Which anthropometric indices best predict metabolic disorders in Taiwan?

Objectives: The aim of this survey was to assess the association between the three simple anthropometric indices (body mass index [BMI], waist-to-hip ratio (WHR), and waist circumference [WC]) and various metabolic disorders, and to identify which indices can best predict metabolic disorders in Taiwan.

Methods: A cross-sectional hospital-based survey was carried out from January to December 2000. We retrospectively analyzed the medical records of all the patients receiving periodic health examination with scheduled items at China Medical University Hospital. Anthropometric indices, metabolic profiles, and abdominal sonography were performed. A total of 746 people aged 19 to 87 were recruited as subjects for this study. The t test, [chi square] test, and stepwise multivariate logistic regression were used.

Results: The subjects included 44.5% women and 55.5% men, with a mean age of 50.74 [+ or -] 12.68 years. After controlling for the other covariables, stepwise multivariate logistic regression showed considerable statistical significance between overall obesity and hypertension, hypertriglyceridemia, hyperuricemia, and fatty liver. There is also statistical significance between abdominal obesity and abnormal ratio of total cholesterol HDL (high-density lipoprotein), hyperuricemia, and fatty liver. None of the indices revealed any association with hyperglycemia, hypercholesterolemia, abnormal low-density lipoprotein, or abnormal HDL.

Conclusions: BMI and WC can best predict some metabolic disorders. For practical reasons, the combined measurement of BMI and WC is the simple and inexpensive anthropometric index for primary health care settings in the routine physical examinations of adults. We hope this study can establish the background data for further investigation on the epidemiology of anthropometric indices in Taiwan.

Key Words: anthropometric indices, body mass index, metabolic disorders, waist circumference, waist-to-hip ratio

**********

Obesity is a frequently found disorder in the world and remains a major critical public health problem in industrialized countries (1-5) because of its association with multiple chronic disorders, especially various metabolic disorders, including hypertension, dyslipidemia, insulin resistance, hyperinsulinemia, diabetes mellitus/hyperglycemia, hyperuricemia, gallbladder disease, osteoarthritis, some cancers, and increased mortality. (6-13) Obesity is characterized by an excess of fat tissue relative to lean body mass. (14) However, it is relatively difficult and complex to measure body fat; although imaging methods, such as computed tomography (CT), are available to document total body fat and its distributions, (15,16) anthropometric measurements are considered a useful aid in clinical practice. To date, little is known about the relationship between the anthropometric indices and various metabolic disorders in Taiwan. On the other hand, it is usually surprisingly expensive to measure these metabolic disorders in primary health care settings. Undoubtedly, if we can find an easily obtained anthropometric index as a predicting tool for various metabolic disorders, primary care physicians can develop policies and programs to promote healthy body status. Thus, various metabolic disorders also can be avoided.

The three most common anthropometric indices, including body mass index (BMI) indicating overall or generalized obesity, and waist-to-hip ratio (WHR) and waist circumference (WC) indicating abdominal or central obesity, are easily calculated and rather inexpensive tools in primary health care settings. Therefore, we conducted a cross-sectional study to investigate the interrelationship between these three simple anthropometric indices and various metabolic disorders and to demonstrate which index can display the greatest correlation with metabolic disorders. Furthermore, early intervention of obesity can be undertaken smoothly.

Materials and Methods

From January to December 2000, a cross-sectional hospital-based study was performed. We retrospectively analyzed the medical records of all the patients receiving periodic health examination with scheduled items at the Department of Family Medicine of China Medical University Hospital in Taichung City, Taiwan. A total of 746 subjects were included for further analysis. These subjects were grossly healthy and, in most cases, without obvious serious disease. Information was collected about each subject's history of systemic diseases and medication use and complete physical examinations were performed by the doctors at the Department of Family Medicine. An abdominal sonography was performed for each subject by the gastroenterologist using a high-resolution real-time machine (Toshiba Sonolayer SSA-270A, convex type 3.5-MHz transducer; Toshiba, Tochigi-Ken, Japan). Fatty liver was diagnosed according to the international criteria. (17, 18)

