Hostname: page-component-586b7cd67f-r5fsc Total loading time: 0 Render date: 2024-11-26T09:20:26.610Z Has data issue: false hasContentIssue false
  • English
  • Français

Optimal cut-off values and population means of waist circumference in different populations

Published online by Cambridge University Press:  20 July 2010

Zhiqiang Wang ,
Jun Ma  and
Damin Si
Zhiqiang Wang*
Affiliation:
School of Medicine, Centre for Chronic Disease, University of Queensland, Brisbane, Australia
Jun Ma
Affiliation:
Institute of Child and Adolescent Health, Peking University Health Sciences Center, Beijing, People's Republic of China
Damin Si
Affiliation:
School of Medicine, Centre for Chronic Disease, University of Queensland, Brisbane, Australia
*
*Corresponding author: Dr Zhiqiang Wang, fax +61 7 3346 4812, email [email protected]
Rights & Permissions [Opens in a new window]

Abstract

Abdominal obesity is a risk factor for cardiometabolic disease, and has become a major public health problem in the world. Waist circumference is generally used as a simple surrogate marker to define abdominal obesity for population screening. An increasing number of publications solely rely on the method that maximises sensitivity and specificity to define ‘optimal’ cut-off values. It is well documented that the optimal cut-off values of waist circumference vary across different ethnicities. However, it is not clear if the variation in cut-off values is a true biological phenomenon or an artifact of the method for identifying optimal cut-off points. The objective of the present review was to assess the relationship between optimal cut-offs and population waist circumference levels. Among sixty-one research papers, optimal cut-off values ranged from 65·5 to 101·2 cm for women and 72·5 to 103·0 cm for men. Reported optimal cut-off values were highly correlated with population means (correlation coefficient: 0·91 for men and 0·93 for women). Such a strong association was independent of waist circumference measurement techniques or the health outcomes (dyslipidaemia, hypertension or hyperglycaemia), and existed in some homogeneous populations such as the Chinese and Japanese. Our findings raised some concerns about applying the sensitivity and specificity approach to determine cut-off values. Further research is needed to understand whether the differences among populations in waist circumference were genetically or environmentally determined, and to understand whether using region-specific cut-off points can identify individuals with the same absolute risk levels of metabolic and cardiovascular outcomes among different populations.

Keywords

Waist circumferenceOptimal cut-off pointsAbdominal obesitySensitivitySpecificity
Type
Review Article
Information
Nutrition Research Reviews , Volume 23 , Issue 2 , December 2010 , pp. 191 - 199
Copyright
Copyright © The Authors 2010

Introduction

Excess abdominal fat is associated with an increased risk of cardiometabolic disease(Reference Klein, Allison and Heymsfield1). However, precise measurement of abdominal fat content requires the use of expensive radiological imaging techniques. Therefore, waist circumference is often used as a surrogate marker of abdominal fat mass, because waist circumference correlates with abdominal fat mass and is associated with cardiometabolic disease risk(Reference Klein, Allison and Heymsfield1). Although there is a continuous association between waist circumference and the risk of cardiometabolic disease, a cut-off point is often determined for defining abdominal (or central) obesity for population screening(Reference Lear, James and Ko2, Reference Qiao and Nyamdorj3). The identification of waist circumference cut-off points is critical for both clinical care and public health research. Recently, an increasing number of research papers have been published to define optimal waist circumference cut-off values in different populations(Reference Lear, James and Ko2, Reference Qiao and Nyamdorj3). Most of those papers solely rely on the receiver-operating characteristic (ROC) curve method to maximise sensitivity and specificity to define ‘optimal’ cut-off values. It is well documented that optimal cut-off values vary across different ethnicities. Such a variation in waist circumference cut-off values may be explained by ethnic differences in visceral adipose tissue and in the relationships between waist circumference and visceral adipose tissue(Reference Carroll, Chiapa and Rodriquez4, Reference Desilets, Garrel and Couillard5). The reported cut-off values also vary substantially within some relatively homogeneous populations such as the Chinese and Japanese. There has been a recommendation to use region-specific cut-off values(Reference Qiao and Nyamdorj3). However, the huge variation among different regions within one ethnic group raised a question whether such a variation in cut-off values is a true biological phenomenon or an artifact of the widely used approach of maximising both sensitivity and specificity for identifying optimal cut-off points.

In the present review, we focused on assessing the relationship between the reported optimal cut-off values and population means of waist circumference. This relationship is essential for understanding why different optimal cut-off values of waist circumference have been reported in different studies. It is also useful for comparing the prevalence of abdominal obesity among different regions and for monitoring changes in the prevalence of abdominal obesity over time in the same population. The techniques for measuring waist circumference vary in the literature; so are the outcome measurements for defining waist circumference cut-off values. We further examined if the relationship between the cut-off values and population means of waist circumference was independent of waist circumference measurement methods and the cardiometabolic outcomes.

Methods

Literature search strategy and inclusion criteria

Papers were included according to the following criteria. First, we searched the following strings on the PubMed Medline: ‘waist circumference’ AND ‘cut-off OR cutoff OR cut point’ AND language as English (eng). A total of 304 citations published from 1999 to 2009 were retrieved for possible inclusion. Second, we (Z. W. and D. S.) reviewed all abstracts and full papers if available to identify articles that defined waist circumference cut-off values in adults. Studies of children and adolescents and those that did not define waist circumference cut-off values were excluded. A total of seventy-five papers that defined waist circumference cut-off values in adults were included. Third, ten studies without obtainable population waist circumference means were excluded. Fourth, four more studies were excluded due to small sample size (n < 100). Therefore, sixty-one studies were included in the present review.

