Health risks among Japanese men with moderate body mass index


BACKGROUND: Despite the westernized lifestyle, most Japanese are not prominently obese. As their obesity may be obscured, we want to further explore central fat distribution and health risks among this population.

METHODS: 2668 men (body mass index (BMI) 20–<26.4 kg/m2) were identified from 3343 men (BMI 14.0–37.7, median of waist‐to‐height ratio (W/Ht) 0.50) who underwent routine health examinations. They were divided into four groups: (1) BMI 20–<24, W/Ht<0.50; (2) BMI 20–<24, W/Ht≥0.50; (3) BMI 24–<26.4, W/Ht<0.50; and (4) BMI 24–<26.4, W/Ht≥0.50. The metabolic risks and physical activities were compared.

RESULTS: W/Ht≥0.50 comprised 35% of the subjects of BMI 20–<24, 91% of BMI 24–<26.4, only 1% of BMI<20 and up to 99% of BMI≥26.4. After adjusting for age and smoking and with group 1 as the reference group, odds ratios (ORs) were significantly higher in both groups 2 and 4 for the risk of hypertension (1.42, 1.98), hyperglycaemia (1.49, 1.78), hypertriglyceridaemia (1.95, 2.56), low HDL cholesterol (1.87, 2.20), hyperuricaemia (2.00, 2.51) and fatty liver (2.57, 5.64) and additionally in group 4 for hypercholesterolaemia (1.35). ORs in group 3 were only significantly higher for risk of hypertension (1.86) and fatty liver (1.89). Significantly lower frequencies of regular physical activity were noted in the higher W/Ht groups (0.75, 0.47, 1.10, 0.52 days per week, from groups 1–4, respectively).

CONCLUSIONS: Measurement of waist circumference may be a preliminary method for the survey of people at higher risk of lifestyle‐related disorders in Japanese men, especially among those with moderate BMI.


There is increasing data to suggest that coronary heart disease and its risk factors are more closely related to excessive central fat accumulation than total body obesity,1,2,3,4,5,6 therefore, a simple and practical index of central fat distribution is important. Waist circumference has been reported to be a simple index of central obesity by some investigators, however, the waist circumference proposed by the above investigators as indexes for weight control in men are over 90 cm and/or over 100 cm from the study populations of high mean body mass index (BMI).7,8,9 Therefore, those people may already know of the necessity of weight loss before any measurement of waist circumference is made.

It has been reported that the metabolic conditions which predispose to obesity among normal weight individuals (the so called ‘metabolically obese’) are common in the general population.10 The conditions are even more common among the Japanese. Even with the rapid westernization of lifestyle in Japanese, the prevalence of obesity in the Japanese population is far less than in Western populations.11 the same situation may be also present in many other oriental countries. Mean BMI among Japanese with normal glucose tolerance, impaired glucose tolerance, or diabetes was reported to be less than 25 kg/m2 for all.12 On the other hand, people who exercise more frequently and who have more muscle mass may also have a higher BMI. Consequently, there are defects in using the BMI for assessing inappropriate or excessive fat accumulation.13 Since waist‐to‐height ratio has been reported to be a useful obesity index14,15,16,17,18,19,20,21 and an independent predictor of multiple coronary risk factors when waist‐to‐height ratio and BMI were compared,14,15,16,17 we analysed routine health examination data to determine whether this simple index better reflects lifestyle and lifestyle‐related disorders among such subjects. Because weight loss is difficult and rebound often occurs in obese people, we believe the best weapon against obesity is to educate the people to modify their lifestyles as soon as possible before they become prominently obese.


Subjects and data collection

The subjects were 3343 men (age, 21–85 y; mean, 48.5; BMI, 14–37.7 kg/m2; mean 23.1) who underwent routine health examinations at Toranomon Hospital, Tokyo from April 1994 to March 1995. Most of the examinees were office workers. Since the distribution of our examinees according to BMI values was <1% for BMI greater than 30, we used the definition of the Japan Society for the Study of Obesity to classify them. According to this definition: underweight=BMI<20; normal range=BMI 20–<24; overweight=BMI 24–<26.4; obese=BMI≥26.4.22 Using this definition, only 9.5% of the examinees were classified as ‘obese’. The majority of the subjects (79.8%) had BMIs within the 20–<26.4 range.