Blood pressure was measured using a mercury sphygmomanometer in the sitting position. BMI was measured as follows: weight (kg)/height ([m.sup.2]). WC was measured as the minimum circumference with the tape positioned between xiphoid process and the umbilicus at the end of a normal expiration. (19) Hip circumference (HC) was measured with the tape positioned at the level of the symphysis pubis and the greatest gluteal protuberance. (20) WHR was measured as follows: WC (cm)/HC (cm). Two definitions of obesity were made. Overall or generalized obesity was defined as BMI [greater than or equal to]27. (21) Abdominal or central obesity was defined as WC [greater than or equal to]90 cm for men and [greater than or equal to]80 cm for women (21) or WHR [greater than or equal to]0.95 for men and [greater than or equal to]0.92 for women (greater than 90th percentile of this study population). Venous blood samples were obtained in the morning after a 12-hour overnight fast. A number of biochemical markers, such as total cholesterol (TC), triglyceride (TG), high-density lipoprotein cholesterol (HDL) fasting glucose, and uric acid, were analyzed by a biochemical autoanalyzer (Hitachi 736-15, Tokyo, Japan) at the Department of Clinical Laboratory of China Medical Uiversity Hospital within 4 hours of collection. Hypercholesterolemia was defined as TC [greater than or equal to]200 mg/dL or subjects with hyperlipidemic history. (22,23) Hypertriglyceridemia was defined as TG [greater than or equal to]200 mg/dL or subjects with hyperlipidemic history. (22,23) Abnormal HDL level was defined as HDL less than 35 mg/dL. (22,23) If TG was less than 400 mg/dL, low-density lipoprotein cholesterol (LDL) was calculated by the Friedewald formula: LDL = TC - (HDL + TG/5). (24) Abnormal LDL level was defined as LDL [greater than or equal to]130 mg/dL. (22,23) The ratio of TC/HDL > 5 was defined as abnormal. (25) Hyperglycemia was defined as fasting glucose [greater than or equal to]110 mg/dL or subjects with diabetic history. (26) Subjects were considered to have hypertension if the average of both hands' readings exceeded 140 mm Hg systolically and/or 90 mm Hg diastolically or subjects had a hypertensive history. (27) Hyperuricemia was defined as serum uric acid [greater than or equal to]7.0 mg/dL in men and [greater than or equal to]6.5 mg/dL in women. (28)

Statistical analysis was performed by the aid of an SAS package (version 6.12; SAS Institute Inc., Cary, NC). The t test, [chi square] test, and stepwise multivariate logistic regression were used for statistical analysis. P < 0.05 was considered statistically significant.

Results

There were 332 women (44.5%) and 414 men (55.5%) included in the study with a mean age of 50.74 [+ or -] 12.68 years (range, 19-87 years). Means and standard deviation of anthropometric indices and risk variables are shown in Table 1. Male subjects showed considerably higher levels of WC, WHR, blood pressure, TG, TC/HDL, and uric acid and lower levels of HDL.

The metabolic variables of the [chi square] test by sex and anthropometric indices are shown in Table 2. There is considerable statistical significance between overall obesity and all metabolic variables, except hypercholesterolemia. There is considerable statistical significance between abdominal obesity and all metabolic variables, except abnormal HDL level.

The results of stepwise multivariate logistic regression by anthropometric indices are demonstrated in Table 3. After controlling for the other covariables, there is considerable statistical significance between overall obesity and hypertension (odds ratio [OR] = 1.86, 95% confidence interval [CI] = 1.19-2.89, P < 0.01); hypertriglyceridemia (OR = 2.22, 95% CI = 1.20-4.11, P < 0.05); hyperuricemia (OR = 2.26, 95% CI = 1.47-3.48, P < 0.001); and fatty liver (OR = 2.27, 95% CI = 1.47-3.50, P < 0.001). That is, people with overall obesity are likely to have some metabolic disorders. There is considerable statistical significance between higher WC and an abnormal ratio of TC/HDL (OR = 2.08, 95% CI = 1.24-3.47, P < 0.01); hyperuricemia (OR = 1.68, 95% CI = 1.10-2.56, P< 0.05); and fatty liver (OR = 3.44, 95% CI = 2.27-5.24, P < 0.001). There is considerable statistical significance between higher WHR and fatty liver (OR = 1.76, 95% CI = 1.15-2.69, P < 0.01). That is, people with abdominal obesity are likely to have hyperuricemia and fatty liver. None of the indices were closely associated with hyperglycemia, hypercholesterolemia, abnormal LDL, or abnormal HDL.

Discussion

The adverse effects of excess body fat are associated and increased with obesity. (6-13,20) Body fat distribution is more important than excess body weight in its relationship to cardiovascular disease. (29,30) Various indices for measuring body fat distribution, including BMI, WHR, and WC, have been proposed and have been shown to be related to cardiovascular risk factors. (20,29,31,32) Furthermore, early recognition of obesity and appropriate intervention should never be overlooked.

