Hostname: page-component-78c5997874-g7gxr Total loading time: 0 Render date: 2024-11-14T23:00:27.618Z Has data issue: false hasContentIssue false

Comparison of body fat estimation using waist:height ratio using different ‘waist’ measurements in Australian adults

Published online by Cambridge University Press:  01 November 2008

Masaharu Kagawa*
Affiliation:
ATN Centre for Metabolic Fitness, School of Human Movement Studies, Institute of Health and Biomedical Innovation, Queensland University of Technology, 60 Musk Avenue, Kelvin Grove, Brisbane, Qld 4059, Australia Institute of Health and Biomedical Innovation, Queensland University of Technology, Brisbane, Australia
Nuala M. Byrne
Affiliation:
ATN Centre for Metabolic Fitness, School of Human Movement Studies, Institute of Health and Biomedical Innovation, Queensland University of Technology, 60 Musk Avenue, Kelvin Grove, Brisbane, Qld 4059, Australia Institute of Health and Biomedical Innovation, Queensland University of Technology, Brisbane, Australia
Andrew P. Hills
Affiliation:
ATN Centre for Metabolic Fitness, School of Human Movement Studies, Institute of Health and Biomedical Innovation, Queensland University of Technology, 60 Musk Avenue, Kelvin Grove, Brisbane, Qld 4059, Australia Institute of Health and Biomedical Innovation, Queensland University of Technology, Brisbane, Australia
*
*Corresponding author: Dr Masaharu Kagawa, fax +61 7 3138 6030, email m.kagawa@qut.edu.au
Rights & Permissions [Opens in a new window]

Abstract

The objective of the present study was to determine differences in predicting total and regional adiposity using the waist:height ratio (WHtR) calculated using different ‘waist’ measurements. Body composition of ninety-five males and 121 female Australian adults (aged 20 years and above) was measured using dual-energy X-ray absorptiometry. The WHtR was calculated using: (1) the narrowest point between the lower costal border and the top of the iliac crest (WHtR-W), and (2) at the level of the umbilicus (WHtR-A). Relationships between calculated WHtR and measured body composition, such as percentage body fat (%BF) and percentage trunk fat (%TF) were determined. Values obtained from WHtR-A were significantly greater than WHtR-W in both groups (P < 0·05). While no correlation differences between WHtR-W and WHtR-A in relation to body composition variables were observed, females showed significantly lower correlation with lean mass compared with BMI. Regression analyses showed that neither WHtR had an age influence on %TF estimation. Estimated %BF and %TF were comparable for both WHtR and also with estimated values using a BMI of 25 kg/m2. Sensitivity of excess %BF and %TF increased by using WHtR-A, particularly in females. In conclusion, the umbilicus measurement may be better than using the narrowest site in the WHtR calculation, particularly in females. To improve the screening ability of the WHtR and make comparisons between studies easier there may be a need to standardise the measurement location. Further studies are recommended to confirm the findings across different ethnic groups.

Type
Full Papers
Copyright
Copyright © The Authors 2008

Being overweight or obese is a contributing factor to the metabolic syndrome(Reference Alberti, Zimmet and Shaw1, Reference Chew, Gan and Watts2), increases the risk of CVD, type 2 diabetes mellitus, and also a number of cancers(Reference Alberti and Zimmet3Reference Grundy, Brewer and Cleeman5). Previous studies have suggested that individuals with a large accumulation of body fat in the abdominal region are at greater risk of development of the metabolic syndrome(Reference Despres and Lemieux6Reference Pi-Sunyer8).

Waist circumference (WC) is a clinically viable technique that has been employed to determine abdominal fat deposition(Reference Bei-Fan9Reference Welborn, Dhaliwal and Bennett11). While WC cannot provide a precise quantification of fat deposition in the region, it is time- and cost-efficient and positively correlated with visceral abdominal fat accumulation obtained from both magnetic resonance imaging and computer tomography (CT) scans(Reference Janssen, Heymsfield, Allison, Kotler and Ross12Reference Brambilla, Bedogni, Moreno, Goran, Gutin, Fox, Peters, Barbeau, De Simone and Pietrobelli14). However, because WC is an absolute value, the measurement is affected by the size of trunk which varies according to age, sex and ethnicity(Reference James15). Although WC has been accepted by international organisations such as the International Diabetes Federation as a diagnostic criteria of metabolic complications(Reference Alberti, Zimmet and Shaw1), its cut-off values are not always applicable to the entire population (for example, to children with possible metabolic syndrome risks). In addition, ‘waist’ circumference has a number of different definitions, including ‘the halfway point between the lower border of the ribs and the iliac crest’(Reference Dalton, Cameron, Zimmet, Shaw, Jolley, Dunstan and Welborn10), ‘at the narrowest point between the lower costal (10th rib) border and the top of the iliac crest’(Reference Marfell-Jones, Olds, Stewart and Carter16), and ‘at the level of umbilicus’ (which is more appropriately classified as the abdominal circumference (AC))(17). The presence of multiple definitions of the WC provides for both inconsistencies in the measurement protocol and subsequent difficulty in comparisons between studies. A recent study reported that the ‘waist’ measurement taken at the narrowest point reflects CVD risks better compared with at the umbilicus level in females(Reference Willis, Slentz, Houmard, Johnson, Duscha, Aiken and Kraus18). However, the study showed no significant correlation differences between ‘waist’ sites and the vast majority of metabolic biomarkers as well as visceral adipose tissue (VAT) obtained from CT. This indicates a further need to determine if either ‘waist’ site is superior to the other, in order to standardise the ‘waist’ measurement protocol.

