Differential obesity indices identify the metabolic syndrome in Black men and women in Cape Town: the CRIBSA study

Journal of Public Health Vol. 38, No. 1, pp. 175 182 doi:10.1093/pubmed/fdu115 Advance Access Publication January 24, 2015 Differential obesity indices identify the metabolic syndrome in Black men and women in Cape Town: the CRIBSA study N. Peer 1,2, K. Steyn 3, N. Levitt 3,4 1 Non-communicable Diseases Research Unit, South African Medical Research Council, Durban, 4001, South Africa 2 Department of Medicine, University of Cape Town, Cape Town, 8001, South Africa 3 Chronic Disease Initiative for Africa, Department of Medicine, University of Cape Town, Cape Town, 8001, South Africa 4 Division of Diabetic Medicine and Endocrinology, Department of Medicine, University of Cape Town, Cape Town, 8001, South Africa Address correspondence to Nasheeta Peer, E-mail: nasheeta.peer@mrc.ac.za ABSTRACT Aims To determine the obesity indices, specifically waist circumference (WC), that identified 2 other metabolic syndrome (MS) components (2009 criteria) in 25- to 74-year-old Africans in Cape Town. Methods Data were collected from a cross-sectional sample by administered questionnaires, clinical measurements and biochemical analyses. The obesity cut points were estimated by the Youden Index. Logistic regression analyses determined whether obesity cut points identifying 2 MS components occurred at true inflection points. Results Among the 1099 participants, the calculated cut points and 95% confidence intervals (CI) were: men, WC 83.9 cm (81.6 86.2), waistto-hip ratio (WHR) 0.89 (0.87 0.90), waist-to-height ratio (WHtR) 0.50 (0.48 0.52) and body mass index (BMI) 24.1 kg/m 2 (22.0 26.1); women, WC 94.0 cm (92.6 95.3), WHR 0.85 (0.83 0.87), WHtR 0.59 (0.57 0.60) and BMI 32.1 kg/m 2 (29.7 34.6). Raised WC was significantly associated with 2 MS components in men: WC 84.0 93.9 cm (odds ratio (OR): 3.19, 95% confidence interval (CI): 1.73 5.85) and WC 94.0 cm (OR: 8.50, 95% CI: 4.44 16.25) compared with WC,84.0 cm, and in women: WC 80.0 93.9 cm (OR: 2.93, 95% CI: 1.32 6.54) and WC 94.0 cm (OR: 5.33, 95% CI: 2.40 11.85) compared with WC,80.0 cm. In the logistic model with BMI for women, obesity (OR: 3.60, 95% CI: 1.82 7.10) but not overweight (P ¼ 0.063) was significantly associated with 2 MS components. Conclusions Obesity cut points for Africans should be re-evaluated and adjusted accordingly. Keywords Black, BMI, obesity, obesity threshold, South Africa, urban, waist circumference Introduction The concept of the metabolic syndrome (MS) arose from the frequent clustering of diabetes, hypertension and dyslipidaemia and is characterized by variable combinations of central obesity and alterations in blood pressure (BP), glucose and lipid metabolisms. 1,2 The relevance of the MS is that a combination of even modest elevations in BP, plasma lipids and blood glucose confers substantial excess risk for diabetes, cardiovascular disease (CVD) and death, and a diagnosis of the MS warns of extremely adverse metabolic states. 3,4 Furthermore, diagnosing the MS extends the concept of cardiovascular risk with the identification of individuals at high risk of developing diabetes, hypertension and premature CVD. 5 7 Over the past few decades, there have been several definitions of the MS using various diagnostic criteria; the most commonly cited include those proposed initially by the WHO (1998, 1999), the National Cholesterol Education Program Third Adult Treatment Panel (ATPIII) (2001, 2004) and more recently, the International Diabetes Federation (IDF) (2005). 8 10 In 2009, in an attempt to unify the criteria, an additional definition was proposed as a harmonized Joint Interim Statement ( JIS) by several organizations. 2 Notwithstanding the various definitions, central obesity, dyslipidaemia, hyperglycaemia and elevated BP are the key characteristics. 11 13 N. Peer, Specialist Scientist K. Steyn, Professor in the Department of Medicine N. Levitt, Head of Diabetic Medicine and Endocrinology # The Author 2015. Published by Oxford University Press on behalf of Faculty of Public Health. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. 175

