Abstract

Anthropometric indicators of abdominal obesity are associated with cardiovascular risk factors, such as type 2 diabetes, hypertension, and dyslipidemia. Controversy remains regarding the best anthropometric indices for cardiovascular risk. Waist-to-height ratio has been reported to be an effective predictor of metabolic risks and it may be a better measure of relative fat distribution amongst subjects of different age and statures. Bangladeshi data lack in this perspective. To determine waist-to-height ratio of Bangladeshi adults along with its variation with socio-economic status, cross-sectional studies were conducted in 2002 and 2003. Data were collected through interviewing and measuring height and waist circumference of 22,995 adult males and females of an urban (Mirpur, Dhaka City) and rural area (Kaliganj sub-district). The mean waist-to-height ratio of 0.48 significantly varied with socio-demographic variables and it was markedly higher in females, older age groups, urban residents and the better educated. Urban residents, females, older people, better educational status, the non-paid and married individuals were more likely to have high waist-to-height ratio (³0.5). High waist-to-height ratio levels using sex-specific cut-offs were more common in females, urban residents, Christians, older individuals, married, the better educated and the non-paid. Age and locality were identified as best predictors in males and females, respectively.

Ibrahim Med. Coll. J. 2010; 4(2): 49-58

Key words: Waist-to-height ratio, adult, Bangladeshi

Address for Correspondence:Dr. Meerjady Sabrina Flora, Associate Professor of Epidemiology, National Institute of Preventive and Social Medicine, Mohakhali, Dhaka, Bangladesh. e-mail: [email protected]

Introduction

Bangladesh is having a double burden of health problems; the occurrence of non-communicable diseases is also increasing in addition to existence and emergence of infectious diseases. Moreover, it faces nutrition transition with over-nutrition and under-nutrition occurring simultaneously. While about a quarter of rural, and lower class urban people have chronic energy deficiency; the prevalence of obesity in the upper and middle class urban people is between 9-11%.1 It is being increasingly recognised that central, rather than general obesity, is likely to coexist with type 2 diabetes and lead to complications including cardiovascular diseases. Although its importance is acknowledged, no unified definition exists for central obesity; several anthropometric indexes such as waist circumference (WC), waist-hip ratio (WHR), waist-to-height ratio (WHtR), conicity index (Cindex) etc, are being used.2 These anthropometric indices are associated with cardiovascular risk factors, such as type 2 diabetes, hypertension, and dyslipidemia. However, controversy remains regarding the best anthropometric indices for cardiovascular disease (CVD) risk.3 WC was the main variable used as a measure of central obesity as it is much simpler and more practical to use and because it associates more strongly with cardio-vascular diseases and is a better predictor of future risk of metabolic diseases.4 However, WC measurement has been criticized for not taking into account differences in body height, and the WHtR value is a better predictor of cardiovascular risk factors.3 The ratio of waist circumference to height may be a superior measure for women as well as men5 and a simple index for measuring coronary risk. Waist circumference reflects abdominal obesity, and height is relatively constant in adults and can be used to compensate for variations in frame size.6 WHtR has been reported to be an effective predictor of metabolic risks and it may be a better measure of relative fat distribution amongst subjects of different age and statures.7 The index, especially for women, is a better indicator for predicting obesity-related CVD risk factors than other indices.8

Although it was thought previously that there might be a sex difference, waist-to-height ratio has been reported to have closer values between men and women than body mass index (BMI) and WC.5 The distribution of the ratio is broadly similar in both sexes, mean values being only slightly higher in men than women (0.54 ± 0.06 versus 0.51 ± 0.07). Therefore the same boundary value may be applied to both men and women. One particular advantage of using WHtR might be that ‘unisex’ action levels could be specified.5 A cut-off of 0.5 of WHtR has been considered as a simple and effective index to identify overweight and normal weight Japanese with higher metabolic risk.7 Lin et al. (2002) suggested WHtR cut-offs of 0.48 and 0.45 in men and women, respectively, as appropriate for defining high risk in Taiwan.8

Collection of good quality national data on different indicators of central obesity is needed. So far data available in this regard, are mostly on WC and WHR. In an earlier publication of this study the Cindex of Bangladeshi population is described.9 The current attempt is to explore WHtR. A small scale study, which was done on rural adults only, provide mean WHtR data of adult male (0.43 ± 0.04) and female (0.44 ± 0.05) Bangladeshi.10 Therefore, the current study attempted to find out the WHtR of rural as well as urban adults from a large sample.

