Abstract
Obesity, particularly central adiposity, is a well-established risk factor for cardiovascular disease (CVD). Waist circumference (WC) is measured in numerous epidemiologic studies as a relatively simple indicator of central adiposity. However, recently, investigators have considered a measure that takes height into consideration Waist-to-Height Ratio (WHtR) as a more sensitive predictor of CVD. A limited number of studies have examined the association between various measures of central adiposity and obesity with CVD, but there is a dearth of information on this topic focused specifically on African American adults. Given the high rates of cardiovascular disease and metabolic risk factors in this population, it is important to develop validated, easy-to-measure indicators of CVD risk for clinical use. Data from 4,758 African American adults participating in the baseline visit of the Jackson Heart Study with available risk factor data were examined, with three measures of body habitus [body mass index (BMI), WC and WHtR] and five CVD risk factors (HDL- and LDL-cholesterol, triglycerides, diabetes and hypertension), the latter also categorized into multiple (2+) risk factors present. C-statistics for WC, BMI, and WHtR were computed and compared for each model to assess their discriminant abilities. WHtR was a stronger correlate of HDL cholesterol, triglycerides, diabetes, hypertension and multiple risk factors compared to BMI, and was a stronger corrleate of HDL-cholesterol when compared to WC. These data indicate that, for African American adults, WHtR may be more appropriate measure to identify those at elevated risk for CVD.
Keywords: Obesity, Cardiovascular Disease, Waist Circumference, African Americans
Introduction
Obesity is a well-established risk factor for cardiovascular diseases (CVD) [1]. Evidence suggests that centrally located body fat may be a more important contributor to CVD risk than measures that do not take central adiposity, such as BMI, into consideration [2–3]. Waist circumference (WC) and its corollary, waist-to-hip ratio (WHR) are used in epidemiologic studies as relatively simple to measure indicators of central adiposity and been shown to be more reflective of overall adiposity relative to other measures [4]. However, recently, investigators have considered a measure that takes height into consideration (Waist-to-Height Ratio, WHtR) as a more sensitive predictor of CVD [5 – 12]. A 2008 report from the World Health Organization (WHO) provided a summary of research to date on WC and WHtR as indicators of chronic disease risk. This report indicated the need to consider age, sex and race/ethnic differences in measures of central adiposity [13].
Among those studies that have examined the predictability of measures of central adiposity on CVD risk, specifically WHtR, few have focused specifically on African American adults. The WHO report [12] did not have reference to this population. Given the high rates of CVD and metabolic risk factors for CVD in this population [1], it is important to develop validated, easy-to-measure indicators of CVD risk for clinical use. This analysis used data from a large epidemiologic study of CVD in African Americans to compare the predictive ability of three measures of adiposity, BMI, WC and WHtR for assessing CVD risk. We used baseline data and compared the associations of these composition measures to traditional CVD risk factors and composites of these indicators by each of these body habitus measures using logistic regression and c-statistics. We hypothesized that WHtR will be more strongly related to CVD risk factors among African Americans relative to the other body habitus measures.
Methods
Overview
The Jackson Heart Study (JHS) has been described in detail elsewhere [14 – 15]. Briefly, the JHS is the largest longitudinal epidemiologic study of CVD in African Americans. The JHS enrolled 5,306 participants aged 21–84 years at the time of the baseline assessment (2000–2004) from the Jackson, Mississippi metropolitan area (Hinds, Madison, and Rankin counties). Participants were recruited from four different sources: 1) previous participants in the Atherosclerosis Risk in Communities (ARIC) study (30% of JHS cohort); 2) family members of JHS participants (28%); 3) random selection from the three counties (17%); and 4) community volunteers (25%). The goal of the JHS was to examine factors that influence the development of CVD in African American men and women to learn how to prevent this group of diseases in this population. JHS participants provided information on demographic, socioeconomic, and lifestyle variables, as well as medication use. Certified technicians and nurses conducted clinic interviews. Two subsequent in-person follow up visits have been completed since baseline (2005–2008 and 2009–2013). Data for this analysis come from the baseline visit, with the analysis limited to the 4,758 participants with data for all the variables considered.