176 JOURNAL OF PUBLIC HEALTH For a diagnosis of the MS by the JIS criteria, three out of five abnormal findings, which include raised waist circumference (WC), raised triglycerides, low high-density lipoprotein cholesterol (HDL-C), raised BP and raised fasting glucose, are required. 2 Notably, although the use of population-specific cut points is recommended for the WC criteria, these have not been defined for Africans because of a lack of appropriate data and are currently the same as that for Caucasians. However, studies conducted in Black populations in South Africa suggest 14 16 that these WC criteria likely require modification. Considering that the WC measurement is the cheapest MS component to determine, it is important to identify the optimal WC cut point to predict the presence of this syndrome to enable correct diagnosis and management. This is particularly true in resource-constrained settings such as South Africa and in view of the high and rising prevalence of the MS components in the country. 17 The Cardiovascular Risk in Black South Africans (CRIBSA) study thus aimed to determine the obesity indices, in particular WC, to identify the MS, its individual components and other dyslipidaemia in urban-dwelling Black men and women in Cape Town. Methodology Study population and sampling procedure In 2008/09, 25- to 74-year-old men and women living in five predominantly Black residential areas (Langa, Guguletu, Crossroads, Nyanga and Khayelitsha) in Cape Town were sampled for this cross-sectional study. Individuals who were unable to give consent, on tuberculosis treatment, on antiretroviral therapy, cancer patients having received treatment within the last year, bedridden, pregnant or lactating, or resident in Cape Town for,3 months were excluded from participating. The sampling procedure has been described in detail previously 17 and included a three-stage cluster sampling stratified by area and housing type. Quotas, pre-specified by age and gender categories, were calculated using the 2001 census. Data collection Trained fieldworkers administered questionnaires pertaining to socio-demographic data and the relevant medical history. Clinical assessments included three BP measurements taken at 2-min intervals using an Omron BP monitor with an appropriately sized cuff after the participant had been seated for 5 min. The average of the second and third BP measurements was used in the analysis. Three trained staff, including two nurses, obtained the anthropometric measurements. Weight, measured to the nearest 0.5 kg, was determined with each participant barefoot and in light clothing using a calibrated digital scale. Height was measured with a stadiometer to the nearest 0.1 cm. A flexible, non-elastic tape measure held parallel to the floor was used to determine waist and hip circumferences to the nearest 0.1 cm. The waist measurement was taken as the smallest circumference between the xiphisternum and the umbilicus on expiration while standing. Hip measurements were taken at the maximum posterior protuberance of the buttocks. 18 Following an overnight 10 h fast, blood samples were drawn for glucose and lipid estimations. A standard oral glucose tolerance test was thereafter administered, and blood samples were taken 120 min later. 19 Definitions The cut points for the MS components by the JIS criteria are as follows: WC: men 94 cm, women 80 cm, triglycerides:.1.7 mmol/l, HDL-C: men,1.0 mmol/l, women,1.3 mmol/l, BP 130/85 mmhg or the use of antihypertensive medication, and fasting glucose 5.6 mmol/l or on diabetes treatment. For the purpose of this study, two out of four components determined the MS with the WC component excluded. Other dyslipidaemia components were defined as follows 20 : total cholesterol (TC).5 mmol/l, HDL-C/TC ratio,20% and calculated low-density lipoprotein cholesterol (LDL-C) using the Friedewald equation.3.0 mmol/l. 21 Standardized international criteria identified overweight (25 29.9 kg/m 2 ) and obesity (30 kg/m 2 ) by body mass index (BMI) and raised waist-to-hip ratio (WHR) (.1.0 in men and.0.85 in women). 22 Raised waist-to-height ratio (WHtR) was defined as.0.5. 23 Statistical analysis Data analyses were conducted using STATA 12. Descriptive statistics, including crude prevalence, were calculated using the weights based on the sample design and adjusted for the realized sample. Univariate analyses are presented as mean values and standard deviations for the continuous data and as percentages for the categorical data. The area under the receiver operating characteristic (ROC) curve and the corresponding 95% confidence intervals (95% CI) estimated the optimum WC cut-off values for the diagnosis of individual MS components and other dyslipidaemia variables. The cut points of obesity indices for identifying the presence of the MS were estimated using the Youden Index and are presented with the associated sensitivity and specificity values. The Youden Index ( J) is the value that has the highest

