Figures
Abstract
Background
Excess adiposity is associated with cardiovascular disease (CVD) risk factors such as hypertension, diabetes mellitus and dyslipidemia. Amongst the various measures of adiposity, the best one to help predict these risk factors remains contentious. A novel index of adiposity, the Body Adiposity Index (BAI) was proposed in 2011, and has not been extensively studied in all populations. Therefore, the purpose of this study is to compare the relationship between Body Mass Index (BMI), Waist Circumference (WC), Waist-to-Hip Ratio (WHR), Waist-to-Height Ratio (WHtR), Body Adiposity Index (BAI) and CVD risk factors in the local adult population.
Methods and Findings
This is a cross sectional study involving 1,891 subjects (Chinese 59.1% Malay 22.2%, Indian 18.7%), aged 21–74 years, based on an employee health screening (2012) undertaken at a hospital in Singapore. Anthropometric indices and CVD risk factor variables were measured, and Spearman correlation, Receiver Operating Characteristic (ROC) curves and multiple logistic regressions were used. BAI consistently had the lower correlation, area under ROC and odd ratio values when compared with BMI, WC and WHtR, although differences were often small with overlapping 95% confidence intervals. After adjusting for BMI, BAI did not further increase the odds of CVD risk factors, unlike WC and WHtR (for all except hypertension and low high density lipoprotein cholesterol). When subjects with the various CVD risk factors were grouped according to established cut-offs, a BMI of ≥23.0 kg/m2 and/or WHtR ≥0.5 identified the highest proportion for all the CVD risk factors in both genders, even higher than a combination of BMI and WC.
Citation: Lam BCC, Koh GCH, Chen C, Wong MTK, Fallows SJ (2015) Comparison of Body Mass Index (BMI), Body Adiposity Index (BAI), Waist Circumference (WC), Waist-To-Hip Ratio (WHR) and Waist-To-Height Ratio (WHtR) as Predictors of Cardiovascular Disease Risk Factors in an Adult Population in Singapore. PLoS ONE 10(4): e0122985. https://doi.org/10.1371/journal.pone.0122985
Academic Editor: Pedro Tauler, University of the Balearic Islands, SPAIN
Received: November 27, 2014; Accepted: February 26, 2015; Published: April 16, 2015
Copyright: © 2015 Lam et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited
Data Availability: All relevant data are within the paper.
Funding: The authors have no support or funding to report.
Competing interests: The authors have declared that no competing interests exist.
Introduction
The prevalence of obesity in the world has risen to epidemic proportions [1]. This is of major concern as excess adiposity is strongly associated with cardiovascular disease (CVD) risk factors such as hypertension, diabetes mellitus and dyslipidemia [2,3]. Hence, a simple and effective measure of adiposity is needed for risk assessment in order to guide appropriate management and develop preventive strategies. However, this ‘best’ measure of adiposity to help predict these CVD risk factors remains contentious despite years of research.
Body Mass Index (BMI) is the widely used measure of obesity. However, the BMI is unable to differentiate between lean mass and fat mass, and hence, it is limited by differences in body adiposity for a given BMI across age, gender and ethnicity [4]. For example, the current definition of obesity based on BMI (BMI ≥30 kg/m2) may actually underestimate obesity among non-Caucasian populations, especially Asians [5]. In addition, the BMI does not consider body fat distribution, which is an important limitation since there are suggestions that the metabolic complications of obesity are more closely related to visceral adiposity than overall adiposity [2]. Hence, other measures of adiposity, which consider body fat distribution, like waist circumference (WC), waist-to-hip ratio (WHR) and waist-to-height ratio (WHtR) have been developed and studied. WC has been proposed to be the best amongst these measures, with excellent correlation with abdominal imaging and high association with CVD risk factors, especially diabetes [2,6,7]. However, WC does not account for differences in height, therefore, potentially over- and under-evaluating risk for tall and short individuals respectively [8]. Consequently, several researchers independently proposed the WHtR as an alternative to WC. This ratio has been shown to be a good indicator of abdominal adiposity, similar to WC [9] and recent systematic reviews and meta-analyses have supported the use of WHtR as a better predictor of CVD risk factors [8,10–12].
