not, an OGTT was carried out to diagnose diabetes.
Anthropometric measurements
Anthropometric measurements included height, weight, WC and HC. BMI was calculated as weight (kg) divided by height squared (m2). WHtR was calculated as height (cm) divided by WC (cm). WHR was calculated as HC (cm) divided by WC (cm). Two nurses completed each of these measurements; one took the measurements, the other recording the readings. Height was measured to the nearest 0.1 cm with an Iowa Height Board, and weight to the nearest 0.1 kg using a balance beam metric scale. No adjustments were made for the weight of the gown, underwear or socks that were worn during the examination. Circumferences were measured using a non‐stretchable tape. WC was measured midway between the inferior margin of the last rib and the crest of the ileum in a horizontal plane, and HC was measured at the greater trochanters.
OGTT
After an overnight fasting for at least 10 h, a standard 75‐g glucose solution was given, and venous blood samples were drawn at 0 min and 120 min (2 h) to identify diabetes and prediabetes (impaired glucose tolerance and/or impaired fasting glucose). Venous blood samples collected from the antecubital vein were put into vacuum tubes containing sodium fluoride in 4°C ice boxes and were analyzed within 3 h. Plasma glucose was measured using the glucose oxidase method, and serum lipids were measured using an automated biochemical instrument (Beckman CX‐7 Biochemical Autoanalyzer; Beckman, Brea, CA, USA).
Blood pressure measurements
Physicians carried out blood pressure measurements using an American Heart Association protocol1. After 5 min of rest in a sitting position, systolic and diastolic pressures were measured from the participant's right arm using a standard mercury sphygmomanometer. Two successive measurements were carried out with at least 1‐min interval in between. The mean measurements were put in the records.
Diagnostic criteria
According to the modified ATPIII definition for MetS3, patients must be diagnosed as having at least three of the following five factors: (i) obesity (defined as WC ≥90 cm in men or ≥80 cm in women in the Asia–Pacific region); (ii) triglycerides ≥1.7 mmol/L (150 mg/dL); (iii) high density lipoprotein (HDL) ≤1.03 mmol/L (40 mg/dL) in men or ≤1.29 mmol/L in women; (iv) systolic blood pressure (SBP) ≥130 mm Hg, or diastolic blood pressure (DBP) ≥85 mmHg; and (v) fasting plasma glucose ≥5.6 mmol/L (100 mg/dL). According to the new IDF definition8 for the Chinese population, for a person to be defined as having MetS, he/she must be diagnosed as having central obesity (defined as WC ≥90 cm in men or ≥80 cm in women) plus any two of the following four factors: (i) triglycerides ≥1.7 mmol/L (150 mg/dL), or specific treatment for this lipid abnormality; (ii) HDL ≤1.03 mmol/L (40 mg/dL) in men or ≤1.29 mmol/L in women, or specific treatment for this lipid abnormality; (iii) SBP ≥130 mmHg or DBP ≥85 mmHg, or treatment for previously diagnosed hypertension; and (iv) fasting plasma glucose ≥5.6 mmol/L (100 mg/dL) or previously diagnosed type 2 diabetes2. According to the CDS definition for MetS, patients must be diagnosed as having at lest three of the following four factors: (i) BMI ≥25.0 kg/m2; (ii) fasting plasma glucose ≥6.1 mmol/L, 2‐h post‐loading plasma glucose ≥7.8 mmol/L or previously diagnosed type 2 diabetes; (iii) SBP ≥140 mmHg or DBP ≥90 mmHg, or treatment for previously diagnosed hypertension; (iv) triglycerides ≥1.7 mmol/L or HDL ≤0.9 mmol/L in men or ≤1.0 mmol/L in women9.
Statistical analysis
Descriptive analysis was used to show distributive features of the study population. Continuous variables are presented as ‘mean (standard deviation),’ categorical variables are given as ‘point estimate (95% confidence interval).’ Intersex and intersubgroup differences were examined by t‐test or χ2‐test, as appropriate to variables.