The basic survey questions included ‘How often do you engage in regular physical activity lasting 30 minutes or more?’ and ‘What is your current smoking status?’ Frequency of regular physical activity was calculated as 0 if they did not engage in regular physical activity at least once a week for 30 min or more.23 The sedentary group accounted for the majority of the subjects. Tennis and jogging were the most frequent activities in the physically active groups.

Waist circumference was measured at the umbilical level with the subjects standing and breathing normally and waist‐to‐height ratio was calculated.14,15,16,17 Blood was drawn in the fasting state. Blood glucose, serum triglyceride, cholesterol, HDL cholesterol and uric acid concentrations were measured using enzymatic methods. Fatty liver was diagnosed by ultrasound.

A total of 2668 subjects within the moderate BMI range (20–<26.4) were divided into four groups according to the BMI categories and whether the value of waist‐to‐height ratio (W/Ht) was less than or greater than/equal to the median value (0.50) of 3343 persons during the same period: group 1, BMI 20–<24, W/Ht<0.50; group 2, BMI 20–<24, W/Ht≥0.50; group 3, BMI 24–<26.4, W/Ht< 0.50; group 4, BMI 24–<26.4, W/Ht≥0.50.

Statistical analysis

Probabilities of significant differences were compared using the chi‐square test for proportions. Odds ratios for lifestyle-related disorders with reference to group 1, adjusted by age and current smoking status, were calculated using JMP software. Probabilities of significant differences for quantitative values of unadjusted and age and smoking adjusted data were compared by SAS software.


Relationship between BMI and waist‐to‐height ratio

There was a strong correlation between BMI and waist‐to‐height ratio (r=0.87, P<0.0001), however, when using waist‐to‐height ratio 0.50 as a cut‐off point, waist‐to‐height ratio ≥0.50 comprised 35% of the subjects with a BMI of 20–<24 and 91% of subjects with a BMI of 24–<26.4, whereas it comprised only 1% of the subjects with a BMI of <20 and up to 99% of the subjects with a BMI of ≥26.4. Waist‐to‐height ratio ≥0.50, on the other hand, comprised no subjects with a BMI of <19.3 and 100% of the subjects with a BMI of >28.3 (Figure 1).

Figure 1

Relationship between body mass index and waist-to-height ratio (r=0.87, waist-to-height ratio=0.166+0.015 body mass index, P<0.0001).

Clinical features and prevalence of lifestyle-related disorders by BMI and waist‐to‐height ratio

The groups with a waist‐to‐height ratio ≥0.50 were a little older. The differences for the prevalence of lifestyle‐related disorders were significant among the four groups in all items, including hypertension, hyperglycaemia, hypertriglyceridaemia, hypercholesterolaemia, low HDL cholesterol level, hyperuricaemia and fatty liver (Table 1).

Table 1 Clinical features and prevalence of lifestyle-related disorders in four groups according to BMI and waist-to-height ratio

Odds ratios for lifestyle‐related disorders by BMI and waist‐to‐height ratio

After adjusting for age and smoking status and with group 1 as the reference group, odds ratios were significantly higher in both groups 2 and 4 for the risk of hypertension, hyperglycaemia, hypertriglyceridaemia, low HDL cholesterol, hyperuricaemia and fatty liver and additionally in group 4 for hypercholesterolaemia. Odds ratios in group 3 were only significantly higher for risk of hypertension and fatty liver (Table 2).

