Relationship of total body fatness and five anthropometric indices in Chinese aged 20–40 years: different effects of age and gender

Article metrics

Abstract

Objectives:

We aim to evaluate the ethnic-specific relationship of total fat mass and anthropometric indices in Chinese.

Design:

Cross-section study.

Setting:

This study was performed at the College of Life Sciences, Hunan Normal University, P.R. China.

Subjects and method:

To increase our understanding of the relationship of total fat mass and anthropometric indices in Chinese, 793 females and 1091 males aged 20–40 years were randomly recruited from Changsha city of P. R. China. Hip circumference (HC) and waist circumference (WC) were measured using standardized equipments, and other three anthropometric indices of body mass index (BMI), waist-to-hip ratio (WHR), and conicity index (CI) were calculated using weight, height, HC and WC. Total body fatness (TBF) in kg was measured using a Hologic QDR 4500 W dual-energy X-ray absorptiometry (DEXA) scanner.

Results:

There was an increasing trend of TBF, %TBF (percent total body fatness) and the five anthropometric indices in successively older age groups. Compared with females, males generally had high average BMI, WC, HC, WHR and CI, but had low average TBF and %TBF. Except for some correlations in 25–29 years age groups, TBF and %TBF were significantly correlated with five anthropometric indices with the Pearson's correlation coefficients ranging from 0.07 to 0.87. Principal component analysis (PCA) was performed to form four principal components (PCs) that interpreted over 99% of the total variation of the five related anthropometric indices in all age groups, with over 53% of the total variation accounted for by the PC1. Multiple regression analyses showed that four PCs combined explained a greater variance (R2=55.2–80.8%) in TBF than did BMI alone (R2=40–74.9%).

Conclusion:

Our results suggest that there is an increasing trend of total fat mass and five anthropometric indices with aging; that age and sex have the important effects on influencing the correlations of TBF and the studied anthropometric indices; and that the accuracy of predicting the TBF using five anthropometric indices is higher than using BMI alone.

Main

Obesity, defined as an excess of body fat, is an increasing health problem. Excess body fat mass is strongly associated with metabolic and vascular diseases (Han et al., 1996; Lindsay et al., 2001; Thomas et al., 2004). However, direct and accurate measurement of adiposity relies on complex technologies such as dual energy X-ray absorptiometry (DEXA), resulting in relatively high cost. Anthropometric indices (such as body mass index (BMI), waist circumference (WC), hip circumference (HC), waist-to-hip ratio (WHR) and conicity index (CI)) are commonly used to assess fat mass because they are relatively straightforward and cheap to obtain (Taylor et al., 2000; Lindsay et al., 2001; Janssen et al., 2002). BMI is the commonly used index in prediction of total body fat (TBF). However, its ability in accurately predicting TBF remains controversial because BMI cannot distinguish fat mass and lean mass, and, more importantly, the accuracy of its prediction is prone to be disturbed by the differences of sample in adipose distribution and body build (Frankenfield et al., 2001; Janssen et al., 2002). WC, HC, WHR and CI are four useful and accurate tools to evaluate the distribution of fat mass or the central adiposity (Taylor et al., 2000; Janssen et al., 2002). Additionally, relatively few data on the prediction of fat mass with anthropometric indices were accumulated in Chinese (Davies et al., 2001; Lanham et al., 2001; Chang et al., 2003; Jia et al., 2003). Furthermore, from these limited data in Chinese, the accuracy of the prediction was still inconsistent (Chang et al., 2003). Hence, the present study was performed to assess the ability of the combined use of the above five anthropometric indices to predict TBF in a very large Chinese population.

As there are ethnic-specific energy expenditure and the body build among various ethnic populations (Deurenberg et al. 1998; Duncan et al. 2004; Rush et al. 2004), the data on the relationship between anthropometric indices and fat mass accumulated in Caucasians may not be suitable for Chinese. Compared with Caucasians, few data both on the relationship between anthropometric indices and body fat measured by sophisticated measures, such as DEXA, were accumulated in Chinese (Ko et al., 2001; Zhang et al., 2004). Therefore, it is necessary to perform the ethnic-specific relationship in Chinese.