As an alternative to the WC, the waist:height ratio (WHtR; waist (cm)/height (cm); also called the index of central obesity) has been suggested as a potentially useful index to determine abdominal fat deposition(Reference Bosy-Westphal, Geisler, Onur, Korth, Selberg, Schrezenmeir and Müller19Reference Parikh, Joshi, Menon and Shah22). As the WHtR adjusts for the height of an individual it has been suggested that the same cut-off point can be used to screen for health risks in different populations that vary in age and sex(Reference Ashwell and Hsieh23, Reference McCarthy and Ashwell24). Previous studies using blood assays have suggested that the cut-off point of 0·5 may be appropriate to determine metabolic complications in both adults and children(Reference Hsieh, Yoshinaga and Muto21, Reference Parikh, Joshi, Menon and Shah22, Reference Lin, Lee, Chen, Lo, Hsia, Liu, Lin, Shau and Huang25Reference Schneider, Glaesmer, Klotsche, Bohler, Lehnert, Zeiher, März, Pittrow, Stalla and Wittchen28). On the other hand, studies that examined a relationship between WHtR and actual body composition of the study groups are limited(Reference Parikh, Joshi, Menon and Shah22, Reference Kagawa, Hills and Binns29). Furthermore, there has been no study that determined the impact of using different WC procedures in the calculation of the WHtR. For overweight and obese individuals who are at greater metabolic complication risks, WC using some definitions may be inappropriate as it is common to have no identifiable narrowing in their trunk. If the WHtR has the potential to be used as a universal screening method for abdominal obesity there is a need to determine the most appropriate WC site that best reflects one's body composition as indicated in percentage body fat (%BF) and abdominal fat values in each sex.

The present study aimed to compare WHtR values obtained using two commonly used WC measurement approaches and their relationships with total body and trunk fat deposition using a dataset of Australian adults whose body composition was measured by dual-energy X-ray absorptiometry (DEXA). The present study used trunk fat instead of abdominal fat or VAT. This is because trunk fat is also associated with a number of metabolic biomarkers, such as TAG and cholesterol in both adults and children(Reference Teixeira, Sardinha, Going and Lohman30, Reference Van Pelt, Evans, Schechtman, Ehsani and Kohrt31), indicating its usefulness in assessing risk of obesity-related metabolic complications associated with abdominal obesity. Also, trunk fat can be assessed by DEXA, a more convenient and economical method compared with VAT assessed using CT or magnetic resonance imaging. Therefore, clarification of the relationship between WHtR and trunk fat accumulation, as well as its association with %BF, may increase the future application of WHtR in prevention strategies.

Methods

A body composition database of adult Australian males (n 95) and females (n 121) aged above 20 years was used in the present study. Participants were volunteers who participated in body composition assessment studies using DEXA at the School of Human Movement Studies, Queensland University of Technology. Overweight and obese individuals were also included if they were not medicated and had no medical history that influenced their daily lifestyle including diet and physical activity levels. Studies were approved by the Human Research Ethics Committee of Queensland University of Technology and adhered to the principles of medical research established by the National Health and Medical Research Council(32). Each participant signed a written informed consent form in which the purpose of the study was explained and the confidentiality of results guaranteed.

Dual-energy X-ray absorptiometry

The DEXA method is based on a three-compartment model that differentiates bone, lean and soft tissues from attenuation of two X-ray beams. The method provides consistent results with other commonly used techniques and is considered as one of the reliable methods to estimate %BF(Reference Heymsfield, Lichtman, Baumgartner, Wang and Kamen33Reference Glickman, Marn, Supiano and Dengel36). Whole-body and trunk lean and fat tissues were determined using DEXA measures (DPX-L; Lunar Radiation Corp., Madison, WI, USA). All scans were analysed with ADULT software, version 3.6 (Lunar Radiation Corp.) which provides the total mass, ratio of soft tissue attenuations, and bone mineral mass for the isolated regions. The ratio of soft tissue attenuation for each region was used to divide bone mineral-free tissue of the extremities into fat and lean components. From the obtained values, fat and lean mass in the trunk region and %BF were determined. In addition, the proportion of fat in the trunk region (%TF =  trunk fat/(trunk lean + trunk fat) × 100) was calculated to determine sex differences in fat accumulation pattern. All measurements were conducted at the School of Human Movement Studies, Queensland University of Technology, by an accredited technician.