As in previous reports, WC is used to assess abdominal obesity and is most highly correlated with cardiovascular risk factors. (32-36) However, in our study, WC only showed an association with TC/HDL, hyperuricemia, and fatty liver. Thus, relationships among body fat distribution and metabolic disorders may differ in different populations. That is, Western results may not be applicable to Taiwan when assessing obesity. On the other hand, WHR only showed an association with fatty liver. Indeed, WC alone can be used instead of WHR as an assessment of abdominal obesity, which was stated by Ledoux et al. (31) According to the above statement, it is not necessary to measure WHR in screening and counseling about obesity. On the other hand, our study revealed a considerable association between overall obesity and hypertension and hypertriglyceridemia, which was in agreement with previous reports. (2,9-12,31) Because of the limitation of a cross-sectional study, the cause-and-effect relationship between overall obesity and hypertension or hypertriglyceridemia needs additional detailed investigation.

Furthermore, BMI or WC demonstrated considerable association with hyperuricemia and fatty liver, which strongly indicated a hypothesis that obesity, regardless of overall or abdominal obesity, might play an important role in the pathogenesis of hyperuricemia and fatty liver.

There were several limitations in this study. First is a bias in sampling population. A hospital-based study could not be extrapolated to the general population. Second, the sample size was too small, so the subjects really could not be representative of the Taiwanese general population. A larger research population is necessary for a more accurate association between anthropometric indices and metabolic disorders. Therefore, while not explicitly novel, the addition of this research to the literature is important, especially considering the unique population of Taiwan.

As stated in previous reports, (31,32) the measurement of WC is not easy to standardize, so it is prone to variations. Thus, careful interpretation should be kept in mind. While BMI may not show a significant association with all metabolic disorders, it is an easily obtained index of overall obesity and is well associated with some metabolic disorders. In conclusion, BMI and WC can really predict some metabolic disorders. For practical use, we propose that the combined measurement of BMI and WC is the best anthropometric index to be incorporated into the routine physical examinations of adults. Undoubtedly, if we can periodically measure BMI and WC, early intervention for obesity can be made.

Table 1. Means and standard deviation of anthropometric indices and
metabolic variables by sex (a)

                                                                    P
Variable                     Female               Male            value

BMI (kg/[m.sup.2])      24.11[+ or -]4.14    24.65[+ or -]3.50    0.0591
WC (cm)                 83.75[+ or -]11.09   88.12[+ or -]8.77    0.0001
WHR                      0.87[+ or -]0.08     0.91[+ or -]0.06    0.0001
Systolic pressure      122.88[+ or -]19.23  129.46[+ or -]17.06   0.0001
  (mm Hg)
Diastolic pressure      76.26[+ or -]10.08   81.45[+ or -]10.66  <0.0001
  (mm Hg)
Fasting sugar (mg/dL)  107.05[+ or -]38.26  104.18[+ or -]33.98   0.2852
Total cholesterol      200.44[+ or -]41.86  200.63[+ or -]38.15   0.9482
  (mg/dL)
Triglyceride (mg/dL)   101.40[+ or -]62.44  119.55[+ or -]61.85   0.0001
LDL (mg/dL)            123.57[+ or -]36.81  128.58[+ or -]33.20   0.0540
HDL (mg/dL)             56.59[+ or -]13.70   48.14[+ or -]12.09   0.0001
TC/HDL                   3.69[+ or -]0.97     4.35[+ or -]1.11    0.0001
Uric acid (mg/dL)        5.47[+ or -]1.35     6.59[+ or -]1.35   <0.0001

(a) LDL, low-density lipoprotein cholesterol; HDL, high-density
lipoprotein cholesterol; TC/HDL, ratio of total cholesterol to
high-density lipoprotein cholesterol.