Table 2 Odds ratio for lifestyle-related disorders with reference to group 1, adjusted for age and smoking

Regular physical activity by BMI and waist‐to‐height ratio

From the unadjusted data, there were significant differences in the frequencies of regular physical activity among the groups (P<0.0001). The frequencies of regular physical activity were significantly higher in groups 1 and 3 than groups 2 or 4, and there were no significant differences between groups 1 and 3 or groups 2 and 4, although the frequencies of regular physical activity tended to be highest in group 3 (Figure 2, left side). After adjusting for age and smoking, the comparisons of regular physical activity among the groups were similar to the unadjusted data, and the regular physical activity was about 1.6 times for group1 vs 2 and 2.1 times for group 3 vs 4 (Figure 2, right side).

Figure 2

Frequency of regular physical activity (mean±s.e.) among the four groups. Left side: unadjusted groups; right side: age and smoking adjusted groups. Overall difference by one‐way ANOVA: P<0.0001 for both unadjusted and age and smoking adjusted data. Inter‐group difference by Tukey–Kremer's method: unadjusted groups—P<0.0001 for 1 vs 2; P=0.0935 for 1 vs 3; P=0.0009 for 1 vs 4; P=0.0002 for 2 vs 3; P>0.2 for 2 vs 4; P=0.0011 for 3 vs 4. Age and smoking adjusted groups: P<0.0001 for 1 vs 2; P=0.0894 for 1 vs 3; P=0.0007 for 1 vs 4; P=0.0003 for 2 vs 3; P>0.2 for 2 vs 4; P=0.0010 for 3 vs 4. W/Ht indicates waist‐to‐height ratio.


In a survey of US adults (1988–1994), approximately 33% of men were overweight (BMI≥27.8)24 In contrast, only 9.5% of Japanese men examined had a BMI≥26.4 in our study. However, there was still possibility of higher health risks associated with excessive central fat accumulation among the remaining subjects with moderate range BMI because many of them had a lack of regular physical activity.23

In our previous studies, waist‐to‐height ratio was the single independent variable to multiple coronary risk factors according to the results of multiple regression analysis of the waist‐to‐height ratio and BMI.14,15,17 On the other hand, in a cross‐sectional study, waist‐to‐height ratio and the prevalence of fatty liver, but not BMI, were found to be significantly lower in the groups that exercised two or more times per week than in the sedentary group.23 Since central fat distribution is more closely associated with inappropriate lifestyle and related disorders, waist‐to‐height ratio may be a practical method for surveying such disorders. We used a waist‐to‐height ratio of 0.50 as a cut‐off value to define central fat distribution, not only because it was the median and mean of our study subjects, and because coronary risk factors were significantly higher above it while waist‐to‐height ratio of the people who exercised on 2 days or more were significantly lower than it,14,15,16,23 but also this measurement is simple and convenient. Another benefit stems from the fact that, as a mathematical index, BMI is difficult to explain to examinees and to relate it to weight that must be lost.25

Despite the difference of BMI, group 3 was similar to group 1 in many aspects, including higher frequency of regular physical activity and lower prevalences of lifestyle‐related disorders, especially for the disorders associated with sedentary life such as hyperglycaemia, hypertriglyceridaemia and low HDL cholesterol level. The higher BMI of group 3 might be partly due to larger muscle mass which suggests group 3 should be treated differently from group 4. However, group 3 had higher prevalences of hypertension and fatty liver than group 1. This may be due to either slightly higher waist‐to‐height ratios (although still below 0.50) of group 3, corresponding to larger body size, or maybe the effect of larger lean body mass on hypertension.26 Although the relative population of group 3 was small in this study, the population may be larger among people who do more muscular work. The real health risks of group 3 compared to those of group 1 requires further exploration.

According to our study, up to 99% of the subjects whose BMI was ≥26.4 had a waist‐to‐height ratio≥0.50 and 99% of the subjects whose BMI was <20 had a waist‐to‐height ratio<0.50. It suggests that both the indexes of weight and waist may be corresponding for the definition of obesity in these ranges. However, there was variability in waist‐to‐height ratios among the subjects with BMI 20–<26.4. Therefore, measurement of waist‐to‐height ratio may be more practical among Japanese men of moderate BMI range (BMI 20–<26.4) as a way of identifying higher risk people.