Waist circumference measurements

Different measurement methods for waist circumference were used in the literature, and those methods were categorised into the following four groups: (1) midway between the bottom of the lower rib and the top of the iliac crest; (2) at the umbilicus; (3) at the narrowest point between the umbilicus and xiphoid process; (4) other methods or unspecified.

Health outcomes

Outcome measures included hyperglycaemia (impaired fasting glucose, impaired glucose tolerance and diabetes), hypertension, dyslipidaemia (high total cholesterol, high TAG, high LDL-cholesterol, and low HDL-cholesterol), the metabolic syndrome, CHD, CVD, elevated visceral fat and overall mortality.

The definitions for some outcomes varied among studies. Most outcomes were defined according to the International Diabetes Federation, the National Cholesterol Education Program Adult Treatment Panel III, the Chinese Diabetes Society, the Japanese Committee of the Criteria for Metabolic Syndrome and the American Diabetes Association. Specifically, the commonly used definitions for hypertension, dyslipidaemia and hyperglycaemia are as follows:

  • Hypertension – systolic blood pressure (SBP) ≥ 140 mmHg or diastolic blood pressure (DBP) ≥ 90 mmHg; or SBP ≥ 130 mmHg or DBP ≥ 85 mmHg.

  • Dyslipidaemia – total cholesterol ≥ 5·2 mmol/l or ≥ 6·2 mmol/l.

  • TAG – ≥ 1·7 mmol/l or ≥ 2·3 mmol/l.

  • HDL-cholesterol – < 0·9 mmol/l, < 1·0 mmol/l, or < 1·29 mmol/l.

  • LDL – ≥ 3·5 mmol/l, or ≥ 4·14 mmol/l.

  • Hyperglycaemia and diabetes – fasting plasma glucose ≥ 5·6 mmol/l, ≥ 6·1 mmol/l, or ≥ 7·0 mmol/l; 2 h plasma glucose ≥ 7·8 mmol/l, or ≥ 11·1 mmol/l.

Most studies presented one waist circumference cut-off value for detecting a clustering of multiple outcomes, such as one or more metabolic risk factors, or two or more metabolic or CVD risk factors. If a study presented multiple cut-off values for individual outcomes instead of a summary cut-off value, a mean cut-off value was calculated. If the sex-specific waist circumference means of the study sample were not presented in the original article, we derived the mean waist circumference using the data provided if possible.

Methods of optimising sensitivity and specificity

Optimal cut-off points are commonly determined according to sensitivity and specificity values. Different methods were used to optimise sensitivity and specificity values. Those methods were categorised into four groups: (1) a waist circumference point with the maximum of the sum of sensitivity and specificity; (2) a point with the shortest distance on the receiver-operating characteristic (ROC) curve, where the distance is ; (3) a point where sensitivity equals specificity; (4) the details were not specified although a sensitivity and specificity approach was mentioned in the paper.

Data analysis to explore the association between cut-off values and population means of waist circumference

Cut-off values were plotted against corresponding population means of waist circumference. Pearson correlation coefficients between cut-off values and population means were calculated for each sex. We also calculated subgroup correlation coefficients by: (1) the methods of optimising sensitivity and specificity; (2) the measuring methods of waist circumference; (3) health outcome measurements; (4) the representativeness of study samples. Correlation coefficients for two well-studied homogeneous populations, the Chinese and Japanese, were calculated separately. All analyses were conducted using Stata 10 (StataCorp LP, College Station, TX, USA)(6).