Materials and methods

This cross-sectional study was undertaken in an urban (Mirpur, Dhaka City) and rural area (Kaliganj sub-district) in 2002 and 2003. Every alternate household which fulfilled the selection criteria (at least one male and one female ³18 years were available), were recruited. A total of 22,995 adult males and females were interviewed. Anthropometric measurements were taken using validated equipment based on standard procedures.11 A pre-tested structured questionnaire printed in Bangla was used for data collection. Verbal consent was obtained from every respondent and interviews were held in a private place. Ethical clearance was obtained from the Institutional Ethical Committee.

Subjects were measured wearing minimal attire. All the equipment was checked regularly to minimise random errors. Height was measured to the nearest 0.1 cm with a specially constructed wooden height stand to which a plastic measuring tape was attached. The subject stood without shoes or head gear (cap, ribbon etc) in an upright posture with their head in the Frankfurt plane. Subjects were asked to keep their heels close together with their hands hanging freely by their side, palms facing inwards. The horizontal blade of the stadiometer was gently placed on the crown of the head to take the measurement. A flexible plastic tape was used to measure waist circumference, accurate up to the nearest 0.1cm. Waist circumference was measured at the level mid way between the lowest rib margin and the superior iliac crest on the mid-axillary line in a horizontal plane. The subjects stood erect with abdomen relaxed, the arms at the side and feet together and breathing normally.

The analyses were carried out primarily using the Statistical Package for Social Sciences (SPSS) version 14.0. Univariate statistical tests used to determine the association between exposure and outcome variables included Student t-test and c2 test. A result was considered significant at a p value level <0.05 but given the large sample sizes a more stringent cut-off of p<0.01, or less, was usually used. In addition because a number of statistical tests were conducted, the Bonferroni correction (a/K, where a is the p value & K is the number of tests used) was used. Effects of exposure variables were also assessed after adjusting for other variables by multivariate analyses.

Results

The mean (SD) waist-to-height ratio was 0.48 (0.07), but there was considerable variation in relation to socio-demographic status (Table 1). Age showed a curvilinear association (3rd order polynomial) with WHtR; WHtR gradually increased with age, ending in a plateau in the 40-69 age group and falling slightly in the 70+ group; after correcting for sex the trend was more pronounced with greater increments between each age group. Females had higher WHtR than males, before and after, controlling for age effects. Urban residents had, on average, a higher WHtR while Hindus, unmarried individuals and manual labourers had, on average, a lower WHtR. There was a general upward trend in mean WHtR with improvement in educational status.

Table 1: Waist-to-Height Ratio in Relation to the Socio-demographic Variables

To see the effect of each socio-demographic variable after controlling for other socio-demographic variables, sequential multiple regression analyses were undertaken. The model was significant (F = 259.7; p<0.001) and explained 19.3% variation in WHtR. Strong influences of sex and locality with WHtR remained after adjustment for the other socio-demographic variables and the association with education was more marked. Females, urban residents, married individuals and the non-paid had, on average, higher WHtR than their counterparts, while younger individuals, Muslims, non-educated respondents had lower WHtR, on average.

Waist-to-height ratio was categorised as normal and high based on a cut-off of 0.5 for both sexes. Overall 32% of the sample were found in the high category although the percentages varied widely by socio-demographic variable (Table 2). Females and urban residents were almost twice more likely to have high WHtR. The proportion of high WHtR increased with age up to 40-49 years, then gradually decreased with advancing age. The proportion also increased with educational attainment. Manual labourers, unmarried individuals and Hindus were less likely to have a high WHtR.

Table 2: Wais- to-Height Ratio Categories Using a Common Cut-off (both sex) in Relation to the Socio-demographic Variables

Sequential logistic regression analyses were undertaken to see the effect of each socio-demographic variable on WHtR levels after correcting for the other socio-demographic variables. The analyses revealed significant associations with all socio-demographic variables except for religion (Table 3). The odds ratio showed that urban residents were 2.3 times and females 3.5 times more likely to have a higher WHtR than their counterparts. The likelihood of high WHtR increased with age and educational attainment. The non-paid and married individuals were more likely to have high WHtR levels than other occupations and marital groups. When all the variables were entered together into a binary logistic regression analysis the model was highly significant (c2 = 4014.8; p<0.001; Nagelkerke R2 = .225); overall 73.5% of WHtR level was correctly classified but there was imbalance in the model with 89.9% of normal WHtR correctly predicted but only 38.5% of high WHtR. A forward logistic regression analysis revealed that the most significant predictors of WHtR levels were occupation and locality.