Dependent Variables
Blood samples were collected using standard procedures and analyzed at a central laboratory (University of Minnesota) [16]. Fasting glucose and lipids were measured on a Vitros 950 or 250, Ortho-Clinical Diagnostics analyzer (Raritan, NJ) in accordance with the College of American Pathologists Proficiency Testing Program (Carpenter). Type 2 diabetes status was defined according to the 2010 American Diabetes Association guidelines: 1) physician diagnosis, 2) use of diabetes medication, 3) fasting glucose ≥ 126 mg/dl, or 4) A1C ≥ 6.5% [17]. Blood pressure was measured using the Omron HEM907XL automated BP monitor. Two resting systolic and diastolic blood pressures were measured at 5-minute intervals and the average was used for the current analysis.
CVD risk factors were considered as continuous variables and were categorized along clinically relevant thresholds. Triglycerides were dichotomized into a low group consisting of < 150 mg/dL and a high group. LDL-cholesterol was dichotomized into a low group consisting of < 160 mg/dL and a high group. HDL-cholesterol was dichotomized using sex specific values with a low group consisting of males with HDL-cholesterol ≤ 35 mg/dL or females with an HDL ≤ 45 mg/dL and a high group. Diabetes was defined as current use of insulin or oral antidiabetic agent, fasting glucose ≥ 126 mg/dL, or HbA1C ≥ 6.5%. Hypertension was defined as having a systolic blood pressure of ≥ 140 mmHg or a diastolic blood pressure ≥ 90 or a self-reported prescription of blood pressure medications. CVD risk factors were defined as having low HDL, high LDL, high triglycerides, having diabetes, and having hypertension. These risk factors were summed into a risk count score. Risk count score was dichotomized into < 2 risk factors and greater.
Independent Variables
Weight was measured to the nearest 0.1 kilogram and height to the nearest centimeter. Body mass index (BMI) was calculated as weight divided by the square of the height. WC was measured to the nearest centimeter as the average of two readings at the umbilicus with the patient upright. WC was dichotomized based on the National Cholesterol Education Program Adult Treatment Panel III cutoff points as ≤102 cm or >102 cm for men, and ≤88 cm or >88 cm for women. BMI was categorized into Underweight, Normal, Overweight, and Obese categories corresponding to BMI levels of < 18.5, 18.5 – 25, 25 – <30, ≥30 kg/m2. Underweight and normal weight were combined in the analyses since the number of underweight participants was relatively small. As has been used in other studies [6], waist-to-height ratio (WHtR) was calculated as waist / height and was dichotomized into ≤ 0.5 and greater.
Statistical Analysis
General linear models were used to examine the association between demographic and health characteristics and each of the three body habitus measures (Table 2). Logistic regression analyses were performed to examine the correlation of WC, BMI, and WHtR with known CVD risk factors, including low HDL, high LDL, high triglycerides, diabetes status, and hypertension status. For each CVD risk factor, three separate logistic regression models were fit using WC, BMI, or WHtR as a predictor while adjusting for age and gender. C-statistics for WC, BMI, and WHtR were computed and compared based on U-statistics theory for each risk factor to assess their discriminant abilities [18]. Similar analyses were conducted for the dichotomized risk score. All analyses were performed on SAS v9.4 (SAS Inc., Cary, NC).
Table 2.
Comparison of Demographic and Health Characteristics by Body Habitus Characteristics
| Characteristic | Under-weight/Normal | Over-weight | Obese | WC Low | WC High | WHtR Low | WHtR High |
|---|---|---|---|---|---|---|---|
| Sex (%) | |||||||
| Female | 378 | 843 | 1821 | 758 | 2284 | 402 | 2640 |
| (54.6%)* | (54.4%)* | (72.3%)* | (42.6%)* | (76.7%)* | (55.8%)* | (65.4%)* | |
| Male | 314 | 706 | 696 | 1023 | 693 | 319 | 1397 |