Table 2. Metabolic variables of [chi square] test by sex and
anthropometric indices (a)

                                       HTN      GLU      CHO
Variable                               (%)      (%)      (%)

Sex
  F                                    25.00**  22.59*   47.29
  M                                    35.02**  16.18*   49.03
BMI (kg/[m.sup.2])
  <27                                  26.36**  16.84**  46.43
  [greater than or equal to]27         46.20**  27.22**  55.06
WC (cm)
  F <80 (M <90)                        22.25**  12.06**  42.63**
  F[greater than or equal to]80        38.87**  26.01**  53.89**
    (M[greater than or equal to]90)
WHR
  F <0.92 (M <0.95)                    25.44**  15.55**  44.35**
  F[greater than or equal to]0.92      46.67**  30.00**  60.56**
    (M[greater than or equal to]0.95)

                                         TG     A-LDL    A-HDL
Variable                                 (%)     (%)      (%)

Sex
  F                                     8.13    37.65*    3.61*
  M                                     9.66    46.86**   8.45*
BMI (kg/[m.sup.2])
  <27                                   6.46**  40.31*    5.27*
  [greater than or equal to]27         18.35**  51.90*   10.13*
WC (cm)
  F <80 (M <90)                         5.36**  37.53**   4.83
  F[greater than or equal to]80        12.60**  47.99**   7.77
    (M[greater than or equal to]90)
WHR
  F <0.92 (M <0.95)                     7.07**  39.93**   5.30
  F[greater than or equal to]0.92      15.00**  51.67**   9.44
    (M[greater than or equal to]0.95)

                                       A-TC/HDL   UA        FL
Variable                                (%)       (%)       (%)

Sex
  F                                    10.24**   20.78**   30.42**
  M                                    24.64**   39.13**   44.20**
BMI (kg/[m.sup.2])
  <27                                  14.63**   25.68**   30.10**
  [greater than or equal to]27         31.65**   50.63**   67.72**
WC (cm)
  F <80 (M <90)                        12.06**   24.40**   21.98**
  F[greater than or equal to]80        24.40**   37.53**   54.16**
    (M[greater than or equal to]90)
WHR
  F <0.92 (M <0.95)                    14.66**   28.27**   30.74**
  F[greater than or equal to]0.92      29.44**   39.44**   61.11**
    (M[greater than or equal to]0.95)

(a) HTN, hypertension; GLU, hyperglycemia; CHO, hypercholesterolemia;
TG, hypertriglyceridemia; A-LDL, abnormal low-density lipoprotein
cholesterol; A-HDL, abnormal high-density lipoprotein cholesterol;
A-TC/HDL, abnormal ratio of total cholesterol to high-density
lipoprotein cholesterol; UA, hyperuricemia; FL, fatty liver.
*p < 0.05, **p < 0.01.

Table 3. Metabolic variables of stepwise multivariate logistic
regression by anthropometric indices (a)


                                    HTN          GLU          CHO
                                    OR           OR           OR
Variable                           (95%)        (95%)        (95%)

Age (yr)                            1.06***      1.04***      1.03***
                                   (1.05-1.08)  (1.02-1.06)  (1.02-1.04)
Sex
  (F as reference)
  M                                 2.10***      0.76         1.16
                                   (1.46-3.03)  (0.51-1.12)  (0.85-1.58)
BMI
  (<27 as reference)
  [greater than or equal to]27      1.86**       1.37         1.12
                                   (1.19-2.89)  (0.84-2.21)  (0.74-1.70)
WC
  (F <80, M <90 as
    reference)
  F[greater than or equal to]80
    (M[greater than or equal to]
    90)                             1.54         1.63         1.18
                                   (0.99-2.39)  (0.99-2.67)  (0.81-1.71)
WHR
  (F <0.92, M <0.95
    as reference)
  F[greater than or equal to]0.92
    (M[greater than or equal to]
    0.95)                           1.25         1.28         1.39
                                   (0.81-1.90)  (0.80-2.03)  (0.93-2.07)

                                    TG           A-LDL        A-HDL
                                    OR            OR           OR
Variable                           (95%)         (95%)        (95%)

Age (yr)                            0.99         1.03***      1.01
                                   (0.97-1.02)  (1.01-1.04)  (0.99-1.04)
Sex
  (F as reference)
  M                                 1.25         1.63**       2.69**
                                   (0.73-2.14)  (1.19-2.22)  (1.34-5.41)
BMI
  (<27 as reference)
  [greater than or equal to]27      2.22*        1.29         1.46
                                   (1.20-4.11)  (0.86-1.95)  (0.68-3.11)
WC
  (F <80, M <90 as
    reference)
  F[greater than or equal to]80
    (M[greater than or equal to]
    90)                             1.62         1.32         1.45
                                   (0.80-3.28)  (0.90-1.93)  (0.66-3.20)
WHR
  (F <0.92, M <0.95
    as reference)
  F[greater than or equal to]0.92
    (M[greater than or equal to]
    0.95)                           1.47         1.07         1.25
                                   (0.79-2.73)  (0.72-1.59)  (0.59-2.62)

