Distinct phenotypic characteristics of normal-weight adults at risk of developing cardiovascular and metabolic diseases

Abishek Stanley; John Schuna; Shengping Yang; Samantha Kennedy; Moonseong Heo; Michael Wong; John Shepherd; Steven B Heymsfield

doi:10.1093/ajcn/nqaa194

. 2020 Jul 20;112(4):967–978. doi: 10.1093/ajcn/nqaa194

Distinct phenotypic characteristics of normal-weight adults at risk of developing cardiovascular and metabolic diseases

Abishek Stanley ¹, John Schuna ², Shengping Yang ³, Samantha Kennedy ⁴, Moonseong Heo ⁵, Michael Wong ⁶, John Shepherd ⁷, Steven B Heymsfield ^8,^✉

PMCID: PMC7762762 PMID: 32687153

ABSTRACT

Background

The normal-weight BMI range (18.5–24.9 kg/m²) includes adults with body shape and cardiometabolic disease risk features of excess adiposity, although a distinct phenotype developed on a large and diverse sample is lacking.

Objective

To identify demographic, behavioral, body composition, and health-risk biomarker characteristics of people in the normal-weight BMI range who are at increased risk of developing cardiovascular and metabolic diseases based on body shape.

Methods

Six nationally representative waist circumference index (WCI, weight/height^0.5) prediction formulas, with BMI and age as covariates, were developed using data from 17,359 non-Hispanic (NH) white, NH black, and Mexican-American NHANES 1999–2006 participants. These equations were then used to predict WCI in 5594 NHANES participants whose BMI was within the normal weight range. Men and women in each race/Hispanic-origin group were then separated into high, medium, and low tertiles based on the difference (residual) between measured and predicted WCI. Characteristics were compared across tertiles; P values for significance were adjusted for multiple comparisons.

Results

Men and women in the high WCI residual tertile, relative to their BMI and age-equivalent counterparts in the low tertile, had significantly lower activity levels; higher percent trunk and total body fat (e.g. NH white men, X ± SE, 25.3 ± 0.2% compared with 20.4 ± 0.2%); lower percent appendicular lean mass (skeletal muscle) and bone mineral content; and higher plasma insulin and triglycerides, higher homeostatic model assessment of insulin resistance (e.g. NH white men, 1.45 ± 0.07 compared with 1.08 ± 0.06), and lower plasma HDL cholesterol. Percent leg fat was also significantly higher in men but lower in women. Similar patterns of variable statistical significance were present within sex and race/ethnic groups.

Conclusions

Cardiometabolic disease risk related to body shape in people who are normal weight according to BMI is characterized by a distinct phenotype that includes potentially modifiable behavioral health risk factors.

Keywords: body shape, body composition, obesity, waist circumference, chronic disease

Introduction

Controversy surrounds the concept of “normal” body weight based on BMI, weight/height² (1). Not only is BMI modestly correlated with adiposity level (2), a key marker of health risks, but mortality studies identify a wide range of relative weights consistent with optimum health and longevity (3). To compensate for BMI limitations, some overweight and obesity guidelines include waist circumference (WC)as an additional anthropometric measure of a person's health risks (4). WC relates to health risks through its empirical correlations with visceral adipose tissue and ectopic lipids, known mediators of adiposity's adverse metabolic and cardiovascular effects (5).

As currently applied, single WC cut-points identifying men and women (e.g. >102 cm and >88 cm, respectively) who are at high risk of developing chronic cardiovascular and metabolic diseases are incorporated into guidelines without consideration for between-individual differences in stature, age, and in some cases, race and ethnicity (4–6). To overcome these limitations, a recent study reported a waist-circumference index (WCI) incorporating adjustment for height (7) analogous to BMI; that is, WCI (waist circumference/height^0.5) is a height-independent measure of adult body shape. When used in tandem with BMI and age in race and Hispanic origin-specific prediction models, a high WCI identifies people in the overweight and obese range with a distinct phenotype (7): high percent total body and trunk fat and low percent skeletal muscle mass. Subjects who have this phenotype also have a high cardiovascular and metabolic disease risk profile as assessed by laboratory and blood pressure studies. The current study aim was to test the hypothesis that this previously reported phenotype is also present in people whose BMI is within the normal weight range (18.5–24.9 kg/m²).

Methods

Study design

Participants in the 1999–2006 NHANES had body shape and composition evaluated using a combination of anthropometric and imaging methods as previously reported (8, 9). Participants were excluded if they were aged <18 y, amputees, pregnant, not in NHANES categories of non-Hispanic (NH) white, NH black, or Mexican American, and in subgroup analyses if they were taking medications known to influence their test results. The overall disposition of participants in the multiple analyses is summarized in Supplementary Methods I (Figure 1, Table 1A–C). All participants as outlined in the figure and tables were included in analyses and no outlying data was removed.

Algorithm used to classify people in the normal-weight BMI range according to their “residual” waist circumference index, calculated as the difference between their measured and predicted waist circumference index. Predicted values for WCI were derived from sex and race/Hispanic origin-specific regression models developed on a nationally representative sample of the US noninstitutionalized civilian population (Supplementary Table 4). People in the high WCI_R tertile are defined as having the metabolically obese, normal weight phenotype. WCI, waist circumference index; WCI_R, waist circumference index residual; MONW, metabolically obese, normal weight.

TABLE 1.