Results

As described previously, sixty-one studies were included in the present review (see Appendix 1)(Reference Aekplakorn, Kosulwat and Suriyawongpaisal7–Reference Ye, Bao and Hou67). Those articles reported optimal cut-off values of waist circumference for different populations including Australians(Reference Chen, Peeters and Magliano16, Reference Huxley, Barzi and Lee31), Brazilians(Reference Almeida, Almeida and Araújo8, Reference Barbosa, Lessa and de Almeida Filho13, Reference Gus, Cichelero and Moreira25, Reference Peixoto Mdo, Benicio and Latorre Mdo48), Canadians(Reference Dobbelsteyn, Joffres and MacLean20), Chinese(Reference Bao, Lu and Wang12, Reference Ho, Lam and Janus29, Reference Huang, Lee and Lee30, Reference Ko, Chan and Cockram35–Reference Ko and Tang37, Reference Lin, Lee and Chen40, Reference Tseng, Chong and Chan62, Reference Wang, Wu and Song64–Reference Ye, Bao and Hou67), French(Reference Balkau, Sapinho and Petrella11), Guadeloupeans(Reference Foucan, Hanley and Deloumeaux24), Indians (Asia)(Reference Deshmukh, Gupta and Dongre19, Reference Mohan, Deepa and Farooq42, Reference Snehalatha, Viswanathan and Ramachandran59), Iranians(Reference Delavari, Forouzanfar and Alikhani18, Reference Esteghamati, Ashraf and Rashidi22, Reference Hadaegh, Zabetian and Sarbakhsh26), Iraqis(Reference Mansour and Al-Jazairi41), Jamaicans(Reference Sargeant, Bennett and Forrester54), Japanese(Reference Doi, Ninomiya and Hata21, Reference Hara, Matsushita and Horikoshi28, Reference Ito, Nakasuga and Ohshima32, Reference Kashihara, Lee and Kawakubo33, Reference Lee, Kawakubo and Mori38, Reference Narisawa, Nakamura and Kato43, Reference Oka, Kobayashi and Yagi46, Reference Sakurai, Takamura and Miura52, Reference Sato, Asayama and Ohkubo55, Reference Shimajiri, Imagawa and Kokawa57, Reference Shiwaku, Anuurad and Enkhmaa58, Reference Tabata, Yoshimitsu and Hamachi61), Koreans(Reference Baik10, Reference Han, Park and Kim27, Reference Kim, Kim and Park34, Reference Lee, Park and Kim39, Reference Park, Choi and Lee47), Mexicans(Reference Alonso, Munguia-Miranda and Ramos-Ponce9, Reference Berber, Gomez-Santos and Fanghanel14, Reference Sanchez-Castillo, Velazquez-Monroy and Berber53), Mongolians(Reference Shiwaku, Anuurad and Enkhmaa58), New Zealand Maoris(Reference Rush, Crook and Simmons51), Singaporeans(Reference Pua, Lim and Ang49, Reference Pua and Ong50), Swedish(Reference Nilsson, Hedberg and Jonason45), Thais(Reference Aekplakorn, Kosulwat and Suriyawongpaisal7, Reference Narksawat, Podang and Punyarathabundu44), Tongans(Reference Craig, Colagiuri and Hussain17), Tunisians(Reference Bouguerra, Alberti and Smida15), Turkish(Reference Uzunlulu, Oguz and Aslan63), and African-Americans and White Americans(Reference Flegal23, Reference Stevens, Couper and Pankow60). Among sixty-one studies, thirty-eight (62 %) used the maximum of the sum of sensitivity and specificity to identify an optimal cut-off point (Table 1). Most studies (fifty-five out of sixty-one; 90 %) were cross-sectional, and six studies were of cohort design. Of the studies, twenty-four (39 %) were based on representative samples. The most commonly used technique to measure waist circumference was to measure the midway between the bottom of the lower rib and the top of the iliac crest (twenty-five out of sixty-one; 41 %).

Table 1 Some characteristics of the sixty-one cited papers

ROC, receiver-operating characteristic.

Of the studies, thirteen presented optimal waist circumference cut-off values based on a single outcome, and most studies (n 48) reported optimal cut-off values for a clustering of multiple outcomes. In addition to the overall optimal cut-off values, some studies reported optimal cut-off values for detecting individual components of the metabolic syndrome or CVD risk factors. Commonly reported cut-off values were for hyperglycaemia (n 29), hypertension (n 25) and dyslipidaemia (n 20).

Means and cut-off values of waist circumference

The mean values of waist circumference varied substantially among different studies and so did the optimal cut-off values, ranging from 72·5 to 103·0 cm for men and from 65·5 to 101·2 cm for women. The minimum and the maximum cut-off values differed by as much as 30·5 cm for men and 35·7 cm for women. The optimal cut-off values were highly correlated with the population mean values (Fig. 1). The cut-off values increased with the increasing population means. The correlation coefficient was 0·91 (95 % CI 0·86, 0·95) for men and 0·93 (95 % CI 0·89, 0·96) for women.

Fig. 1 Optimal waist circumference cut-off values and population means by sex: ○, female (r 0·93); △, male (r 0·91). (- - -), Line of identity.

The Chinese and Japanese are considered two relative homogeneous groups genetically. Although their mean values and optimal cut-off values were generally smaller than those in non-Asian populations, the correlation coefficient remained high, particularly in Chinese women (r 0·96; 95 % CI 0·88, 0·99) and Japanese women (r 0·92; 95 % CI 0·72, 0·98). The optimal cut-off values among studies within each ethnic and sex group still differed by 9·1 to 18·5 cm (Fig. 2).

Fig. 2 Optimal waist circumference cut-off values and population means of (a) Chinese by sex: ⋄, female (r 0·96); △, male (r 0·84). Optimal waist circumference cut-off values and population means of (b) Japanese by sex: ⋄, female (r 0·92); △, male (r 0·58). (- - -), Line of identity.

The method of measuring waist circumference had little impact on the association between waist circumference cut-off values and population means. Correlation coefficients were calculated separately according to the waist circumference measurement method. The correlation coefficients were 0·96 (95 % CI 0·91, 0·99), 0·88 (95 % CI 0·59, 0·97) and 0·92 (95 % CI 0·76, 0·97) for men whose waist circumferences were measured using (1) the midway between the bottom of the lower rib and the top of the iliac crest, (2) at the umbilicus and (3) at the narrowest point between the umbilicus and xiphoid process, respectively. The corresponding values for women were 0·96 (95 % CI 0·92, 0·98), 0·91 (95 % CI 0·65, 0·98) and 0·86 (95 % CI 0·64, 0·95). In addition, the relationship between cut-off values and population means existed in both cross-sectional (r 0·91, 95 % CI 0·85, 0·95 for men; r 0·92, 95 % CI 0·87, 0·96 for women) and cohort (r 0·93, 95 % CI 0·57, 0·99 for men; r 0·98, 95 % CI 0·88, 0·99 for women) studies.