To obtain the optimal cut‐off point for WC, BMI and WHR in predicting the presence of two or more risk factors for MetS (excluding WC to avoid self‐correlation), we chose the point on the receiver operating characteristic (ROC) curve that represented the largest sum of sensitivity and specificity10, or equivalently, of the Youden index: sensitivity + specificity − 1. The area under the ROC curve (AUC) was used as a general measure of discrimination of a predictor. To assess whether the difference in the areas under two ROC curves is of statistical significance, we used the Z‐test described in basic principles of ROC analysis11.
All analyses were carried out separately for men and women. The level for statistical significance was set at 0.05 (two‐sided). The Statistical Analysis System, version 9.2 (SAS Institute Inc., Raleigh, NC, USA) was used for all statistical analyses.
Results
Study population characteristics
The mean age of the study population was 52.9 ± 8.1 years. Participants aged >60 years made up 19.08% of the study population. Compared with women, men had a significantly higher height, weight, WC, WHR, SBP and DBP, and FPG and TG concentration, but smaller WHtR, lower 2‐h post‐loading plasma glucose, total cholesterol and HDL cholesterol concentrations (P < 0.05 for all comparisons; Table 1).
Table 1.
Mean levels of various characteristics by sex
| Variables | Men (n = 4,194) | Women (n = 5,885) | t | P |
|---|---|---|---|---|
| Age (years) | 53.22 ± 8.04 | 52.6 ± 8.11 | 3.75 | 0.0002 |
| Weight (kg) | 67.18 ± 9.93 | 58.98 ± 8.79 | 42.79 | <0.0001 |
| Height (cm) | 166.23 ± 6.26 | 155.74 ± 5.73 | 85.81 | <0.0001 |
| BMI (kg/m2) | 24.31 ± 3.93 | 24.32 ± 3.86 | 0.13 | 0.8998 |
| Waist circumference (cm) | 82.84 ± 9.23 | 79.95 ± 8.7 | 15.86 | <0.0001 |
| Hip circumference (cm) | 93.2 ± 6.51 | 93.16 ± 6.59 | 0.33 | 0.7397 |
| WHR | 0.89 ± 0.07 | 0.86 ± 0.06 | 22.77 | <0.0001 |
| WHtR | 0.50 ± 0.06 | 0.51 ± 0.06 | 12.89 | <0.0001 |
| Systolic blood pressure (mmHg) | 129.29 ± 18.04 | 128.15 ± 18.74 | 3.09 | 0.002 |
| Diastolic blood pressure (mmHg) | 79.35 ± 10.38 | 77.36 ± 9.85 | 9.71 | <0.0001 |
| FPG (mmol/L) | 5.97 ± 1.48 | 5.83 ± 1.35 | 4.62 | <0.0001 |
| 2hPPG (mmol/L) | 7.46 ± 3.43 | 7.87 ± 3.17 | 6.16 | <0.0001 |
| Total cholesterol | 4.25 ± 1.03 | 4.33 ± 1.01 | 3.65 | 0.0003 |
| Triglycerides | 1.68 ± 1.44 | 1.51 ± 1.1 | 6.18 | <0.0001 |
| HDL‐C | 1.17 ± 0.32 | 1.23 ± 0.3 | 9.64 | <0.0001 |
Normal data are given as the mean ± standard deviation; skewed data are given as the geometric mean with 95% confidence intervals in parentheses. 2hPPG, 2‐h post‐loading plasma glucose; BMI, body mass index; DBP, diastolic blood pressure; FPG, fasting plasma glucose; HDL‐C, high‐density lipoprotein cholesterol; SBP, systolic blood pressure; WHR, waist‐to‐hip ratio; WHtR, waist‐to‐height ratio.