Waist‐to‐height ratio is potentially confounded by height. Both BMI and waist‐to‐height ratio were positively correlated with all of the disorders mentioned, while height was only weakly negatively correlated with hypertension (−0.05, P=0.006) and hyperglycaemia (−0.07, P=0.0007) in the men with a moderate BMI. The prediction of coronary risk factors by waist‐to‐height ratio might be due to the joint effects of central obesity and height for some risk factors. There are reports that the relative mortality and risk of cardiovascular disease are high in relatively short populations27,28,29 If this is true, then height not only has the role of adjusting for stature, but helps this index better reflect the outcome of the disease. In any event, further study is necessary to clarify this point.

Accumulation of central fat may be due to energy imbalance from overnutrition, aging, genetic factors or hormone disorders, and may be exaggerated by a sedentary life. Prevalence of lifestyle‐related disorders, physical inactivity and probable overnutrition are very common in Japanese men. Our studies suggest that measurement of waist circumference may be a preliminary method to reflect lifestyle‐related disorders and may be particularly important for most Asians, who have a lower BMI than Caucasians.


  1. 1

    Vague J . The degree of masculine and differentiation of obesities Am J Clin Nutr 1956 4:: 20–34.

  2. 2

    Lasson B, Svädsudd K, Welin L, Wilhelmsen L, Bjöntrop P, Tibblin G . Abdominal adipose tissue distribution, obesity, and risk of cardiovascular disease and death: 13 year follow up of participants in the study of men born in 1913 Br Med J 1984 288: 1401–1404.

  3. 3

    Fujioka S, Matsuzawa Y, Tokunaga K, Tarui S . Contribution of intra‐abdominal fat accumulation to the impairment of glucose and lipid metabolism in human obesity Metabolism 1987 36: 54–59.

  4. 4

    Deprés J‐P . Lipoprotein metabolism in visceral obesity Int J Obes Relat Metab Disord 1991 15: 45–52.

  5. 5

    Blair D, Habicht J‐P, Sims EAH, Sklwester D, Abraham S . Evidence for an increased risk for hypertension with centrally located body fat and the effect of race and sex on this risk Am J Epidemiol 1984 119: 526–539.

  6. 6

    Kaplan NM . The deadly quartet. Upper‐body obesity, glucose intolerance, hypertriglyceridemia, and hypertension Arch Intern Med 1989 149: 1514–1520.

  7. 7

    Pouliot M‐C, Després J‐P, Lemieux S, Moorjani S, Bouchard C, Tremblay A et al.Waist circumference and abdominal saggital diameter. Best simple anthropometric indexes of abdominal visceral adipose tissue accumulation and related cardiovascular risk in men and women Am J Cardiol 1994 73: 460–468.

  8. 8

    Lean MEJ, Han TS, Morrison CE . Waist circumference as a measure for indicating need for weight management Br Med J 1995 311: 158–161.

  9. 9

    Lean MEJ, Han TS, Seidell JC . Impairment of health and quality of life in people with large waist circumference Lancet 1998 351: 853–856.

  10. 10

    Ruderman N, Chisholm D, Pi‐Sunyer Xavier, Schneider S . Perspective in diabetes. The metabolically obese, normal‐weight individuals revised Diabetes 1998 47: 699–713.

  11. 11

    Yoshiike Y, Matsumara M, Zaman MM, Yamaguchi M . Descriptive epidemiology of body index in Japanese adults in a representative sample from the National Nutrition Survey 1990–1994 Int J Obes Relat Metab Disord 1995 19:: 585–589.

  12. 12

    Kosaka K, Kuzuya T, Yoshinaga H, Hagura R . A prospective study of health check examinees for the development of non‐insulin‐dependent diabetes mellitus: relationship of the incidence of diabetes with the initial insulinogenic index and degree of obesity Diabetic Med 1998 22: 684–687.