OBESITY INDICES FOR METABOLIC SYNDROME IN BLACK MEN AND WOMEN 177 sum of sensitivity and specificity, i.e. J ¼ maximum fsensitivity þ specificity21g. 24 Sensitivity is the true-positive rate, i.e. the percentage of participants with the MS risk who were correctly identified by the anthropometric index cut points. Specificity is the true-negative rate, i.e. the percentage of participants without the MS risk who were correctly identified by the anthropometric index cut points. Survey logistic regression analyses, adjusted for age, determined whether the obesity cut-off values, obtained from the ROC curves to identify the MS, occurred at true inflection points. Any significant increases in the risk for MS at the obesity cut points were observed. There were three categories for WC; the cut points determined in this study represent the lower WC limit in men (84.0 cm) and the upper WC limit in women (94.0 cm). Standard cut-off values were used to categorize overweight and obesity by BMI, because the 95% CI for the calculated BMI cut points overlapped the international criteria. Each obesity measure was entered in a gender-specific model because of differences in obesity cut points. The University of Cape Town s Research and Ethics Committee approved the study. All participants signed informed consent. Results Of the 1303 individuals invited to participate, 187 did not respond and 1116 were examined, 17 of whom were excluded from the final analyses, because they did not fulfil the selection criteria. The realized study sample, with a response rate of 86%, comprised 1099 participants: 392 men and 707 women. The number of men and women, respectively, per age category were as follows: 25 34 years: 121 and 235; 35 44 years: 99 and 163; 45 54 years: 86 and 168; 55 64 years: 58 and 92; and 65 74 years: 28 and 49. As shown in Table 1, women generally had a worse cardiometabolic profile than men. The prevalence of BMI 25 kg/m 2 was 82.8% in women and 28.9% in men while WC.80 cm in the former was 86.0% and WC.94 cm in the latter was 20.1%. Dysglycaemia and most dyslipidaemia parameters were higher in women than in men, but raised triglycerides and BP were more prevalent in the latter. Among men and women with dysglycaemia, 20.8 and 27.7% (P ¼ 0.371), respectively, were on diabetes treatment. Antihypertensive agents were taken by 16.0% of men and 43.9% of women (P, 0.001) with raised BP. The ROC curve analyses that determined the obesity cut points for diagnosing the MS and the sensitivity and specificity of each of these cut points are presented in Table 2. In men, the 95% CI for the cut points do not include the currently recommended cut point values: WC: 83.9 cm, 95% CI: 81.6 86.2 and WHR: 0.89, 95% CI: 0.87 0.90. However, the cut point for BMI (24.1 kg/m 2, 95% CI: 22.0 26.1) in men approximated the overweight value of 25 kg/m 2. The accuracy of the estimated cut points for the area under ROC curve in men was satisfactory at.0.70. In women, the cut points for WC (94.0 cm, 95% CI: 92.6 95.3) and WHtR (0.59, 95% CI: 0.57 0.60) were higher than the currently recommended values (Table 2). The 95% CI for the BMI cut point (32.1 kg/m 2, 95% CI: 29.7 34.6) in women included the obesity threshold of 30 kg/m 2. The accuracy of the estimated cut points in women (0.63 0.69) was lower than in men. The prevalence of the MS by the JIS criteria (three out of five abnormal findings) was 16.5% (95% CI: 12.7 21.2) and 43.5% (95% CI: 39.2 47.9) in men and women, respectively. If the WC criteria were adjusted to 84 cm in men and 94 cm in women, MS prevalence would increase to 23.3% (95% CI: 18.7 28.5) in men and decrease to 34.6% (95% CI: 30.4 39.0) in women. Table 3 presents the area under the ROC curve that determined the WC cut points for the individual MS components and other dyslipidaemia variables along with the sensitivity and specificity for each of these cut-off values. In men, the WC cut-off values (83.8 90.1 cm) to identify the individual MS components and other dyslipidaemia variables were lower than the 94 cm currently being used. In contrast, the cut-off values in women, which ranged from 94.9 to 100.1 cm, were much higher than the 80 cm WC cut point currently recommended. The accuracy of the estimated cut point in both genders was the highest for dysglycaemia (0.76 0.78) and the lowest for HDL-C (0.57), as well as TC (0.55) in women. The obesity cut points, identified from ROC curve analyses and which diagnosed the MS, were further investigated to determine whether they occurred at true inflection points for MS risk and are presented in Table 4. In men, WC greater than that determined in this study was associated with an increased odds for the MS: WC 84.0 93.9 cm (odds ratio (OR): 3.19, 95% CI: 1.73 5.85) and WC 94.0 cm (OR: 8.50, 95% CI: 4.44 16.25) compared with WC,84.0 cm. Compared with BMI, 25 kg/m 2, being overweight (P, 0.001) or obese (P, 0.001) in men was significantly associated with the MS. Compared with women with WC,80 cm, those with WC 80 93.9 cm had an almost 3-fold greater likelihood of the MS (P ¼ 0.009), while those with WC 94 cm had a 5-fold higher risk (P, 0.001) (Table 4). The odds for the MS in women who were overweight was not significantly different from those with BMI, 25 kg/m 2 (P ¼ 0.063); however, the odds were significantly higher for obese women (OR: 3.60, 95% CI: 1.82 7.10, P, 0.001).