Anthropometry

Anthropometric measurements included in the analyses were height, weight and three circumferences (waist, abdominal and hip). WC was measured at the narrowest point between the lower costal border and the top of the iliac crest. AC was measured at the level of the umbilicus and the hip circumference was measured at the greatest posterior protuberance. From the anthropometric measurements, BMI and WHtR using WC (WHtR-W) and AC (WHtR-A) were calculated. In addition, hip circumference measurements were used to compare sex differences in fat distribution pattern.

All statistical analyses were conducted using the SPSS (version 14.0.0, 2005; SPSS, Inc., Chicago, IL, USA) statistical package. The independent t test was used to assess sex differences in body composition. Relationships between body composition obtained from DEXA and WHtR using different ‘waist’ values were assessed using Spearman's correlation coefficients. Observed correlation coefficients were compared using a test for equal correlations, which is a ratio that uses Fisher's Z transformation in the numerator and the square root of the sum of the variances in the denominator(Reference Cohen and Cohen37). The generalised linear modelling analyses were conducted using %BF and %TF as dependent variables, sex as a fixed factor, BMI, WHtR using different ‘waist’ measures and age as covariate factors. Results were presented together with adjusted R 2 (R 2adj) and standard error of estimates. In addition, cross-tabulation was conducted in order to determine sensitivities and specificities of each index using %BF and %TF values using 20 % for males and 30 % for females as cut-off points. The cut-off points were decided based on a previous study that stated that proliferation of adipose tissue cells begins at these values in adults(Reference Huenemann, Hampton, Shapiro and Behnke38) and a study that indicated the possible comparability between %TF and %BF of individuals(Reference Kagawa, Hills and Binns29).

Results

Results of body composition assessments using DEXA and anthropometry are shown in Table 1. While males were significantly (P < 0·01) greater in body size (i.e. height and body mass) and upper-body circumferences (i.e. waist and umbilicus), females displayed significantly greater body fat deposition (%BF and %TF). Both sexes showed significant differences in WC and AC values (P < 0·01) but females showed greater difference between the two circumferences (6·6 cm) compared with males (3·0 cm). Males also showed significantly (P < 0·01) greater WHtR-W compared with females but no sex difference was observed in BMI or WHtR-A. In the present study 71·6 % of males and 60·3 % of females had BMI values equal or greater than 25 kg/m2. Also 71·6 % of males and 50·4 % of females had WHtR-W equal or greater than 0·5. Using WHtR-A, this proportion increased to 82·1 % in males and 70·2 % in females.

Table 1 Physical characteristics of male and female subjects

(Mean values, ranges and standard deviations)

WHtR-W, waist:height ratio using waist circumference; WHtR-A, waist:height ratio using abdominal circumference; %BF, percentage body fat; DEXA, dual-energy X-ray absorptiometry; %TF, percentage trunk fat.

*  Mean value is significantly different from that for males (P < 0·01).

For details of procedures, see Methods.

Table 2 presents a comparison of the WHtR values obtained from WC and AC as well as correlations between anthropometric indices (i.e. WHtR and BMI) and body composition results obtained from DEXA. A comparison of two WHtR values showed that the WHtR values calculated from AC were significantly (P < 0·05) greater than the values for WC in both sexes and the difference was greater in females. Both BMI and WHtR using different ‘waist’ circumferences correlate significantly (P < 0·05) with body composition variables in both sexes. Although no significant differences in correlations were obtained from WHtR-A and WHtR-W in both sexes, males tended to show greater correlations using WHtR-A compared with WHtR-W and the reverse was evident in females. Males showed significant (P < 0·05) correlation differences between BMI and WHtR only for body mass, whereas females showed significant (P < 0·05) differences for body mass, trunk fat mass and %BF. The study also showed lower correlations with trunk and total lean mass in WHtR compared with BMI in both sexes and correlations obtained from BMI and WHtR-A were significantly (P < 0·05) different in females (trunk lean mass: r (BMI) 0·499, r (WHtR-A) 0·276; total lean mass: r (BMI) 0·537, r (WHtR-A) 0·229). The results may indicate that WHtR is a good ‘fat-sensitive’ index and may be a useful screening tool.

Table 2 Associations between body composition results and waist:height ratio (WHtR) and BMI of male and female subjects§

(Mean values and standard deviations for Results and Spearman's correlation coefficients)

WHtR-W, WHtR using waist circumference; WHtR-A, WHtR using abdominal circumference.

*  Mean value is significantly lower compared with that for WHtR-A (P < 0·01).

 Significant correlation (P < 0·01).

 Correlation is significantly different from the correlation obtained from BMI (P < 0·05).

§ For details of subjects and procedures, see Table 1 and Methods.