The current study was undertaken with three major aims: (1) evaluating the effect of age and sex on TBF, anthropometric indices and their relationship; (2) investigating the correlation between TBF and anthropometric indices; (3) evaluating the ability of predicting TBF using five anthropometric indices and developing eight age- and gender-specific equations to predict TBF by the above five anthropometric indices in Chinese.

Materials and methods

Subjects

The project was approved by the Research Administration Department of Hunan Normal University. These samples were from a large population that was randomly recruited for genetic studies aimed at searching for genes underlying peak bone mass variation in Chinese. We first gave an advertisement in the surrounding areas of Hunan Normal University located in the central south area of P. R. China. After the subjects signed the informed-consent document, a questionnaire was given to obtain the subject's information on age, sex, medical history, family history, female history, physical activity, alcohol use, dietary habits and smoking history. We adopted the exclusion criteria detailed by Deng et al. (2003) to screen and recruit ‘healthy’ subjects. Briefly, patients with chronic diseases and conditions, which may potentially affect bone mass have been excluded from the study. These diseases/conditions included chronic disorders involving vital organs (heart, lung, liver, kidney and brain), serious metabolic diseases (diabetes, hypo- and hyperparathyroidism, and hyperthyroidism, etc.), other skeletal diseases (Paget's disease, osteogenesis imperfecta, and rheumatoid arthritis, etc.), chronic use of drugs affecting bone metabolism (corticosteroid therapy and anti-convulsant drugs), and malnutrition conditions (chronic diarrhea, and chronic ulcerative colitis, etc.), etc. Finally, a total of 1884 healthy Chinese volunteers (793 females and 1091 males aged 20–40 years), who were measured for both bone mineral density and TBF, were included in the present study.

Measurement

Weight and height were measured using standardized equipment. BMI (kg/m2) was calculated as weight (kg) divided by square of height (m2). WC and HC were measured with an anthropometric tape over light clothing, with measurement of WC and HC at the minimum circumference between the iliac crest and the rib cage, and at the maximum protuberance of the buttocks. Then WHR was calculated as WC divided by HC. The CI evaluated the fat distribution without the requirement of HC (Valdez et al. 1993; Taylor et al. 2000), which was calculated as WC/[0.109 × square root of (weight/height)], where WC and height were measured in meter and weight was measured in kilograms. TBF in kilograms was measured using a Hologic QDR 4500 W dual-energy X-ray absorptiometry (DEXA) scanner (Hologic Corp., Waltham, MA). The percentage of total fat mass (%TBF) was calculated as fat mass/(fat mass+lean mass+bone mineral content). The coefficient of variation (CV) of total body fat, obtained from 30 individuals repeatedly measured two times, of the DEXA measurements was 0.38%.

Statistical analysis

Statistical analyses were performed with SAS package (SAS Institute Inc., Cary, NC, USA). The differences of the studied phenotypes between males and females were evaluated by the Student's t-test. To investigate the effects of age, we divided the individuals into four 5-years age groups (20–24, 25–29, 31–34, and 35–39 years) in both males and females. The Pearson's correlation coefficients were used to investigate the linear correlation of TBF and %TBF with five anthropometric indices. As BMI, WC, HC, WHR and CI were five highly related anthropometric indices, to avoid the disturbance of the colinearity when simultaneously modeling the five indices in multiple regression analysis for predicting TBF, a principal component analysis (PCA) was performed to form four principal components (PCs) accounting for the variation of the five anthropometric indices as well as Eigenvalues of matrix and Eigenvectors (Green 1978; Vapnik 1998). Subsequently, the PC values were calculated by Eigenvalues of matrix and Eigenvectors, and then they were used to estimate their regression coefficients and the proportion of the variance (R2) of TBF predicted by these combined PCs by multiple regression analysis. Regression analyses were also conducted to determine whether four PCs combined explained a greater variance (R2) of TBF than did BMI alone, and also to investigate the correlation coefficient of the measured TBF and the predicted TBF by four PCs or BMI. Finally, we developed eight age- and gender- specific equations to predict TBF by the measured values of five anthropometric indices. The equations were calculated as follows:

where RCPCs are the regression coefficients for four PCs, respectively; EPCs BMI, EPCs WC, EPCs HC, EPCs WHR and EPCs CI are the corresponding Eigenvectors of BMI, WC, HC, WHR and CI for four PCs; SBMI, SWC, SHCI, SWHR and SCI are the corresponding standard values [(measured value−mean measured value)/mean measured value] for five anthropometric index.

Results

Age was significantly correlated with TBF, %TBF, and five anthropometric indices either among the whole males or among the total females, but was not significantly associated with them in each 5-years age group in both males and females (data not shown). As shown in Table 1, there was an increasing trend of TBF, %TBF and five anthropometric indices in successively older age groups (For example, the average TBF was 13.2±3.31, 13.3±3.23, 14.6±4.20 and 16.7±3.66 in four age groups in females.). Compared with females, the males generally had larger average BMI, WC, HC, WHR and CI, but had lower average TBF and %TBF (Table 1). Except for age, statistically significant differences between males and female were found for the other studied indices by Student's t-test (P<0.0001).

Table 1 Basic characteristics of the studied sample

Correlations between TBF and five anthropometric indices were generally higher than those between %TBF and the studied anthropometric indices in each age group (Table 2). So the following analyses were focused on the prediction of TBF with five anthropometric indices, because the high correlations resulted in more accuracy of prediction. Except for some correlations in 25–29 years age groups, TBF and %TBF were significantly correlated with five anthropometric indices with the correlation coefficients ranging from 0.07 to 0.87 (Table 2). In each age group, there was a decreasing trend of correlations between TBF and BMI, WC, HC, WHR and CI, generally with the highest correlation coefficients between TBF and BMI (For instance, the correlation coefficients between TBF and BMI, WC, HC, WHR and CI in 20–24 years age female group were 0.82, 0.71, 0.69, 0.34 and 0.22.). There were no evident differences of correlations among different age groups in the males, but in females in the 25–29 years age group, the correlation coefficients were evidently smaller than those in other age female groups. In addition, in the same age group, there was intuitively gender-specific correlation, especially in 25–29 years age group.

Table 2 Correlations of TBF and %TBF with anthropometric indices

As shown in Tables 3 and 4, the four PCs interpreted over 99% of the total variation of five relative anthropometric indices by PCA, with over 53% of the total variation accounted by PC1. The PC1 is highly and positively correlated with all the indices in all age and gender groups according to their Eigenvectors, but the other three PCs are inconsistently related with the studied indices. Regression analyses (Table 5) showed that the four PCs combined explained a greater variance (R2=55.2–80.8%) of TBF than did BMI alone (R2=40–74.9%) (e.g., in the 20–24 years group, the proportions were 66.6% by BMI, and 74.9% by the four PCs.). Although the four PCs combined or BMI alone were strong correlates of TBF, about 20–60% of the variance in TBF remained unexplained by the four PCs combined or BMI alone. Figure 1 intuitively demonstrated that the simple correlation coefficients between TBF and the predicted TBF by combined PCs or BMI in regression analysis were very high in both females and males at all the age groups (r=0.64–0.9), whereas the correlations of TBF and the predicted TBF by the four PCs combined were higher than those of TBF and the predicted TBF by BMI alone, indicating that using the four PCs predict more precise TBF than using BMI alone. We also developed eight age- and gender- specific equations to predict TBF using measured values of five anthropometric indices (Table 6), which are detailed in Statistical analysis. To investigate the external validation of predicting TBF, we determined the equations to predict body fat in half of the randomly chosen sample and tested the prediction equations in the other half. The prediction equations in half of the randomly chosen sample were very similar with those in whole sample (data not shown), and then the predicted TBF using these equations in the other half were highly correlated with the measured TBF (r=0.73–0.86).