| (45.4%)* | (45.6%)* | (27.7%)* | (57.4%)* | (23.3%)* | (44.2%)* | (34.6%)* | |
| Age (Mean, SD) | 55.1 | 56.3 | 54.3 | 53.6 | 55.9 | 50.9 | 55.8 |
| (14.7)* | (12.8)* | (12.3)* | (13.4)* | (12.5)* | (13.7)* | (12.6)* | |
| Age Group (Years, %) | |||||||
| <45 | 192 | 341 | 629 | 517 | 645 | 261 | 901 |
| (27.7%)* | (22.0%)* | (25.0%)* | (29.0%)* | (21.7%)* | (36.2%)* | (22.3%)* | |
| 45 – < 65 | 304 | 796 | 1361 | 885 | 1576 | 343 | 2118 |
| (43.9%)* | (51.4%)* | (54.1%)* | (49.7%)* | (52.9%)* | (47.6%)* | (52.5%)* | |
| ≥65 | 196 | 412 | 527 | 379 | 756 | 117 | 1018 |
| (28.3%)* | (26.6%)* | (20.9%)* | (21.3%)* | (25.4%)* | (16.2%)* | (25.2%)* | |
| Triglycerides, mg/dL (Mean, SD) | 83.9 | 99.8 | 108.2 | 90.2 | 109.0 | 76.4 | 106.5 |
| (47.9)* | (54.1)* | (55.5)* | (50.9)* | (55.6)* | (41.0)* | (55.5)* | |
| Triglycerides, mg/dL (%) | |||||||
| Low | 633 | 1319 | 2067 | 1599 | 2420 | 683 | 3336 |
| (91.5%)* | (85.2%)* | (82.1%)* | (89.8%)* | (81.3%)* | (94.7%)* | (82.6%)* | |
| High | 59 | 230 | 450 | 182 | 557 | 38 | 701 |
| (8.5%)* | (14.9%)* | (17.9%)* | (10.2%)* | (18.7%)* | (5.3%)* | (17.4%)* | |
| Total Cholesterol, mg/dL (Mean, SD) | 196.4 | 200.3 | 198.8 | 197.0 | 200.1 | 194.7 | 199.7 |
| (41.9) | (39.9) | (38.8) | (40.4)* | (39.1)* | (39.7)* | (39.6)* | |
| LDL-Cholesterol, mg/dL (Mean, SD) | 121.4 | 128.3 | 127.0 | 125.5 | 127.3 | 121.3 | 127.6 |
| (37.7)* | (36.9)* | (36.0)* | (37.2) | (36.3) | (36.0)* | (36.6)* | |
| LDL-Cholesterol, mg/dL (%) | |||||||
| Low | 587 | 1259 | 2088 | 1487 | 2447 | 618 | 3316 |
| (84.8%) | (81.3%) | (83.0%) | (83.5%) | (82.2%) | (85.7%)* | (82.1%)* | |
| High | 105 | 290 | 429 | 294 | 530 | 103 | 721 |
| (15.2%) | (18.7%) | (17.0%) | (16.5%) | (17.8%) | (14.3%)* | (17.9%)* | |
| HDL-Cholesterol, mg/dL (Mean, SD) | 58.2 | 52.0 | 50.2 | 53.4 | 51.0 | 58.1 | 50.8 |
| (17.4)* | (14.2)* | (13.5)* | (15.5)* | (14.0)* | (16.6)* | (14.0)* | |
| HDL-Cholesterol, mg/dL (%) | |||||||
| High | 602 | 1254 | 1781 | 1528 | 2109 | 648 | 2989 |
| (87.0%)* | (81.0%)* | (70.8%)* | (85.8%)* | (70.8%)* | (89.9%)* | (74.0%)* | |
| Low | 90 | 295 | 736 | 253 | 868 | 73 | 1048 |
| (13.0%)* | (19.0%)* | (29.2%)* | (14.2%)* | (29.2%)* | (10.1%)* | (26.0%)* | |
| Diabetes (%) | |||||||
| No | 655 | 1319 | 1903 | 1630 | 2247 | 697 | 3180 |
| (94.7%)* | (85.2%)* | (75.6%)* | (91.5%)* | (75.5%)* | (96.7%)* | (78.8%)* | |
| Yes | 37 | 230 | 614 | 151 | 730 | 24 | 857 |
| (5.3%)* | (14.8%)* | (24.4%)* | (8.5%)* | (24.5%)* | (3.3%)* | (21.2%)* | |
| Hypertension, mmHg (%) | |||||||
| No | 371 | 668 | 862 | 916 | 985 | 431 | 1470 |
| (53.5%)* | (43.2%)* | (34.2%)* | (51.4%)* | (33.1%)* | (59.8%)* | (36.4%)* | |
| Yes | 321 | 881 | 1655 | 865 | 1992 | 290 | 2567 |
| (46.4%)* | (56.9%)* | (65.8%)* | (48.6%)* | (66.9%)* | (40.2%)* | (63.6%)* | |
| Total Risk Factors (Mean, SD) | 0.9 | 1.2 | 1.5 | 1.0 | 1.6 | 0.7 | 1.5 |
| (0.9)* | (1.0)* | (1.1)* | (0.9)* | (1.1)* | (0.8)* | (1.1)* | |
| Risk Factor Categories (%) | |||||||
| 0–1 | 544 | 1006 | 1347 | 1346 | 1551 | 602 | 2295 |
| (78.6%)* | (64.9%)* | (53.5%)* | (75.6%)* | (52.1%)* | (83.5%)* | (56.8%)* | |
| 2+ | 148 | 543 | 1170 | 435 | 1426 | 119 | 1742 |
| (21.4%)* | (35.1%)* | (46.5%)* | (24.4%)* | (47.9%)* | (16.5%)* | (43.2%)* | |
Comparisons are within each category of the body composition measure; P < 0.05
Results
Table 1 displays demographic and health characteristics of the study population at the baseline visit stratified by gender. The mean age of the sample was approximately 55 years, with the majority being in the 45 – 64 year age range. Approximately 15 – 20% had high triglycerides and high LDL-cholesterol. More women had low HDL-cholesterol compared to men (27.0% vs. 17.5%). Diabetes was slightly more prevalent among women compared to men (19.4% vs. 17.0%), as well as hypertension (61.6% vs. 57.3%). Men and women had on average one risk factor, and about 36% of men and 41% of women had at least two risk factors present.
