OBJECTIVE: To evaluate the associations of body size and fat distribution with incidence of coronary heart disease (CHD) in Chinese women.
DESIGN: Population-based, prospective cohort study.
SUBJECTS: A total of 67 334 women aged 40–70 y, who had no prior history of CHD, stroke, and cancer at study recruitment.
MEASUREMENTS: Weight, standing and sitting heights, circumferences of waist and hip, and ratios of the anthropometric measurements. Outcome: incidence of CHD (non-fatal myocardial infarction (MI) or fatal CHD).
RESULTS: After a mean follow-up of 2.5 y (168 164 person-years), there were 70 incident cases of CHD (49 non-fatal MIs and 21 CHD deaths). Body mass index (BMI), waist circumference (WC), waist-to-hip ratio (WHR), waist-to-standing height ratio (WHtR), waist-to-sitting height ratio (WsHtR), and conicity index were all positively associated with the risk of CHD. With the exception of WHR, all other anthropometric indexes only predicted the risk of CHD among women ≤55 y of age at enrollment. The relative risks (RRs) between extreme tertiles of BMI were 9.0 (95% CI, 2.0–41.5; P for trend=0.002) for younger women vs 1.3 (0.6–3.0; P for trend=0.83) for older women. Similarly, the RRs for WC, WHtR, WsHtR, and conicity index were 6.1 (1.8–21.4) vs 1.9 (0.6–5.4), 9.4 (2.6–33.8) vs 1.2 (0.5–3.1), 15.2 (3.3–69.1) vs 1.0 (0.4–2.5), and 7.8 (2.2–28.0) vs 0.9 (0.4–2.3) for the young and elderly, respectively. In contrast, the RR for WHR was 3.2 (1.1–9.1) for the young and 2.9 (1.0–8.4) for the elderly.
CONCLUSIONS: WHR was positively associated with the risk of CHD in both younger and older women, while other anthropometrics, including BMI, were related to CHD risk primarily among younger women.
Coronary heart disease (CHD) is one of the most detrimental health outcomes associated with obesity.1 Overall adiposity, usually measured by body mass index (BMI), and abdominal adiposity, usually assessed by waist-to-hip ratio (WHR), have been associated with an increased risk of CHD.2,3,4,5,6 Other indicators of abdominal obesity such as waist circumference (WC)7,8,9,10 and waist circumference-to-height ratio (WHtR)11,12 have also been shown to be associated with cardiovascular risk.
Cumulative evidence indicates that there are ethnic differences in the relationship between BMI and body composition, and between indicators of abdominal adiposity and the actual amount of visceral fat.13,14,15 Therefore, estimates of disease risk for a given level of an anthropometric indicator may differ in different study populations. The majority of epidemiologic studies examining the relationship between anthropometric measures and incident CHD are conducted in the Western society. Therefore, evaluation of these relationships in non-western populations is needed. The purpose of this study was to examine the risk of CHD in relation to anthropometric measures of body size and fat distribution among Chinese women using prospective data collected in the Shanghai Women's Health Study between 1997 and 2002.
The Shanghai Women's Health Study is a population-based prospective cohort study conducted among Chinese women who were 40–70 y of age and living in seven urban communities of Shanghai. Of 81 271 eligible women identified from the Shanghai residential registry, 75 322 (92.7%) were recruited to study and completed the baseline survey between 1997 and 2000. The reasons for nonparticipation included refusal (n=2047, 3.0%), absence during the baseline recruitment period (n=2073, 2.6%), and some health-related problems such as mental disorder, hearing, and speaking problems (n=1469, 1.8%). After exclusion of those who were later found to be outside of the study's age range, the final cohort consisted of 74 943 women. The baseline survey was conducted at participants' homes by trained interviewers using a structured questionnaire designed to collect information on demographic characteristics, diet and lifestyle habits, and medical history. Anthropometric measurements were also taken. The study participants are followed through biennial in-person interviews.
For the present study, we excluded women who were pregnant or reported a history of CHD, stroke, or cancer at the time of the baseline survey (n=7409). We also excluded those with missing data on any anthropometric measurement (n=31) and those who were lost to follow-up immediately after initial entry (n=169). After these exclusions, a total of 67 334 women remained for this analysis.