The relationships between body composition variables such as %BF and %TF, and anthropometric indices such as BMI and WHtR using different ‘waist’ measurements were determined using the generalised linear modelling analyses (Table 3). Although R 2adj observed from the prediction equations using BMI were slightly higher than the values obtained from the equations using WHtR, the observed values were comparable. In addition, while %BF estimation using BMI and %TF using WHtR were only influenced by sex, %TF estimation using BMI and %BF estimation using WHtR were influenced by both sex and age of participants. Using the mean age for each sex, %BF and %TF were estimated as 24·1 and 26·1 % in males and 35·8 and 33·7 % in females at the BMI of 25 kg/m2. Similarly, %BF and %TF at a BMI of 30 kg/m2 were 31·8 and 34·6 % respectively for males and 43·5 and 42·4 % respectively for females. These results indicate that %TF and %BF values were relatively consistent and it may be possible to use %TF as an indication of health risk alternative to %BF.

Table 3 Proposed prediction equations for percentage total body fat (%BF) and percentage trunk fat (%TF) using BMI, waist:height ratio using waist circumference (WHtR-W) and WHtR using abdominal circumference (WHtR-A)*

R 2adj, adjusted R 2; see, standard error of estimates.

* For details of subjects and procedures, see Table 1 and Methods.

For males, ‘Sex’ = 1; for females, ‘Sex’ = 0.

Comparison of prediction equations using different WHtR suggests that WHtR-A showed higher correlations in both %BF (R 2adj 0·689 for WHtR-W and 0·719 for WHtR-A) and %TF (R 2adj 0·654 for WHtR-W and 0·680 for WHtR-A). Using the WHtR cut-off point of 0·5 and average age for the study groups, %BF and %TF were estimated to be 23·3–24·7 % and 25·2–26·8 % respectively for males and 34·6–38·3 % and 32·4–36·4 % respectively in females. These estimated %BF and %TF values from the WHtR equations with the cut-off point of 0·5 were comparable with the values calculated from the equations using the BMI cut-off value of 25 kg/m2. Equations using WHtR-W estimated greater %BF and %TF values than the equations using WHtR-A in both sexes. Furthermore, proposed equations showed a tendency that estimated %TF values at given cut-off points for BMI and WHtR to be greater than %BF values in males, whereas %BF of females were estimated to be greater than %TF. This may be associated with a sex difference in body fat distribution.

In order to determine differences in sensitivity and specificity using WHtR-W and WHtR-A at the cut-off point of 0·5, and also to compare the results with BMI at different cut-off points, cross-tabulation was conducted using %BF and %TF of 20 and 30 % as cut-off points for males and females, respectively (Table 4). In comparison with WHtR-W, WHtR-A showed higher sensitivity but lower specificity for both %BF and %TF regardless of sex. An increase in sensitivity by choosing WHtR-A is greater in females (difference: 21·5 % for %BF and 21·3 % for %TF) than their male counterparts (difference: 10·2 % for %BF and 9·8 % for %TF). On the other hand, changes in specificity were comparable (range 12·5–15·4 %). This suggests that WHtR-A may improve screening of females with a considerable level of body fat accumulation in general as well as in the trunk region. In comparison with BMI, WHtR-W showed comparable levels of sensitivity and specificity with the BMI using 25 kg/m2 as a cut-off point in males but showed lower sensitivity and higher specificity in females. When WHtR-A was used, both groups showed greater sensitivities and lower specificities than BMI with 25 kg/m2 for both %BF and %TF.

Table 4 Sensitivity and specificity of the BMI and the waist:height ratio (WHtR) at given percentage total body fat (%BF) and percentage trunk fat (%TF) in Australian males and females*

WHtR-W, WHtR using waist circumference; WHtR-A, WHtR using abdominal circumference; TP, true positives; FN, false negatives; TN, true negatives; FP, false positives.

* For details of subjects and procedures, see Table 1 and Methods.

Discussion

Previous studies have suggested that the WHtR is a useful anthropometric index to assess abdominal obesity as an alternative to WC(Reference Ho, Lam and Janus26, Reference Hsieh and Muto27) for individuals varying in age, sex and ethnicity as it adjusts for body size (i.e. height)(Reference Ashwell and Hsieh23). However, no universal or standard definition of the ‘waist’ circumference that is appropriate to be used in the calculation of WHtR has been proposed to date. The present study was the first study that compared WHtR values using different ‘waist’ circumferences and determined their usefulness as a screening tool.

Defining ‘waist’ as the narrowest point or at the umbilicus level, the present study found differences in WHtR values in both sexes. In comparison with WHtR-W, WHtR-A was greater in both groups, indicating a significant impact of the measurement site to the WHtR value. Females showed a greater difference between AC and WC values than males, which was consistent with findings in a recent study(Reference Willis, Slentz, Houmard, Johnson, Duscha, Aiken and Kraus18). Because the sex difference in circumference was smaller in AC compared with WC, the calculated WHtR-A was not significantly different between sexes. The observed results may indicate that WC will be affected by a sex difference in body fat distribution pattern compared with AC and therefore it is possible that WHtR using WC may fail to estimate the maximum amount of fat accumulation in the trunk region in females.