Demographic characteristics of the normal-weight NHANES sample¹

		Men				Women
Characteristic	Tertile	NHW	NHB	MA	Total	NHW	NHB	MA	Total
N¹		466–467	241–242	225	932–934	563	160–161	208–209	931–933
Age, years	Low	42.2 ± 1.0	39.6 ± 1.1^a	32.4 ± 1.2	38.1 ± 0.6^ab	48.0 ± 0.6^a	40.7 ± 1.2^a	36.8 ± 1.1	41.9 ± 0.6^a
	Medium	40.5 ± 0.8	35.6 ± 1.0^b	32.6 ± 0.9	36.2 ± 0.5^a	42.5 ± 0.7^b	35.5 ± 1.4^b	33.2 ± 1.0	37.1 ± 0.6^b
	High	42.2 ± 0.8	38.5 ± 1.2^ab	33.7 ± 1.1	38.2 ± 0.7^b	43.4 ± 1.0^b	40.6 ± 1.3^a	35.1 ± 1.6	39.7 ± 0.7^c
Height, cm	Low	176.3 ± 0.4^a	175.4 ± 0.5^a	169.1 ± 0.6	173.6 ± 0.3^a	163.2 ± 0.3^a	161.3 ± 0.6^a	158.0 ± 0.6	160.9 ± 0.3^a
	Medium	177.5 ± 0.5^a	177.3 ± 0.6^ab	169.5 ± 0.6	174.8 ± 0.3^b	163.7 ± 0.3^ab	163.6 ± 0.6^b	158.8 ± 0.5	162.0 ± 0.3^b
	High	179.0 ± 0.4^b	178.5 ± 0.5^b	170.4 ± 0.5	176.0 ± 0.3^c	164.1 ± 0.3^b	164.5 ± 0.5^b	158.8 ± 0.6	162.5 ± 0.2^b
Weight, kg	Low	71.0 ± 0.5^a	69.4 ± 0.6^a	64.5 ± 0.7	68.3 ± 0.4^a	59.0 ± 0.3^a	58.3 ± 0.6^a	56.5 ± 0.4	57.9 ± 0.2^a
	Medium	71.3 ± 0.6^a	69.5 ± 0.6^ab	65.2 ± 0.7	68.7 ± 0.4^a	59.1 ± 0.3^a	59.0 ± 0.6^a	56.9 ± 0.4	58.4 ± 0.3^a
	High	73.1 ± 0.4^b	71.5 ± 0.5^b	66.7 ± 0.6	70.4 ± 0.3^b	60.7 ± 0.3^b	61.0 ± 0.5^b	56.8 ± 0.6	59.5 ± 0.3^b
BMI, kg/m²	Low	22.8 ± 0.1	22.5 ± 0.1	22.5 ± 0.2	22.6 ± 0.1^ab	22.1 ± 0.1^a	22.3 ± 0.2	22.6 ± 0.1	22.4 ± 0.1^ab
	Medium	22.6 ± 0.1	22.1 ± 0.1	22.6 ± 0.1	22.4 ± 0.1^a	22.0 ± 0.1^a	22.0 ± 0.2	22.5 ± 0.1	22.2 ± 0.1^a
	High	22.8 ± 0.1	22.4 ± 0.1	22.9 ± 0.1	22.7 ± 0.1^b	22.5 ± 0.1^b	22.5 ± 0.2	22.5 ± 0.2	22.5 ± 0.1^b
WC, cm	Low	82.4 ± 0.3^a	77.2 ± 0.4^a	79.4 ± 0.4^a	79.7 ± 0.2^a	75.1 ± 0.2^a	73.0 ± 0.4^a	75.5 ± 0.3^a	74.5 ± 0.2^a
	Medium	85.8 ± 0.3^b	79.6 ± 0.4^b	83.3 ± 0.4^b	82.9 ± 0.2^b	79.4 ± 0.2^b	77.7 ± 0.5^b	80.0 ± 0.3^b	79.0 ± 0.2^b
	High	91.3 ± 0.3^c	86.1 ± 0.5^c	87.9 ± 0.5^c	88.4 ± 0.2^c	85.9 ± 0.3^c	84.3 ± 0.5^c	85.4 ± 0.5^c	85.2 ± 0.2^c
WCI, cm^0.5	Low	6.21 ± 0.02^a	5.83 ± 0.03^a	6.11 ± 0.03^a	6.05 ± 0.01^a	5.88 ± 0.02^a	5.75 ± 0.03^a	6.01 ± 0.03^a	5.88 ± 0.01^a
	Medium	6.45 ± 0.02^b	5.98 ± 0.03^b	6.40 ± 0.03^b	6.28 ± 0.02^b	6.20 ± 0.02^b	6.08 ± 0.04^b	6.35 ± 0.02^b	6.21 ± 0.01^b
	High	6.83 ± 0.02^c	6.45 ± 0.04^c	6.73 ± 0.03^c	6.67 ± 0.02^c	6.71 ± 0.02^c	6.58 ± 0.04^c	6.78 ± 0.04^c	6.69 ± 0.02^c
Education, %	Low	90.6 ± 1.6^a	62.5 ± 4.8	48.1 ± 4.7	67.0 ± 2.3	93.6 ± 1.1^a	83.1 ± 2.0^a	57.2 ± 5.8	78.0 ± 2.0^a
	Medium	83.5 ± 3.3^ab	59.6 ± 4.2	40.0 ± 5.1	61.0 ± 2.4	90.0 ± 1.5^ab	71.4 ± 5.0^ab	55.2 ± 5.2	72.2 ± 2.2^ab
	High	82.7 ± 2.4^b	64.3 ± 4.5	34.1 ± 3.8	60.4 ± 2.3	86.1 ± 1.7^b	61.6 ± 3.7^b	57.0 ± 4.9	68.2 ± 2.2^b
Smoking, %	Low	25.5 ± 2.5^a	37.1 ± 3.2	15.3 ± 3.5	26.0 ± 1.9	17.0 ± 1.9^a	18.2 ± 2.8	7.2 ± 2.5	14.2 ± 1.4^a
	Medium	28.5 ± 2.5^ab	38.5 ± 3.3	22.8 ± 4.5	29.9 ± 2.2	19.7 ± 1.8^a	16.7 ± 3.1	13.3 ± 3.7	16.6 ± 1.5^ab
	High	36.8 ± 2.7^b	32.1 ± 4.1	14.9 ± 2.4	27.9 ± 2.0	28.2 ± 2.5^b	21.6 ± 3.8	10.1 ± 2.7	20.0 ± 1.9^b
Activity, %	Low	78.1 ± 2.1^a	62.9 ± 3.9^ab	50.7 ± 3.7^a	63.9 ± 2.0^a	71.1 ± 2.6^a	58.2 ± 5.1^a	65.7 ± 4.5^a	65.0 ± 2.6^a
	Medium	67.7 ± 2.4^b	64.6 ± 3.2^a	40.3 ± 4.7^ab	57.6 ± 2.2^a	69.0 ± 2.1^a	53.6 ± 4.4^a	47.4 ± 4.3^b	56.7 ± 2.2^a
	High	60.0 ± 2.6^b	50.4 ± 4.4^b	34.4 ± 4.6^b	48.3 ± 2.4^b	60.1 ± 2.6^b	38.8 ± 4.1^b	47.9 ± 4.4^b	48.9 ± 2.3^b

Open in a new tab

Results are mean ± SE. Sample sizes given as ranges; details are presented in Supplementary Table 1A. Values with different superscript letters within the same cell and column are significantly different. Pairwise comparisons across tertiles for each variable were performed with a Bonferroni correction (P <0.05/3 = 0.0167). Education level is expressed as the percent of participants who responded as having at least a high school or General Education Development degree. Smoking is expressed as the percentage of participants who responded “yes” to the question “do you currently smoke at least a cigarette a day?” Activity levels are reported as the percent of participants engaging in a minimum of 150 min of moderate to vigorous physical activity weekly. MA, Mexican American, NHB, non-Hispanic black; NHW, non-Hispanic white; WC, waist circumference; WCI, waist circumference index.

The first analysis stage involved the development and validation of 6 WCI prediction models based on BMI and age for men and women across the NH white, NH black, and Mexican-American NHANES participants. The residual WCI (WCI_R) was then calculated for each normal-weight (BMI, ≥18.5–24.9) participant as the difference between their actual and predicted WCI value. People with a high WCI_R thus had a relatively large WC for their sex, race and Hispanic origin, BMI, and age.

Normal-weight participants within each sex and race and Hispanic origin group were next allocated into tertiles based on WCI_R: actual value greater than, like, or smaller than predicted for their BMI and age. The mean (±SE) values for demographic variables, body composition, blood studies, and blood pressure were then compared across the high and low WCI_R tertiles. The algorithm for identifying people in the 3 WCI_R tertiles is summarized in Figure 1.