Table 3 Eigenvalues of the correlation matrix and eigenvectors in the PCA for five anthropometric indices in females
Table 4 Eigenvalues of the correlation matrix and eigenvectors in the PCA for five anthropometric indices in males
Table 5 The variation (R2) of TBF by four PCs or BMI, and the regression coefficients of four PCs
Figure 1
figure1

Simple correlation coefficients between TBF and the predicted TBF by four PCs using multiple regression analysis: (a) in females, (b) in males.

Table 6 Predicted TBF calculated by the raw values of five anthropometric indices using principal component analysis

Discussion

Our major findings in this study are that there is an increasing trend of total fat mass and five anthropometric indices in successively older age groups; that age and sex have the important effects on influencing the relationship of TBF and the studied anthropometric indices; and that the accuracy of predicting the TBF using five anthropometric indices is higher than using BMI alone.

Compared with Caucasians, few data on the relationship of anthropometric indices and body fat mass were reported in Chinese. A conclusion from a meta-analysis (Deurenberg et al., 1998) has shown that the differences in energy balance as well as in body build may result in their population-specific relationship. Therefore, to increase our understanding on the ethnic-specific correlation and prediction of body fat mass with five anthropometric indices and to develop eight age- and gender- specific equations for prediction of TBF in Chinese, we performed the present study in a very large population of both sexes.

Most of the previous studies have concentrated on the prediction of %TBF with anthropometric indices (Goran et al., 1996; Pietrobelli et al., 1998; Lindsay et al., 2001), whereas the present study focused on the prediction of TBF. Two reasons may justify our choice. First, the simple correlations of anthropometric indices are greater with TBF than with %TBF in our large Chinese sample of two sexes, which are in agreement with those from previous studies in Caucasians (Goran et al., 1996; Pietrobelli et al., 1998; Lindsay et al., 2001), and thus the high correlations resulted in more accuracy of prediction. Second, TBF is regarded as an absolute measure that reflects overall size as well as adiposity and %TBF serves as a relative measure of adiposity; %TBF is of clinical importance as a generally accepted criterion of diagnosing obesity (Heyward et al., 1992; Blew et al., 2002; Eto et al., 2004); moreover, %TBF is easily calculated from an estimate of total body fat and total weight. Therefore, the prediction of TBF in the present study is the basis of predicting %TBF. Based on the above reasons, we focused on investigating the prediction of TBF, rather than %TBF, with anthropometric indices.

Our results support earlier work indicating that age has a differential effect on body fat mass and five anthropometric indices (Horber et al. 1997). Our results are similar with those from previous studies (Horber et al., 1997; Jackson et al., 2002) that there is an increasing trend of total fat mass and five anthropometric indices in successively older age groups. Potential reason is that individuals tend to gain fatness with aging because of substantial increase of adipocyte size and number (Evans and Campbell, 1993). Therefore, to try our best to decrease the disturbance of age, it is appropriate to divide our sample into four 5-years age groups in both sexes, because age was significantly correlated with total fat mass and five anthropometric indices in all sample, but was not associated with them in each divided age group.

Our results indicate that the four PCs from five anthropometric indices are better predictors of TBF than is BMI alone. The evidence may come from that the combined use of the four PCs values substantially increases the explained variance (R2) of TBF, and that the simple correlation coefficients are higher between TBF and the predicted TBF by the four PCs than between TBF and the predicted TBF by BMI alone (Figure 1). BMI is the one most commonly and widely used tool for assessing body fatness. However, use of the BMI predictor of body fatness has been criticized (Smalley et al., 1990; Roubenoff et al., 1995), because of a relatively low correlation coefficient between BMI and fat mass, and low proportion of fat mass variance explained by BMI from multiple regression analysis (Frankenfield et al., 2001). Adipose distribution may partially disturb the accuracy of assessing body fatness by BMI. Four indices of WC, HC, WHR and CI are useful and accurate tools to evaluate the distribution of fat mass or the central adiposity (Taylor et al., 2000). However, those indices are highly related and thus they cannot simply be modeled in multiple regression analysis to predict TBF. Therefore, a PCA has been employed in this study with an attempt to extract useful information from five related indices (BMI, WC, HC, WHR and CI). To the best of our knowledge, this study is the first effort to simultaneously use five anthropometric indices to predict TBF in Chinese adults of both sexes.