In comparison with WHtR, the BMI showed higher correlations with body mass and trunk fat mass. However, the BMI was also highly correlated with total and trunk lean mass compared with WHtR in both sexes. The results were consistent with a previous study(Reference Kagawa, Hills and Binns29). This suggests that the WHtR is a more ‘fat-sensitive’ index compared with the BMI and has the potential to reduce misclassification of individuals as has been reported for BMI(Reference Ross, Crawford, Kerr and Ward39). In addition, the present study showed that both BMI and WHtR have comparable (i.e. non-significant) correlations with %TF in both sexes. Considering the ‘fat-sensitive’ nature and comparable correlations for %BF using BMI, WHtR may be a useful alternative to screen for obesity, particularly in individuals with excessive fat accumulation in the trunk region.

The proposed equations that examined relationships between BMI, WHtR, %BF and %TF indicated that %TF at given BMI cut-off points (i.e. 25 and 30 kg/m2) were consistent to %BF values that were estimated at the same BMI cut-off points. The present study also showed that a WHtR cut-off point of 0·5 provided comparable %BF and %TF values to BMI estimated at the cut-off point of 25 kg/m2. These results may indicate the potential usefulness of %TF as an alternative indicator of excessive body fat accumulation. Further, although %TF is different from abdominal fat deposition or VAT accumulation by definition, it has been suggested that trunk fat is associated with metabolic biomarkers(Reference Teixeira, Sardinha, Going and Lohman30, Reference Van Pelt, Evans, Schechtman, Ehsani and Kohrt31). Considering the existing literature that provides evidence that WHtR is a useful indicator of abdominal obesity and related metabolic complications(Reference Lin, Lee, Chen, Lo, Hsia, Liu, Lin, Shau and Huang25Reference Hsieh and Muto27, Reference Sargeant, Bennett, Forrester, Cooper and Wilks40), %TF may be strongly associated with abdominal visceral fat accumulation. In addition, unlike BMI, an estimation of %TF using WHtR was not influenced by age. The results were consistent with a previous study(Reference Ashwell and Hsieh23), which suggested that WHtR can be applied to screening the general population across a wide age range using the same cut-off point (i.e. 0·5). Compared with BMI which needs to consider different cut-off points according to the age of participants(Reference Cole, Bellizzi, Flegal and Dietz41), WHtR may be a convenient option in epidemiological screening to identify individuals at risk.

When the cut-off point of 0·5 was used, the WHtR-W estimated greater %TF values compared with the WHtR-A in both sexes. Also, WHtR-A showed slightly higher R 2adj and smaller standard error of estimates in relation to %BF and %TF compared with WHtR-W. This may be simply because WHtR-W takes longer for the ratio to reach 0·5 as it uses WC measured at the narrowest point between the costal rib and iliac crest. A difference in ‘waist’ measurement sites in WHtR calculations also impacted on sensitivity and specificity to screen individuals with considerable fat deposition. Using %BF and %TF of 20 % in males and 30 % in females as cut-off points, the present study showed higher sensitivities but lower specificities using WHtR-A compared with WHtR-W in both sexes. While males showed comparable levels of sensitivity and specificity between BMI of 25 kg/m2 and WHtR-W, females showed lower sensitivity in WHtR-W compared with the BMI. This suggests that the application of the WHtR-W will disadvantage females and be unable to identify those who require improvement in health status. Lower specificity in WHtR-A compared with the other indices indicated a greater risk to misclassify healthy individuals as at-risk by applying the index alone. Therefore it is recommended that WHtR-A be combined with detailed assessments, including blood assays, to avoid unnecessary medical prescriptions if using the index as a clinical diagnostic tool. However, the high sensitivity of WHtR-A indicated that it is more suitable as a general population screening tool to identify potentially at-risk individuals who should seek further assessment or be provided with advice on lifestyle modifications that do not involve medical treatment.

Although no correlation differences were found between body fat accumulation and the WHtR using WC and AC, potential benefits of using AC to calculate WHtR in females have been observed. The result was inconsistent with a previous study that suggested the usefulness of WC to assess CVD risk(Reference Willis, Slentz, Houmard, Johnson, Duscha, Aiken and Kraus18). However, the previous study did not show significant differences between WC and AC in correlations with VAT. Because no other research has compared the appropriateness of different ‘waist’ circumferences for screening, it is difficult to draw definitive conclusion from these outcomes. However, the present study did show generally higher correlations with body composition variables in males and an increased ‘fat-sensitive’ nature of WHtR in females. Also, WHtR-A showed a considerable improvement in sensitivity in detecting at-risk individuals, particularly females, compared with WHtR-W and BMI. These results suggest that WHtR using AC may be a better option to identify at-risk Australian adults.