  13. 13

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

  14. 14

    Hsieh SD, Yoshinaga H . Abdominal fat distribution and coronary heart disease risk factors in men—waist/height ratio as a simple and useful predictor Int J Obes Relat Metab Disord 1995 19: 585–589.

  15. 15

    Hsieh SD, Yoshinaga H . Waist/height ratio as a simple and useful predictor of coronary heart disease risk factors in women Intern Med 1995 34: 1147–1152.

  16. 16

    Hsieh SD, Yoshinaga H . Is there any difference in coronary heart disease risk factors and prevalence of fatty liver in subjects with normal body mass index having different physiques? Tohuku J Exp Med 1995 177: 223–231.

  17. 17

    Hsieh SD, Yoshinaga H . The relationships between various obesity indices and coronary heart risk factors. The comparisons of waist/height ratio to other obesity indices J Japan Human Dry Dock 1996 11: 26–29 (in Japanese).

  18. 18

    Lee JS, Aoki K, Kawakubo K, Gunji A . A study on indices of body fat distribution for screening for obesity J Occup Health 1995 37: 9–18.

  19. 19

    Ashwell M, Lejeune S, McPherson K . Ratio of waist circumference to height may be better indicator of need for weight management Br Med J 1996 312: 377.

  20. 20

    Cox BD, Whichelow MJ, Ashwell M, Prevost AT, Lejeune SRE . Association of anthropometric indices with elevated blood pressure in British adults Int J Obes Relat Metab Disord 1997 21: 674–680.

  21. 21

    Ashwell M, Cole TJ, Dixon AK . Ratio of waist circumference to height is a strong predictor of intra‐abdominal fat Br Med J 1996 313: 559–560.

  22. 22

    Japan Society for the Study of Obesity . Obesity: the guidance of diagnosis and treatment. Ishiyaku: Tokyo, 1993, p18 (in Japanese).

  23. 23

    Hsieh SD, Yoshinaga H, Muto T, Sakurai Y . Regular physical activity and coronary risk factors in Japanese men Circulation 1998 97: 661–665.

  24. 24

    National Center for Health Statistics and National Center for Chronic Disease Prevention and Health Promotion . Update: prevalence of overweight among children, adolescent and adults‐United States, 1988–1994 JAMA 1997 277 (14): 1111.

  25. 25

    National Institutes of Health Consensus Development Panel on the Health Implications of Obesity . Health implications of obesity, national institutes of health consensus development conference statement Ann Intern Med 1985 103 (6 pt 2): 1073–1077.

  26. 26

    Weinsier RL, Norris DJ, Birch R, Bernstein RS, Wang J, Yang M‐U et al.The relative contribution of body fat and fat pattern to blood pressure level Hypertension 1985 7: 578–585.

  27. 27

    Waaler Ht . Height, weight and mortality—the Norwegian experience Acta Med Scand 1984 679 (Suppl): 1–56.

  28. 28

    Hebert PR, Rich‐Ewards JW, Manson JE, Ridker PM, Cook NR, O'Connor GT et al.Height and incidence of cardiovascular disease in male physicians Circulation 1993 88 (part 1): 1437–1443.

  29. 29

    Parker DR, Lapane KL, Lasater TM, Carleton RA . Short stature and cardiovascular disease among men and women from two southeastern New England communities Int J Epidemiol 1998 27: 970–975.

Download references


We thank Mitsui Chemical Industries Ltd, for their financial assistance in this study and Miss Miyuki Suzuki for her secretarial assistance in the preparation of this manuscript.

Author information

Correspondence to SD Hsieh.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Hsieh, S., Yoshinaga, H., Muto, T. et al. Health risks among Japanese men with moderate body mass index. Int J Obes 24, 358–362 (2000).

Download citation


  • moderate BMI
  • waist‐to‐height ratio
  • health risks

Further reading