Table 1.
Selected Demographic and Health Characteristics of Study Participants By Sex
| Characteristic | Male (N = 1716) | Female (N = 3042) | ||
|---|---|---|---|---|
| Mean (SD) | N (%) | Mean (SD) | N (%) | |
| Age (Mean, SD) | 54.3 (13.0) | 55.5 (12.8) | ||
| Age Group (Years, %) | ||||
| <45 | 449 (26.2) | 713 (23.4) | ||
| 45 –< 65 | 884 (51.5) | 1577 (51.8) | ||
| ≥65 | 383 (22.3) | 752 (24.7) | ||
| Triglycerides, mg/dL (Mean, SD) | 108.2 (59.5) | 98.4 (51.3) | ||
| Triglycerides, mg/dL (%) | ||||
| Low | 1396 (81.4) | 2623 (86.2) | ||
| High | 320 (18.6) | 419 (13.8) | ||
| Total Cholesterol, mg/dL (Mean, SD) | 196.1 (39.3) | 200.5 (39.7) | ||
| LDL-Cholesterol, mg/dL (Mean, SD) | 128.4 (36.9) | 125.6 (36.4) | ||
| LDL-Cholesterol, mg/dL (%) | ||||
| Low | 1407 (82.0) | 2527 (83.1) | ||
| High | 309 (18.0) | 515 (16.9) | ||
| HDL-Cholesterol, mg/dL (Mean, SD) | 46.1 (12.7) | 55.2 (14.6) | ||
| HDL-Cholesterol, mg/dL (%) | ||||
| High | 1415 (82.5) | 2222 (73.0) | ||
| Low | 301 (17.5) | 820 (27.0) | ||
| Diabetes (%) | ||||
| No | 1424 (83.0) | 2453 (80.6) | ||
| Yes | 292 (17.0) | 589 (19.4) | ||
| Hypertension, mmHg (%) | ||||
| No | 733 (42.7) | 1168 (38.4) | ||
| Yes | 983 (57.3) | 1874 (61.6) | ||
| Total Risk Factors (Mean, SD) | 1.3 (1.0) | 1.4 (1.1) | ||
| Risk Factor Categories (%) | ||||
| 0–1 | 1093 (63.7) | 1804 (59.3) | ||
| ≥2 | 623 (36.3) | 1238 (40.7) | ||
| Waist Circumference, cm (Mean, SD) | 101.1 (15.0) | 100.2 (16.7) | ||
| High Waist Circumference (%) | ||||
| Low | 1023 (59.6) | 758 (24.9) | ||
| High | 693 (40.4) | 2284 (75.1) | ||
| BMI, kg/m2 (Mean, SD) | 29.8 (6.1) | 32.8 (7.6) | ||
| BMI Categories | ||||
| Underweight | 13 (0.8) | 11 (0.4) | ||
| Normal | 301 (17.5) | 367 (12.1) | ||
| Overweight | 706 (41.1) | 843 (27.7) | ||
| Obese | 696 (40.6) | 1821 (59.9) | ||
| Waist-Height Ratio (Mean, SD) | 0.6 (0.1) | 0.6 (0.1) | ||
| Waist-Height Ratio (%) | ||||
| ≤0.5 | 319 (18.6) | 402 (13.2) | ||
| >0.5 | 1397 (81.4) | 2640 (86.8) | ||
With regard to body habitus measures, the mean WC and mean BMI for men and women was 101.3 and 100.4 cm, and 29.9 and 32.8 kg/m2, respectively. Women were far more likely than men to have a higher WC (75.1% vs. 40.4%). Similarly, far more women than men were classified as obese (59.9% vs. 40.5%). The mean WHtR was similar for men and women (0.6), but slightly more women than men had a WHtR above the recommended cutoff point level (86.8% vs. 81.4%).