Data sources/study population

The baseline characteristics of the 6 NHANES sex and race/Hispanic origin groups are summarized in Supplementary Methods I, Table 2 for all participants and in Supplementary Table 3 for the normal-weight group defined by BMI. Overall, there were 17,359 total participants with 5594 classified as having a normal weight after exclusions. Of the normal-weight group, there were 1399 NH white, 725 NH black, and 675 Mexican-American men. Similarly, there were 1689 NH white, 481 NH black, and 625 Mexican-American women. The NHANES protocol was approved by the Institutional Review Board of the National Center for Health Statistics, CDC. All participants provided written informed consent.

TABLE 2.

Body composition of the waist circumference index residual tertiles¹

		Men				Women
Characteristic	Tertile	NHW	NHB	MA	Total	NHW	NHB	MA	Total
N		424–446	232–238	216–222	878–905	524–539	154–159	204–206	882–904
Total Fat	Low	20.4 ± 0.2^a	17.2 ± 0.3^a	20.3 ± 0.4^a	19.3 ± 0.2^a	33.3 ± 0.2^a	31.5 ± 0.6^ab	34.7 ± 0.4	33.2 ± 0.3^a
	Medium	22.2 ± 0.2^b	18.5 ± 0.3^b	22.5 ± 0.3^b	21.1 ± 0.2^b	33.9 ± 0.2^a	31.8 ± 0.4^a	35.1 ± 0.5	33.6 ± 0.2^a
	High	25.3 ± 0.2^c	22.6 ± 0.4^c	24.6 ± 0.3^c	24.1 ± 0.2^c	35.3 ± 0.3^b	33.3 ± 0.5^b	35.7 ± 0.4	34.8 ± 0.2^b
Trunk Fat	Low	9.2 ± 0.1^a	6.9 ± 0.1^a	9.2 ± 0.2^a	8.5 ± 0.1^a	13.3 ± 0.2^a	11.6 ± 0.3^a	15.1 ± 0.3^a	13.4 ± 0.2^a
	Medium	10.5 ± 0.1^b	7.6 ± 0.2^b	10.8 ± 0.2^b	9.6 ± 0.1^b	14.3 ± 0.2^b	12.2 ± 0.3^a	15.7 ± 0.4^ab	14.1 ± 0.2^b
	High	12.4 ± 0.2^c	10.2 ± 0.3^c	12.1 ± 0.2^c	11.6 ± 0.1^c	16.1 ± 0.2^c	14.3 ± 0.3^b	17.1 ± 0.4^b	15.8 ± 0.2^c
Leg Fat	Low	7.2 ± 0.1^a	6.5 ± 0.2^a	7.0 ± 0.2^a	6.9 ± 0.1^a	14.5 ± 0.1^a	14.5 ± 0.3^a	13.8 ± 0.2^a	14.3 ± 0.1^a
	Medium	7.6 ± 0.1^b	7.0 ± 0.1^a	7.5 ± 0.1^a	7.4 ± 0.1^b	14.0 ± 0.1^b	14.1 ± 0.2^ab	13.6 ± 0.3^ab	13.9 ± 0.1^a
	High	8.5 ± 0.1^c	8.2 ± 0.2^b	8.0 ± 0.1^b	8.2 ± 0.1^c	13.4 ± 0.1^c	13.4 ± 0.2^b	12.8 ± 0.2^b	13.2 ± 0.1^b
ALST	Low	34.2 ± 0.2^a	37.5 ± 0.2^a	34.0 ± 0.2^a	35.2 ± 0.1^a	26.2 ± 0.1^a	28.9 ± 0.3^a	25.4 ± 0.2^a	26.8 ± 0.1^a
	Medium	33.1 ± 0.1^b	36.4 ± 0.2^b	32.6 ± 0.2^b	34.0 ± 0.1^b	26.0 ± 0.1^a	28.5 ± 0.2^a	25.0 ± 0.3^ab	26.5 ± 0.1^a
	High	31.2 ± 0.1^c	34.0 ± 0.3^c	31.4 ± 0.2^c	32.2 ± 0.1^c	25.2 ± 0.2^b	27.5 ± 0.3^b	24.3 ± 0.2^b	25.7 ± 0.1^b
BMC	Low	3.75 ± 0.02^a	4.17 ± 0.03^a	3.74 ± 0.04^a	3.89 ± 0.02^a	3.46 ± 0.02^a	3.67 ± 0.05^ab	3.48 ± 0.03	3.54 ± 0.02^a
	Medium	3.59 ± 0.02^b	4.08 ± 0.04^a	3.58 ± 0.04^b	3.75 ± 0.02^b	3.44 ± 0.02^a	3.68 ± 0.04^a	3.42 ± 0.04	3.52 ± 0.02^a
	High	3.44 ± 0.02^c	3.82 ± 0.04^b	3.47 ± 0.03^b	3.58 ± 0.02^c	3.27 ± 0.03^b	3.51 ± 0.05^b	3.37 ± 0.07	3.38 ± 0.03^b

Open in a new tab

Results are mean ± SE percentage of body weight. Sample sizes given as ranges; details are presented in Supplementary Table 1B. Values with different superscript letters within the same cell and column are significantly different. Pairwise comparisons across tertiles for each variable were performed with a Bonferroni correction (P <0.05/3 = 0.0167). Separate models for each body composition variable were fitted for each of the 5 NHANES imputed data sets. ALST, appendicular lean soft tissue; BMC, bone mineral content; MA, Mexican American; NHB, non-Hispanic black; NHW, non-Hispanic white; WC, waist circumference index; WCI, waist circumference index.

Three demographic and behavioral variables were evaluated in the current study, education, smoking status, and activity levels (Supplementary Methods II). Education level is reported as the percentage of participants who responded as having at least a high school or General Education Development degree (10). Smoking is reported as the percentage of participants who responded “yes” to the question “do you currently smoke at least a cigarette a day?” (11). Activity levels are reported as the percentage of participants engaging in a minimum of 150 minutes of moderate to vigorous physical activity weekly (12).

Measurements

Anthropometry

Body weight, height, and WC were measured according to standard NHANES procedures and are summarized in Supplementary Methods II (13, 14). BMI was calculated as body weight/height² (kg/m²) and WCI as WC/height^0.5 (cm/cm^0.5 or cm^0.5).

Body composition

Each participant completed a whole-body DXA scan (15) as outlined in Supplementary Methods II. The following were measured and reported as a percentage of body weight in the current study: total body, trunk, and leg fat mass; appendicular lean soft tissue (ALST) mass (sum of lean soft tissue mass in the arms and legs); and total bone mineral content. The ALST component is a surrogate measure of total body skeletal muscle mass (16) and bone mineral content is a measure of total body bone mass.

Avatar image development

Three-dimensional avatars of men and women were generated as an aid to visualizing the typical body shape of people in the high and low WCI_R tertiles using image prediction models as previously reported (17) and detailed in Supplementary Methods II.

Health risk factors

Blood chemistries and blood pressure were measured according to standard NHANES procedures (18) as reported in Supplementary Methods III. Participants who were fasting overnight (≥8 h) and were not taking medications for diabetes were included in plasma glucose, insulin, and HOMA-IR analyses. HOMA-IR was calculated as fasting insulin × fasting glucose/405 (18). Participants who were fasting overnight and were not taking lipid-lowering drugs were included in the cholesterol (total, HDL, and LDL) and triglyceride analyses. Participants who were not taking antihypertensive medications were included in the analyses of systolic and diastolic blood pressure. Blood pressure was measured using standard procedures as outlined in Supplementary Methods III and in the NHANES operations manual (19).