We assessed the relationships between population means and cut-off values of waist circumference separately for three commonly reported single outcomes (Fig. 3). Correlation coefficients were 0·93 (95 % CI 0·86, 0·97), 0·94 (95 % CI 0·85, 0·98) and 0·94 (95 % CI 0·85, 0·98) in men, and 0·96 (95 % CI 0·91, 0·98), 0·92 (95 % CI 0·80, 0·97) and 0·97 (95 % CI 0·94, 0·99) in women, for hyperglycaemia, dyslipidaemia and hypertension, respectively.

Fig. 3 Optimal waist circumference cut-off values and population means of (a) males by outcome: ○, hyperglycaemia (r 0·93); △, dyslipidaemia (r 0·94);+, hypertension (r 0·94). Optimal waist circumference cut-off values and population means of (b) females by outcome: ○, hyperglycaemia (r 0·96); △, dyslipidaemia (r 0·92);+, hypertension (r 0·97). (- - -), Line of identity.

The representativeness of the study samples had little impact on the relationship between population means and cut-off values. The correlation coefficients were 0·97 (95 % CI 0·93, 0·99) for men and 0·94 (95 % CI 0·86, 0·97) for women in studies with representative samples and 0·89 (95 % CI 0·79, 0·94) for men and 0·91 (95 % CI 0·83, 0·95) for women in those with convenience samples.

Discussion

The optimal cut-off values of waist circumference vary substantially across different populations. Importantly, the optimal cut-off values determined using sensitivity and specificity values also vary considerably among studies within relatively homogeneous ethnic groups such as the Chinese and Japanese. We found that such a variation was mainly driven by the population waist circumference levels. The cut-off values linearly increase with increasing population means. The strong relationship is independent of waist circumference measurement techniques regardless of whether the health outcome is hypertension, dyslipidaemia, hyperglycaemia or a cluster of multiple outcomes.

Our findings raised some concerns about applying the sensitivity and specificity approach to determine an optimal cut-off value. The so-called optimal cut-off is a point that maximises both sensitivity and specificity. To achieve this, there will always be some ‘optimised’ numbers of participants above and below the cut-off point. This will make the cross-population comparison of the prevalence of abdominal obesity difficult. For example, the optimal cut-off values for Chinese women in two studies are 74·7 cm(Reference Ho, Lam and Janus29) and 82·05 cm(Reference Wang, Wu and Song64), respectively. Although the difference in waist circumference means between two study populations is as high as 7·7 cm (75·3 v. 83·0 cm), the two populations have a similar prevalence of ‘abdominal obesity’ (53 %) if applying each cut-off value to its own population. Therefore, the true problem of abdominal obesity in the population with a higher waist circumference is masked by a higher cut-off value, unless such a difference is mainly determined by genetic background.

There are an increasing number of studies using the sensitivity and specificity approach to determine optimal cut-off values in recent years. We should be cautious about interpreting and applying those optimal cut-off values. Because the sensitivity and specificity approach produces different optimal cut-off values for the regions with different levels of waist circumference within one population, it has been suggested that a region-specific cut-off value should be considered(Reference Qiao and Nyamdorj3). Our question is: Should a population with higher waist circumference levels have higher cut-off values? To answer this question, further investigation is needed. First, efforts should be made to understand the causes of regional variations in population waist circumference levels. If the regional variation in waist circumference levels is mainly genetically determined, the use of region-specific cut-off values can be justified. A recent study by the DECODE group (Diabetes Epidemiology: Collaborative analysis of Diagnostic criteria in Europe) has shown that given the same waist circumference cut-off value, the prevalence of diabetes varies among ethnic groups, and the Europeans need a higher cut-off than Asians to obtain the same prevalence of diabetes(Reference Nyamdorj, Pitkaniemi and Tuomilehto68). This indicates that genetic differences play a major role among those populations. However, if such a variation is mainly due to the differences in diet and physical activity among regions, a uniform cut-off value across regions within a population is preferred. Generally, Asians with lower mean waist circumference have lower cut-off values. However, the huge variation in the waist circumference mean values among different regions in China accompanied by the rapid increase in population means in recent years(Reference Zhou, Wu and Zhao69) suggests that the nutrition, physical activity and lifestyle factors may have contributed to regional variation. Calculating absolute risk corresponding to different waist circumference cut-off values in different regions will provide some evidence on this issue. Regardless of population waist circumference levels, if the absolute risks corresponding to a specific waist circumference point are the same in different regions, a uniform cut-off value is warranted.

Since most studies are cross-sectional, more cohort studies in this area have been encouraged(Reference Lear, James and Ko2, Reference Huxley, Mendis and Zheleznyakov70). However, if the approach of maximising sensitivity and specificity is applied to cohort data, the strong relationship between estimated cut-off values and population means remains, as indicated by the high correlation coefficient among cohort studies in the present review.

The current trend of mainly relying on maximising sensitivity and specificity to determine cut-off points may over-simplify the complexity for defining waist circumference cut-off values. Future research should focus on searching and applying methods alternative to the widely used method that maximises sensitivity and specificity. Perhaps, multiple approaches should be applied. One alternative is to calculate an absolute risk for the health outcome of interest. Although further research is needed on how to identify a cut-off value according to absolute risks, cut-off values based on multiple outcomes should be identified when the absolute risks reach a certain level that is considered to be high enough to take action. Further understanding of genetic and lifestyle factors contributing to the regional variation in waist circumference will also be useful. Most studies cited in the present review investigate cut-off values in a single population in isolation. Further research is needed to study different populations using the same sample selection criteria, measurements and analytical techniques.