The present study found that WHtR values vary depending on the definition of ‘waist’ used. In order to optimise the screening ability of WHtR, it may be important to standardise its calculation protocol, especially the definition of the ‘waist’ circumference. To date, there has been a trend to accept AC as the method of choice to measure ‘waist’ and the cut-off points proposed from different ‘waist’ measurements have been listed together as a diagnostic criteria for metabolic complications(Reference Alberti, Zimmet and Shaw1). However, a measurement at the umbilicus is technically a measurement of the ‘abdomen’ and the present findings suggest that its application may be beneficial to identify individuals at risk of obesity, particularly due to trunk fat deposition. Therefore, it is recommended to differentiate the umbilicus circumference from other ‘waist’ measurements and apply it to calculate the ‘abdominal:height ratio’ instead of WHtR-A.

In the present study each circumference was only measured once. Therefore, it is not possible to calculate the technical error of measurement (TEM) for each site. Anthropometry is largely affected by the skill of the anthropometrist responsible for the measurements. Given that %CV (sd/mean × 100) of WC and AC were consistent in both groups (about 11 % for males and about 13 % for females), it is unlikely there was a considerable difference in TEM between the measured sites. However, future research should record either duplicate or triplicate measurements to enable calculation of TEM. In addition, the present study was conducted using Australian adults who were mainly Caucasians. Future research should include different ethnic groups to confirm the present findings. Furthermore, the present study was unable to clarify the impact of the measured site to the actual VAT. A previous study only used a single CT scan which was taken at the L4 pedicle and did not compare two scans that reflect each circumference site(Reference Willis, Slentz, Houmard, Johnson, Duscha, Aiken and Kraus18). In order to elucidate the influence of measurement site on the actual VAT, it may be essential to conduct research to compare WHtR using different circumferences and VAT determined from CT scans taken at the precise level of each circumference measure.

Conclusions

The present study showed that WHtR using different ‘waist’ definitions correlated highly with %TF as well as %BF in both sexes. However, the study indicated that WHtR using the umbilical measurement increases the ‘fat-sensitive’ nature of the index and also increases the sensitivity of identifying at-risk individuals compared with WHtR using the narrowest circumference, particularly in females. As a result, it may be better to use AC to calculate WHtR for early screening purposes in Australian adults. In order to make the index more generalisable, including being comparable between studies, standardisation of ‘waist’ measurement and differentiation of umbilicus measurement from other ‘waist’ measurements is recommended. Future research is recommended to confirm the present findings across different ethnic groups.

Acknowledgements

The authors would like to express gratitude to Miss Rachel Colley, Mr Andrew Hunt, Dr Jarrod Meerkin, Mr Darren Roffey and Ms Connie Wishart for their assistance in data collection.

All co-authors contributed to preparation of the manuscript. A. D. H. and N. M. B. were responsible for supervision of the projects in which the data used in the present study were collected. M. K. was responsible for data analyses and a preparation of the manuscript. The manuscript has been revised by all co-authors and there are no conflicts of interest. In addition, the study was funded through a project grant awarded by the National Health and Medical Research Council (Canberra, Australia).