                                   A-TC/HDL       UA           FL
                                      OR          OR           OR
Variable                            (95%)        (95%)        (95%)

Age (yr)                            1.02         1.01         0.99
                                   (0.99-1.03)  (0.99-1.02)  (0.98-1.01)
Sex
  (F as reference)
  M                                 3.54***      2.82***      2.59***
                                   (2.26-5.55)  (1.98-4.03)  (1.81-3.71)
BMI
  (<27 as reference)
  [greater than or equal to]27      1.62         2.26***      2.27***
                                   (0.99-2.65)  (1.47-3.48)  (1.47-3.50)
WC
  (F <80, M <90 as
    reference)
  F[greater than or equal to]80
    (M[greater than or equal to]
     90)                            2.08**       1.68*        3.44***
                                   (1.24-3.47)  (1.10-2.56)  (2.27-5.24)
WHR
  (F <0.92, M <0.95
    as reference)
  F[greater than or equal to]0.92
    (M[greater than or equal to]
    0.95)                           1.36         0.95         1.76**
                                   (0.84-2.20)  (0.62-1.46)  (1.15-2.69)

(a) OR, odds ratio: 95% CI, 95% confidence interval.
*p < 0.05, **p < 0.01, ***p < 0.001.

Accepted August 19, 2002.

Copyright [c] 2004 by The Southern Medical Association

0038-4348/04/9706-0578

References

1. Ginsberg-Fellner F, Jagenforf LA, Carmel H, et al. Overweight and obesity in preschool children in New York City. Am J Clin Nutr 1981; 34:2236-2241.

2. Van Itallie TB. Health implications of overweight and obesity in the United States. Ann Intern Med 1985;103:983-988.

3. Chou P, Liao MJ, Kuo HS, et al. Program description and preliminary health survey data in Kin-Hu, Kinmen. Zhonghua Yi Xue Za Zhi (Taipei) 1993;52:241-248.

4. Melchionda N, Enzi G, Caviezel F, et al. Epidemiology of obesity in the elderly: CNR multicentric study in Italy. Diabetes Res Clin Pract 1990; 10:11-16.

5. DeVore PA. Assessment of nutritional status and obesity in elderly patients as seen in general medical practice. South Med J 1993;86:1008-1010.

6. Pi-Sunyer FX. Health implications of obesity. Am J Clin Nutr 1991;53: 1595-1603.

7. Kannel WB. Risk stratification in hypertension: New insights from the Framingham Study. Am J Hypertens 2000;13:3-10.

8. Miccoli R, Ceraudo AM, Manfredi SG, et al. Atherogenic dyslipidemia, metabolic syndrome, and cardiovascular risk. Cardiologia 1999;44:885-99.

9. Lin CC, Li TC, Lai SW, et al. Epidemiology of obesity in elderly people. Yale J Biol Med 1999;72:385-391.

10. Lai SW, Tan CK, Ng KC. Epidemiology of hyperuricemia in the elderly. Yale J Biol Med 2001;74:151-157.

11. Lin CC, Li TC, Lai SW, et al. Hypertriglyceridemia and the related factors in middle-aged adults in Taiwan. Mid Taiwan J Med 2001;6:1-6.

12. Ng KC, Lin CC, Lai SW, et al. Obesity and the related factors in middle-aged adults in Chung-Hsing village in Taiwan. Mid Taiwan J Med 2001;6:35-39.

13. Tan CC, Ng KC, Lai SW, et al. The results of preventive services for adults at a medical center in Taichung. Mid Taiwan J Med 2001;6:233-237.

14. Chen W. Childhood obesity in Taiwan. Zhonghua Min Guo Xiao Er Ke Yi Xue Hui Za Zhi 1997;38:438-442.

15. Ko GT, Chan JC, Woo J, et al. Simple anthropometric indexes and cardiovascular risk factors in Chinese. Int J Obes Relat Metab Disord 1997;21:995-1001.

16. Ferland M, Despres JP, Tremblay A, et al. Assessment of adipose tissue distribution by computed axial tomography in obese women: Association with body density and anthropometric measurements. Br J Nutr 1989;61:139-148.

17. Lin DY, Sheen IS, Chiu CT, et al. Clinical significance of ultrasonographic fatty liver in asymptomatics: Analysis of 1040 check-up subjects. J Med Ultrasound 1993;1:165-171.