In conclusion, the cut-off values determined using the sensitivity and specificity approach are highly correlated with population waist circumference levels. This can be problematic for comparing the prevalence of abdominal obesity across different populations as well as for monitoring the time trend in the same population, particularly when the population variation in waist circumference is due to differences in diet and lifestyle factors. Further research is required to examine alternatives including methods that use absolute risk levels to define waist circumference cut-off values in different populations.

Acknowledgements

The present study was financially supported by the National Health and Medical Research Council of Australia (NHMRC; no. 511013 to Z. W.).

Z. W. proposed the project, performed data analysis and drafted the manuscript. D. S, J. M. and Z. W. contributed to the study design and data collection, and revised the article critically and approved the final version.

The authors declare that there are no conflicts of interest.

Appendix 1 Studies defining waist circumference cut-off values in adult populations

* Methods for optimising sensitivity and specificity: M1, maximum of sensitivity and specificity; M2, shortest distance on receiver-operating characteristic (ROC) curve; M3, sensitivity = specificity; M4, other methods or unspecified.

† Outcome measurements: MF, multiple factors for CVD or the metabolic syndrome; VFA, visceral fat area; DM, diabetes mellitus; MFT, mesenteric fat thickness; CRF, cardiorespiratory fitness; HOMA-IR, homeostatic model assessment of insulin resistance; PBF, elevated percentage body fat.

‡ Waist circumference measurements: W1, midway between the bottom of the lower rib and top of the iliac crest; W2, the narrowest point between the umbilicus and xiphoid process; W3, at the umbilicus; W4, other methods and unspecified.

Footnotes

* Methods for optimising sensitivity and specificity: M1, maximum of sensitivity and specificity; M2, shortest distance on receiver-operating characteristic (ROC) curve; M3, sensitivity = specificity; M4, other methods or unspecified.

† Outcome measurements: MF, multiple factors for CVD or the metabolic syndrome; VFA, visceral fat area; DM, diabetes mellitus; MFT, mesenteric fat thickness; CRF, cardiorespiratory fitness; HOMA-IR, homeostatic model assessment of insulin resistance; PBF, elevated percentage body fat.

‡ Waist circumference measurements: W1, midway between the bottom of the lower rib and top of the iliac crest; W2, the narrowest point between the umbilicus and xiphoid process; W3, at the umbilicus; W4, other methods and unspecified.

