Author(s): Okosun IS, Liao Y, Rotimi CN, Choi S, Cooper RS
Abstract Share this page
Abstract Waist circumferences (WC) >/=102 cm for men and >/=88 cm for women have been proposed by an expert panel as cut-points for identifying increased risk for the development of obesity comorbidities for most adults. The aim of this investigation was to examine the predictive values of these WC cut-points for hypercholesterolemia, low concentration of high (HDL-C), and high concentration of low (LDL-C) density lipoprotein cholesterol, hypertriglyceridemia, type 2 diabetes, and hypertension in overweight American adults. Data from NHANES III were utilized for the analysis. Predictive abilities were determined by calculating sensitivity, specificity, positive (PV+) and negative (PV-) predictive values in overweight subjects with BMI 25-29.9 kg/m(2). Sensitivity of WC cut-point was stronger for high LDL-C compared to other risk factors with the highest values recorded in the 40-59 and 60-69 year age groups in men and women, respectively. PV+ of WC cut-points for dyslipidemia, type 2 diabetes, and hypertension were low in men compared to women. PV+ tended to increase with age, from 19-39, 40-59 to 60-90 year age groups in Whites, Blacks, and Hispanic men. In men, the highest PV+ were recorded for hypertriglyceridemia in the 60-90 years old groups, with values of 71.6\%, 52.5\%, and 43.3\% in Whites, Blacks, and Hispanics, respectively. The CVD risk factor associated with the highest PV+ in women was diabetes with values of 97.2\% in Whites and 88.9\% in Blacks, and hypertriglyceridemia with a value of 93.8\% in the 17-39 year age group in Hispanics. Among Black men 40-59 years of age, only 32\% of a population of overweight hypertensives were detected by the WC cut-points, and among Black women, 40-59 years of age, only 54\% were detected. Given the low sensitivity of these cut-points for detecting hypertension, one of the major co-morbidities of obesity, these cut-points failed to provide adequate evidence for the use of WC in determining or evaluating patients as to co-morbid states. We recommend further studies to determine a set of specific cut-points associated with increased risk of CVD in different population groups.
This article was published in J Clin Epidemiol
and referenced in Journal of Diabetes & Metabolism