Table 3: Socio-demographic Predictors of Waist-to-Height Ratio Categories Using a Common Cut-off: Sequential Logistic Regression Analysis Adjusted for the Other Socio-demographic Variables

WHtR was also categorised using sex-specific cut-offs (male 0.48 and female 0.45) and half of the sample were found to have high WHtR. Considerable heterogeneity was observed in the WHtR categories in relation to the socio-demographic variables (Table 4). Females, urban residents, the better educated, older individuals, widows/divorcees, the non-paid and Christians were more likely to have a high WHtR.

Table 4: Waist-to-Height Ratio Categories Using Sex-specific Cut-offs in Relation to the Socio-demographic Variables

After correcting for the other socio-demographic variables by sequential logistic regression analyses, an association of WHtR with each socio-demographic variable remained. The odds ratio presented in Table 5 revealed that females were almost 8 times more likely to have a higher WHtR than males. High WHtR was more likely to occur in urban residents, Christians, older individuals, married, the better educated and the non-paid. The model correctly predicted 64.2% of the normal and 76.6% of the high WHtR and overall 70.5% were correctly predicted by the model (c2 = 5643.0; p<0.001; Nagelkerke R2 = .292). A forward logistic regression analysis found that sex and age group were the best predictors of WHtR categories. The analyses were repeated for each sex separately and in males, age and educational status, and in females, locality and age, were the best predictors of WHtR categories.

Table 5: Socio-demographic Predictors of Waist-to-Height Ratio Categories Using Sex-specific Cut-offs: Sequential Logistic Regression Analysis Adjusted for the Other Socio-demographic Variables

Table 6: Comparison of Mean BMI, WC, WHtR and Cindex between the Current and Other Asian Studies 

Discussion

Vague was the first to observe that women with android obesity had a high prevalence of diabetes and atherosclerosis.12 Subsequent studies have shown that abdominal obesity, as measured by the waist circumference or related indexes, is associated with the subsequent development of type 2 diabetes13-16 and ischemic heart disease17-19 as well as with risk factors for CVD.20

The waist-to-height ratio was first used in the Framingham Study21 and subsequently other studies7, 22 have concluded that this ratio is more strongly associated with CVD risk factors than the body mass index (BMI; in kg/m2). In addition, waist-to-height ratio may be simpler to use and the same cutoff (e.g., 0.5) could possibly be used to identify adverse measures of waist-to-height ratio among both children and adults,23,24 which would simplify the expression of obesity-related disease risk. However, relatively few studies have examined the relation of waist-to-height ratio to CVD risk factors, and it is important to examine these associations in other data.

The advantages of WHtR were listed by Hsieh et al. (2003): “(1) closer agreement of values between men and women at all ages; (2) more accurate tracking of fat distribution and accumulation by age; (3) closer correlation with morbidity index for coronary risk factors; (4) more comprehensive identification of overweight individuals and those of normal weight facing higher risks (5) greater simplicity, in that a single rule (keep your waist circumference below half your height) may be applied both for men and women, enabling busy physicians and other professionals to screen and counsel examinees who face higher metabolic risks during physical examinations”. In this way, the index can serve as a ‘second stethoscope’.25

Hsieh & Muto (2005) explained the practicality of this ratio for screening non-obese people at a higher risk by: (i) existence of higher correlation coefficient between WHtR and the sum of coronary risk factors other than anthropometric indicators; (ii) height had a negative independent effect on the sum of coronary risk factors; (iii) WHtR of 0.5 identified more people at risk and had higher sensitivity in identification of clustering of coronary risk factors than other proposed anthropometric indices in both genders.7 Yasmin & Masci-Taylor (2000) suggested that WHtR might be an important indicator in predicting risk and could be used routinely for purposes of health education and in large scale epidemiologic studies.26 WHtR may be globally applicable as well, as the index may be effective in observation of fat distribution and related metabolic risks from childhood to old age.25 People who have a prominently large WC might have needs for reducing WC for health risks irrespective of their height. But short people with moderate WC should be more attentive than tall people with similar WC.27