In 2011, the Body Adiposity Index (BAI) was proposed [13]. This is a composite index based on hip circumference and height (BAI = hip circumference (cm) divided by (height (m))1.5 minus 18), and was developed with the intention that this index would provide a direct estimate of percentage (%) body adiposity. From a population study of Mexican Americans, it was found that hip circumference (positive correlation) and height (negative correlation) were the most correlated variables with dual energy X-ray absorptiometry (DEXA)-derived % body adiposity. The final formula was derived to predict DEXA-derived % body adiposity, thus overcoming the limitation of the BMI in differentiating between fat and lean mass. This new index was then validated in a separate study of African Americans, and hence was suggested as a ‘better index of body adiposity by Bergman et al. [13]. However, validation (of the BAI) studies done in various populations with different ethnicities have shown consistently that the BAI tends to overestimate adiposity at lower percentage body fat (%BF) and underestimate adiposity at higher %BF [14]. Additionally, although the BAI seemed to correlate with DEXA-derived % adiposity better than the BMI when males and females were considered together, these studies showed that this was no longer the case when stratified by gender [14]. Nonetheless, excess adiposity as determined by %BF would theoretically predict cardiovascular disease and its risk factors better than indirect indices.
As mentioned, the best adiposity measure to help predict CVD risk factors has remained contentious. Although there are systematic reviews and meta-analyses that consistently support the case for WHtR, these reviews and meta-analyses tend to have more Asians than Caucasians [10,12], with subgroup analyses showing more positive results for WHtR in the Asian group than the Caucasian group [12]. This would be consistent with the findings by individual studies in Western populations, showing that WC is the better adiposity measure in predicting CVD risk factors [15–17] though some have continued to stress the importance of BMI [18,19]. Additionally, a systematic review and meta-analysis done exclusively on the Caucasian population, which included WHtR as a comparator, concluded that WC was more associated with CVD risk factors, and therefore recommended the use of WC in the clinic and in research studies [20].
Thus, further research is warranted in population groups and ethnicities where the various anthropometric measures including the recently proposed BAI have not been extensively analyzed and compared. Therefore, the purpose of this study is to compare the relationship between various anthropometric measures (BMI, WC, WHR, WHtR and the BAI) and CVD risk factors (i.e. hypertension, diabetes mellitus and dyslipidemia) in an adult population in Singapore, a South-East Asian country. The aims of the study are, firstly, to evaluate the performance of the BAI as a measure of adiposity, and secondly, to evaluate if WHtR is indeed a better predictor of CVD risk factors in this population.
Methods
Study Subjects
This was a retrospective study based on existing data, which were derived from standard procedures during an annual employee health screening conducted in 2012, and de-identified prior to data analysis. These subjects were employees from Khoo Teck Puat Hospital (KTPH), a regional hospital in Singapore. Only those who were 21 years old and above, and of Chinese, Malay or Indian ethnicity were considered for this study, yielding a sample of 1,945 subjects. Of these, subjects with incomplete and/or unverifiable data for the health screening questionnaire or study measurements were excluded (n = 54), resulting in a final sample of 1,891 subjects for analysis. No new data were collected (thus, there was no consent taking process in this study), and the data were collected and analyzed in a manner that subjects cannot be identified, directly or through identifiers linked to the subjects. Hence, this research involved no more than minimal risk to the individual, and the study proceeded only after obtaining approval from the National Healthcare Group Domain Specific Review Board (NHG DSRB, Reference Number: 2013/00598).
Health Screening Questionnaire
During the health screening, conducted in KTPH, participants were required to complete a questionnaire to determine their demographic characteristics and past medical history of hypertension, diabetes and dyslipidemia, including treatments, if any, for these conditions. ‘Smokers’ were defined as those who were currently smoking either regularly or occasionally and ‘regular exercise’ was defined as 150 minutes of moderate-intensity aerobic activity a week or 60 minutes of vigorous-intensity aerobic activity a week.