Prevalence of different definitions of mets and the components
Based on different criteria, the prevalence of MetS in the study population was 22.36%, 28.53% and 18.67% (23.78%, 28.76% and 19.37% adjusted for age), respectively, for ATPIII, IDF and CDS. For almost all age groups, the prevalence of MetS was significant higher in women than in men according to the ATPIII and IDF definitions, and showed no difference between men and women according to the CDS criteria. The components could be classified into four groups (Table 2). In group 1, ATPIII and IDF had the same definitions for obesity. The prevalence showed no significant difference between ATPIII/IDF and CDS (χ2 = 0.8035, P = 0.3701). There were more women diagnosed with obesity (central obesity) than men under the ATPIII and IDF criteria. However, except for CDS, there was no significant difference found by sex. In groups 2–4, there was no significant difference between ATPIII and IDF criteria, because the two definitions have the same cut‐offs, but the participants diagnosed and treated were included. Compared with ATPIII and IDF criteria, the prevalence under the CDS definition was much lower in groups 2–4. The prevalence of low HDL cholesterol levels in women were higher than men, whereas high fasting glucose and hypertriglyceride prevalence were much lower in women.
Table 2.
Prevalence of components of metabolic syndrome in men and women
| Male | Female | Total | χ2 | P | |
|---|---|---|---|---|---|
| Group 1: | |||||
| Central obesity (ATPIII and IDF) | 966 (23.03) | 2,970 (50.47) | 3,936 (39.05) | 774.3677 | <0.0001 |
| BMI ≥25 kg/m2 (CDS) | 1,619 (38.60) | 2,255 (38.32) | 3,874 (38.44) | 0.0841 | 0.7719 |
| Group 2: | |||||
| Hypertriglyceride (ATPIII) | 1,345 (32.07) | 1,616 (27.46) | 2,961 (29.38) | 25.0838 | <0.0001 |
| Low HDL‐C (ATPIII) | 1,480 (35.29) | 3,512 (59.68) | 4,992 (49.53) | 582.6813 | <0.0001 |
| Hypertriglyceride or under treatment (IDF) | 1,364 (32.52) | 1,642 (27.90) | 3,006 (29.82) | 24.9868 | <0.0001 |
| Low HDL‐C or under treatment (IDF) | 1,506 (35.91) | 3,533 (60.03) | 5,039 (50.00) | 570.1271 | <0.0001 |
| Dyslipidemia (CDS) | 1,817 (43.32) | 2,431 (41.31) | 4,248 (42.15) | 4.0793 | 0.0434 |
| Group 3: | |||||
| High fasting glucose (ATPIII) | 2,107 (50.24) | 2,650 (45.03) | 4,757 (47.20) | 26.6589 | <0.0001 |
| High fasting glucose or under treatment (IDF) | 2,130 (50.79) | 2,682 (45.57) | 4,812 (47.74) | 26.6770 | <0.0001 |
| High blood glucose or under treatment (CDS) | 1,682 (40.10) | 2,448 (41.60) | 4,130 (40.98) | 2.2550 | 0.1332 |
| Group 4: | |||||
| High blood pressure (ATPIII) | 2,068 (49.31) | 2,738 (46.53) | 4,806 (47.68) | 7.6054 | 0.0058 |
| High blood pressure or under treatment (IDF) | 2,292 (54.65) | 3,002 (51.01) | 5,294 (52.53) | 13.0005 | 0.0003 |
| High blood pressure or under treatment (CDS) | 1,721 (41.03) | 2,238 (38.03) | 3,959 (39.28) | 9.2771 | 0.0023 |
| Metabolic syndrome | |||||
| Defined by ATPIII | 632 (15.07) | 1,622 (27.56) | 2,254 (22.36) | 220.1149 | <0.0001 |
| Defined by IDF | 754 (17.98) | 2,122 (36.06) | 2,876 (28.53) | 392.5301 | <0.0001 |
| Defined by CDS | 772 (18.41) | 1,110 (18.86) | 1,882 (18.67) | 0.3328 | 0.564 |
Data show the number of participants in each group, with percentages given in parentheses. ATPIII, National Cholesterol Education Program Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults with definition for Asia–Pacific region; BMI, body mass index; CDS, Chinese Diabetes Society; IDF, International Diabetes Federation.