Adult anthropometric data in Bangladesh cover weight and BMI and most nutrition research has focused on under-nutrition, particularly among women and children. BMI does not give any indication of the distribution of weight in the body. Anthropometric indicators of abdominal obesity estimate the amount of visceral fat tissue which, in turn, is associated with a higher risk of development of cardiovascular diseases. So, waist circumference, waist–to-height ratio as well as conicity index were also used in order to provide some notion of central obesity. However, WC, a much studied indicator and Cindex did not show high predictive accuracy. The current study focused on this important but relatively less used indicator. The overall mean WHtR in the current study was 0.48 mean which is comparable with other Asian studies as shown in Table 6 with mean values in the current study within the Asian range. The average WHtR was higher in females suggesting consistency with the findings of Sayeed et al.10 The overall WHtR seems higher than the previous Bangladeshi data because that study was done in rural samples which were lower than urban residents. The current study supports this idea showing higher WHtR in urban residents. Overall 19% of the variation in WHtR were explained by socio-demographic variables, and the main predictors were locality, age, and education.

In studies conducted in Brazil by Pitanga & Lessa28 and Almeida et al.29 suggested 0.53 and 0.55, respectively as the best cut-off point for WHtR. The cut-off point for Mexican women was similar to that which ranged from 0.53 to 0.535 for WHtR to discriminate type-2 diabetes, hypertension and dyslipidemias. Asian studies show lower cut-off values. A cut-off of 0.5 of WHtR was found in Japanese7 and 0.48 and 0.45 in Taiwanese men and women,8 the cut-off for women was proposed at 0.50 in China30 and 0.48 in Singapore.31

The current study used both Japanese7 and Taiwanese8 cut-offs to detect high WHtR. About 32% and half of the samples respectively were identified as high WHtR using cut-off of 0.5 for both sex and sex-specific cut-off. Urban residents, females, older people, better educational status, the non-paid were more likely to have a high WHtR. Occupation followed by locality (for WHtR ³0.5) and sex followed by age (for WHtR ³0.48 for men and ³0.45 for women) were the best predictors. Age and locality were identified as best predictors in males and females, respectively. No such data were available to compare with.

One third to half of the adults were with high WHtR. Specific preventive measure is required to prevent metabolic diseases. The variation in the cut-off value across different population recommends further study to identify cut-off values for the Bangladeshi population. For a national cut-off point the socio-demographic differentials need to be considered.

Acknowledgements

The authors are grateful to the Department for International Development (DFID), United Kingdom, Board of Graduate Studies, the University of Cambridge, The British Federation of Women Graduates Charitable Foundation, The Charles Wallace Bangladesh Trust, and Churchill College, the University of Cambridge for their support.

References

1.   Rahman M, Rahman M, Flora MS, Akter SFU, Hossain S & Mascie-Taylor CGN. Behavioural risk factors of non-communicable diseases in Bangladesh. NIPSOM and WHO, Dhaka 2006.

2.   Mamtani MR & Kulkarni HR. Predictive performance of anthropometric indexes of central obesity for the risk of type 2 Diabetes. Arch Med Res 2005; 36: 581-589.

3.   Park SH, Choi SJ, Lee KS & Park HY. Waist circumference and waist-to-height ratio as predictors of cardiovascular disease risk in Korean adults. Circ J 2009; 73: 1643–1650.

4.   Ford ES, Mokdad AH & Giles WH. Trends in waist circumference among US adults. Obes Res 2003; 11: 1223-1231.

5.   Ashwell M & Lejeune S. Ratio of waist circumference to height may be better indicator of need for weight management (letter). BMJ 1996; 312: 377-378.

6.   Hsieh SD & Yoshinaga H. Waist/Height ratio as a simple and useful predictor of coronary heart disease risk factors in women. Int Med 1995; 34: 1147-1152.

7.   Hsieh SD & Muto T. The superiority of waist-to-height ratio as an anthropometric index to evaluate clustering of coronary risk factors among non-obese men and women. Prev Med 2005; 40: 216-220.

8.   Lin WY, Lee LT, Chen CY, Lo H, Hsia HH, Liu IL et al. Optimal cut-off values for obesity: using simple anthropometric indices to predict cardiovascular risk in taiwan. Int J Obes 2002; 26: 1232-1238.

9.   Flora MS, Mascie-Taylor CGN & Rahman M. Conicity index of adult Bangladeshi population and their socio-demographic characteristics. Ibrahim Medical College Journal 2009; 3: 1-8.

10.Sayeed MA, Mahtab H, Latif ZA, Khanam PA, Ahsan KA, Banu A et al. Waist-to-height ratio is a better obesity index than body mass index and waist-to-hip ratio for predicting diabetes, hypertension and lipidemia. Bangladesh Med Res Counc Bull 2003; 29: 1-10.