Statistical analyses

The NHANES applies a complex multistage sampling strategy since random sampling is not feasible from the entire US population that incudes subgroups of people who are institutionalized. Therefore, to increase the representativeness of the sample at the individual subject level, probability sampling weights are applied considering survey nonresponse, oversampling, poststratification, and sampling errors. More details on the NHANES sampling strategy can be found at the CDC website (20, 21).

All parameter estimates were obtained using R (version 3.6.1; R Foundation for Statistical Computing) and its associated “survey” package (22, 23) to yield nationally representative parameter estimates while accounting for the complex, multistage probability design of NHANES. The R survey package is 1 of the software platforms recommended for analyzing NHANES data. All models included appropriate sample weights to account for noncoverage, nonresponse, and oversampling of some groups. Sample strata and weights were adjusted in the same manner as in the SAS SURVERYMEANS and SURVEYREG procedures (23). SEs were derived using Taylor series linearization.

Descriptive statistics characterizing the study cohort included categorical variables summarized as frequencies and continuous variables as means and SEs. The WCI prediction models were fitted using regression analysis in a complex sample framework. After restricting the sample to normal-weight participants (BMI, ≥18.5–24.9), additional regression models were fitted in a complex samples framework to evaluate heterogeneity across WCI_R tertiles within each sex by race/ethnicity combination for a variety of demographic, anthropometric, body composition, and cardiometabolic variables. Models included WCI tertile designation (high, medium, or low), race/Hispanic origin (NH white, NH black, Mexican American), and their interaction as independent predictors. Graphical diagnostics of model residuals for plasma triglyceride analyses demonstrated evidence of heterogeneous variance. As such, a variance stabilizing transformation (natural logarithm) was employed prior to triglyceride-related analyses and all parameter estimates were later back transformed to their original scale to facilitate interpretation of results.

All analyses were stratified by sex and race/Hispanic origin combinations. For each variable, pairwise comparisons across tertiles were performed with a Bonferroni correction (P <0.05/3 = 0.0167).

Specific to NHANES body composition analyses using DXA data, separate models were fitted for each of the 5 imputed data sets. Location and variance estimates were averaged across models while adhering to Rubin's rules (24) using the “MIcombine” function. Critical t-statistic values were derived using Barnard and Rubin's method (24) for penalizing Df as a function of the data's missingness. The complete-data Df used in this derivation was 59 (number of primary sampling units – number of sampling strata).

Results

WCI prediction models

BMI was strongly correlated with WCI for total men and women (R²s, 0.839 and 0.824, both P <0.001) and for each of the 6 sex and race/Hispanic origin groups (Supplementary Figure 2). Age added along with BMI to the WCI prediction models for each race and Hispanic origin group (Supplementary Table 4) with R²s ranging from 0.86 in Mexican-American women to 0.93 in NH black men; all models were significant at P <0.001. The prediction equations shown in Supplementary Table 4 were used to calculate WCI_R; participants were then sorted into the high, medium, and low WCI_R shape tertiles.

Characteristics of the study population

Demographic

The demographic characteristics across the high, medium, and low WCI_R tertiles are summarized for men and women in the 3 race/Hispanic origin groups in Table 1. The overall sample had an average age of ∼40 y with NH white men and women older than their NH black and Mexican-American counterparts, in that order. People in the high WCI_R tertile were, on average, ∼1–2 cm taller than those in the corresponding low tertile. Body weight paralleled height across the tertiles such that average BMI was similar at ∼22 to 23 in all 6 sex and race/Hispanic origin groups. As expected, WC decreased from the high to low WCI_R tertiles. For example, among NH white men with the same age and BMI, the mean WC decreased from 91.3 cm in the high tertile to 85.8 cm and 82.4 cm in the medium and low WCI_R tertiles, respectively.

The percentage of people with General Education Development degrees decreased from NH white to NH black and Mexican-American NHANES participants. The mean education level was lower in the high versus low WCI_R tertile groups for all 6 sex and race/Hispanic origin groups, although statistical significance was inconsistent.

The percentage of participants who smoked ≥1 cigarette a day was highest in NH black men (∼32–37%) and lowest in Mexican-American women (∼7–13%). The pattern of smoking across WCI_R tertiles was like that observed for education, with people in the high tertiles tending to have higher mean smoking percentages than those in the low WCI_R tertiles, although between-group statistical significance was inconsistent. Women in the high WCI_R group had significantly less education and higher smoking rates compared with their low tertile counterparts.

The percentage of participants engaging in a minimum of 150 min of moderate to vigorous physical activity weekly was highest among the NH white men and women with lower levels in NH black and Mexican-American men and women. A consistent across-WCI_R tertile gradient in activity was present among men and women in the 3 race/Hispanic-origin groups: people in the low WCI_R tertile had the highest reported activity levels and those in the high tertile had the lowest activity levels. The differences in activity levels between the high and low tertile groups were all significant, including for total men and women, except for NH black men.

Body composition

The results of body composition estimates across the high, medium, and low WCI_R tertiles are summarized for men and women in the 3 race/Hispanic origin groups in Table 2. Total body and trunk percent fat decreased from high to low tertiles in all 6 sex and race/Hispanic origin groups. The differences in total and trunk percent fat between the high and low tertile groups were all significant except for total percent fat in the NH black and Mexican-American women. A similar pattern was present for leg percent fat in the 3 groups of men, although the opposite trend was present in women: leg percent fat decreased in the groups of women from low to high WCI_R tertiles. The differences in leg percent fat between the high and low tertile groups of women were all significant. Men thus had greater total and regional adiposity in the high WCI_R tertile compared with their low tertile counterparts. Women in the high WCI_R tertile also had greater relative total and trunk adiposity but lower leg adiposity than their low WCI_R counterparts.

All 6 sex and race/Hispanic origin groups and all men and all women in the high WCI_R tertiles had significantly smaller percentages of ALST and bone mineral content than those in the low tertile except for bone mineral content in the NH black and Mexican-American women whose values were in the same direction but were not significant. Not including the aforementioned exceptions, people with a high WCI_R were thus characterized by smaller proportions of their body weight as skeletal muscle and bone than those with low residual WCIs.

The main morphologic and body composition features of high- and low-tertile representative NH white men and women are shown as the digital avatars in Figure 2. The mean age, weight, height, and WC values for the 4 respective groups were used in generating the images with manifold regression analysis (8, 17). Notable high-low tertile differences in body shape and composition, as shown in the bar graphs, are present with the hypothetical people all the same race/Hispanic origin, age, and BMI (∼22).