Optimal cut‐off levels of obesity indices for three MetS definitions
The AUCs, cut‐off points, sensitivity, and specificity of BMI, WC, WHR and WHtR in different definitions of MetS by sex are presented in Table 3 for comparison. Because WC and BMI are the key factors for ATPIII, IDF and CDS, respectively, we set them as the references by different definitions, respectively. In the ATPIII definition, the AUC of BMI was significantly larger than WC in both men and women, AUC of WHR and WHtR in women were larger and smaller than WC, respectively. In the IDF definition, AUC of BMI in men and WHtR in women were larger than WC, and WHR in women was smaller than the reference. In the CDS definition, only AUC of WC and WHR in men were smaller than BMI, and the other indices had no significant difference with the reference.
Table 3.
Areas under the receiver operating cha"racteristic curve, 95% confidence interval, optimal cut‐off value, sensitivity, specificity, and Youdon Index of body mass index, waist circumference, waist‐to‐hip ratio and waist‐to‐height ratio associated with different definitions of metabolic syndrome by sex
| Men | Women | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| AUC | 95% CI | Z‐score | Cut‐off point | Sensitivity | Specificity | AUC | 95% CI | Z‐score | Cut‐off point | Sensitivity | Specificity | |
| ATPIII | ||||||||||||
| WC | 0.645 | 0.627–0.662 | Ref | 84 | 0.583 | 0.63 | 0.639 | 0.625–0.654 | Ref | 80.2 | 0.596 | 0.612 |
| BMI | 0.674 | 0.657–0.691 | 4.009** | 23.7 | 0.741 | 0.524 | 0.652 | 0.638–0.667 | 2.193* | 23.8 | 0.683 | 0.549 |
| WHR | 0.633 | 0.615–0.650 | 1.706 | 0.875 | 0.727 | 0.469 | 0.627 | 0.613–0.642 | 2.421* | 0.867 | 0.543 | 0.65 |
| WHtR | 0.649 | 0.631–0.666 | 1.064 | 0.493 | 0.689 | 0.532 | 0.647 | 0.632–0.661 | 2.808** | 0.506 | 0.679 | 0.545 |
| IDF | ||||||||||||
| WC | 0.696 | 0.680–0.712 | Ref | 83 | 0.605 | 0.692 | 0.666 | 0.652–0.680 | Ref | 81 | 0.521 | 0.735 |
| BMI | 0.715 | 0.699–0.731 | 2.874** | 23.9 | 0.666 | 0.663 | 0.676 | 0.662–0.690 | 1.735 | 23.8 | 0.649 | 0.616 |
| WHR | 0.691 | 0.675–0.707 | 0.779 | 0.883 | 0.657 | 0.631 | 0.65 | 0.636–0.665 | 3.316** | 0.856 | 0.598 | 0.632 |
| WHtR | 0.702 | 0.686–0.718 | 1.758 | 0.493 | 0.67 | 0.641 | 0.673 | 0.659–0.687 | 2.532* | 0.511 | 0.612 | 0.655 |
| CDS | ||||||||||||
| BMI | 0.677 | 0.660–0.694 | Ref | 23.8 | 0.737 | 0.535 | 0.663 | 0.648–0.678 | Ref | 23.9 | 0.691 | 0.549 |
| WC | 0.652 | 0.634–0.670 | 3.375** | 82 | 0.725 | 0.504 | 0.653 | 0.638–0.668 | 1.6 | 81 | 0.576 | 0.651 |
| WHR | 0.644 | 0.626–0.662 | 3.366** | 0.897 | 0.594 | 0.621 | 0.654 | 0.639–0.669 | 1.075 | 0.864 | 0.61 | 0.627 |
| WHtR | 0.666 | 0.648–0.684 | 1.951 | 0.502 | 0.646 | 0.605 | 0.668 | 0.653–0.683 | 0.955 | 0.527 | 0.572 | 0.684 |
All areas under the receiver operating characteristic curve were statistically significant. *P < 0.05; ** P < 0.01. ATPIII, US National Cholesterol Education Program Adult Treatment Panel III; AUC, area under receiver operating characteristic curve; BMI, body mass index; CDS, Chinese Diabetes Society; CI, confidence interval; IDF, International Diabetes Federation; ROC, receiver operating characteristic; WC, wait circumference; WHR, waist‐to‐hip ratio; WHtR, waist‐to‐height ratio.