11.Lohman TG, Roche AF & Martorell R, eds. Anthropometric standardization reference manual. champaign, illinois: human kinetic books, 1988.

12.Vague J. The degree of masculine differentiation of obesities:afactordeterminingpredispositiontodiabetes, atherosclerosis, gout, and uric calculous disease. Am J Clin Nutr 1956; 4: 20–34.

13.Stevens J, Couper D, Pankow J, Folsom AR, Duncan BB, Nieto FJ et al. Sensitivity and specificity of anthropometrics for the prediction of diabetes in a biracial cohort. Obes Res 2001; 9: 696–705.

14.Tulloch-Reid MK, Williams DE, Looker HC, Hanson RL & Knowler WC. Do measures of body fat distribution provide information on the risk of type 2 diabetes in addition to measures of general obesity? Comparison of anthropometric predictors of type 2 diabetes in Pima Indians. Diabetes Care 2003; 26: 2556–2561.

15.Wang Y, Rimm EB, Stampfer MJ, Willett WC & Hu FB. Comparison of abdominal adiposity and overall obesity in predicting risk of type 2 diabetes among men. Am J Clin Nutr 2005; 81: 555–563.

16.Schulze MB, Heidemann C, Schienkiewitz A, Bergmann MM, Hoffmann K & Boeing H. Comparison of anthropometric characteristics in predicting the incidence of type 2 diabetes in the EPIC-potsdam study. Diabetes Care 2006; 29: 1921–1923.

17.Folsom AR, Kushi LH, Anderson KE, Mink PJ, Olson JE, Hong CP et al. Associations of general and abdominal obesity with multiple health outcomes in older women: the iowa women’s health study. Arch Intern Med 2000; 160: 2117–2128.

18.Yarnell JW, Patterson CC, Thomas HF & Sweetnam PM. Central obesity: predictive value of skinfold measurements for subsequent ischaemic heart disease at 14 years follow-up in the Caerphilly Study. Int J Obes Relat Metab Disord 2001; 25: 1546–1549.

19.Rimm EB, Stampfer MJ, Giovannucci E, Ascherio A, Spiegelman D, Colditz GA et al. Body size and fat distribution as predictors of coronary heart disease among middle-aged and older US men. Am J Epidemiol 1995; 141: 1117–1127.

20.Despres JP. Is visceral obesity the cause of the metabolic syndrome? Ann Med 2006; 38: 52–63.

21.Higgins M, Kannel W, Garrison R, Pinsky J & Stokes J III. Hazards of obesity–the framingham experience. Acta Med Scand Suppl 1988; 723: 23–36.

22.Bosy-Westphal A, Geisler C, Onur S, Korth O, Selberg O, Schrezenmeir J et al. Value of body fat mass vs anthropometric obesity indices in the assessment of metabolic risk factors. Int J Obes 2006; 30: 475– 483.

23.Ashwell M & Hsieh SD. 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 2005; 56: 303–307.

24.McCarthy HD & Ashwell M. 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 (Lond) 2006; 30: 988 –992.

25.Hsieh SD, Yoshinaga H & Muto T. Waist-to-height ratio, a simple and practical index for assessing central fat distribution and metabolic risk in Japanese men and women. In J Obes 2003; 27: 610-616.

26.Yasmin & Mascie-Taylor CGN. Adiposity indices and their relationship with some risk factors of coronary heart disease in middle-aged Cambridge men and women. Ann Hum Biol 2000; 27: 239-248.

27.Hsieh SD & Yoshinaga H. Do people with similar waist circumference share similar health risks irrespective of height? Tohoku J Exp Med 1999; 188: 55-60.

28.Pitanga FJG & Lessa I. Sensitivity and specificity of the conicity index as a coronary risk predictor among adults in salvador, Brazil. Rev Bras Epidemiol 2004; 7: 259-269.

29.Almeida RT, Almeida MMG & Araújo TM. Abdominal obesity and cardiovascular risk: performance of anthropometric indexes in women. Arq Bras Cardiol 2009; 92: 345-350.

30.Ko GTC, Chan JCN, Cockram CS & Woo J. Prediction of hypertension, diabetes, dyslipidaemia or albuminuria using simple anthropometric indexes in Hong Kong Chinese. Int J Obes 1999; 23: 1136-1142.

31.Pua YH & Ong PH. Anthropometric indices as screening tools for cardiovascular risk factors in Singaporean women. Asia Pac J Clin Nutr 2005; 14: 74-79.