Health Screening Measurements
Well-trained examiners measured anthropometric indices, with participants wearing only minimal clothing with no footwear during measurements. Weight and height were measured, with the subject standing, to the nearest 0.1 kg and 1 cm, respectively, using a strain gauge scale, which was also equipped with a stadiometer (Seca 769, Hamburg, Germany). BMI was calculated in the standard way: weight (kg) divided by square of height (m). Waist and hip circumferences were measured to the nearest 0.1 cm using a flexible metric measuring tape with the subject in a standing position. Waist circumference was measured around the abdomen at the level of the umbilicus. Hip circumference was measured at the level of the maximum extension of the buttocks posteriorly in a horizontal plane. WHR was calculated as waist circumference (cm) divided by hip circumference (cm), while WHtR was calculated as waist circumference (cm) divided by height (cm). BAI was calculated as proposed by Bergman et al. [13]: hip circumference (cm) divided by (height (m))1.5 minus 18. Resting blood pressure (BP) was measured after five minutes in a seated position with an automated BP monitor using the oscillometric method (Omron HEM-7211, Kyoto, Japan). Any systolic BP found to be ≥140mmHg and/or diastolic BP ≥90mmHg was verified with a standard mercury sphygmomanometer after a two-minute interval.
Laboratory Analysis
Blood samples were collected on-site, with the subjects having fasted for at least 8 hours, and sent to an internationally certified laboratory located within the hospital. Fasting blood glucose (FBG) was measured quantitatively by the enzymatic reference method with hexokinase [21] while total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C) and triglycerides (TG) were assessed using standard enzymatic colorimetric method [22,23]. Low-density lipoprotein cholesterol (LDL-C) was estimated indirectly using the Friedwald formula [24]: LDL-C = TC-(HDL-C+(TG/5)) for subjects with TG levels <400 mg/dl. All samples were analyzed with an auto-analyzer (cobas 501, Roche, Basel, Switzerland).
Definition of CVD Risk Factors
Hypertension was defined as having one or more of the following: (1) a systolic BP ≥140 mmHg, (2) a diastolic BP ≥90 mmHg, (3) physician-diagnosed hypertension and (4) use of antihypertensive medication. Diabetes mellitus was defined as having one or more of the following: (1) FBG ≥126 mg/dl, (2) physician-diagnosed diabetes mellitus and (3) use of oral hypoglycemic agents. The cut-off points for dyslipidemia were plasma TC ≥240 mg/dl and/or use of medications to lower blood cholesterol for high TC, TG ≥200 mg/dl for high TG, HDL-C <40 mg/dl for low HDL-C, and LDL-C ≥160 mg/dl and/or use of medications to lower blood cholesterol for high LDL-C [25].
Statistical Analysis
All statistical analyses were performed using SPSS version 21 (SPSS Inc., Chicago, IL, USA). Continuous variables were tested for normality using the Kolmogorov-Smirnov test. Comparisons between males and females were performed using two independent samples t-test or the Mann-Whitney U test (as appropriate) for continuous and the chi-square test for categorical data. The relationship between anthropometric measures and CVD risk factors was first examined using Spearman’s correlation analysis. Receiver Operating Characteristic (ROC) analyses were then used to calculate the area under ROC curves (AUC) between each CVD risk factor and anthropometric measure, adjusted for age, gender, ethnicity, smoking status and physical activity status. The ROC was also used to identify cut-off values that best balanced sensitivity and specificity for the anthropometric measure with regards to the specific CVD risk factor, with respective sensitivity and specificity values reported. If multiple cut-off values were generated, the highest value is the one presented. Additionally, sensitivity and specificity based on established cut-off values for the various anthropometric measures were also explored using the ROC curves. Next, multiple logistic regression was used to evaluate the association between CVD risk factor and each standard deviation (SD) increase of the anthropometric measure adjusted for age, gender, ethnicity, smoking status and physical activity status. The effect of central obesity on overall obesity was also examined using multivariate models. Finally, in an attempt to delineate the clinical utility of BMI, WC and WHtR, grouping patients based on established cut off for BMI (≥23.0 kg/m2) [5], WC (≥80 cm for females, ≥90 cm for males) [26], and WHtR (≥0.5) [8], was done. All analyses were two-tailed, and a P value of <0.05 was considered to indicate statistical significance.
Results
Characteristics of the Study Subjects
The basic characteristics of the study population, stratified by gender, are shown in Table 1. 78.1% (n = 1,476) of the subjects were females, and the mean age of the study population was 35.7 years, with a mean BMI of 23.2 kg/m2 (minimum: 13.8 kg/m2, maximum: 40.9 kg/m2). The majority of the subjects was Chinese (n = 1,118, 59.1%) with the rest being Malay (n = 420, 22.2%) or Indian (n = 353, 18.7%). Males tended to have higher mean height, weight, waist circumference, hip circumference, BMI, WHR, and WHtR, but lower mean BAI (all P<0.001).