All anthropometric measurements, including weight, standing and sitting heights, and circumferences of waist and hip, were carried out according to a standard protocol by trained interviewers who were retired medical professionals. Study participants were asked to wear light indoor clothing during the measurements. Waist circumference was measured at 2.5 cm above the umbilicus and hip circumference at the level of maximum width of the buttocks with the subject in a standing position. Weight was measured to the nearest 0.1 kg. A digital weight scale was used and calibrated every 6 months. Heights and circumferences were measured to the nearest 0.1 cm. All measurements were taken twice. A tolerance limit of 1 kg was set for weight measurement and 1 cm for height and circumference measurements. A third measurement was taken if the difference of the first two measurements was greater than the tolerance limit. The average of the two closest measurements was used in the analyses. From these measurements, the following variables were created: BMI: weight in kilograms divided by the square of height in meters, WHR: waist circumference divided by hip circumference, WHtR: waist circumference divided by standing height, waist circumference-to-sitting height ratio (WsHtR): waist circumference divided by sitting height. The conicity index, an indicator of abdominal adiposity, was calculated as follows:16 waist circumference/(0.109) , where the waist and height are in meters and weight in kilograms.
The primary endpoint for this study was incident CHD, including both non-fatal myocardial infarction (MI) and CHD death, that occurred after the baseline survey. Cases of non-fatal MI were identified through in-person follow-up interviews conducted approximately 2 y after the baseline survey. Medical records were sought for all self-reported cases of MI and reviewed by physicians who were unaware of the participant's exposure status. The diagnosis of MI was confirmed using the World Health Organization criteria (ie, symptoms plus either diagnostic electrocardiographic changes or elevation of cardiac enzyme levels).17 When diagnostic information in the medical record was incomplete or unavailable, and confirmatory information was obtained solely through personal interview, the case was classified as possible MI.
Deaths were ascertained through reports by the next of kin and linkage with vital statistical registry kept at the Shanghai Center for Disease Control and Prevention. Virtually all cohort members (99.7%) were successfully followed for their vital status by means of the biennial home visits in conjunction with administrative data. The underlying causes of death were established by reviewing death certificates and medical records whenever possible, and by interviewing the next of kin. Deaths were presumed to be attributed to CHD if the underlying cause of death on the death certificate was recorded as an International Classification of Diseases, Ninth Revision (ICD-9) code of 410–414, or CHD was suggested to be the most plausible cause of death based on interviews, but no relevant medical records were available. All possible MI cases (n=11) and presumed CHD deaths (n=14) were included in this analysis after assuring that the results would not vary materially by inclusion or exclusion of these cases.
Pearson correlation coefficients were estimated for the anthropometric variables. Person-years of follow-up were calculated for each participant from the date of the baseline survey to the date of the end point, death, or the follow-up survey, whichever came first. Subjects were categorized according to the tertile distribution of the baseline anthropometric measurements among the entire cohort. The lowest tertile was used as the reference group. Incidence rates were calculated by dividing the number of events by the person-years of follow-up in each category. Cox proportional hazards model was employed to compute the relative risks (RRs) of CHD associated with various anthropometric measures and their 95% confidence intervals (CIs), and to adjust for potential confounders. Variables adjusted for included age, cigarette smoking (ever smoked at least one cigarette per day for 6 months), alcohol consumption (ever drank at least three times a week for 6 months), exercise (at least once a week for 3 months and usually sweating), attended education level, family income, menopausal status, postmenopausal hormone use, oral contraceptive use, season of recruitment, and intakes of fat, fiber, as well as soy protein, which was found to be an important determinant of CHD risk in this population.18 Considering that hypertension, diabetes, and dyslipidemia are in the causal pathway between obesity and CHD, we did not control for these factors in our primary analyses. Tests for linear trend across the tertiles of anthropometric measurements were performed by using the median value for each category and modeling them as continuous variables. We also conducted analyses stratified by age to examine the association between measures of obesity and risk of CHD among younger and older women. All P-values are two-sided. Statistical analyses were performed using SAS version 8.2 (SAS Institute, Cary, NC, USA).
Table 1 shows the baseline characteristics of study population across tertiles of WHR and BMI. Women with either higher WHR or BMI were older, more likely to have low education and low income, and be post-menopausal. They were also more likely to be a smoker, an oral contraceptive or postmenopausal hormone user. The prevalence of hypertension and diabetes increased markedly with increasing WHR and BMI. Both WHR and BMI were associated with intakes of fat, fiber, and soy protein. The correlations between various anthropometric measures of obesity are presented in Table 2. WC, WHtR, and WsHtR were all highly correlated with weight and BMI, with the Pearson correlation coefficients varying from 0.61 to 0.82; WHR and conicity index showed much weaker correlations with weight and BMI (r=0.27–0.45).