With the aim of predicting TBF by a simple method, we develop eight age- and gender-specific equations using the measured values of five anthropometric indices in the present study, which have practical importance for routine clinical evaluation. Moreover, the validation of predicting TBF is very high, because the correlation coefficients between TBF and the predicted TBF by the four PCs using multiple regression analysis range from 0.75–0.9 (Figure 1). The combined use of five anthropometric indices has certain advantage over assessment of fat mass using DEXA. Measurement of anthropometric indices are cheaper and technically easier than assessment of adiposity. Therefore, it is of very important clinical implication to accurately assess fat mass using simple anthropometric indices, especially in a developing country like China, because the direct measurement of adiposity incurs relatively high cost. This is the primary motive in performing the present study in a Chinese population.

We have confirmed that the BMI and other four anthropometric indices are highly correlated with TBF or TBF% (Pietrobelli et al., 1998; Lindsay et al., 2001) with important effects of gender and age influencing their relationship. However, the correlation coefficients in our Chinese sample are generally smaller than those in Caucasians (Pietrobelli et al., 1998; Lindsay et al., 2001). For example, Lindsay et al. (2001) found that the linear correlation coefficients between BMI and percent fat were above 0.8, but these ranged from 0.45 to 0.75 in this Chinese sample. The differences in body build and fat mass distribution may account for the above observation. Body build tends to vary among different ethnic groups (Duncan et al., 2004). Generally, relative sitting height tends to be higher in Asian ethnic groups, which may result in a relatively low BMI (Deurenberg et al., 1999; Gurrici et al., 1999). Furthermore, when adjusting for body build, most of the ethnic-specific differences were eliminated associated with %TBF prediction equations for bioelectrical impedance analysis (BIA) in Chinese, Malay, and Indian Singaporeans (Deurenberg et al., 2002). Race differences in fat distribution are also evident. There is evidence that Asian children and adolescents have a greater central fat mass than that of Europeans and other ethnic groups (Duncan et al., 2004). When body size (weight and height) was controlled, He et al. (2002) inferred that Asian girls had greater relative truncal or central fat mass. Therefore, these differences in body build and fat distribution may result in the difference in relationship between anthropometric indices and TBF. Additionally, the differences in body build and fat mass distribution may also explain the gender-specific relationship between anthropometric indices and TBF. These results support our belief that gender- and race-specific definitions of adiposity might be required.

However, the study has some limitations. First, although DEXA is becoming increasingly popular for the measurement of soft tissue composition as well as bone mineral density, it cannot distinguish between visceral and subcutaneous fat mass. Second, the present results from a normal healthy population may not be applied to an overweight and obese population or other patients with chronic diseases and conditions. How to apply the present data to these populations awaits further studies. Third, the samples are aged between 20 and 40 years. Therefore, these results may have inability to generalize to older or younger populations.

In conclusion, the present study explored the complex relationship between the five anthropometric indices and total body fatness in a very large population of both sexes. The obtained results add to our understanding of the relationship of anthropometric indices and adiposity, as well as the effects of age and sex on their relationship.

References

  1. Blew RM, Sardinha LB, Milliken LA, Teixeira PJ, Going SB, Ferreira DL et al. (2002). Assessing the validity of body mass index standards in early postmenopausal women. Obes Res 10, 799–808.

  2. Chang CJ, Wu CH, Chang CS, Yao WJ, Yang YC, Wu JS et al. (2003). Low body mass index but high percent body fat in Taiwanese subjects: implications of obesity cutoffs. Int J Obes Relat Metab Disord 27, 253–259.