Table 2 shows the distribution of the selected demographic characteristics according to the categories of BMI, WC and WHtR. The proportion of female participants increased from low- to high-risk categories. For BMI, the overweight group has the highest mean age, and for the WC and WHtR, participants in the highest-risk categories were older. In general, across the health characteristics, the high-risk body habitus categories corresponded with higher CVD risk, and the proportion of participants with 2+ CVD risk factors was highest in the highest-risk body habitus categories for all measures.
Table 3 shows age- and sex-adjusted c-statistic differences between WHtR and BMI values, and between WHtR and WC values, for each of the five CVD risk factors individually, and for the dichotomous compilation of risk factors (<2, ≥2). For the comparison between WHtR and BMI, the c-statistic value was significantly higher (p < 0.0001) between triglycerides, diabetes, hypertension, and having ≥2 risk factors. For the comparison between WHtR and WC, the WHtR c-statistic was similar between the two measures for most risk factors, and was significantly greater only for HDL-cholesterol.
Table 3.
Comparison of Demographic and Health Characteristics by Continuous Body Habitus Measures (Mean, Standard Deviation)
| Characteristics | BMI | WC | WHtR |
|---|---|---|---|
| Sex | |||
| Female | 32.8 (7.6)* | 100.2 9 (16.7) | 0.61 (0.10)* |
| Male | 29.8 (6.1) | 101.1 (15.0) | 0.57 (0.08) |
| Age Group (Years) | |||
| <45 | 32.4 (8.4)* | 98.7 (18.1)* | 0.58 (0.11)* |
| 45 – 64 | 31.9 (7.1)* | 101.0 (15.7) | 0.60 (0.10)* |
| ≥65 | 30.6 (6.3) | 101.3 (14.8) | 0.61 (0.09) |
| Triglycerides | |||
| Low | 31.67 (7.4) | 99.7 (16.4) | 0.59 (0.10) |
| High | 32.5 (6.1) | 104.7 (13.7) | 0.62 (0.08) |
| LDL-Cholesterol | |||
| Low | 31.8 (7.4) | 100.4 (16.5) | 0.60 (0.10) |
| High | 31.5 (6.3) | 100.9 (14.4) | 0.60 (0.09) |
| HDL-Cholesterol | |||
| High | 31.1 (7.0)* | 99.1 (15.9)* | 0.59 (0.10)* |
| Low | 33.8 (7.6) | 105.0 (15.9) | 0.63 (0.10) |
| Diabetes | |||
| No | 31.1 (7.1)* | 98.5 (15.6)* | 0.58 (0.09)* |
| Yes | 34.6 (7.3) | 109.2 (15.4) | 0.65 (0.10) |
| Hypertension | |||
| No | 30.5 (7.0)* | 96.3 (15.6)* | 0.57 (0.09)* |
| Yes | 32.6 (7.3) | 103.3 (15.9) | 0.61 (0.10) |
| Risk Factor Categories | |||
| 0 – 1 | 30.7 (7.1)* | 97.2 (15.7)* | 0.58 (0.09)* |
| 2+ | 33.3 (7.2) | 105.6 (15.5) | 0.63 (0.10) |
Bottom row for each characteristic was used as the reference group
P<0.05
Discussion
CVD is a major public health issue, and African Americans are disproportionately impacted by this condition [1]. Obesity is a well-recognized risk factor for CVD [1]. Thus, tools are needed to adequately identify African Americans at increased risk for CVD in the clinical setting. Traditionally, BMI has been used as a surrogate for more invasive measures of body composition and obesity status, but concerns have been raised that this measure does not accurately take into consideration central adiposity, which in and of itself is a risk factor for CVD, leading for a call to validate measures such as WC and WHtR to be used to assess CVD risk [13].