As for CVD risk factor variables, males tended to have higher median FBG (P<0.001) and TG level (P<0.001), higher mean systolic and diastolic BP (both P<0.001), TC level (P = 0.012), and LDL-C level (P<0.001) but lower mean HDL-C level (P<0.001). Correspondingly, the percentages of males with diabetes mellitus, hypertension, high TC, high TG, low HDL-C and high LDL-C were all significantly higher than the percentages in females (all P<0.05).
Correlations between Anthropometric Measures and CVD Risk Factor Variables
The Spearman’s correlation coefficients between the various anthropometric measures with the CVD risk factor variables, stratified by gender, are shown in Table 2. All the anthropometric measures correlated significantly with CVD risk factor variables, with BAI consistently having the lowest correlation coefficients when both genders were considered together. In this same scenario, measures of central adiposity, namely WC and WHtR, correlated best with the various CVD risk factor variables, with WC correlating the best for five out of the seven CVD risk factor variables (FBG, systolic and diastolic BP, TG level and HDL-C level). When genders were considered separately, measures of central adiposity (WC and WHtR) continued to correlate the best with CVD risk factor variables in males, while BMI (measure of overall adiposity) became the best correlated for three out of the seven CVD risk factor variables (systolic and diastolic BP, and HDL-C level) in females with WHtR correlating the best for the rest of the variables.
Association of Various Anthropometric Measures and CVD Risk Factors using ROC Curve Analyses
The area under ROC curves (AUC) between each CVD risk factor and anthropometric measure, after adjusting for age, gender, ethnicity, smoking status and physical activity status, are shown in Table 3. WC had the largest AUC for four out of the six CVD risk factors (diabetes mellitus, high TC, high TG, and high LDL-C), while BMI had the largest AUC for the other two (hypertension and low HDL-C). However, the differences in the AUC for the various anthropometric measures were often small with overlapping 95% confidence intervals (CIs).
The optimal cut-off points that best balanced sensitivity and specificity for the various CVD risk factors of BMI, WC and WHtR, in males and females, are shown in Table 4a. The cut off values in males ranged from 23.5 kg/m2 to 23.9 kg/m2 for BMI, 84.5 cm to 88.5 cm for WC and 0.479 to 0.501 for WHtR; and in females, from 23.5 kg/m2 to 27.7 kg/m2 for BMI, 79.5 cm to 83.5 cm for WC and 0.460 to 0.531 for WHtR. When cut off points were based on recommended cut-offs for the Asian population for BMI (high risk: BMI ≥27.5 kg/m2) and WC (≥80 cm for females, ≥90 cm for males), and the proposed cut off of 0.5 for WHtR, WC and WHtR (measures of central adiposity) had better sensitivities compared to BMI, while BMI (of 27.5 kg/m2) had the highest specificities, in both genders (Table 4b). Comparing between WC (≥80 cm for females, ≥90 cm for males) and WHtR (0.5) only, WHtR in general had slightly better sensitivities (except for low HDL-C in males and females, and low LDL-C in males), while WC had generally better specificities (except for low HDL-C in males and females, and low LDL-C in males).
Odds Ratios of CVD Risk Factors by Anthropometric Measures
Multivariate-adjusted odds ratios (ORs) of each CVD risk factor with each SD increase of the anthropometric measure are shown in Tables 5 and 6. In the first model, the ORs were adjusted for age, gender, ethnicity, smoking status and physical activity status, and this showed that each SD increase of WC had the highest ORs for all the CVD risk factors amongst all the anthropometric measures (Table 5a), with the OR of high TG being significantly higher for each SD increase of WC (2.78, 95% CI: 2.13–3.61) than for each SD increase of WHR (1.60, 95% CI: 1.25–2.05). However, the rest of the ORs were not considered significantly different from each other due to the overlapping 95% confidence intervals. In the second model, adjustment for BMI was added into the model, and this showed that each SD increase of WC continued to have significant ORs for diabetes mellitus, high TC, high TG and high LDL-C, with WC having the highest OR for all except high LDL-C (WHtR had the highest OR) (Table 5b). Comparing the ORs between WC and WHtR in this model, they are not significantly different except for high TG where each SD increase of WC still had a significant OR (2.00, 95% CI: 1.21–3.30) whereas each SD increase of WHtR became non-significant (1.61, 95% CI: 0.99–2.63).