Three-dimensional representation of non-Hispanic white men (upper panel) and women (lower panel) who are in the low and high WCI_R tertiles and a bar graph showing the high-low differences (Δ, %) in fat and lean mass components. The average values for age, body weight, height, and WC were used from the men and women in the low and high WCI_R tertiles to produce these avatars with manifold regression analysis as previously reported (17). The main feature depicted is the relatively large difference in central body volume even though all 4 avatars have the same BMI of ∼22 kg/m² and an age of ∼40 y. The directional differences in body composition, calculated as the average high-low percentage for each component between the high and low WCI_R groups, are shown in the 2 panels; positive differences are in red and negative differences are in blue. Skeletal muscle and bone mass are the anatomic representations of measured appendicular lean soft tissue mass and bone mineral content, respectively. All high-low body composition differences are statistically significant. ALST, appendicular lean soft tissue mass; BMC, bone mineral content; SM, skeletal muscle; WCI_R, waist circumference index residual.

Health risk biomarkers

The health risk biomarkers organized according to WCI_R tertiles are summarized in Table 3. About 10% of participants overall were removed from the analyses because they were not fasting or were taking medications that influenced metabolic markers, lipid concentrations, or blood pressure (Supplementary Table 1C and D). Plasma triglycerides tended to be lower and HDL cholesterol concentrations tended to be higher in the NH black participants compared with the other 2 race/ethnic groups. Systolic blood pressures were also higher in the NH black men and women.

TABLE 3.

Health risk factors of the waist circumference index residual tertiles¹

		Men				Women
Characteristic	Tertile	NHW	NHB	MA	Total	NHW	NHB	MA	Total
N²		145–411	66–225	77–217	290–578	174–489	51–139	71–194	305–822
Glucose, mg/dL	Low	94.9 ± 0.9	92.0 ± 1.1^ab	95 ± 1.0	94.0 ± 0.6^a	90.8 ± 0.5	86.3 ± 1.2	93.3 ± 1.3	90.1 ± 0.6
	Medium	96.8 ± 1.4	90.6 ± 0.9^a	101.1 ± 4.4	96.1 ± 1.5^ab	90.3 ± 0.6	86.5 ± 1.1	91.8 ± 1.6	89.5 ± 0.6
	High	96.4 ± 1.4	95.9 ± 1.8^b	101.9 ± 3.5	98.1 ± 1.4^b	92.1 ± 0.8	93.0 ± 3.5	93.0 ± 1.2	92.7 ± 1.2
Insulin, uU/mL	Low	4.5 ± 0.2^a	4.5 ± 0.3^a	5.8 ± 0.5^ab	4.9 ± 0.2^a	4.5 ± 0.2^a	4.9 ± 0.4^a	6.1 ± 0.4^a	5.2 ± 0.2^a
	Medium	5.2 ± 0.2^b	5.3 ± 0.3^ab	5.3 ± 0.3^a	5.3 ± 0.1^a	4.9 ± 0.2^a	5.7 ± 0.5^ab	6.0 ± 0.5^ab	5.5 ± 0.3^a
	High	6.0 ± 0.3^b	6.1 ± 0.3^b	7.3 ± 0.6^b	6.5 ± 0.3^b	5.9 ± 0.3^b	6.9 ± 0.6^b	7.7 ± 0.6^b	6.8 ± 0.3^b
HOMA-IR	Low	1.08 ± 0.06^a	1.02 ± 0.08^a	1.38 ± 0.13^ab	1.16 ± 0.05^a	1.04 ± 0.06^a	1.06 ± 0.09^a	1.44 ± 0.12	1.18 ± 0.05^a
	Medium	1.26 ± 0.06^ab	1.21 ± 0.07^ab	1.33 ± 0.08^a	1.27 ± 0.04^a	1.10 ± 0.05^a	1.26 ± 0.13^ab	1.36 ± 0.13	1.24 ± 0.07^a
	High	1.45 ± 0.07^b	1.46 ± 0.09^b	1.86 ± 0.16^b	1.59 ± 0.07^b	1.40 ± 0.09^b	1.65 ± 0.19^b	1.79 ± 0.14	1.61 ± 0.08^b
TG,³ mg/dL	Low	89.4 ± 4.6^a	71.1 ± 4.4	103.3 ± 8.4	86.9 ± 3.3^a	86.1 ± 3.0^a	60.3 ± 5.5^ab	102.7 ± 6.5^a	81.1 ± 3.1^a
	Medium	100.7 ± 6.6^ab	74.0 ± 4.7	88.8 ± 6.1	87.2 ± 3.3^a	93.2 ± 3.4^ab	58.9 ± 4.0^a	78.3 ± 5.6^b	75.5 ± 2.4^a
	High	114.2 ± 6.4^c	85.9 ± 8.5	108.2 ± 9.7	102.0 ± 5.5^b	111.3 ± 7.7^b	79.2 ± 5.4^b	103.2 ± 6.7^a	96.9 ± 3.7^b
Total-C, mg/dL	Low	188.0 ± 3.1^a	185.6 ± 3.9	177.0 ± 5.4^a	183.5 ± 2.3^a	203.2 ± 2.4	185.3 ± 4.6	187.0 ± 4.4	191.8 ± 1.9^a
	Medium	190.4 ± 3.6^ab	180.3 ± 5.3	188.1 ± 5.7^ab	186.3 ± 3.1^ab	195.7 ± 3.1	171.1 ± 6.9	176.5 ± 3.2	181.1 ± 2.9^b
	High	199.8 ± 3.4^b	178.9 ± 5.4	204.3 ± 6.3^b	194.3 ± 3.3^b	195.9 ± 5.1	183.4 ± 9.1	181.9 ± 3.7	187.1 ± 3.2^ab
LDL-C, mg/dL	Low	116.1 ± 3.6	107.2 ± 3.3	101.9 ± 4.4^a	108.4 ± 2.2	115.8 ± 2.7	107.0 ± 6.1	114.4 ± 4.0^a	112.4 ± 2.2
	Medium	115.6 ± 3.1	109.3 ± 5.3	117.9 ± 6.1^b	114.3 ± 3.0	112.4 ± 2.9	102.5 ± 7.4	99.9 ± 3.7^b	104.9 ± 2.9
	High	120.8 ± 3.9	105.0 ± 6.3	121.0 ± 5.1^b	115.6 ± 3.0	113.7 ± 4.4	105.8 ± 7.8	107.1 ± 5.0^ab	108.8 ± 3.3
HDL-C, mg/dL	Low	54.4 ± 1.2^a	64.6 ± 2.3^a	54.9 ± 1.7	58.0 ± 1.2^a	68.4 ± 1.1^a	69.1 ± 2.7^a	59.1 ± 1.4^ab	65.5 ± 1.0^a
	Medium	50.9 ± 1.2^b	57.2 ± 2.3^ab	51.7 ± 1.7	53.3 ± 1.0^b	60.8 ± 1.2^b	58.9 ± 2.0^b	61.3 ± 1.9^a	60.3 ± 1.1^b
	High	51.8 ± 1.9^ab	56.0 ± 2.3^b	51.0 ± 1.3	52.9 ± 1.1^b	59.1 ± 1.2^b	60.1 ± 3.4^ab	54.2 ± 1.7^b	57.8 ± 1.3^b
SBP, mmHg	Low	118.8 ± 0.7	122.7 ± 1.2	115.0 ± 1.0	118.8 ± 0.6	115.7 ± 0.7^a	115.1 ± 1.7	111.1 ± 1.2	114.0 ± 0.7^a
	Medium	117.2 ± 0.7	119.8 ± 1.0	115.9 ± 1.2	117.6 ± 0.6	112.8 ± 0.8^b	115.0 ± 1.8	108.4 ± 0.9	112.1 ± 0.8^b
	High	120.3 ± 0.8	122.1 ± 1.1	115.9 ± 0.8	119.4 ± 0.6	114.4 ± 0.8^a	117.9 ± 1.6	113.1 ± 1.8	115.1 ± 0.8^a
DBP, mmHg	Low	67.9 ± 0.7^a	72.0 ± 0.9	66.7 ± 0.7	68.8 ± 0.5	68.5 ± 0.5	66.9 ± 1.2	66.5 ± 1.0	67.3 ± 0.5
	Medium	69.6 ± 0.5^ab	69.7 ± 1.0	65.0 ± 1.0	68.1 ± 0.5	68.5 ± 0.6	69.8 ± 1.6	66.4 ± 0.6	68.2 ± 0.7
	High	71.5 ± 0.6^b	72.3 ± 0.9	66.0 ± 1.1	69.9 ± 0.5	69.3 ± 0.5	68.5 ± 1.3	67.7 ± 1.2	68.5 ± 0.6