Under three different definitions, there was a very small variation found in men and women for the cut‐off points of BMI (range 23.7–23.9). In contrast, the cut‐off points of the other obesity indices had relatively significant sex differences, and cut‐off levels of those indices for men were higher than women, except for WHtR.
Discussion
There has been a rapid urbanization in China during the past two decades. The transition to urbanized lifestyles, especially in terms of dietary habits, have inevitably led to an increase in the prevalence of obesity‐related diseases, such as diabetes, hypertension and MetS. A cross‐sectional survey carried out in 2005 in east China showed that 12.7% of men and 10.1% of women in an urban area, compared with 1.7% of men and 9.7% of women in a rural area, had MetS12. Another survey carried out in 2010 in Zhejiang Province of China showed that the prevalence of MetS was 21.48% (18.03% for men and 25.13% for women), defined by modified ATPIII criteria13. In the present study, the prevalence in men (15.07%) was lower than the average level of the whole province, but prevalence in women (27.56%) was higher using the same criteria, and they were both higher than the prevalence reported from Qingpu, Shanghai14. According to the IDF definition, the prevalence in women (36.06%) was much higher than the research carried out in Taiwan15. However, the situation is complicated by there being many criteria for the diagnosis of MetS, which always creates difficulties in comparisons with different definitions. The major difference between definitions is the measures for obesity and their cut‐off value. The present study evaluated and compared the extent to which four body stature measures, WC, BMI, WHR and WHtR, are able to predict two or more non‐adipose components (without WC or BMI) of MetS when defined by the ATPIII, IDF and CDS definitions, respectively.
The study was a community‐based, cross‐sectional survey with a large Chinese population. We used ROC analysis to address the issue of predictive ability. BMI and WC are two commonly accepted anthropometric indices for predicting MetS. Our main finding was that BMI is the best indicator to discriminate MetS, with an optimal cut‐off of 24 kg/m2 for both women and men, which is lower than the guidelines provided by WHO16. Nguyen et al.17 found that optimal BMI cut‐offs were 23–24, 21–22.5, and 20.5–21 for Chinese, Indonesian and Vietnamese adults, respectively. Another study, carried out using data obtained from more than 220,000 urban adults from mainland China, showed that the optimal cut‐off values to discriminating cardiovascular disease risk factors for men and women were approximately 24.0 and 23.0 kg/m2 for BMI, 85.0 and 75.0 cm for WC, and 0.50 and 0.48 for WHtR, respectively18. Thus, the appropriate BMI cut‐off values to detect the presence of multiple metabolic risk factors in the Chinese population might be lower than 25 kg/m2. The results of the present study were consistent with those results.
A study in a Korean population suggested that the optimal WC values were 84–86 cm for men and 78–80 cm for women to detect multiple cardiovascular risk factors19. Ko et al.20 determined that in a Chinese population, WC of 84.6 cm in men and 75.7 cm in women were the optimal cut‐off values to predict high mesenteric fat thickness with ROC analysis. The appropriate cut‐off values of WC in our study were similar to those of these previous studies.
There is a controversial issue of whether specific values measuring central fat distribution more accurately indicate health risk than BMI or WC21, 22, 23. Although some studies report that WHR identifies patients with abdominal obesity, we believe WC might be a more practical measure of abdominal fat mass and total body fat, and is more closely correlated with abdominal adipose tissue than WHR. Yang et al 24 reported that the AUC of WC and BMI for obesity were high in both sexes and subjects aged 20–45 years, and those of WHR were a little lower in the central south of China. According to the results of the present study, we consider that WHR in both sexes is a much weaker predictor; our optimal cut‐offs (0.89 in men, 0.86 in women) are close to those suggested by the WHO and close to Yang et al. (0.86 in men and 0.79 in women)24.