Finally, when adjustment for WC or WHtR was added into the model, each SD increase of BMI still had significant ORs for hypertension and low HDL-C (Table 6).
Proportion of Subjects Identified Based on Established/Proposed Cut Offs for BMI, WC and WHtR for Each CVD Risk Factor Group
The proportion of those who would have been identified based on established cut offs for BMI (≥23.0 kg/m2), WC (≥80 cm for females, ≥90 cm for males), and WHtR (≥0.5), singly or in combination, are shown in Table 7. Based on a WHtR of ≥0.5, the proportion of those who would have been identified was consistently higher than that for a WC of ≥80 cm in females, and WC ≥90 cm in males, for all CVD risk factors. When a combination of measures were included in the evaluation, a BMI of ≥23.0 kg/m2 and/or WHtR ≥0.5 identified the highest proportion for all the CVD risk factors in both genders, even higher than a combination of BMI and WC.
Discussion
The first aim was to evaluate the utility of the recently proposed BAI as a measure of adiposity and the analyses reveal a few observations with regards to this. Firstly, when both genders were considered together, BAI consistently correlated the poorest with the various CVD risk factor variables amongst all the anthropometric measures being studied. Although BAI seemed to correlate better when stratified by gender, the correlation coefficients were generally lower than for WC and WHtR in males, and lower than for BMI and WHtR in females.
Secondly, based on adjusted AUCs between each CVD risk factor and the anthropometric measure, the BAI had consistently lower AUC values than BMI, WC and WHtR, although the differences were often small, with overlapping 95% confidence intervals. Thirdly, after adjusting for BMI, BAI did not further increase the odds of CVD risk factors unlike WC and WHtR, suggesting that BAI behaved similarly to BMI and hence, BAI has no additional value after taking BMI into account. This is consistent with the fact that although BAI attempts to give an estimation of %BF, it does not distinguish the distribution of that body adiposity, and thus would function as an overall measure of adiposity, like the BMI.
Hence, this study purports that the utility of the BAI would be that of an overall measure of adiposity like the BMI, and there is no suggestion that BAI would be better than the well-established BMI. Moreover, with validation studies consistently showing that the BAI tends to over- or underestimate adiposity at extreme ends of %BF, and that the BAI does not correlate with dual energy X-ray absorptiometer (DEXA)-derived % adiposity better than BMI when stratified by gender [14], the BAI is unlikely to be a better overall measure of adiposity than the BMI.
The second aim of this study was to evaluate if the WHtR is indeed a better predictor of CVD risk factors such as hypertension, diabetes mellitus and dyslipidemia in an adult population in Singapore, as suggested by recent systematic reviews and meta-analyses [8,10–12]. From the initial analyses of this study, it would seem that WC and WHtR (measures of central adiposity) are better than BMI (measure of overall adiposity) based on correlation coefficients, estimated AUC and OR values, with WC generally having the higher estimated values compared to WHtR. This is consistent with a number of studies concluding that measures of central adiposity such as WC is more closely associated with CVD risk factors than BMI [15,16,27–29], suggesting that the complications of obesity are more closely related to the distribution rather than the absolute degree of adiposity per se. However, it must be noted that the 95% confidence intervals of these AUC and OR values were all overlapping in this study, and hence, BMI, WC and WHtR could all be considered comparable in their association with CVD risk factors, which is consistent with some of the other studies as well [30–33].
When further analyses such as additional multiple regression models to investigate the effect of central adiposity on overall adiposity were done, it was found that each SD increase of WC continued to have significant ORs for diabetes mellitus, high TC, high TG and high LDL-C, after adjusting for BMI. This was similarly observed for WHtR (except for high TG where the OR became non-significant), suggesting that measures of central adiposity continue to increase the odds of these CVD risk factors, independent of BMI. As for hypertension and low HDL, the ORs for each SD increase of WC/WHtR became non-significant after accounting for BMI, suggesting that the association of WC and WHtR with hypertension and low HDL may not be independent of BMI. This is confirmed when WC or WHtR was included in the model, and each SD increase of BMI continued to have significant ORs, that is, the increase in odds is now independent of WC or WHtR. Other studies have made similar observations [34,35] and this is understood to be due to the BMI better reflecting body volume and mass, which is associated with blood viscosity and blood volume, and hence more closely related to BP, while measures of central adiposity are good indicators of visceral adiposity and hence more closely associated with diabetes [34].