During 168 164 person-years of follow-up, 70 incident cases of CHD (49 non-fatal MI cases and 21 deaths from CHD) were documented. After adjustment for age, the measures of total and regional adiposity were all positively associated with risk of CHD. Further adjustment for other CHD risk factors and socioeconomic status resulted in little changes in the risk estimates. The multivariate-adjusted RRs comparing extreme tertiles were 2.1 (95% CI, 1.1–4.2) for weight, 2.4 (1.2–5.0) for BMI, 3.1 (1.4–7.2) for WC, 3.0 (1.4–6.3) for WHR, 3.0 (1.3–6.8) for WHtR, 3.1 (1.4–7.0) for WsHtR, and 2.4 (1.1–5.3) for conicity index (Table 3). Additional analyses were performed to control for adiposity effects for variables that assess fat distribution, and vice versa. After adjusting for WHR, the RR for the upper tertile of BMI was reduced by about 29% and was no longer significant. When BMI was adjusted for, the RRs associated with fat distribution were attenuated by roughly 16–20% and the associations with WHR and WsHtR remained significant. Additional adjustment for a history of hypertension and diabetes in the multivariate model attenuated the RR for BMI to 1.7 (95% CI, 0.8–3.7), the RR for WHR to 2.4 (1.1–5.0), and the RR for WsHtR to 2.3 (1.0–5.2).
It was observed in this cohort that weight gain increased with increasing age until age 55 y and declined thereafter, whereas WHR increased constantly with aging.19 We conducted additional analyses to investigate whether age modifies the associations between major measures of obesity and risk of CHD (Table 4). Among women aged 55 y or under, although the cardiac events were few, BMI was significantly associated with an increased risk of CHD in a dose-response manner, even after controlling for WHR. The RR derived from the multivariate model including WHR was 7.2 (95% CI, 1.5–35.7; P for trend=0.01) for the highest vs the lowest tertile of BMI. For women older than 55 y, however, the BMI was a poor predictor of CHD risk (RR=0.9; 95% CI, 0.4–2.2; P for trend=0.53). A similar association pattern was observed for all measures of abdominal adiposity with the exception of WHR. Among younger women, after controlling for BMI, the RRs for WHtR, WsHtR, and conicity index were 5.6 (95% CI, 1.0–30.8), 11.5 (1.8–74.2), and 5.2 (1.4–19.2), respectively; the corresponding RRs among the elderly were 0.9 (0.3–3.2), 0.8 (0.2–2.4), and 0.9 (0.4–2.2), respectively. The WC was not predictive for either younger or older women after controlling for BMI. Before adjusting for BMI, WHR predicted the risk of CHD equally well for both younger and older women. After adjusting for BMI, the WHR remained a strong and also the only significant predictor among older women. The RR between the extreme tertiles of WHR was 3.0 (95% CI, 1.0–8.9) for older women vs 1.7 (0.5–5.1) for younger women.
In this population-based prospective cohort study of Chinese women, we found that measures of total and central adiposity were both strongly and positively associated with risk of CHD. After adjustment for each other, the indicators of central adiposity appeared to be more strongly associated with CHD risk than did BMI. Their effect, however, was influenced by age. BMI, WHtR, WsHtR, and conicity index were all strongly associated with CHD risk among women ≤55 y of age, but not among those >55 y of age. However, among older women, WHR was the only independent anthropometric index studied that predicts the risk of CHD. These data indicate that a high level of adiposity increases the risk of CHD at all ages, but the predictive value of different obesity measurements may vary with age.