In this analysis, we examined the associations of three measures of body habitus, BMI, WC and WHtR with traditional CVD risk factors in a large, well-characterized sample of African American adults. We found that WHtR was more strongly related to CVD risk than BMI, was similar to WC in terms of most CVD risk factors, and was more strongly related to HDL-cholesterol compared to both BMI and WC. Our findings are similar to those of other studies on this topic [5 – 12]. In their systematic review, Browning and colleagues found that the majority of studies in their review favored WHtR and WC as stronger predictors of cardiometabolic risk compared to BMI, with c-statistic values of 0.704, 0.693 and 0.671, respectively. The studies in their analysis included populations of Caucasian, Asian and Central American descent, and affirmed the 0.5 cutoff for WHtR as the optimal level of assessing CVD and diabetes risk [19]. It is also worth noting that in their cross-sectional analyses examining associations between CVD risk factors and obesity indices, similar to our design, most of the studies in this systematic review adjusted for BMI in the WC and WHtR logistic models.
There is some evidence that the relationship between measures of adiposity and CVD risk differs between whites and African Americans. Katzmarzyk and colleagues [20] found that BMI, body adiposity index, WC, WHtR and waist-to-hip ratio were also significantly and positively associated with all-cause mortality in whites, but WC was the only indicator that significantly predicted mortality in African Americans. However, a recent study by Kazlauskaite and colleagues showed that WHtR is similarly effective as a CVD risk screening tool for midlife Hispanic, Chinese, Japanese, White and African American women [21].
This study has a number of strengths that should be acknowledged. First, we analyzed data from a large, population-based study of African American adults, a population at significant risk for CVD. Second, this is, to our knowledge, the largest study of the relationship of measures of body habitus and CVD risk among African American adults. Second, this study included well- validated measures of traditional CVD risk factors. Finally, our sample size allowed us to assess these relationships across various demographic characteristics, as well as aggregate CVD risk factors to determine the association of body habitus with “clusters” of risk factors.
Despite these strengths, there are several limitations to this analysis. First, we assessed the body habitus/CVD risk factor relationship using cross-sectional data, so we are unable to assign causality to these observed relationships. We also did not examine the relationship between these measures of body habitus and CVD events. This study only included African Americans, so we are not able to make comparisons within this study to other racial/ethnic groups. Finally, the JHS only included an assessment of central adiposity with a measure of central adiposity by measuring the circumference at the point of the umbilicus.
In summary, this study provides evidence that measures of central adiposity, particularly WHtR, may be more appropriate for assessing CVD risk in African American adults. Given the simple public health message that your waist should be no more than half your height [19, 22], WHtR potentially provides an easy-to-use screening tool for the assessment of CVD risk in a clinical setting. Further research is needed to assess the longitudinal risk associated with measures of central adiposity in this population.
Table 4.
Age & Gender -Adjusted C-Statistic Comparisons Between Body Habitus Measures and CVD Risk Factors
| c-statistic Difference between WHtR and BMI (P-Value) |
c-statistic Difference Between WHtR and WC (P-Value) |
BMI c-statistic | WC c-statistic | WHTR c-statistic | |
|---|---|---|---|---|---|
| HDL-Cholesterol | −0.0129 | 0.0050 | 0.6224 | 0.6403 | 0.6354 |
| (p=0.0002) | (p=0.0095) | ||||
| LDL- Cholesterol | 0.0007 | −0.0005 | 0.5411 | 0.5399 | 0.5403 |
| (p=0.7952) | (p=0.5423) | ||||
| Triglycerides | −0.0199 | −0.0005 | 0.5969 | 0.6163 | 0.6168 |
| (p<.0001) | (p=0.8161) | ||||
| Diabetes | −0.0204 | 0.0012 | 0.7216 | 0.7432 | 0.7420 |
| (p<.0001) | (p=0.5442) | ||||
| Hypertension | −0.0047 | −0.0005 | 0.7555 | 0.7597 | 0.7602 |
| (p=0.0120) | (p=0.5890) | ||||
| Multiple Risk Factors | −0.0181 | 0.0007 | 0.6778 | 0.6966 | 0.6959 |
| (p<.0001) | (p=0.6456) |
Acknowledgments
Funding: This manuscript was approved by the Jackson Heart Study Publications and Presentations Committee. The Jackson Heart Study is supported by contracts HHSN268201300046C, HHSN268201300047C, HHSN268201300048C, HHSN268201300049C, HHSN268201300050C from the NHBLI and the NIMHD. This secondary analysis was funded by NHLBI Grant (R01 HL117285-01, Bertoni, PI).
The authors thank the participants and data collection staff of the Jackson Heart Study. The views expressed in this manuscript are those of the authors and do not necessarily represent the views of the National Heart, Lung, and Blood Institute; the National Institutes of Health; or the U.S. Department of Health and Human Services.