Hence, this study suggests that using a combination of measures, one that includes a measure of general adiposity and a measure of central adiposity, would be more appropriate in the identifying of CVD risk factors, and this is supported by the observation that a BMI of ≥23.0 kg/m2 and/or WHtR ≥0.5 yielded the highest proportion who would be identified amongst those with the various CVD risk factors. For example, relying on a WC of ≥80 cm for females and ≥90cm for males would have identified 74.5% (70/94) of the people (males and females) with diabetes, while a combination of BMI ≥23.0 kg/m2 and/or WHtR ≥0.5 would have identified 92.6% (87/94), even more than a combination of BMI ≥23.0 kg/m2 and/or WC ≥80 cm for females, and ≥90 cm for males (86.2%, 81/94). This is consistent with this study’s finding that WHtR ≥0.5 was found to be generally more sensitive (based on ROC analysis) than WC ≥80 cm in females, and WC ≥90 cm in males, an attribute that is more desirable for a screening test since a diagnostic test (measurement of blood pressure or blood test) would still be required to confirm the diagnosis and rule the disease in. Thus, WHtR may have better utility than WC as a measure of obesity to screen for CVD risk factors.
In addition, unlike other indices of abdominal adiposity, a universal cut-off value for WHtR has been suggested in several studies, obviating the need for age-, sex- and ethnic-specific cut-off values [8,36]. In this study of an adult population in Singapore, the optimal cut off value found was 0.479 to 0.501 for males and 0.460 to 0.531 for females, approximating to the 0.5 proposed in several studies [8,36]. As an individual’s height is relatively fixed, the WHtR would allow individuals to have individualized cut offs for waist circumference while affording a simple public health message of getting the public to visit their doctors for a medical evaluation for CVD risk factors if their waist circumference is more than half their height, especially if their BMI is ≥23.0 kg/m2.
This study has important limitations. First, this study population (employees from a hospital) may not be entirely representative of the general adult Asian population in Singapore. Second, while the sample size is believed to be adequate for analysis as a whole, the small numbers of Malay and Indian prevented analyses stratified by ethnic groups. Hence, an approach to adjust for ethnic groups was used instead. We believe this approach allows for an overall conclusion in a multi-ethnic population, yet takes into account the effect of ethnicity on the relationship between measures of obesity and CVD risk factors. Third, age, gender, ethnicity, smoking and physical activity status have been associated with the development of CVD risk factors [26,37,38]. While these have been accounted for by the adjustment for these factors in the analysis, other factors known to influence the development of CVD risk factors such as dietary habits and family history have not been accounted for. Fourth, the findings of this cross-sectional study are not conclusive evidence of a causal relation of the various anthropometric measures with CVD risk factors. Further studies of a prospective nature would be required to confirm the findings of this study.
In conclusion, BAI may function as a measure of overall adiposity but it is unlikely to be better than BMI. While BMI, WC and WHtR seem comparable in their association with CVD risk factors, a combination of BMI and WHtR could have the best clinical utility in identifying patients with CVD risk factors in an adult population in Singapore.
Acknowledgments
We acknowledge the support of the Clinical Research Unit, Khoo Teck Puat Hospital, Singapore. We also want to thank the Outreach team from the Department of Family and Community Medicine, Khoo Teck Puat Hospital, Singapore, for the assistance rendered in acquiring the data and providing the necessary support where required in the conducting of this study.
Author Contributions
Conceived and designed the experiments: BCCL MTKW. Performed the experiments: BCCL MTKW. Analyzed the data: BCCL GCHK CC. Contributed reagents/materials/analysis tools: BCCL MTKW SJF. Wrote the paper: BCCL GCHK CC SJF MTKW. Supervision of study: BCCL MTKW SJF.