18. Lin SC, Shih SC, Kao CR, et al. Prevalence of antibodies to hepatitis C virus in patients with nonalcoholic fatty liver. Zhonghua Yi Xue Za Zhi (Taipei) 1995;56:80-85.

19. Woo J, Swaminathan R, Cockram C, et al. Association between serum uric acid and some cardiovascular risk factors in a Chinese population. Postgrad Med J 1994;70:486-491.

20. Macdonald SM, Reeder BA, Chen Y, et al. Obesity in Canada: A descriptive analysis. Canadian Heart Health Surveys Research Group. CMAJ 1997;157:S3-S9.

21. Department of Health, Taiwan. Available at: http://www.doh.gov.tw/dohenglish/index.asp. Accessed on April 29, 2003.

22. Summary of the second report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel II). JAMA 1993;269:3015-3023.

23. Summary of the national guidelines for the diagnosis and management of lipid disorders in Taiwan: The experts panel. Zhonghua Yi Xue Za Zhi (Taipei) 1995;55:1-14.

24. Friedwald WT, Levy RI, Fredrickson DS. Estimation of the concentration of low-density lipoprotein cholesterol in plasma without use of the preparative ultracentrifuge. Clin Chem 1972;18:499-502.

25. Paccaud F, Schluter-Fasmeyer V, Wietlisbach V, et al. Dyslipidemia and abdominal obesity: An assessment in three general populations. J Clin Epidemiol 2000;53:393-400.

26. Report of the expert committee on the diagnosis and classification of diabetes mellitus. Diabetes Care 2000;23(Suppl 1):S4-S19.

27. Subcommittee of WHO/ISH Mild Hypertension Liaison committee. Summary of 1993 World Health Organization-International Society of Hypertension guidelines for the management of mild hypertension. BMJ 1993;307:1541-1546.

28. Saggiani F, Pilati S, Targher G, et al. Serum uric acid and related factors in 500 hospitalized subjects. Metabolism 1996;45:1557-1561.

29. Ledoux M, Lambert J, Reeder BA, et al. Correlation between cardiovascular disease risk factors and simple anthropometric measures: Canadian Heart Health Surveys Research Group. CMAJ 1997;157:S46-S53.

30. Pihl E, Jurimae T. Cardiovascular disease risk factors in males with normal body weight and high waist-to-hip ratio. J Cardiovasc Risk 2001;8:299-305.

31. Ledoux M, Lambert J, Reeder BA, et al. A comparative analysis of weight-to-height and waist-to-hip circumference indices as indicators of the presence of cardiovascular disease risk factors. Canadian Heart Health Surveys Research Group. CMAJ 1997;157:S32-S38.

32. Reeder BA, Senthilselvan A, Despres JP, et al. The association of cardiovascular disease risk factors with abdominal obesity in Canada. Canadian Heart Health Surveys Research Group. CMAJ 1997;157:S39-S45.

33. Jee SH, Kim MT, Lee SY, et al. Segregation analysis of waist circumference, hip circumference and waist-to-hip ratio in the Korean Nationwide Family Study. Int J Obes Relat Metab Disord 2002;26:228-233.

34. Lakka HM, Lakka TA, Tuomilehto J, et al. Abdominal obesity is associated with increased risk of acute coronary events in men. Eur Heart J 2002;23:706-713.

35. Lai SW, Tan CC, Ng KC. Hepatic effects in hyper lipidemic patients. Mid Taiwain J Med 2002;7:160-164.

36. Lin CH, Lai SW, Liu CS. Prevalence of hypercholesterolemia and its related factors in middle-aged Taiwanese adults--a hospital-based study. Mid Taiwan J Med 2003;8:85-90.

RELATED ARTICLE: Key Points

* Obesity remains a major critical public health problem in industrialized countries because of its association with various metabolic disorders.

* It is relatively difficult to quantify obesity.

* The three most common anthropometric indices, including body mass index, waist-to-hip ratio, and waist circumference are easily calculated tools to evaluate obesity in primary health care settings.

* We conducted a cross-sectional study to investigate which index can display the greatest correlation with metabolic disorders.

Shih-Wei Lai, MD, and Kim-Choy Ng, MD

From the Departments of Community Medicine and Emergency Medicine, China Medical University Hospital, Taichung, Taiwan.

Reprint requests to Kim-Choy Ng, MD, Department of Emergency Medicine. China Medical University Hospital, No. 2, Yuh-Der Road, Taichung 404, Taiwan. Email: [e-mail]