178 JOURNAL OF PUBLIC HEALTH Table 1 Mean + SD and prevalence of cardio-metabolic risk factors in men and women Men Women Number 392 707 Mean + SD Anthropometry Height (m) 1.7 + 0.1 1.6 + 0.1 Weight (kg) 69.6 + 15.8 84.0 + 21.1 Waist circumference (cm) 85.7 + 13.8 97.5 + 15.0 Hip circumference (cm) 95.6 + 10.3 114.8 + 15.6 Body mass index (kg/m 2 ) 24.0 + 5.3 33.1 + 8.1 Waist-to-hip ratio 0.89 + 0.1 0.85 + 0.1 Waist-to-height ratio 0.50 + 0.1 0.61 + 0.1 Blood pressure (mmhg) Systolic 131.5 + 22.4 123.0 + 23.2 Diastolic 82.5 + 13.4 81.7 + 13.2 Glucose (mmol/l) Fasting 5.2 + 2.0 5.5 + 2.8 2h 6.4 + 3.1 7.2 + 4.7 Lipids Total cholesterol (TC) (mmol/l) 4.3 + 1.1 4.5 + 1.2 Low-density lipoprotein cholesterol (LDL-C) (mmol/l) 2.9 + 0.9 3.1 + 1.0 High-density lipoprotein cholesterol (HDL-C) (mmol/l) 1.3 + 0.5 1.1 + 0.4 Triglycerides (mmol/l) 1.2 + 1.0 1.1 + 0.9 HDL-C : TC ratio 29.4 + 10.2 26.0 + 9.5 Prevalence, % Anthropometry Body mass index 25 kg/m 2 28.9 82.8 Waist circumference: men.94 cm, women.80 cm 20.1 86.0 Waist-to-hip ratio: men.1.0, women.0.85 9.2 47.6 Waist-to-height ratio.0.5 37.6 85.3 Hypertension BP 130/85 mmhg or on hypertension treatment 51.2 46.5 Diabetes Fasting glucose 5.6 mmol/l or on diabetes treatment 15.4 21.7 Dyslipidaemia TC.5 mmol/l 25.2 23.1 HDL-C: men,1.0 mmol/l, women,1.3 mmol/l 38.8 74.7 Triglycerides.1.7 mmol/l 17.9 11.5 LDL-C.3.0 mmol/l 37.8 47.0 HDL-C : TC,20% 14.4 18.9 Smoke daily or occasionally 57.8 9.8 Alcohol use 67.5 26.9 Mean + SD are reported for the study sample and not adjusted for the population. Discussion Main finding of this study This is the first population-based study in South Africa that identified the MS risk by multiple obesity indices in an African population. The current recommended obesity indices cut points in African populations, in particular WC, do not concur with the findings in this sample of Black men and women in Cape Town. The IDF WC value of 94 cm in Sub-Saharan African men is markedly higher than the WC cut point of

OBESITY INDICES FOR METABOLIC SYNDROME IN BLACK MEN AND WOMEN 179 Table 2 Cut points of obesity indices for identifying the presence of 2 other components of the metabolic syndrome estimated by the Youden Index with the associated sensitivity and specificity Area under ROC curve (95% CI) Cut point (95% CI) Sensitivity (95% CI) Specificity (95% CI) Men Waist circumference (cm) 0.77 (0.72 0.81) 83.9 (81.6 86.2) 0.74 (0.64 0.85) 0.69 (0.60 0.76) Waist-to-hip ratio 0.74 (0.69 0.79) 0.89 (0.87 0.90) 0.73 (0.63 0.83) 0.65 (0.55 0.74) Waist-to-height ratio 0.76 (0.71 0.81) 0.50 (0.48 0.52) 0.68 (0.54 0.82) 0.73 (0.59 0.86) Body mass index (kg/m 2 ) 0.72 (0.67 0.78) 24.1 (22.0 26.1) 0.61 (0.45 0.76) 0.73 (0.57 0.90) Women Waist circumference (cm) 0.68 (0.64 0.71) 94.0 (91.9 96.0) 0.74 (0.68 0.80) 0.57 (0.51 0.62) Waist-to-hip ratio 0.63 (0.59 0.68) 0.85 (0.83 0.87) 0.60 (0.50 0.70) 0.61 (0.50 0.72) Waist-to-height ratio 0.69 (0.66 0.73) 0.59 (0.57 0.60) 0.76 (0.68 0.85) 0.57 (0.51 0.63) Body mass index (kg/m 2 ) 0.66 (0.62 0.70) 32.1 (29.7 34.6) 0.63 (0.50 0.77) 0.60 (0.49 0.71) Waist circumference not included in definition of 2 metabolic syndrome (MS) components. Table 3 Characteristics of ROC curves used for