References
- 1. WHO (2000) Obesity: preventing and managing the global epidemic. Report of a WHO consultation. World Health Organ Tech Rep Ser 894: i–xii, 1–253. pmid:11234459
- 2. Cornier MA, Despres JP, Davis N, Grossniklaus DA, Klein S, Lamarche B, et al. (2011) Assessing adiposity: a scientific statement from the American Heart Association. Circulation 124: 1996–2019. pmid:21947291
- 3. Guh DP, Zhang W, Bansback N, Amarsi Z, Birmingham CL, Anis AH (2009) The incidence of co-morbidities related to obesity and overweight: a systematic review and meta-analysis. BMC Public Health 9: 88. pmid:19320986
- 4. Jackson AS, Stanforth PR, Gagnon J, Rankinen T, Leon AS, Rao DC, et al. (2002) The effect of sex, age and race on estimating percentage body fat from body mass index: The Heritage Family Study. Int J Obes Relat Metab Disord 26: 789–796. pmid:12037649
- 5. WHO (2004) Appropriate body-mass index for Asian populations and its implications for policy and intervention strategies. The Lancet 363: 157–163. pmid:14726171
- 6. Nyamdorj R, Qiao Q, Lam TH, Tuomilehto J, Ho SY, Pitkaniemi J, et al. (2008) BMI compared with central obesity indicators in relation to diabetes and hypertension in Asians. Obesity (Silver Spring) 16: 1622–1635. pmid:18421260
- 7. InterAct C, Langenberg C, Sharp SJ, Schulze MB, Rolandsson O, Overved K, et al. (2012) Long-term risk of incident type 2 diabetes and measures of overall and regional obesity: the EPIC-InterAct case-cohort study. PLoS Med 9: e1001230. pmid:22679397
- 8. Browning LM, Hsieh SD, Ashwell M (2010) A systematic review of waist-to-height ratio as a screening tool for the prediction of cardiovascular disease and diabetes: 0.5 could be a suitable global boundary value. Nutr Res Rev 23: 247–269. pmid:20819243
- 9. Ashwell M, Cole TJ, Dixon AK (1996) Ratio of waist circumference to height is strong predictor of intra-abdominal fat. Bmj 313: 559–560. pmid:8790002
- 10. Lee CM, Huxley RR, Wildman RP, Woodward M (2008) Indices of abdominal obesity are better discriminators of cardiovascular risk factors than BMI: a meta-analysis. J Clin Epidemiol 61: 646–653. pmid:18359190
- 11. Ashwell M, Gunn P, Gibson S (2012) Waist-to-height ratio is a better screening tool than waist circumference and BMI for adult cardiometabolic risk factors: systematic review and meta-analysis. Obes Rev 13: 275–286. pmid:22106927
- 12. Savva SC, Lamnisos D, Kafatos AG (2013) Predicting cardiometabolic risk: waist-to-height ratio or BMI. A meta-analysis. Diabetes Metab Syndr Obes 6: 403–419. pmid:24179379
- 13. Bergman RN, Stefanovski D, Buchanan TA, Sumner AE, Reynolds JC, Sebring NG, et al. (2011) A better index of body adiposity. Obesity (Silver Spring) 19: 1083–1089. pmid:21372804
- 14. Lam BC, Lim SC, Wong MT, Shum E, Ho CY, Bosco JI, et al. (2013) A method comparison study to validate a novel parameter of obesity, the body adiposity index, in Chinese subjects. Obesity (Silver Spring) 21: E634–639. pmid:23630126
- 15. Janssen I, Katzmarzyk PT, Ross R (2004) Waist circumference and not body mass index explains obesity-related health risk. Am J Clin Nutr 79: 379–384. pmid:14985210
- 16. Menke A, Muntner P, Wildman RP, Reynolds K, He J (2007) Measures of adiposity and cardiovascular disease risk factors. Obesity (Silver Spring) 15: 785–795. pmid:17372330
- 17. Brenner DR, Tepylo K, Eny KM, Cahill LE, El-Sohemy A (2010) Comparison of body mass index and waist circumference as predictors of cardiometabolic health in a population of young Canadian adults. Diabetol Metab Syndr 2: 28. pmid:20459858
- 18. Huerta JM, Tormo MJ, Chirlaque MD, Gavrila D, Amiano P, Arriola L, et al. (2013) Risk of type 2 diabetes according to traditional and emerging anthropometric indices in Spain, a Mediterranean country with high prevalence of obesity: results from a large-scale prospective cohort study. BMC Endocr Disord 13: 7. pmid:23388074
- 19. Schneider HJ, Glaesmer H, Klotsche J, Bohler S, Lehnert H, Zeiher AM, et al. (2007) Accuracy of anthropometric indicators of obesity to predict cardiovascular risk. J Clin Endocrinol Metab 92: 589–594. pmid:17105840
- 20. van Dijk SB, Takken T, Prinsen EC, Wittink H (2012) Different anthropometric adiposity measures and their association with cardiovascular disease risk factors: a meta-analysis. Neth Heart J 20: 208–218. pmid:22231153
- 21.