  3. Davies PS, Lanham DA, Stead MA, Tsang K (2001). Body mass index and the prediction of percentage body fat in Australian Chinese women. Ann Hum Biol 28, 467–470.

  4. Deng FY, Liu MY, Li MX, Lei SF, Qin YJ, Zhou Q et al. (2003). Tests of linkage and association of the COL1A2 gene with bone phenotypes’ variation in Chinese nuclear families. Bone 33, 614–619.

  5. Deurenberg P, Deurenberg Yap M, Wang J, Lin FP, Schmidt G (1999). The impact of body build on the relationship between body mass index and percent body fat. Int J Obes Relat Metab Disord 23, 537–542.

  6. Deurenberg P, Deurenberg-Yap M, Schouten FJ (2002). Validity of total and segmental impedance measurements for prediction of body composition across ethnic population groups. Eur J Clin Nutr 56, 214–220.

  7. Deurenberg P, Yap M, van Staveren WA (1998). Body mass index and percent body fat: a meta analysis among different ethnic groups. Int J Obes Relat Metab Disord 22, 1164–1171.

  8. Duncan E, Schofield G, Duncan S, Kolt G, Rush E (2004). Ethnicity and body fatness in New Zealanders. N Z Med J 117, U913.

  9. Eto C, Komiya S, Nakao T, Kikkawa K (2004). Validity of the body mass index and fat mass index as an indicator of obesity in children aged 3–5 year. Physiol Anthropol Appl Human Sci 23, 25–30.

  10. Evans WJ, Campbell WW (1993). Sarcopenia and age-related changes in body composition and functional capacity. J Nutr 123, 465–468.

  11. Frankenfield DC, Rowe WA, Cooney RN, Smith JS, Becker D (2001). Limits of body mass index to detect obesity and predict body composition. Nutrition 17, 55–56.

  12. Goran MI, Driscoll P, Johnson R, Nagy TR, Hunter G (1996). Cross-calibration of body-composition techniques against dual-energy X-ray absorptiometry in young children. Am J Clin Nutr 63, 299–305.

  13. Green PE (1978). Analyzing Multivariate Data. The Dryden Press: Hinsdale, Illinois.

  14. Gurrici S, Hartriyanti Y, Hautvast JG, Deurenberg P (1999). Differences in the relationship between body fat and body mass index between two different Indonesian ethnic groups: the effect of body build. Eur J Clin Nutr 53, 468–472.

  15. Han TS, van Leer EM, Seidell JC, Lean ME (1996). Waist circumference as a screening tool for cardiovascular risk factors: evaluation of receiver operating characteristics (ROC). Obes Res 4, 533–547.

  16. He Q, Horlick M, Thornton J, Wang J, Pierson Jr RN, Heshka S et al (2002). Sex and race differences in fat distribution among Asian, African–American, and Caucasian prepubertal children. J Clin Endocrinol Metab 87, 2164–2170.

  17. Heyward VH, Cook KL, Hicks VL, Jenkins KA, Quatrochi JA, Wilson WL (1992). Predictive accuracy of three field methods for estimating relative body fatness of nonobese and obese women. Int J Sport Nutr 2, 75–86.

  18. Horber FF, Gruber B, Thomi F, Jensen EX, Jaeger P (1997). Effect of sex and age on bone mass, body composition and fuel metabolism in humans. Nutrition 13, 524–534.

  19. 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.

  20. Janssen 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, 683–688.

  21. Jia WP, Lu JX, Xiang KS, Bao YQ, Lu HJ, Chen L (2003). Prediction of abdominal visceral obesity from body mass index, waist circumference and waist–hip ratio in Chinese adults: receiver operating characteristic curves analysis. Biomed Environ Sci 16, 206–211.

  22. Ko GT, Tang J, Chan JC, Sung R, Wu MM, Wai HP et al. (2001). Lower BMI cut-off value to define obesity in Hong Kong Chinese: an analysis based on body fat assessment by bioelectrical impedance. Br J Nutr 85, 239–242.