the identification of waist circumference cut-off values for the diagnosis of metabolic syndrome components and other dyslipidaemia in men and women Area under ROC curve (95% CI) Waist circumference cut-off value in cm (95% CI) Sensitivity (95% CI) Specificity (95% CI) Men BP 130/85 mmhg or on hypertension treatment 0.70 (0.65 0.75) 79.8 (75.7 83.8) 0.73 (0.62 0.85) 0.60 (0.48 0.71) Fasting glucose 5.6 mmol/l or on diabetes treatment 0.76 (0.70 0.82) 88.4 (82.1 94.6) 0.67 (0.54 0.80) 0.73 (0.60 0.86) Total cholesterol.5 mmol/l 0.64 (0.57 0.70) 87.7 (81.9 93.4) 0.53 (0.37 0.70) 0.70 (0.54 0.87) HDL-C,1.0 mmol/l 0.57 (0.51 0.63) 83.8 (76.6 90.9) 0.55 (0.41 0.69) 0.58 (0.40 0.75) Triglycerides.1.7 mmol/l 0.74 (0.68 0.80) 87.7 (80.9 94.4) 0.68 (0.55 0.81) 0.72 (0.58 0.86) LDL-C.3.0 mmol/l 0.66 (0.61 0.72) 82.4 (78.3 86.8) 0.69 (0.56 0.81) 0.61 (0.50 0.71) HDL-C : TC,20% 0.74 (0.67 0.81) 90.1 (87.2 93.0) 0.69 (0.58 0.81) 0.76 (0.68 0.84) Women BP 130/85 mmhg or on hypertension treatment 0.67 (0.63 0.71) 95.2 (93.6 96.8) 0.67 (0.61 0.74) 0.62 (0.55 0.68) Fasting glucose 5.6 mmol/l or on diabetes treatment 0.78 (0.63 0.72) 100.1 (94.7 105.5) 0.62 (0.47 0.78) 0.65 (0.53 0.76) Total cholesterol.5 mmol/l 0.55 (0.51 0.60) 94.9 (91.6 98.2) 0.65 (0.55 0.75) 0.49 (0.41 0.57) HDL-C,1.3 mmol/l 0.57 (0.52 0.61) 96.1 (92.0 100.2) 0.53 (0.44 0.63) 0.59 (0.48 0.69) Triglycerides.1.7 mmol/l 0.61 (0.55 0.67) 97.9 (93.3 10.5) 0.64 (0.50 0.78) 0.56 (0.44 0.68) LDL-C.3.0 mmol/l 0.57 (0.53 0.62) 94.9 (91.9 97.9) 0.62 (0.52 0.71) 0.52 (0.45 0.60) HDL-C : TC,20% 0.64 (0.60 0.69) 99.2 (95.4 102.9) 0.60 (0.49 0.71) 0.61 (0.51 0.71) 84 cm shown in this study. On the other hand, the WC threshold that identified the MS in women was 94 cm, markedly higher than the 80 cm cut point currently used. The odds for the MS in men with intermediate raised WC (84.0 93.9 cm) was comparable to that for overweight by BMI (25 29.9 kg/m 2 ). The WC threshold of 94 cm in women corresponded with a BMI cut point in the obesity range for identifying 2 MS components. The fairly narrow range of WC cut points identified in men and women for the detection of individual MS components and other dyslipidaemia variables demonstrates that the risk for these conditions increased at similar levels of abdominal adiposity. This may suggest a common aetiological pathway where there may be a level of visceral adiposity above which CVD risk increases. 15 That the accuracy of the estimated cut point was the highest for raised glucose, among the individual MS

180 JOURNAL OF PUBLIC HEALTH Table 4 Associations of obesity indices with the presence of 2 metabolic syndrome components in men and women Men Cut points n % Odds ratio 95% confidence interval Women Cut points n % Odds ratio 95% confidence interval Lower limit Upper limit Lower limit Upper limit 2 MS components 392 32.6 707 45.0 Waist circumference (cm),84 203 15.9 1.00,80 81 15.3 1.00 84 93.9 92 42.9 3.19 1.73 5.85 80 93.9 214 35.9 2.93 1.32 6.54 94 92 69.1 8.50 4.44 16.25 94 412 57.1 5.33 2.40 11.85 Waist-to-hip ratio,0.90 231 20.7 1.00,0.85 357 36.2 1.00 0.90 161 53.6 2.91 1.76 4.79 0.85 350 54.7 1.49 1.08 2.03 Waist-to-height ratio 0.50 228 18.1 1.00,0.60 335 32.8 1.00.0.50 164 56.7 4.23 2.46 7.27 0.60 372 56.6 2.14 1.50 3.06 Body mass index (kg/m 2 ),25 270 22.5 1.00,25 114 22.3 1.00 25 29.9 75 52.0 3.21 1.70 6.08 25 29.9 148 36.2 1.93 0.96 3.87 30 47 67.9 5.14 2.14 12.37 30 445 54.3 3.60 1.82 7.10 Data adjusted for age; waist circumference not included in definition of 2 metabolic syndrome (MS) components. components, was not surprising considering the close link between obesity and diabetes, which has also been reported in this population. 