References

1Klein, S, Allison, DB, Heymsfield, SB, et al. . (2007) Waist circumference and cardiometabolic risk: a consensus statement from Shaping America's Health: Association for Weight Management and Obesity Prevention; NAASO, The Obesity Society; the American Society for Nutrition; and the American Diabetes Association. Am J Clin Nutr 85, 1197–1202.CrossRefGoogle Scholar
2Lear, SA, James, PT, Ko, GT, et al. . (2010) Appropriateness of waist circumference and waist-to-hip ratio cutoffs for different ethnic groups. Eur J Clin Nutr 64, 42–61.CrossRefGoogle ScholarPubMed
3Qiao, Q & Nyamdorj, R (2009) The optimal cutoff values and their performance of waist circumference and waist-to-hip ratio for diagnosing type II diabetes. Eur J Clin Nutr 64, 23–29.CrossRefGoogle ScholarPubMed
4Carroll, JF, Chiapa, AL, Rodriquez, M, et al. . (2008) Visceral fat, waist circumference, and BMI: impact of race/ethnicity. Obesity (Silver Spring) 16, 600–607.CrossRefGoogle ScholarPubMed
5Desilets, MC, Garrel, D, Couillard, C, et al. . (2006) Ethnic differences in body composition and other markers of cardiovascular disease risk: study in matched Haitian and White subjects from Quebec. Obesity (Silver Spring) 14, 1019–1027.CrossRefGoogle ScholarPubMed
6StataCorp LP (2007) Stata Statistical Software, release 10 (program). College Station, TX: StataCorp LP.Google Scholar
7Aekplakorn, W, Kosulwat, V & Suriyawongpaisal, P (2006) Obesity indices and cardiovascular risk factors in Thai adults. Int J Obes (Lond) 30, 1782–1790.CrossRefGoogle ScholarPubMed
8Almeida, RT, Almeida, MM & Araújo, TM (2009) Abdominal obesity and cardiovascular risk: performance of anthropometric indexes in women. Arq Bras Cardiol 92, 345–350, 362–367, 375–380.Google ScholarPubMed
9Alonso, AL, Munguia-Miranda, C, Ramos-Ponce, D, et al. . (2008) Waist perimeter cutoff points and prediction of metabolic syndrome risk. A study in a Mexican population. Arch Med Res 39, 346–351.CrossRefGoogle Scholar
10Baik, I (2009) Optimal cutoff points of waist circumference for the criteria of abdominal obesity. Circ J 73, 2068–2075.CrossRefGoogle ScholarPubMed
11Balkau, B, Sapinho, D, Petrella, A, et al. . (2006) Prescreening tools for diabetes and obesity-associated dyslipidaemia: comparing BMI, waist and waist hip ratio. The D.E.S.I.R. Study. Eur J Clin Nutr 60, 295–304.CrossRefGoogle ScholarPubMed
12Bao, Y, Lu, J, Wang, C, et al. . (2008) Optimal waist circumference cutoffs for abdominal obesity in Chinese. Atherosclerosis 201, 378–384.CrossRefGoogle ScholarPubMed
13Barbosa, PJ, Lessa, I, de Almeida Filho, N, et al. . (2006) Criteria for central obesity in a Brazilian population: impact on metabolic syndrome. Arq Bras Cardiol 87, 407–414.Google Scholar
14Berber, A, Gomez-Santos, R, Fanghanel, G, et al. . (2001) Anthropometric indexes in the prediction of type 2 diabetes mellitus, hypertension and dyslipidaemia in a Mexican population. Int J Obes Relat Metab Disord 25, 1794–1799.CrossRefGoogle Scholar
15Bouguerra, R, Alberti, H, Smida, H, et al. . (2007) Waist circumference cut-off points for identification of abdominal obesity among the Tunisian adult population. Diabetes Obes Metab 9, 859–868.CrossRefGoogle ScholarPubMed
16Chen, L, Peeters, A, Magliano, DJ, et al. . (2007) Anthropometric measures and absolute cardiovascular risk estimates in the Australian Diabetes, Obesity and Lifestyle (AusDiab) Study. Eur J Cardiovasc Prev Rehabil 14, 740–745.CrossRefGoogle ScholarPubMed
17Craig, P, Colagiuri, S, Hussain, Z, et al. . (2007) Identifying cut-points in anthropometric indexes for predicting previously undiagnosed diabetes and cardiovascular risk factors in the Tongan population. Obes Res Clin Prac 1, 17–25.CrossRefGoogle ScholarPubMed
18Delavari, A, Forouzanfar, MH, Alikhani, S, et al. . (2009) First nationwide study of the prevalence of the metabolic syndrome and optimal cutoff points of waist circumference in the Middle East: the national survey of risk factors for noncommunicable diseases of Iran. Diabetes Care 32, 1092–1097.CrossRefGoogle ScholarPubMed
19Deshmukh, PR, Gupta, SS, Dongre, AR, et al. . (2006) Relationship of anthropometric indicators with blood pressure levels in rural Wardha. Indian J Med Res 123, 657–664.Google ScholarPubMed
20Dobbelsteyn, CJ, Joffres, MR, MacLean, DR, et al. . (2001) A comparative evaluation of waist circumference, waist-to-hip ratio and body mass index as indicators of cardiovascular risk factors. The Canadian Heart Health Surveys. Int J Obes Relat Metab Disord 25, 652–661.CrossRefGoogle ScholarPubMed
21Doi, Y, Ninomiya, T, Hata, J, et al. . (2009) Proposed criteria for metabolic syndrome in Japanese based on prospective evidence: the Hisayama study. Stroke 40, 1187–1194.CrossRefGoogle ScholarPubMed
22Esteghamati, A, Ashraf, H, Rashidi, A, et al. . (2008) Waist circumference cut-off points for the diagnosis of metabolic syndrome in Iranian adults. Diabetes Res Clin Pract 82, 104–107.CrossRefGoogle ScholarPubMed
23Flegal, KM (2007) Waist circumference of healthy men and women in the United States. Int J Obes (Lond) 31, 1134–1139.CrossRefGoogle ScholarPubMed
24Foucan, L, Hanley, J, Deloumeaux, J, et al. . (2002) Body mass index (BMI) and waist circumference (WC) as screening tools for cardiovascular risk factors in Guadeloupean women. J Clin Epidemiol 55, 990–996.CrossRefGoogle ScholarPubMed