WHtR has received considerable interest, and the results suggest keeping one's waist to less than half one's height25, 26, 27. A Chinese study reported that waist‐to‐stature ratio (or saying WHtR) is the best simple anthropometric indicator in predicting a wide range of cardiovascular risk factors and related health conditions. The optimal waist‐to‐stature ratio cut‐off value was 0.48 for both men and women27. In the present analyses, WHtR cut‐off values were 0.50 and 0.52 in men and women, respectively. We found that it is probably that WHtR is a better predictor than WC in women, but for men WHtR and WC have the same predictive ability. There were some studies that found WHtR to be a better indicator to reflect intra‐abdominal fat distribution than BMI28, 29.
There were some limitations to the present study. First, the study was a cross‐sectional study carried out in a population aged 40 years and older, and children and young people were excluded from the study. The results and interpretation of WHR and WHtR for MetS prediction requires further validation in prospective studies. Second, the accuracy of anthropometric variables as indicators of multiple metabolic risk was not high. Swets30 suggested that 0.5 < AUC < 0.7 indicates that the diagnostic is less accurate. Further studies are required to evaluate the association between these four values and future occurrence of cardiovascular events to define their appropriate cut‐off values in a Chinese population.
Disclosure
The authors declare no conflict of interest.
Acknowledgments
We thank all of the participants and their families. We gratefully acknowledge the Center for Epidemiological Studies and Clinical Trials, Endocrinology Institute of Shanghai University, Rui‐Jin Hospital, and Jiashan Center for Disease Control and Prevention. The study was sponsored by ‘Clinical Medicine Research Special Fund of Chinese Medical Association’ (grant number: 13040530438).
J Diabetes Investig 2016; 7: 594–600
References
- 1. Marchesini G, Forlani G, Cerrelli F, et al WHO and ATPIII proposals for the definition of the metabolic syndrome in patients with Type 2 diabetes. Diabet Med 2004; 21: 383–387. [DOI] [PubMed] [Google Scholar]
- 2. Balkau B, Charles MA. Comment on the provisional report from the WHO consultation. European Group for the Study of Insulin Resistance (EGIR). Diabet Med 1999; 16: 442–443. [DOI] [PubMed] [Google Scholar]
- 3. Detection EPO. Executive Summary of The 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). JAMA 2001; 285: 2486–2497. [DOI] [PubMed] [Google Scholar]
- 4. Society EPoMSoCD . Recommendations on metabolic syndrome of Chinese diabetes society. Chin J Diab 2004; 12: 156–161. [Google Scholar]
- 5. Eckel RH, Grundy SM, Zimmet PZ. The metabolic syndrome. Lancet 2005; 365: 1415–1428. [DOI] [PubMed] [Google Scholar]
- 6. Alberti KG, Zimmet P, Shaw J. The metabolic syndrome–a new worldwide definition. Lancet 2005; 366: 1059–1062. [DOI] [PubMed] [Google Scholar]
- 7. The IDF consensus worldwide definition of the metabolic syndrome. Available from: www.idf.org/webdata/docs/IDF_Metasyndrome_definition.pdf Accessed December 14, 2014.
- 8. Alberti KG, Eckel RH, Grundy SM, et al 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: 1640–1645. [DOI] [PubMed] [Google Scholar]
- 9. Chinese Diabetes Society Metabolic Syndrome Research Group . The suggestions of metabolic syndrome form Chinese Diabetes Society. Zhonghuatangniaobingxuezazhi 2004; 12: 156–161. [Google Scholar]
- 10. Altman DG, Bland JM. Diagnostic tests 3: receiver operating characteristic plots BMJ 1994; 309: 188. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Metz CE. Basic principles of ROC analysis. Semin Nucl Med 1978; 8: 283–298. [DOI] [PubMed] [Google Scholar]
- 12. Gu D, Reynolds K, Wu X, et al Prevalence of the metabolic syndrome and overweight among adults in China. Lancet 2005; 365: 1398–1405. [DOI] [PubMed] [Google Scholar]