Our findings are generally in agreement with those observed in Caucasians. In the Nurses' Health Study, a positive dose–response relationship between BMI and risk of CHD was consistently reported during 8 and 14 y of follow-up among middle-aged women (30–55 y).2,3 In the Health Professionals Follow-up study,4 BMI was found to predict the risk of CHD better than WHR among young subjects, whereas for the elderly WHR was the better predictor. WHR was also noted to be the better predictor for CHD mortality among older Iowa women (aged 55–69 y).20 In contrast to these data, the Physicians' Health Study failed to observe an independent association with risk of CHD for WHR among either middle-aged or older men, but consistently demonstrated a higher predictive ability of BMI in the younger than the older subjects.6
In addition to BMI and WHR, the most commonly used surrogate measures of obesity, other anthropometric measures of obesity have also been suggested as indicators of disease risk. Their usefulness has been debated based on various criteria, including correlations with disease-risk factors, with morbidity and mortality, or with direct measures of body fat.21 WC has been proposed to replace WHR as an indicator of abdominal obesity, because it is highly correlated with visceral fat and easier to measure and interpret.21,22 Several cross-sectional studies comparing WC, BMI, and WHR have shown that WC was the best indicator of cardiovascular risk factors,9,10 but it was no longer a significant predictor after controlling for BMI.10 In our study as well as in the Nurses' Health Study,5 WHR appeared to be a better independent predictor than WC of CHD. In the Iowa Women's Health Study, WHR best predicted total mortality in comparisons with WC and BMI.20 These data suggest that the WHR may provide additional information beyond BMI and WC. Conicity index, another measure of abdominal adiposity, has been claimed to have several advantages, including having a theoretical range, a built-in adjustment of waist circumference for height and weight, and not requiring the hip circumference to assess fat distribution.16 Our study found that among younger women conicity index was a powerful predictor of CHD, but it was not predictive for older women. The Framingham Heart Study23 showed no association between conicity index and CHD among men and women aged 30–62 y. No previous study has investigated WHtR or WsHtR in relation to risk of CHD. But WHtR has been shown to be a useful predictor of multiple CHD risk factors.11,12 In a study of correlations of selected anthropometric variables, including WHtR, WC, BMI, and WHR, with intra-abdominal fat measured by computed tomography, WHtR showed the highest correlation.24 These findings seem to provide some biological explanations for our observations that WHtR and WsHtR were strongly predictive of CHD. More body composition studies are needed to clarify the biologic relevance of these anthropometric indexes, particularly among different age groups. Such information is critical for interpreting the statistical associations between anthropometric measurements and disease risk.
This large population-based prospective study had a very high participation rate (92.7%) and a virtually complete cohort follow-up. It is noteworthy that very few women ever smoked cigarettes and drank alcoholic beverages because of the Chinese culture and thus confounding from these factors, especially smoking, was minimized. Moreover, unlike most previous cohort studies,2,3,4,5,6,20 our study used anthropometric variables that were directly measured by trained medical professional rather than self-reported or self-measured, which eliminated bias-related differential reporting and minimized the measurement error. The short interval between anthropometric measurement and CHD event reduced the effect of fluctuations in body weight and fat distribution over time on the disease association. However, this raised concern about the effect of subclinical illness on the results.4,25 We have excluded subjects who had a history of CHD and other major chronic diseases at baseline from the analysis to minimize the potential bias. If the observed associations were a result of subclinical disease(s), we would expect to see similar associations for both older and younger women. The short length of follow-up also led to a small number of outcomes, which resulted in less precise estimates of RRs. Further maturation of this cohort will expand the number of CHD outcomes and will enable us to exclude women who developed CHD in the period immediately after their baseline measures to minimize possible effects of subclinical disease on these anthropometric outcomes.
In conclusion, our study suggests that most proposed anthropometric measures of obesity in the literature are strongly predictive of risk of CHD among Chinese women; their effect, however, may vary with age.
National Task Force on the Prevention and Treatment of Obesity. Overweight, obesity, and health risk. Arch Intern Med 2000; 160: 898–904.
Manson JE, Colditz GA, Stampfer MJ, Willett WC, Rosner B, Monson RR, Speizer FE, Hennekens CH . A prospective study of obesity and risk of coronary heart disease in women. N Engl J Med 1990; 322: 882–889.
Willett WC, Manson JE, Stampfer MJ, Colditz GA, Rosner B, Speizer FE, Hennekens CH . Weight, weight change, and coronary heart disease in women. JAMA 1995; 273: 461–465.
Rimm EB, Stampfer MJ, Giovannucci E, Ascherio A, Spiegelman D, Colditz GA, Willett WC . 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.
Rexrode KM, Carey VJ, Hennekens CH, Walters EE, Colditz GA, Stampfer MJ, Willett WC, Manson JE . Abdominal adiposity and coronary heart disease in women. JAMA 1998; 280: 1843–1848.
Rexrode KM, Buring JE, Manson JE . Abdominal and total adiposity and risk of coronary heart disease in men. Int J Obes Relat Metab Disord 2001; 25: 1047–1056.