25Gus, M, Cichelero, FT, Moreira, CM, et al. . (2009) Waist circumference cut-off values to predict the incidence of hypertension: an estimation from a Brazilian population-based cohort. Nutr Metab Cardiovasc Dis 19, 15–19.CrossRefGoogle ScholarPubMed
26Hadaegh, F, Zabetian, A, Sarbakhsh, P, et al. . (2009) Appropriate cutoff values of anthropometric variables to predict cardiovascular outcomes: 7.6 years follow-up in an Iranian population. Int J Obes (Lond) 33, 1437–1445.CrossRefGoogle Scholar
27Han, JH, Park, HS, Kim, SM, et al. . (2008) Visceral adipose tissue as a predictor for metabolic risk factors in the Korean population. Diabet Med 25, 106–110.CrossRefGoogle ScholarPubMed
28Hara, K, Matsushita, Y, Horikoshi, M, et al. . (2006) A proposal for the cutoff point of waist circumference for the diagnosis of metabolic syndrome in the Japanese population. Diabetes Care 29, 1123–1124.CrossRefGoogle ScholarPubMed
29Ho, SY, Lam, TH & Janus, ED (2003) Waist to stature ratio is more strongly associated with cardiovascular risk factors than other simple anthropometric indices. Ann Epidemiol 13, 683–691.CrossRefGoogle ScholarPubMed
30Huang, KC, Lee, MS, Lee, SD, et al. . (2005) Obesity in the elderly and its relationship with cardiovascular risk factors in Taiwan. Obes Res 13, 170–178.CrossRefGoogle ScholarPubMed
31Huxley, R, Barzi, F, Lee, CM, et al. . (2007) Waist circumference thresholds provide an accurate and widely applicable method for the discrimination of diabetes. Diabetes Care 30, 3116–3118.Google ScholarPubMed
32Ito, H, Nakasuga, K, Ohshima, A, et al. . (2003) Detection of cardiovascular risk factors by indices of obesity obtained from anthropometry and dual-energy X-ray absorptiometry in Japanese individuals. Int J Obes Relat Metab Disord 27, 232–237.CrossRefGoogle ScholarPubMed
33Kashihara, H, Lee, JS, Kawakubo, K, et al. . (2009) Criteria of waist circumference according to computed tomography-measured visceral fat area and the clustering of cardiovascular risk factors. Circ J 73, 1881–1886.CrossRefGoogle ScholarPubMed
34Kim, HK, Kim, CH, Park, JY, et al. . (2009) Lower waist-circumference cutoff point for the assessment of cardiometabolic risk in Koreans. Diabetes Res Clin Pract 85, 35–39.CrossRefGoogle ScholarPubMed
35Ko, GT, Chan, JC, Cockram, CS, et al. . (1999) Prediction of hypertension, diabetes, dyslipidaemia or albuminuria using simple anthropometric indexes in Hong Kong Chinese. Int J Obes Relat Metab Disord 23, 1136–1142.CrossRefGoogle ScholarPubMed
36Ko, GT, Liu, KH, So, WY, et al. . (2009) Cutoff values for central obesity in Chinese based on mesenteric fat thickness. Clin Nutr 28, 679–683.CrossRefGoogle ScholarPubMed
37Ko, GT & Tang, JS (2007) Waist circumference and BMI cut-off based on 10-year cardiovascular risk: evidence for ‘central pre-obesity’. Obesity (Silver Spring) 15, 2832–2839.CrossRefGoogle ScholarPubMed
38Lee, JS, Kawakubo, K, Mori, K, et al. . (2007) Effective cut-off values of waist circumference to detect the clustering of cardiovascular risk factors of metabolic syndrome in Japanese men and women. Diab Vasc Dis Res 4, 340–345.CrossRefGoogle ScholarPubMed
39Lee, SY, Park, HS, Kim, DJ, et al. . (2007) Appropriate waist circumference cutoff points for central obesity in Korean adults. Diabetes Res Clin Pract 75, 72–80.CrossRefGoogle ScholarPubMed
40Lin, WY, Lee, LT, Chen, CY, et al. . (2002) Optimal cut-off values for obesity: using simple anthropometric indices to predict cardiovascular risk factors in Taiwan. Int J Obes Relat Metab Disord 26, 1232–1238.CrossRefGoogle ScholarPubMed
41Mansour, AA & Al-Jazairi, MI (2007) Cut-off values for anthropometric variables that confer increased risk of type 2 diabetes mellitus and hypertension in Iraq. Arch Med Res 38, 253–258.CrossRefGoogle ScholarPubMed
42Mohan, V, Deepa, M, Farooq, S, et al. . (2007) Anthropometric cut points for identification of cardiometabolic risk factors in an urban Asian Indian population. Metabolism 56, 961–968.CrossRefGoogle Scholar
43Narisawa, S, Nakamura, K, Kato, K, et al. . (2008) Appropriate waist circumference cutoff values for persons with multiple cardiovascular risk factors in Japan: a large cross-sectional study. J Epidemiol 18, 37–42.CrossRefGoogle ScholarPubMed
44Narksawat, K, Podang, J, Punyarathabundu, P, et al. . (2007) Waist circumference, body mass index and health risk factors among middle aged Thais. Asia Pac J Public Health 19, 10–15.CrossRefGoogle ScholarPubMed
45Nilsson, G, Hedberg, P, Jonason, T, et al. . (2008) Waist circumference alone predicts insulin resistance as good as the metabolic syndrome in elderly women. Eur J Intern Med 19, 520–526.CrossRefGoogle ScholarPubMed
46Oka, R, Kobayashi, J, Yagi, K, et al. . (2008) Reassessment of the cutoff values of waist circumference and visceral fat area for identifying Japanese subjects at risk for the metabolic syndrome. Diabetes Res Clin Pract 79, 474–481.CrossRefGoogle ScholarPubMed
47Park, SH, Choi, SJ, Lee, KS, et al. . (2009) Waist circumference and waist-to-height ratio as predictors of cardiovascular disease risk in Korean adults. Circ J 73, 1643–1650.CrossRefGoogle ScholarPubMed
48Peixoto Mdo, R, Benicio, MH, Latorre Mdo, R, et al. . (2006) Waist circumference and body mass index as predictors of hypertension. Arq Bras Cardiol 87, 462–470.Google ScholarPubMed
49Pua, YH, Lim, CK & Ang, A (2006) Categorization of low cardiorespiratory fitness using obesity indices in non-smoking Singaporean women. Obesity (Silver Spring) 14, 1992–1999.CrossRefGoogle ScholarPubMed