  23. Lanham DA, Stead MA, Tsang K, Davies PS (2001). The prediction of body composition in Chinese Australian females. Int J Obes Relat Metab Disord 25, 286–291.

  24. Lindsay RS, Hanson RL, Roumain J, Ravussin E, Knowler WC, Tataranni PA (2001). Body mass index as a measure of adiposity in children and adolescents: relationship to adiposity by dual energy X-ray absorptiometry and to cardiovascular risk factors. J Clin Endocrinol Metab 86, 4061–4067.

  25. Pietrobelli A, Faith MS, Allison DB, Gallagher D, Chiumello G, Heymsfield SB (1998). Body mass index as a measure of adiposity among children and adolescents: a validation study. J Pediatr 132, 204–210.

  26. Roubenoff R, Dallal GE, Wilson PW (1995). Predicting body fatness: the body mass index vs estimation by bioelectrical impedance. Am J Public Health 85, 726–728.

  27. Rush E, Plank L, Chandu V, Laulu M, Simmons D, Swinburn B et al. (2004). Body size, body composition, and fat distribution: a comparison of young New Zealand men of European, Pacific Island, and Asian Indian ethnicities. N Z Med J 117, U1203.

  28. Smalley KJ, Knerr AN, Kendrick ZV, Colliver JA, Owen OE (1990). Reassessment of body mass indices. Am J Clin Nutr 52, 405–408.

  29. Taylor RW, Jones IE, Williams SM, Goulding A (2000). Evaluation of waist circumference, waist-to-hip ratio, and the conicity index as screening tools for high trunk fat mass, as measured by dual-energy X-ray absorptiometry, in children aged 3–19 years. Am J Clin Nutr 72, 490–495.

  30. Thomas GN, Ho SY, Lam KS, Janus ED, Hedley AJ, Lam TH, Hong Kong Cardiovascular Risk Factor Prevalence Study Steering Committee (2004). Impact of obesity and body fat distribution on cardiovascular risk factors in Hong Kong Chinese. Obes Res 12, 1805–1813.

  31. Valdez R, Seidell JC, Ahn YI, Weiss KM (1993). A new index of abdominal adiposity as an indicator of risk for cardiovascular disease. A cross-population study. Int J Obes Relat Metab Disord 17, 77–82.

  32. Vapnik V (1998). Statistical Learning Theory[M]. Wiley: New York.

  33. Zhang Q, Du WJ, Hu XQ, Liu AL, Pan H, Ma GS (2004). The relation between body mass index and percentage body fat among Chinese adolescent living in urban Beijing. Zhonghua Liu Xing Bing Xue Za Zhi 25, 113–116. (in Chinese).

Download references

Acknowledgements

We thank three anonymous reviewers for their comments to improve this paper. The study was partially supported by a key project grant (30230210), a general grant (30470534) from National Science Foundation of China, three projects from Scientific Research Fund of Hunan Provincial Education Department (02A027, 03C226, 04B039), and a grant from Natural Science Foundation of Hunan Province (04JJ1004). Investigator H.W.D. was partially supported by grants from Health Future Foundation of USA, grants of National Health Institute (K01 AR02170-01A2, R01 GM60402 and 5R01 AR050496-02).

Author information

Correspondence to H-W Deng.

Additional information

Guarantor: H-W Deng.

Contributors: SFL was responsible for the data analysis and writing of the manuscript. HWD is principal investigator and contributed to the study design and its implementation. YJL was involved in the revision of the manuscript. Other coauthors participated in sample recruitment, data preparation or manuscript preparation.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Lei, S., Liu, M., Chen, X. et al. Relationship of total body fatness and five anthropometric indices in Chinese aged 20–40 years: different effects of age and gender. Eur J Clin Nutr 60, 511–518 (2006) doi:10.1038/sj.ejcn.1602345

Download citation

Keywords

  • anthropometric index
  • obesity
  • total body fatness
  • percent total body fatness
  • principal component analysis

Further reading