References

1Alberti, KGMM, Zimmet, P & Shaw, J (2005) The metabolic syndrome – a new worldwide definition. Lancet 366, 10591062.CrossRefGoogle ScholarPubMed
2Chew, GT, Gan, SK & Watts, GF (2006) Revisiting the metabolic syndrome. Med J Aust 185, 445449.CrossRefGoogle ScholarPubMed
3Alberti, KG & Zimmet, PZ (1998) Definition, diagnosis and classification of diabetes mellitus and its complications. Part 1: diagnosis and classification of diabetes mellitus provisional report of a WHO consultation. Diabet Med 15, 539553.3.0.CO;2-S>CrossRefGoogle ScholarPubMed
4World Health Organization (1999) Definition, Diagnosis and Classification of Diabetes Mellitus and its Complications: Report of a WHO Consultation. Part 1: diagnosis and classification of diabetes mellitus. Geneva, Switzerland: WHO.Google Scholar
5Grundy, SM, Brewer, HBJ, Cleeman, JI, et al. (2004) Definition of metabolic syndrome: Report of the National Heart, Lung, and Blood Institute/American Heart Association conference on scientific issues related to definition. Circulation 109, 433438.CrossRefGoogle ScholarPubMed
6Despres, J-P & Lemieux, I (2006) Abdominal obesity and metabolic syndrome. Nature 444, 881887.CrossRefGoogle ScholarPubMed
7von Eyben, FE, Mouritsen, E, Holm, J, Montvilas, P, Dimcevski, G, Suciu, G, Helleberg, I, Kristensen, L & von Eyben, R (2003) Intra-abdominal obesity and metabolic risk factors: a study of young adults. Int J Obes 27, 941949.CrossRefGoogle ScholarPubMed
8Pi-Sunyer, FX (2004) The epidemiology of central fat distribution in relation to diease. Nutr Rev 62, S120S126.CrossRefGoogle Scholar
9Bei-Fan, Z (2002) Predictive values of body mass index and waist circumference for risk factors of certain related diseases in Chinese adults: study on optimal cut-off points of body mass index and waist circumference in Chinese adults. Asia Pacific J Clin Nutr 11, S685S693.CrossRefGoogle Scholar
10Dalton, M, Cameron, AJ, Zimmet, PZ, Shaw, JE, Jolley, D, Dunstan, DW, Welborn, TA; AusDiab Steering Committee (2003) Waist circumference, waist-hip ratio and body mass index and their correlation with cardiovascular disease risk factors in Australian adults. J Intern Med 254, 555563.CrossRefGoogle ScholarPubMed
11Welborn, TA, Dhaliwal, SS & Bennett, SA (2003) Waist-hip ratio is the dominant risk factor predicting cardiovascular death in Australia. Med J Aust 179, 580585.CrossRefGoogle ScholarPubMed
12Janssen, I, Heymsfield, SB, Allison, DB, Kotler, DP & Ross, R (2002) Body mass index and waist circumference independently contribute to the prediction of nonabdominal, abdominal subcutaneous, and visceral fat. Am J Clin Nutr 75, 683688.CrossRefGoogle Scholar
13Onat, A, Avci, GS, Barlan, MM, Uyarel, H, Uzunlar, B & Sansoy, V (2004) Measures of abdominal obesity assessed for visceral adiposity and relation to coronary risk. Int J Obes Relat Metab Disord 28, 10181025.CrossRefGoogle ScholarPubMed
14Brambilla, P, Bedogni, G, Moreno, LA, Goran, MI, Gutin, B, Fox, KR, Peters, DM, Barbeau, P, De Simone, M & Pietrobelli, A (2006) Crossvalidation of anthropometry against magnetic resonance imaging for the assessment of visceral and subcutaneous adipose tissue in children. Int J Obes 30, 2330.CrossRefGoogle ScholarPubMed
15James, WPT (2005) Assessing obesity: are ethnic differences in body mass index and waist classification criteria justified? Obes Rev 6, 179181.CrossRefGoogle ScholarPubMed
16Marfell-Jones, M, Olds, T, Stewart, A & Carter, JEL (2006) International Standards for Anthropometric Assessment, 2nd ed.Potchefstroom, South Africa: The International Society for the Advancement of Kinanthropometry.Google Scholar
17Examination Committee of Criteria for ‘Obesity Disease’ in Japan: Japan Society for the Study of Obesity (2002) New criteria for ‘obesity disease’ in Japan. Circ J 66, 987992.CrossRefGoogle Scholar
18Willis, LH, Slentz, CA, Houmard, JA, Johnson, JL, Duscha, BD, Aiken, LB & Kraus, WE (2007) Minimal versus umbilical waist circumference measures as indicators of cardiovascular disease risk. Obesity 15, 753759.CrossRefGoogle ScholarPubMed
19Bosy-Westphal, A, Geisler, C, Onur, S, Korth, O, Selberg, O, Schrezenmeir, J & Müller, MJ (2006) Value of body fat mass vs anthropometric obesity indices in the assessment of metabolic risk factors. Int J Obes (Lond) 30, 475483.CrossRefGoogle ScholarPubMed
20Sakurai, M, Miura, K, Takamura, T, Ota, T, Ishizaki, M, Morikawa, Y, Kido, T, Naruse, Y & Nakagawa, (2006) Gender differences in the association between anthropometric indices of obesity and blood pressure in Japanese. Hypertens Res 29, 7580.CrossRefGoogle ScholarPubMed
21Hsieh, SD, Yoshinaga, H & Muto, T (2003) Waist-to-height ratio, a simple and practical index for assessing central fat distribution and metabolic risk in Japanese men and women. Int J Obes 27, 610616.CrossRefGoogle ScholarPubMed