17 The least accurate of the estimated cut points was for HDL-C with the area under ROC curve of 0.57 in both men and women. This was in accordance with previous reports which found that HDL-C levels appeared to be relatively insensitive to the effects of obesity in Africans. 16 What is already known on this topic Other South African studies have reported a lower WC threshold in men than the current recommendations. 14,16 Motala and colleagues 14 found a WC of 86 cm to be predictive of 2 other MS components in rural Black men. Kalk and colleagues 16 recommended WC cut points of 88 or 90 cm in Black men with diabetes. The WC of 90 cm recommended for South African men of mixed ancestry was also lower than the currently used threshold. 25 The WC threshold that identified 2 MS components in women was higher than the recommended 80 cm cut point in rural Black women (92 cm) 14 and in the Birth to Twenty cohort in Soweto (91.5 cm). 15 Notably, despite the differences among the current and latter studies with regards to age, obesity prevalence and rural urban location, the WC cut points reported in these Black women were similar. The markedly higher WC cut point in Black women highlights ethnic differences in body composition and is likely related to their lower visceral fat levels for a given WC compared with White women. 26 Of interest is that the South African studies that assessed both men and women, i.e. this and the rural South African study, 14 reported a higher WC threshold in Black women compared with men. Studies have found that greater body fat accumulation is required in women before they approximate the visceral adipose levels found in men. 27 One demonstrated that intra-abdominal adiposity in young women was minimal until they accumulated 30 kg of fat elsewhere on the body compared with men where intraabdominal fat increased in a straight line relationship with increasing overall adiposity. Although this may account for the lower WC cut point in Black men compared with their female counterparts, further research examining the lower WC threshold in Black compared with White men is required. Ethnic differences in body fat patterning or sensitivity to the factors affecting insulin resistance may contribute to the population variances in WC thresholds for men. What this study adds The MS risk for overweight (BMI 25 29.9 kg/m 2 ) in men approximates WC of 84.0 93.9 cm. The IDF WC cut point of.94 cm in men may be equated to very high risk in this sample. This contrasts with the WC cut points of 94 cm identified as increased risk and 102 cm as still higher risk in White men. 2 Consideration may be given to reducing the WC cut point in Black men as this may enhance the identification of those at risk for diabetes and CVD.