Tietz NW (1995) Clinical guide to laboratory tests. United States of America: W.B. Saunders Comapny.
- 22. Allain CC, Poon LS, Chan CS, Richmond W, Fu PC (1974) Enzymatic determination of total serum cholesterol. Clin Chem 20: 470–475. pmid:4818200
- 23. Sugiuchi H, Uji Y, Okabe H, Irie T, Uekama K, Kayahara N, et al. (1995) Direct measurement of high-density lipoprotein cholesterol in serum with polyethylene glycol-modified enzymes and sulfated alpha-cyclodextrin. Clin Chem 41: 717–723. pmid:7729051
- 24. Friedewald WT, Levy RI, Fredrickson DS (1972) Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem 18: 499–502. pmid:4337382
- 25.
National Cholesterol Education Program (NCEP) Expert Panel on Detection E, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III) (2002) 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 106: 3143.
- 26. Alberti KG, Zimmet P, Shaw J (2007) International Diabetes Federation: a consensus on Type 2 diabetes prevention. Diabet Med 24: 451–463. pmid:17470191
- 27. Zhu S, Wang Z, Heshka S, Heo M, Faith MS, Heymsfield SB, et al. (2002) 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 76: 743–749. pmid:12324286
- 28. Dong X, Liu Y, Yang J, Sun Y, Chen L (2011) Efficiency of anthropometric indicators of obesity for identifying cardiovascular risk factors in a Chinese population. Postgrad Med J 87: 251–256. pmid:21273363
- 29. Park SH, Choi SJ, Lee KS, Park HY (2009) Waist circumference and waist-to-height ratio as predictors of cardiovascular disease risk in Korean adults. Circ J 73: 1643–1650. pmid:19638708
- 30. He YH, Chen YC, Jiang GX, Huang HE, Li R, Li XY, et al. (2012) Evaluation of anthropometric indices for metabolic syndrome in Chinese adults aged 40 years and over. Eur J Nutr 51: 81–87. pmid:21479941
- 31. Paniagua L, Lohsoonthorn V, Lertmaharit S, Jiamjarasrangsi W, Williams MA (2008) Comparison of waist circumference, body mass index, percent body fat and other measure of adiposity in identifying cardiovascular disease risks among Thai adults. Obesity Research & Clinical Practice 2: 215–223.
- 32. Nyamdorj R, Qiao Q, Soderberg S, Pitkaniemi J, Zimmet P, Shaw J, et al. (2008) Comparison of body mass index with waist circumference, waist-to-hip ratio, and waist-to-stature ratio as a predictor of hypertension incidence in Mauritius. J Hypertens 26: 866–870. pmid:18398327
- 33. Vazquez G, Duval S, Jacobs DR Jr., Silventoinen K (2007) Comparison of body mass index, waist circumference, and waist/hip ratio in predicting incident diabetes: a meta-analysis. Epidemiol Rev 29: 115–128. pmid:17494056
- 34. Hou X, Lu J, Weng J, Ji L, Shan Z, Liu J, et al. (2013) Impact of waist circumference and body mass index on risk of cardiometabolic disorder and cardiovascular disease in Chinese adults: a national diabetes and metabolic disorders survey. PLoS One 8: e57319. pmid:23520466
- 35. Huxley R, James WP, Barzi F, Patel JV, Lear SA, Suriyawongpaisal P, et al. (2008) Ethnic comparisons of the cross-sectional relationships between measures of body size with diabetes and hypertension. Obes Rev 9 Suppl 1: 53–61. pmid:18307700
- 36. Pua YH, Ong PH (2005) Anthropometric indices as screening tools for cardiovascular risk factors in Singaporean women. Asia Pac J Clin Nutr 14: 74–79. pmid:15734711
- 37. Hughes K, Choo M, Kuperan P, Ong CN, Aw TC (1998) Cardiovascular risk factors in relation to cigarette smoking: a population-based survey among Asians in Singapore. Atherosclerosis 137: 253–258. pmid:9622268
- 38. Albarwani S, Al-Siyabi S, Tanira MO (2014) Prehypertension: Underlying pathology and therapeutic options. World J Cardiol 6: 728–743. pmid:25228952