22Parikh, RM, Joshi, SR, Menon, PS & Shah, NS (2007) Index of central obesity – a novel parameter. Med Hypotheses 68, 12721275.CrossRefGoogle ScholarPubMed
23Ashwell, M & Hsieh, SD (2005) 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 56, 303307.CrossRefGoogle ScholarPubMed
24McCarthy, HD & Ashwell, M (2006) A study of central fatness using waist-to-height ratios in UK children and adolescents over two decades supports the simple message – ‘keep your waist circumference to less than half your height’. Int J Obes 30, 988992.CrossRefGoogle ScholarPubMed
25Lin, JS, Lee, LT, Chen, CY, Lo, H, Hsia, HH, Liu, IL, Lin, RS, Shau, WY & Huang, KC (2002) Optimal cut-off values for obesity: using simple anthropometric indices to predict cardiovascular risk factors in Taiwan. Int J Obes 26, 12321238.CrossRefGoogle ScholarPubMed
26Ho, 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, 683691.CrossRefGoogle ScholarPubMed
27Hsieh, SD & Muto, T (2006) Metabolic syndrome in Japanese men and women with special reference to the anthropometric criteria for the assessment of obesity: proposal to use the waist-to-height ratio. Prev Med 42, 135139.CrossRefGoogle Scholar
28Schneider, HJ, Glaesmer, H, Klotsche, J, Bohler, S, Lehnert, H, Zeiher, AM, März, W, Pittrow, D, Stalla, GK, Wittchen, HU; DETECT Study Group (2007) Accuracy of anthropometric indicators of obesity to predict cardiovascular risk. J Clin Endocrinol Metab 92, 589594.CrossRefGoogle ScholarPubMed
29Kagawa, M, Hills, AP & Binns, CW (2007) The usefulness of the waist-to-height ratio to predict trunk fat accumulation in Japanese and Australian Caucasian young males living in Australia. Int J Body Comp Res 5, 5763.Google Scholar
30Teixeira, PJ, Sardinha, LB, Going, SB & Lohman, TG (2001) Total and regional fat and serum cardiovascular disease risk factors in lean and obese children and adolescents. Obes Res 9, 432442.CrossRefGoogle ScholarPubMed
31Van Pelt, RE, Evans, EM, Schechtman, KB, Ehsani, AA & Kohrt, WM (2002) Contributions of total and regional fat mass to risk for cardiovascular disease in older women. Am J Physiol Endocrinol Metab 282, E1023E1028.CrossRefGoogle ScholarPubMed
32National Health and Medical Research Council (1999) National Statement on Ethical Conduct in Research Involving Humans. Canberra, ACT: NHMRC.Google Scholar
33Heymsfield, SB, Lichtman, S, Baumgartner, RN, Wang, J & Kamen, Y (1990) Body composition of humans: comparison of two improved four-compartment models that differ in expense, technical capacity, and radiation exposure. Am J Clin Nutr 52, 5258.CrossRefGoogle Scholar
34Wellens, R, Chumlea, WC, Guo, S, Roche, AF & Reo, NV (1994) Body composition in white adults by dual-energy X-ray absorptiometry, densitometry, and total body water. Am J Clin Nutr 59, 547555.CrossRefGoogle ScholarPubMed
35Van Loan, MD (1998) Estimates of fat-free mass (FFM) by densitometry, dual energy X-ray absorptiometry (DXA), and bioimpedance spectroscopy (BIS) in Caucasian and Chinese-American women. Appl Radiat Isot 49, 751752.CrossRefGoogle ScholarPubMed
36Glickman, SG, Marn, CS, Supiano, MA & Dengel, DR (2004) Validity and reliability of dual-energy X-ray absorptiometry for the assessment of abdominal adiposity. J Appl Physiol 97, 509514.CrossRefGoogle ScholarPubMed
37Cohen, J & Cohen, P (1983) Applied Multiple Regression/ Correlation Analysis for the Behavioral Sciences. Hillsdale, NJ: Eribaum.Google Scholar
38Huenemann, RL, Hampton, MC, Shapiro, LR & Behnke, AR (1966) Adolescent food practices associated with obesity. Fed Proc 25, 410.Google ScholarPubMed
39Ross, WD, Crawford, SM, Kerr, DA & Ward, R (1988) Relationship of the body mass index with skinfolds, girths, and bone breadths in Canadian men and women aged 20–70 years. Am J Phys Anthropol 77, 169173.CrossRefGoogle ScholarPubMed
40Sargeant, LA, Bennett, FI, Forrester, TE, Cooper, RS & Wilks, RJ (2002) Predicting incident diabetes in Jamaica: the role of anthropometry. Obes Res 10, 792798.CrossRefGoogle ScholarPubMed
41Cole, TJ, Bellizzi, MC, Flegal, KM & Dietz, WH (2000) Establishing a standard definition for child overweight and obesity worldwide: international survey. BMJ 320, 12401245.CrossRefGoogle ScholarPubMed
Figure 0

Table 1 Physical characteristics of male and female subjects†(Mean values, ranges and standard deviations)

Figure 1

Table 2 Associations between body composition results and waist:height ratio (WHtR) and BMI of male and female subjects§(Mean values and standard deviations for Results and Spearman's correlation coefficients)

Figure 2

Table 3 Proposed prediction equations for percentage total body fat (%BF) and percentage trunk fat (%TF) using BMI, waist:height ratio using waist circumference (WHtR-W) and WHtR using abdominal circumference (WHtR-A)*

Figure 3

Table 4 Sensitivity and specificity of the BMI and the waist:height ratio (WHtR) at given percentage total body fat (%BF) and percentage trunk fat (%TF) in Australian males and females*