- 13. Ye Z, Cong LM, Ding GQ, et al A survey of the prevalence of diabetes mellitus in adults of Zhejiang province. Zhonghuaneifenmidaixiezazhi 2011; 27: 988–991. [Google Scholar]
- 14. Huang Y, Zhao Z, Li X, et al Prevalence of metabolic syndrome and its association with obesity indices in a Chinese population. J Diabetes 2009; 1: 57–64. [DOI] [PubMed] [Google Scholar]
- 15. Chien KL, Lee BC, Hsu HC, et al Prevalence, agreement and classification of various metabolic syndrome criteria among ethnic Chinese: a report on the hospital‐based health diagnosis of the adult population. Atherosclerosis 2008; 196: 764–771. [DOI] [PubMed] [Google Scholar]
- 16. World Health Organization . Obesity: Preventing and Managing the Global Epidemic: Report of a WHO Consultation on Obesity. Geneva: World Health Organization, 1998. [PubMed] [Google Scholar]
- 17. Nguyen TT, Adair LS, Suchindran CM, et al The association between body mass index and hypertension is different between East and Southeast Asians. Am J Clin Nutr 2009; 89: 1905–1912. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Qiang Z, Yuan H, Shengyong D, et al Optimal cut‐off values of BMI, waist circumference and waist:height ratio for defining obesity in Chinese adults. Br J Nutr 2014; 112: 1735–1744. [DOI] [PubMed] [Google Scholar]
- 19. Baik I. Optimal cutoff points of waist circumference for the criteria of abdominal obesity: comparison with the criteria of the International Diabetes Federation. Circ J 2009; 73: 2068–2075. [DOI] [PubMed] [Google Scholar]
- 20. Ko GT, Liu KH, So WY, et al Cutoff values for central obesity in Chinese based on mesenteric fat thickness. Clin Nutr 2009; 28: 679–683. [DOI] [PubMed] [Google Scholar]
- 21. Zhu S, Wang Z, Heshka S, et al Waist circumference and obesity‐associated risk factors among whites in the third National Health and Nutrition Examination Survey: clinical action thresholds. Am J Clin Nutr 2002; 76: 743–749. [DOI] [PubMed] [Google Scholar]
- 22. Janssen I, Katzmarzyk PT, Ross R. Waist circumference and not body mass index explains obesity‐related health risk. Am J Clin Nutr 2004; 79: 379–384. [DOI] [PubMed] [Google Scholar]
- 23. Huxley R, James WP, Barzi F, et al Ethnic comparisons of the cross‐sectional relationships between measures of body size with diabetes and hypertension. Obes Rev 2008; 9(Suppl 1): 53–61. [DOI] [PubMed] [Google Scholar]
- 24. Yang F, Lv JH, Lei SF, et al Receiver‐operating characteristic analyses of body mass index, waist circumference and waist‐to‐hip ratio for obesity: screening in young adults in central south of China. Clin Nutr 2006; 25: 1030–1039. [DOI] [PubMed] [Google Scholar]
- 25. Ashwell M, Hsieh SD. Six reasons why the waist‐to‐height ratio is a rapid and effective global indicator for health risks of obesity and how its use could simplify the international public health message on obesity. Int J Food Sci Nutr 2005; 56: 303–307. [DOI] [PubMed] [Google Scholar]
- 26. Lee K, Song YM, Sung J. Which obesity indicators are better predictors of metabolic risk?: healthy twin study. Obesity (Silver Spring) 2008; 16: 834–840. [DOI] [PubMed] [Google Scholar]
- 27. Ho SY, Lam TH, Janus ED. Waist to stature ratio is more strongly associated with cardiovascular risk factors than other simple anthropometric indices. Ann Epidemiol 2003; 13: 683–691. [DOI] [PubMed] [Google Scholar]
- 28. Hayashi T, Boyko EJ, McNeely MJ, et al Minimum waist and visceral fat values for identifying Japanese Americans at risk for the metabolic syndrome. Diabetes Care 2007; 30: 120–127. [DOI] [PubMed] [Google Scholar]
- 29. Tong J, Boyko EJ, Utzschneider KM, et al Intra‐abdominal fat accumulation predicts the development of the metabolic syndrome in non‐diabetic Japanese‐Americans. Diabetologia 2007; 50: 1156–1160. [DOI] [PubMed] [Google Scholar]
- 30. Swets JA. Measuring the accuracy of diagnostic systems. Science 1988; 240: 1285–1293. [DOI] [PubMed] [Google Scholar]