OBESITY INDICES FOR METABOLIC SYNDROME IN BLACK MEN AND WOMEN 181 Notably, although the odds for identifying 2 MS components for BMI 25 29.9 kg/m 2 in women was higher than BMI, 25 kg/m 2, this was not significant. Ascribing high risk to Black women in the overweight BMI range should thus perhaps be done with greater circumspection. Interestingly, the lack of a significant association for overweight women, in contrast to intermediate raised WC, may reinforce that central obesity is a better indicator of visceral fat and metabolic risk. The significant association of intermediate raised WC with the MS suggests that there is likely an incremental increase in risk associated with rising WC rather than an absolute cut-off level for high risk. WC 80.0 93.9 cm may therefore be regarded as moderate risk. Limitations of this study The major limitation of this study for determining obesity cut points is the cross-sectional design. This precludes causal associations and can only suggest a long-term relationship between specific obesity indices and CVD risk; longitudinal studies examining the development of these risk factors are required to confirm these findings. Furthermore, the accuracy of the ROC curves, and the sensitivity and specificity of the cut points were not high. Nonetheless, these obesity indices, WC in particular, are extremely useful screening tools. Other limitations include the low sample realization in men (64%) which necessitated higher sampling weights and loss of precision, and BP readings taken on one rather than on at least three separate occasions for the diagnosis of hypertension which may have led to some overestimation of the prevalence. A notable limitation is the exclusion of known HIV-infected individuals from the study. Considering that HIV infection is associated with weight loss, and that the condition and treatment contribute to a higher risk for cardio-metabolic abnormalities, it is possible that the obesity indices for identifying MS in the HIV infected in this population may be lower than these study findings. Conclusion The findings of this study highlight that it may be prudent to caution African men with WC of 84.0 93.9 cm of their greater risk for the MS compared with WC, 84.0 cm. Furthermore, in African women, there appears to be a graded risk for the MS with much higher risk associated with WC 94 cm compared with WC 80.0 93.9 cm. In light of the findings of this and other studies that underscore the discrepancy in the proposed WC cut points for MS risk in Black men and women, there is a need for further research to identify the obesity thresholds to diagnose the MS in this population. Ideally, large prospective randomly selected cohort studies are required for the determination of optimal obesity indices to diagnose increased CVD risk. The data currently available, including that from a single cohort study in South Africa, 28 may provide impetus for the planning of future cohort studies in Sub-Saharan Africa. Acknowledgements The authors acknowledge the participants, fieldworkers, MRC research nurse fieldworkers Debbie Jonathan and Theresa Gogela, fieldwork co-ordinator Erica April, and study manager Serena van Haght. We also thank the City of Cape Town for the provision of the aerial maps. Funding This work was supported by an unrestricted grant from Servier Laboratories (South Africa); the South African Medical Research Council; the Initiative for Cardiovascular Health Research in Developing Countries (IC Health) Foundation Council and Brigham and Women s Hospital, Harvard University. References 1 Summary of the 2007 European Society of Hypertension (ESH) and European Society of Cardiology (ESC) guidelines for the management of arterial hypertension. Vasc Health Risk Manag 2007;3(6): 783 95. 2 Alberti KG, Eckel RH, Grundy SM et al. Harmonizing the metabolic syndrome: a joint interim statement of the international diabetes federation task force on epidemiology and prevention; National Heart, Lung, and Blood Institute; American Heart Association; World Heart Federation; International Atherosclerosis Society; and International Association for the Study of Obesity. Circulation 2009;120(16): 1640 5. 3 Meigs JB. Epidemiology of the metabolic syndrome, 2002. Am J Manag Care 2002;8(11 Suppl):S283 92; quiz S93 6. 4 Wilson PW, D Agostino RB, Parise H et al. Metabolic syndrome as a precursor of cardiovascular disease and type 2 diabetes mellitus. Circulation 2005;112(20):3066 72. 5 Messerli FH, Williams B, Ritz E. Essential hypertension. Lancet 2007;370(9587):591 603. 6 Opie LH. Metabolic syndrome. Circulation 2007;115(3):e32 5. 7 Jee SH, Jo J. Linkage of epidemiologic evidence with the clinical aspects of metabolic syndrome. Korean Circ J 2012;42(6):371 8. 8 Motala AA, Mbanya JC, Ramaiya KL. Metabolic syndrome in sub-saharan Africa. Ethn Dis 2009;19(2 Suppl. 2):S2-8 10. 9 Wong ND. Metabolic syndrome: cardiovascular risk assessment and management. Am J Cardiovasc Drugs 2007;7(4):259 72.