50Pua, YH & Ong, PH (2005) Anthropometric indices as screening tools for cardiovascular risk factors in Singaporean women. Asia Pac J Clin Nutr 14, 74–79.Google ScholarPubMed
51Rush, EC, Crook, N & Simmons, D (2009) Optimal waist cutpoint for screening for dysglycaemia and metabolic risk: evidence from a Maori cohort. Br J Nutr 102, 786–791.CrossRefGoogle ScholarPubMed
52Sakurai, M, Takamura, T, Miura, K, et al. . (2009) Middle-aged Japanese women are resistant to obesity-related metabolic abnormalities. Metabolism 58, 456–459.CrossRefGoogle ScholarPubMed
53Sanchez-Castillo, CP, Velazquez-Monroy, O, Berber, A, et al. . (2003) Anthropometric cutoff points for predicting chronic diseases in the Mexican National Health Survey 2000. Obes Res 11, 442–451.CrossRefGoogle ScholarPubMed
54Sargeant, LA, Bennett, FI, Forrester, TE, et al. . (2002) Predicting incident diabetes in Jamaica: the role of anthropometry. Obes Res 10, 792–798.CrossRefGoogle ScholarPubMed
55Sato, A, Asayama, K, Ohkubo, T, et al. . (2008) Optimal cutoff point of waist circumference and use of home blood pressure as a definition of metabolic syndrome: the Ohasama study. Am J Hypertens 21, 514–520.CrossRefGoogle ScholarPubMed
56Seo, JA, Kim, BG, Cho, H, et al. . (2009) The cutoff values of visceral fat area and waist circumference for identifying subjects at risk for metabolic syndrome in elderly Korean: Ansan Geriatric (AGE) cohort study. BMC Public Health 9, 443.CrossRefGoogle ScholarPubMed
57Shimajiri, T, Imagawa, M, Kokawa, M, et al. . (2008) Revised optimal cut-off point of waist circumference for the diagnosis of metabolic syndrome in Japanese women and the influence of height. J Atheroscler Thromb 15, 94–99.CrossRefGoogle ScholarPubMed
58Shiwaku, K, Anuurad, E, Enkhmaa, B, et al. . (2005) Predictive values of anthropometric measurements for multiple metabolic disorders in Asian populations. Diabetes Res Clin Pract 69, 52–62.CrossRefGoogle ScholarPubMed
59Snehalatha, C, Viswanathan, V & Ramachandran, A (2003) Cutoff values for normal anthropometric variables in Asian Indian adults. Diabetes Care 26, 1380–1384.CrossRefGoogle ScholarPubMed
60Stevens, J, Couper, D, Pankow, J, et al. . (2001) Sensitivity and specificity of anthropometrics for the prediction of diabetes in a biracial cohort. Obes Res 9, 696–705.CrossRefGoogle Scholar
61Tabata, S, Yoshimitsu, S, Hamachi, T, et al. . (2009) Waist circumference and insulin resistance: a cross-sectional study of Japanese men. BMC Endocr Disord 9, 1.CrossRefGoogle ScholarPubMed
62Tseng, CH, Chong, CK, Chan, TT, et al. . (2010) Optimal anthropometric factor cutoffs for hyperglycemia, hypertension and dyslipidemia for the Taiwanese population. Atherosclerosis 210, 585–589.CrossRefGoogle ScholarPubMed
63Uzunlulu, M, Oguz, A, Aslan, G, et al. . (2009) Cut-off values for waist circumference in Turkish population: is there a threshold to predict insulin resistance? Turk Kardiyol Dern Ars 37, Suppl. 6, 17–23.Google Scholar
64Wang, F, Wu, S, Song, Y, et al. . (2009) Waist circumference, body mass index and waist to hip ratio for prediction of the metabolic syndrome in Chinese. Nutr Metab Cardiovasc Dis 19, 542–547.CrossRefGoogle ScholarPubMed
65Woo, J, Ho, SC, Yu, AL, et al. . (2002) Is waist circumference a useful measure in predicting health outcomes in the elderly? Int J Obes Relat Metab Disord 26, 1349–1355.CrossRefGoogle ScholarPubMed
66Yang, F, Lv, JH, Lei, SF, et al. . (2006) 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 25, 1030–1039.CrossRefGoogle ScholarPubMed
67Ye, Y, Bao, Y, Hou, X, et al. . (2009) Identification of waist circumference cutoffs for abdominal obesity in the Chinese population: a 7·8-year follow-up study in the Shanghai urban area. Int J Obes (Lond) 33, 1058–1062.CrossRefGoogle Scholar
68Nyamdorj, R, Pitkaniemi, J, Tuomilehto, J, et al. . (2010) Ethnic comparison of the association of undiagnosed diabetes with obesity. Int J Obes (Lond) 34, 332–339.CrossRefGoogle ScholarPubMed
69Zhou, B, Wu, Y, Zhao, L, et al. . (2001) Relationship of central obesity to cardiovascular risk factors and their clustering in middle aged Chinese populations. Chin J Cardiol 29, 70–73.Google Scholar
70Huxley, R, Mendis, S, Zheleznyakov, E, et al. . (2010) Body mass index, waist circumference and waist:hip ratio as predictors of cardiovascular risk – a review of the literature. Eur J Clin Nutr 64, 16–22.CrossRefGoogle ScholarPubMed
Figure 0

Table 1 Some characteristics of the sixty-one cited papers

Figure 1

Fig. 1 Optimal waist circumference cut-off values and population means by sex: ○, female (r 0·93); △, male (r 0·91). (- - -), Line of identity.

Figure 2

Fig. 2 Optimal waist circumference cut-off values and population means of (a) Chinese by sex: ⋄, female (r 0·96); △, male (r 0·84). Optimal waist circumference cut-off values and population means of (b) Japanese by sex: ⋄, female (r 0·92); △, male (r 0·58). (- - -), Line of identity.

Figure 3

Fig. 3 Optimal waist circumference cut-off values and population means of (a) males by outcome: ○, hyperglycaemia (r 0·93); △, dyslipidaemia (r 0·94);+, hypertension (r 0·94). Optimal waist circumference cut-off values and population means of (b) females by outcome: ○, hyperglycaemia (r 0·96); △, dyslipidaemia (r 0·92);+, hypertension (r 0·97). (- - -), Line of identity.

Figure 4

Appendix 1 Studies defining waist circumference cut-off values in adult populations