Open in a new tab

Results are mean ± SE. Values with different superscript letters within the same cell and column are significantly different. Pairwise comparisons across tertiles for each variable were performed with a Bonferroni correction (P <0.05/3 = 0.0167). Participants were removed from the analyses who were not fasting or who were taking medications known to moderate the specific evaluated outcome measure.

Sample sizes given as ranges; details are presented in Supplementary Table 1c.

Values are back-transformed geometric means and SEs from analytical models employing a natural logarithm transformation. C, cholesterol; DBP, diastolic blood pressure; MA, Mexican American; NHB, non-Hispanic black; NHW, non-Hispanic white; SBP, systolic blood pressure; TG, triglycerides; WC, waist circumference index; WCI, waist circumference index.

The mean plasma concentrations of glucose, insulin, and HOMA-IR were higher in the high versus low WCI_R tertiles; differences were significant for all men and all women except for plasma glucose in the women. The mean plasma triglyceride concentrations were higher in the high versus low WCI_R tertiles; differences were significant for NH white men and women and for total men and women. The mean plasma HDL cholesterol concentrations were lower in the high versus low WCI_R tertiles; differences were significant for NH black men and NH white women and for total men and women. No consistent WCI_R patterns were present for total and LDL cholesterol or for systolic and diastolic blood pressure.

To establish if these health risk biomarker patterns differ after adjusting for physical activity levels and smoking status, the analyses were re-run adjusting for both behavioral measures (Supplementary Table 5). The patterns linking WCI_R with health risk biomarkers shown in Table 3 were largely unchanged as summarized in Supplementary Table 6.

Discussion

An algorithm combining BMI and WCI estimates was used in the current study to identify normal-weight NHANES participants who were at increased risk of developing cardiovascular and metabolic diseases. Using this approach, people in the high WCI_R tertile relative to those in the low tertile, had a remarkably distinct phenotype: low activity levels, increased total body and trunk percent fat, low percent skeletal muscle (i.e. ALST) and bone (i.e. bone mineral content), and an array of unfavorably directed risk markers including elevated plasma insulin and triglycerides, HOMA-IR, and lowered plasma HDL cholesterol concentrations. Moreover, there was a sexual dimorphism in this phenotype, with women in the high WCI_R tertile having relatively less leg fat than their low-WCI_R counterparts; the opposite effect was present in men. Adipose tissue in the thigh region is inversely related to metabolic and cardiovascular disease risk (25), adding to the other risk factors noted in women who were in the high WCI_R tertile. The pattern of demographic features of people with a high WCI_R is summarized in Figure 3.

Pattern of key demographic, body composition, and blood chemistry observations. Boxes in red represent the presence of significant differences between the high and low WCI_R tertiles. Boxes in orange represent the same directional effects as their adjacent red counterparts but the between-group differences were not statistically significant. The direction of high-low (H-L) differences (Δ) is shown in the righthand column. The analyses used in generating the figure are presented in Tables 1–3. ALST, appendicular lean soft tissue; BMC, bone mineral content; HDL-C, plasma HDL cholesterol; NHB, non-Hispanic black; NHW, non-Hispanic white; TG, plasma triglyceride; WCI_R, waist circumference index residual.

At the core of the developed algorithm is WCI, WC adjusted for between-individual differences in height (7). Absolute values of WC, notably single sex-specific cut-points, are often used in identifying patients with metabolic syndrome (6) or who are at risk of developing chronic diseases secondary to overweight and obesity (4). However, as confirmed in the current study, the body-size measure WC is strongly correlated with BMI (Supplementary Figure 2) and age, in addition to height (7), with differences present even after controlling for these 3 variables across sex and race/Hispanic origin groups. Notably, WC adjusted for height as WCI varies independently with BMI and age even within the normal weight range. These factors were controlled for in the current study and presumably helped to identify the distinct phenotypes associated with a high WCI_R.

The present study refines and extends many earlier studies, with smaller samples or limited detection methods that reported components of these phenotypes (Supplementary Literature Review IV). In 1981 and later in 1982, Ruderman and colleagues (26, 27) first described obesity as a spectrum, with some people who are normal weight, at the lower end of the adiposity distribution, having enlarged fat cells and raised plasma insulin concentrations. These people tended to be inactive and had an unfavorable diet. Two decades later Ruderman et al. (28) revisited their concept of the “metabolically obese, normal-weight individual” and further refined their “insulin resistance syndrome” phenotype to include central fat distribution. By then, a parallel knowledge-base had formed around the metabolic significance of central fat distribution as first reported by Vague in 1952 (29) and later by Krotkiewski et al. in 1983 (30). Others have now expanded the metabolically obese, normal-weight phenotype to include increased visceral fat (31–34), insulin resistance, high concentrations of inflammatory cytokines, and dyslipidemia (35–38), different strategies to identify people at-risk (e.g. ranking people who are normal weight by BMI, waist to hip circumference ratio, or WC to height ratio [37, 39–42]), and introducing new nomenclature and acronyms such as hypertriglyceridemic waist, NOW (normal-weight obesity), and TOFI (thin-on-the-outside fat-on-the-inside [43]). The current study consolidates these previous findings and provides insights into potential underlying interacting mechanisms, notably low activity levels and, to a lesser extent, other demographic (education) and behavioral (smoking) characteristics. Genetic (44) and dietary (26) effects may also contribute to the observed phenotypes.

An important finding of the present study is that the identified high WCI_R body shape and composition phenotype in people with a “normal” BMI, is identical to that observed in people who are overweight and obese (7). This phenotype goes well beyond a “large” WC, as is often inferred, and includes greater relative amounts of total body fat, sex-specific regional fat distribution patterns, and smaller relative amounts of skeletal muscle and bone even after controlling for BMI and age. These observations reveal the remarkable heterogeneity in human body shape and composition as well as their relations to chronic disease risk factors.