182 JOURNAL OF PUBLIC HEALTH 10 Ford ES. Prevalence of the metabolic syndrome defined by the International Diabetes Federation among adults in the U.S. Diabetes Care 2005;28(11):2745 9. 11 Taslim S, Tai ES. The relevance of the metabolic syndrome. Ann Acad Med Singapore 2009;38(1):29 33. 12 Ceska R. Clinical implications of the metabolic syndrome. Diab Vasc Dis Res 2007;4(Suppl. 3):S2 4. 13 International Diabetes Federation. The IDF Consensus Worldwide Definition of the Metabolic Syndrome. Belgium: Brussels International Diabetes Federation, 2006. http://www.idf.org/webdata/docs/mets_ def_update2006.pdf (12 March 2014, date last accessed). 14 Motala AA, Esterhuizen T, Pirie FJ et al. The prevalence of metabolic syndrome and determination of the optimal waist circumference cutoff points in a rural South african community. Diabetes Care 2011;34(4):1032 7. 15 Crowther NJ, Norris SA. The current waist circumference cut point used for the diagnosis of metabolic syndrome in sub-saharan African women is not appropriate. PloS ONE 2012;7(11):e48883. 16 Kalk WJ, Joffe BI, Sumner AE. The waist circumference of risk in black South african men is lower than in men of European ancestry. Metab Syndr Relat Disord 2011;9(6):491 5. 17 Peer N, Steyn K, Lombard C et al. Rising diabetes prevalence among urban-dwelling black South Africans. PloS ONE 2012;7(9):e43336. 18 Puoane T, Steyn K, Bradshaw D et al. Obesity in South Africa: the South African demographic and health survey. Obes Res 2002;10(10):1038 48. 19 World Health Organization. Definition, Diagnosis and Classification of Diabetes Mellitus and Its Complications: Report of a WHO Consultation. Geneva, Switzerland: World Health Organization, 1999. 20 South African Medical Association and Lipid and Atherosclerosis Society of Southern Africa Working Group. Diagnosis, management and prevention of the common dyslipidaemias in South Africa clinical guideline, 2000. S Afr Med J 2000;90:164 74, 76 78. 21 Friedewald WT, Levy RI, Fredrickson DS. Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem 1972;18(6):499 502. 22 World Health Organization. Obesity: Preventing and Managing the Global Epidemic: Report of a WHO Consultation. Geneva, Switzerland: World Health Organization, 2000. 23 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(2):247 69. 24 Akobeng AK. Understanding diagnostic tests 3: receiver operating characteristic curves. Acta Paediatr 2007;96(5):644 7. 25 Matsha TE, Hassan MS, Hon GM et al. Derivation and validation of a waist circumference optimal cutoff for diagnosing metabolic syndrome in a South African mixed ancestry population. Int J Cardiol 2013;168(3):2954 5. 26 Sumner AE, Micklesfield LK, Ricks M et al. Waist circumference, BMI, and visceral adipose tissue in white women and women of African descent. Obesity (Silver Spring) 2011;19(3):671 4. 27 Friedl KE. Waist circumference threshold values for type 2 diabetes risk. J Diabetes Sci Technol 2009;3(4):761 9. 28 Ware LJ, Rennie KL, Kruger HS 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(8):900 7.