Data from this analysis were presented in poster format at the 2016 Obesity Society’s Annual Scientific Conference, October 31st – November 4th, 2016.
Footnotes
Compliance with Ethical Standards
Conflicts of Interest: The authors acknowledge that they have no relevant conflicts of interest pertinent to the content of this manuscript.
Ethical Approval: Informed consent was obtained from all individual participants included in this study in accordance with the protocol of the parent study.
Contributor Information
Ronny A. Bell, Department of Public Health, East Carolina University, 600 Moye Boulevard, Mailstop 660, Greenville, NC 27834, 252-744-4065.
Haiying Chen, Department of Biostatistical Sciences, Division of Public Health Sciences, Wake Forest School of Medicine, Medical Center Boulevard, Winston-Salem, NC 27157-1063, 336-716-4431.
Santiago Saldana, Department of Biostatistical Sciences, Division of Public Health Sciences, Wake Forest School of Medicine, Medical Center Boulevard, Winston-Salem, NC 27157-1063, 336-713-1399.
Alain G. Bertoni, Department of Epidemiology and Prevention, Division of Public Health Sciences, Maya Angelou Center for Health Equity, Wake Forest School of Medicine, Medical Center Boulevard, Winston-Salem, NC 27157-1063, 336-716-2824.
Valery S. Effoe, Department of Medicine, Division of General Internal Medicine, Morehouse School of Medicine, 720 Westview Drive S.W., Atlanta, GA 30310.
Kristen G Hairston, Section on Endocrinology and Metabolism, Maya Angelou Center for Health Equity, Wake Forest School of Medicine, Medical Center Boulevard, Winston-Salem, NC 27157-1063, 336-713-7251.
Rita R. Kalyani, Johns Hopkins Center on Aging and Health, Welch Center for Prevention, Epidemiology and Clinical Research, Johns Hopkins School of Medicine, 1830 East Monument Street, Suite 333, Baltimore, MD, 21287.
Arnita F. Norwood, Department of Medicine, University of Mississippi Medical Center, 2500 North State Street, Jackson, MS 39216.
References
- 1.Benjamin EJ, Blaha MJ, Chiuve SE, Cushman M, Das SR, Deo R, et al. Heart disease and stroke statistics—2017 update: a report from the American Heart Association. Circulation. 2017;135:e146–e603. doi: 10.1161/CIR.0000000000000485. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Smith SC, Jr, Haslam D. Abdominal obesity, waist circumference and cardio-metabolic risk: awareness among primary care physicians, the general population and patients at risk—the Shape of the Nations survey. Curr Med Res Opin. 2007;23:29– 47. doi: 10.1185/030079906X159489. [DOI] [PubMed] [Google Scholar]
- 3.Neeland IJ, Turer AT, Ayers CR, Powell-Wiley TM, Vega GL, Garzaneh-Far R, et al. Dysfunctional adiposity and the risk of prediabetes and type 2 diabetes in obese adults. JAMA. 2012;308:1150– 1159. doi: 10.1001/2012.jama.11132. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Swainson MG, Batterham AM, Tsakirides C, Rutherford ZH, Hind K. Prediction of whole-body fat percentage and visceral adipose tissue mass from five anthropometric variables. PLoS ONE. 2017;12:e0177175. doi: 10.1371/journal.pone.0177175. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Ashwell M, Mayhew L, Richardson J, Rickayzen B. Waist-to-height ratio is more predictive of years of life lost than body mass index. PLOS One. 2014;9:e103483. doi: 10.1371/journal.pone.0103483. https://doi:10.1371/journal.pone.0103483. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Ashwell M, Gunn P, Gibson S. Waist-to-height ratio is a better screening tool than waist circumference and BMI for adult cardiometabolic risk factors: systematic review and meta-analysis. Obes Rev. 2012;13:275–286. doi: 10.1111/j.1467-789X.2011.00952.x. [DOI] [PubMed] [Google Scholar]
- 7.Schneider HJ, Friedrich N, Klotsche J, Pieper L, Nauck M, John U, et al. The predictive value of different measures of obesity for incident cardiovascular events and mortality. J Clin Endocrinol Metab. 2010;95:1777–1785. doi: 10.1210/jc.2009-1584. [DOI] [PubMed] [Google Scholar]