Despite recommendations by national, international, and professional organizations to include WC measurements in patient evaluations (45–47), clinical uptake has been slow. Removing clothing, identifying the endorsed measurement site, the need for observer training and skill, and a lack of agreed upon normative values, currently contribute to the limited clinical use of WC measurements, particularly in people who are within the normal BMI range. Nevertheless, the distinct phenotype with high WCI_R in the current study identifies people at risk of developing not only cardiovascular and metabolic diseases, but conditions associated with aging such as sarcopenia (48) and osteoporosis (49), reflecting low skeletal muscle and bone mass, respectively. Moreover, low activity levels, and even smoking, are modifiable risk factors amenable to behavioral management. With an increasing clinical focus on disease prevention (50), people with the distinct high WCI_R phenotype observed in the current study are candidates for close scrutiny of their behavioral and metabolic characteristics. Moreover, the current study identifies linkages between behavioral and cardiometabolic risk factors that can be tested for causality in future prospective proof-of-concept studies.

Given the large number of people across the USA that likely would be classified as having a high WCI_R and the nuances of accurately measuring WC (5, 51), are there other diagnostic approaches that should be considered? Local WC norms developed by highly trained teams might reduce technical measurement errors. Recently developed 3-dimensional optical imaging methods that can be applied in clinical settings reduce human circumference measurement error, generate hundreds of different anthropometric estimates, and offer a new opportunity to expand detection of increased risk body shape phenotypes (17, 52). Advances in deep learning and other similar mathematical methods offer a new opportunity to identify people at increased chronic disease risk using combinations of patient history and body measurements beyond those examined in the current report. Lastly, the current study reveals related demographic, body shape, body composition, and blood marker features of those at an increased risk of developing chronic diseases.

Limitations

The current study was not designed to establish the genetic/molecular mechanisms that underlie the metabolically obese, normal-weight state. Although some of these topics have been explored in previous studies (5), the BMI-WCI approach described in our report can facilitate patient identification for clinical management and future studies. The effect sizes observed for some comparisons in this study were relatively small, a limitation partially overcome by the large NHANES normal-weight sample, particularly the NH white men and women (total ∼3000); less statistical power was present for the smaller NH black and Mexican-American (each ∼1300) groups. Sample sizes were further reduced for blood studies after nonfasting participants and those taking relevant medications were removed from analyses. Recent studies highlight the potential advantages of anthropometric measurements other than WC, such as neck circumference (53), in identifying people who are at risk of developing metabolic complications. Neck circumference is less prone to variability caused by factors such as meal ingestion than is WC (51). The NHANES evaluations were conducted both fasting in the morning and in the early afternoon. We assume that any meal-related effects on WC are randomly distributed across the tertiles. Nevertheless, body size measures such as neck circumference alone or in combination with WC should be evaluated in future studies as in the current report. Limited demographic and behavioral measures were evaluated in the current study and our findings highlight the need for in-depth analyses of these types of outcomes in future studies. A final concern is that within-race/ethnic and sex group across-tertile age heterogeneity was low and caution is therefore in order when extrapolating the current study findings to the general population.

Conclusions

The current study establishes that 3 clinical measurements, body weight, height, and WC; 2 calculated indices therefrom (BMI and WCI); along with a person's sex, age, and race-ethnicity designation can be used to estimate their cardiovascular and metabolic disease risk. Moreover, those individuals in the high WCI_R category are likely inactive and may have other predisposing features that are modifiable through behavioral approaches. Although the characteristics of these phenotypes are well recognized among people who are overweight and obese (7), the current study establishes a practical clinical detection method that identifies those at risk even though they have a “normal” body weight.

Supplementary Material

nqaa194_Supplemental_File

Click here for additional data file.^{(304KB, docx)}

Acknowledgments

The authors’ contributions were as follows— AS, SK, MH, JSh, JSc, and SBH: concept and design; AS, JSc, SY, MH, MW, and SBH: acquisition, analysis, or interpretation of data; AS, SK, MH, JSc, and SBH: drafting of the manuscript; all authors: critical revision of the manuscript for important intellectual content; AS, JSc, SY, and MH: statistical analysis; JSh and SBH: obtained funding; JSh and SBH: administrative, technical, or material support; JSh and SBH: supervision; and all authors: read and approved the final manuscript. SBH is on the Tanita Medical Advisory Board; all other authors report no conflicts of interest.

Notes

This work was partially supported by NIH NORC Center Grants P30DK072476, Pennington/Louisiana, P30DK040561, Harvard, and R01DK109008, Shape UP! Adults.

Data described in the manuscript and code book are publicly and freely available without restriction at https://wwwn.cdc.gov/nchs/nhanes/Default.aspx.

Supplemental Tables 1–6, Supplementary Methods I–III, Supplemental Figures 1 and 2, and the Supplementary Literature Review are available from the “Supplementary data” link in the online posting of the article and from the same link in the online table of contents at https://academic.oup.com/ajcn/.

Abbreviations used: ALST, appendicular lean soft tissue; MA, Mexican American; NH, non-Hispanic; WC, waist circumference; WCI, waist circumference index; WCI_R waist circumference index residual.

Contributor Information

Abishek Stanley, Pennington Biomedical Research Center, Louisiana State University System, Baton Rouge, LA, USA.

John Schuna, College of Public Health and Human Sciences, Oregon State University, Corvallis, OR, USA.

Shengping Yang, Pennington Biomedical Research Center, Louisiana State University System, Baton Rouge, LA, USA.

Samantha Kennedy, Pennington Biomedical Research Center, Louisiana State University System, Baton Rouge, LA, USA.

Moonseong Heo, Department of Public Health Sciences, Clemson University, Clemson, South Carolina, SC, USA.

Michael Wong, University of Hawaii Cancer Center, Honolulu, HI, USA.

John Shepherd, University of Hawaii Cancer Center, Honolulu, HI, USA.

Steven B Heymsfield, Pennington Biomedical Research Center, Louisiana State University System, Baton Rouge, LA, USA.