Pouliot MC, Despres JP, Lemieux S, Moorjani S, Bouchard C, Tremblay A, Nadeau A, Lupien PJ . Waist circumference and abdominal sagittal 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.
Han TS, van Leer EM, Seidell JC, Lean ME . Waist circumference action levels in the identification of cardiovascular risk factors: prevalence study in a random sample. BMJ 1995; 311: 1401–1405.
Dobbelsteyn CJ, Joffres MR, MacLean DR, Flowerdew G . A comparative evaluation of waist circumference, waist-to-hip ratio and body mass index as indicators of cardiovascular risk factors. The Canadian Heart Health Surveys. Int J Obes Relat Metab Disord 2001; 25: 652–661.
Ho SC, Chen YM, Woo JL, Leung SS, Lam TH, Janus ED . Association between simple anthropometric indices and cardiovascular risk factors. Int J Obes Relat Metab Disord 2001; 25: 1689–1697.
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.
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.
Deurenberg P, Deurenberg-Yap M . Differences in body-composition assumptions across ethnic groups: practical consequences. Curr Opin Clin Nutr Metab Care 2001; 4: 377–383.
Park YW, Allison DB, Heymsfield SB, Gallagher D . Larger amounts of visceral adipose tissue in Asian Americans. Obes Res 2001; 9: 381–387.
Deurenberg P, Deurenberg-Yap M, Guricci S . Asians are different from Caucasians and from each other in their body mass index/body fat per cent relationship. Obes Rev 2002; 3: 141–146.
Valdez R, Seidell JC, Ahn YI, Weiss KM . A new index of abdominal adiposity as an indicator of risk for cardiovascular disease. A cross-population study. Int J Obes Relat Metab Disord 1993; 17: 77–82.
Rose GA, Blackburn H . Cardiovascular survey methods. WHO monograph series, No. 58 World Health Organization: Geneva, Switzerland; 1982.
Zhang X, Shu XO, Gao YT, Yang G, Li Q, Li H, Jin F, Zheng W . Soy food consumption is associated with lower risk of coronary heart disease in Chinese women. J Nutr 2003; 133: 2874–2878.
Wen W, Gao YT, Shu XO, Yang G, Li HL, Jin F, Zheng W . Sociodemographic, behavioral, and reproductive factors associated with weight gain in Chinese women. Int J Obes Relat Metab Disord 2003; 27: 933–940.
Folsom AR, Kushi LH, Anderson KE, Mink PJ, Olson JE, Hong CP, Sellers TA, Lazovich D, Prineas RJ . 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.
Molarius A, Seidell JC . Selection of anthropometric indicators for classification of abdominal fatness—a critical review. Int J Obes Relat Metab Disord 1998; 22: 719–727.
Lean ME, Han TS, Morrison CE . Waist circumference as a measure for indicating need for weight management. BMJ 1995; 311: 158–161.
Kim KS, Owen WL, Williams D, Adams-Campbell LL . A comparison between BMI and Conicity index on predicting coronary heart disease: the Framingham Heart Study. Ann Epidemiol 2000; 10: 424–431.
Ashwell M, Cole TJ, Dixon AK . Ratio of waist circumference to height is strong predictor of intra-abdominal fat. BMJ 1996; 313: 559–560.
Willett WC, Dietz WH, Colditz GA . Guidelines for healthy weight. N Engl J Med 1999; 341: 427–434.
This work was supported by NIH grant R01CA70867.
About this article
Cite this article
Zhang, X., Shu, X., Gao, Y. et al. Anthropometric predictors of coronary heart disease in Chinese women. Int J Obes 28, 734–740 (2004) doi:10.1038/sj.ijo.0802634
- coronary heart disease
Current Opinion in Genetics & Development (2018)
Nippon Ronen Igakkai Zasshi. Japanese Journal of Geriatrics (2018)
Nippon Ronen Igakkai Zasshi. Japanese Journal of Geriatrics (2018)
A STUDY ON RELATIONSHIP OF BODY MASS INDEX (BMI) AND WAIST-TO-HIP RATIO (WHR) TO CORONARY ARTERY DISEASE RISK FACTORS IN CASES AND CONTROLS
Journal of Evolution of Medical and Dental Sciences (2017)
Performance of the waist-to-height ratio in identifying obesity and predicting non-communicable diseases in the elderly population: A systematic literature review
Archives of Gerontology and Geriatrics (2016)