- 8.Wohlfahrt P, Somers VK, Cifkova R, Filipovsky J, Seidlerova J, Krajcoviechova A, et al. Relationship between measures of central and general adiposity with aortic stiffness in the general population. Atherosclerosis. 2014;235:625– 631. doi: 10.1016/j.atherosclerosis.2014.05.958. [DOI] [PubMed] [Google Scholar]
- 9.Hori A, Nanri A, Sakamoto N, Kuwahara K, Nagahama S, Kato N, et al. Comparison of body mass index, waist circumference, and waist-to-height ratio for predicting the clustering of cardiometabolic risk factors by age in Japanese workers – Japan Epidemiology Collaboration on Occupational Health Study. Circ J. 2014;78:1160– 1168. doi: 10.1253/circj.cj-13-1067. [DOI] [PubMed] [Google Scholar]
- 10.Ware LJ, Rennie KL, Kruger HS, Kruger IM, Greeff M, Fourie CM, et al. Evaluation of waist-to-height ratio to predict 5-year cardiometabolic risk in sub-Saharan African Adults. Nutr Metab Cardiovasc Dis. 2014;24:900– 907. doi: 10.1016/j.numecd.2014.02.005. [DOI] [PubMed] [Google Scholar]
- 11.Lee CMY, Huxley RR, Wildman RP, Woodward M. Indices of abdominal obesity are better discriminators of cardiovascular risk factors than BMI: a meta-analysis. J Clin Epidemiol. 2008;61:646–653. doi: 10.1016/j.jclinepi.2007.08.012. [DOI] [PubMed] [Google Scholar]
- 12.Khalili S, Hatami M, Hadaegh F, Sheikholeslami F, Azizi F. Prediction of cardiovascular events with consideration of general and central obesity measures in diabetic adults: Results of the 8.4-year follow-up. Metabol Syndrome Related Disord. 2012;10:218–224. doi: 10.1089/met.2011.0070. [DOI] [PubMed] [Google Scholar]
- 13.World Health Organization. Waist Circumference and Waist-Hip Ratio: Report of a WHO Expert Consultation. 2008 Available at: http://apps.who.int/iris/bitstream/10665/44583/1/9789241501491_eng.pdf.
- 14.Fuqua SR, Wyatt SB, Andrew ME, Sarpong DF, Henderson FR, Cunningham MF, et al. Recruiting African-American research participation in the Jackson Heart Study: methods, response rates and sample description. Ethn Dis. 2005;15(4 Suppl 6):S6-18–29. [PubMed] [Google Scholar]
- 15.Sempos CT, Bild DE, Manolio TA. Overview of the Jackson Heart Study: a study of cardiovascular diseases in African American men and women. Am J Med Sci. 1999;317:142– 6. doi: 10.1097/00000441-199903000-00002. [DOI] [PubMed] [Google Scholar]
- 16.Carpenter MA, Crow R, Steffes M, Rock W, Heilbraun J, Evans G, et al. Laboratory, reading center, and coordinating center data management methods in the Jackson Heart Study. Am J Med Sci. 2004;328:131–144. doi: 10.1097/00000441-200409000-00001. [DOI] [PubMed] [Google Scholar]
- 17.American Diabetes Association. Diagnosis and classification of diabetes mellitus. Diabetes Care. 2010;33:S62–S69. doi: 10.2337/dc10-S062. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.DeLong ER, DeLong DM, Clarke-Pearson DL. Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach. Biometrics. 1988 Sep;44:837–45. [PubMed] [Google Scholar]
- 19.Browning LM, Hsieh SD, Ashwell M. A systematic review of waist-to-height ratio as a screening tool for the prediction of cardiovascular disease and diabetes: 0.5 could be a suitable global boundary value. Nutr Res Rev. 2010;23:247– 269. doi: 10.1017/S0954422410000144. [DOI] [PubMed] [Google Scholar]
- 20.Katzmarzyk PT, Mire E, Bray GA, Greenway FL, Heymsfield SB, Bouchard C. Anthropometric markers of obesity and mortality in white and African American adults: the pennington center longitudinal study. Obesity (Silver Spring) 2013;21:1070– 1075. doi: 10.1002/oby.20151. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Kazlauskaite R, Avery-Mamer EF, Li H, Chataut CP, Janssen I, Powell LH, et al. Race/ethnic comparisons of waist-to-height ratio for cardiometabolic screening: The study of women’s health across the nation. Am J Hum Biol. 2017;29 doi: 10.1002/ajhb.22909. Epub 2016 Nov 1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.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 (London) 2006;30:988– 992. doi: 10.1038/sj.ijo.0803226. [DOI] [PubMed] [Google Scholar]