References

1. Heymsfield SB, Cefalu WT. Does body mass index adequately convey a patient's mortality risk?. JAMA. 2013;309(1):87–8. [DOI] [PubMed] [Google Scholar]
2. Shah NR, Braverman ER. Measuring adiposity in patients: the utility of body mass index (BMI), percent body fat, and leptin. PLoS One. 2012;7(4):e33308. [DOI] [PMC free article] [PubMed] [Google Scholar]
3. Aune D, Sen A, Prasad M, Norat T, Janszky I, Tonstad S, Romundstad P, Vatten LJ. BMI and all cause mortality: systematic review and non-linear dose-response meta-analysis of 230 cohort studies with 3.74 million deaths among 30.3 million participants. BMJ. 2016;353:i2156. [DOI] [PMC free article] [PubMed] [Google Scholar]
4. Jensen MD, Ryan DH, Apovian CM, Ard JD, Comuzzie AG, Donato KA, Hu FB, Hubbard VS, Jakicic JM, Kushner RF et al. 2013 AHA/ACC/TOS guideline for the management of overweight and obesity in adults: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines and The Obesity Society. Circulation. 2014;129(25 Suppl 2):S102–38. [DOI] [PMC free article] [PubMed] [Google Scholar]
5. Ross R, Neeland IJ, Yamashita S, Shai I, Seidell J, Magni P, Santos RD, Arsenault B, Cuevas A, Hu FB et al. WC as a vital sign in clinical practice: a Consensus Statement from the IAS and ICCR Working Group on Visceral Obesity. Nat Rev Endocrinol. 2020;16(3):177–89. [DOI] [PMC free article] [PubMed] [Google Scholar]
6. Eckel RH, Grundy SM, Zimmet PZ. The metabolic syndrome. Lancet. 2005;365(9468):1415–28. [DOI] [PubMed] [Google Scholar]
7. Hwaung P, Heo M, Kennedy S, Hong S, Thomas DM, Shepherd J, Heymsfield SB. Optimum WC-height indices for evaluating adult adiposity: an analytic review. Obes Rev. 2020;21(1):e12947. [DOI] [PubMed] [Google Scholar]
8. Heymsfield SB, Peterson CM, Thomas DM, Heo M, Schuna JM Jr. Why are there race/ethnic differences in adult body mass index-adiposity relationships? A quantitative critical review. Obes Rev. 2016;17(3):262–75. [DOI] [PMC free article] [PubMed] [Google Scholar]
9. Heymsfield SB, Peterson CM, Thomas DM, Heo M, Schuna JM Jr., Hong S, Choi W. Scaling of adult body weight to height across sex and race/ethnic groups: relevance to BMI. Am J Clin Nutr. 2014;100(6):1455–61. [DOI] [PMC free article] [PubMed] [Google Scholar]
10. Centers for Disease Control and Prevention (CDC). Version current January 2015. [Internet] Available from: https://wwwn.cdc.gov/nchs/nhanes/2011-2012/DEMO_G.htm#DMDEDUC2 (accessed 15 May, 2020).
11. Centers for Disease Control and Prevention (CDC). Version current September 2017. [Internet] Available from: https://wwwn.cdc.gov/Nchs/Nhanes/2015-2016/SMQ_I.htm (accessed 15 May, 2020).
12. Piercy KL, Troiano RP, Ballard RM, Carlson SA, Fulton JE, Galuska DA, George SM, Olson RD. The physical activity guidelines for Americans. JAMA. 2018;320(19):2020–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
13. Centers for Disease Control and Prevention (CDC). Version current January 2007. [Internet] Available from: https://www.cdc.gov/nchs/data/nhanes/nhanes_07_08/manual_an.pdf (accessed 15 May, 2020).
14. Centers for Disease Control and Prevention (CDC), National Center for Health Statistics (NCHS). Version current 21 February, 2020. [Internet] Available from: https://wwwn.cdc.gov/nchs/nhanes/nhanes3/AnthropometricVideos.aspx (accessed 15 May, 2020).
15. Centers for Disease Control and Prevention (CDC), National Center for Health Statistics (NCHS). Version current 21 February, 2020. [Internet] Available from: https://wwwn.cdc.gov/nchs/nhanes/dxa/dxa.aspx (accessed 15 May, 2020).
16. Kim J, Wang Z, Heymsfield SB, Baumgartner RN, Gallagher D. Total-body skeletal muscle mass: estimation by a new dual-energy X-ray absorptiometry method. Am J Clin Nutr. 2002;76(2):378–83. [DOI] [PubMed] [Google Scholar]
17. Ng BK, Sommer MJ, Wong MC, Pagano I, Nie Y, Fan B, Kennedy S, Bourgeois B, Kelly N, Liu YE et al. Detailed 3-dimensional body shape features predict body composition, blood metabolites, and functional strength: the Shape Up! studies. Am J Clin Nutr. 2019;110(6):1316–26. [DOI] [PMC free article] [PubMed] [Google Scholar]
18. Centers for Disease Control and Prevention (CDC), National Center for Health Statistics (NCHS). Version current 21 February, 2020. [Internet] Available from: https://wwwn.cdc.gov/nchs/nhanes/continuousnhanes/labmethods.aspx?BeginYear = 2011 (accessed 15 May, 2020).
19. Centers for Disease Control and Prevention (CDC). Version current September 2017. [Internet] Available from: https://wwwn.cdc.gov/Nchs/Nhanes/2015-2016/BPX_I.htm (accessed 15 May, 2020).
20. Centers for Disease Control and Prevention (CDC), National Center for Health Statistics (NCHS). Version current 21 February, 2020. [Internet] Available from: https://wwwn.cdc.gov/nchs/nhanes/tutorials/module2.aspx (accessed 15 May, 2020).
21. Centers for Disease Control and Prevention (CDC), National Center for Health Statistics (NCHS). Version current 21 February, 2020. [Internet] Available from: https://wwwn.cdc.gov/nchs/nhanes/tutorials/module3.aspx (accessed 15 May, 2020).
22. Lumley T. Analysis of complex survey samples. J Stat Soft. 2004;9(8):1–19. [Google Scholar]
23. Lumley T. Complex Surveys: A Guide to Analysis Using R. Hoboken, NJ: John Wiley; 2010. [Google Scholar]
24. Barnard J, Rubin D. Small-sample degrees of freedom with multiple imputation. Biometrika. 1999;86(4):948–55. [Google Scholar]
25. Snijder MB, Visser M, Dekker JM, Goodpaster BH, Harris TB, Kritchevsky SB, De Rekeneire N, Kanaya AM, Newman AB, Tylavsky FA et al. Low subcutaneous thigh fat is a risk factor for unfavourable glucose and lipid levels, independently of high abdominal fat. The Health ABC Study. Diabetologia. 2005;48(2):301–8. [DOI] [PubMed] [Google Scholar]
26. Ruderman NB, Berchtold P, Schneider S. Obesity-associated disorders in normal-weight individuals: some speculations. Int J Obes. 1982;6(Suppl 1):151–7. [PubMed] [Google Scholar]
27. Ruderman NB, Schneider SH, Berchtold P. The “metabolically-obese,” normal-weight individual. Am J Clin Nutr. 1981;34(8):1617–21. [DOI] [PubMed] [Google Scholar]
28. Ruderman N, Chisholm D, Pi-Sunyer X, Schneider S. The metabolically obese, normal-weight individual revisited. Diabetes. 1998;47(5):699–713. [DOI] [PubMed] [Google Scholar]
29. Vague J. La Differentiacion Sexuelle Humaine: Ses Incidences en Pathologie. Paris: Masson Editeur; 1953. [Google Scholar]
30. Krotkiewski M, Bjorntorp P, Sjostrom L, Smith U. Impact of obesity on metabolism in men and women. Importance of regional adipose tissue distribution. J Clin Invest. 1983;72(3):1150–62. [DOI] [PMC free article] [PubMed] [Google Scholar]
31. Kelishadi R, Cook SR, Motlagh ME, Gouya MM, Ardalan G, Motaghian M, Majdzadeh R, Ramezani MA. Metabolically obese normal weight and phenotypically obese metabolically normal youths: the CASPIAN Study. J Am Diet Assoc. 2008;108(1):82–90. [DOI] [PubMed] [Google Scholar]
32. Mathew H, Farr OM, Mantzoros CS. Metabolic health and weight: understanding metabolically unhealthy normal weight or metabolically healthy obese patients. Metabolism. 2016;65(1):73–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
33. Shen W, Punyanitya M, Chen J, Gallagher D, Albu J, Pi-Sunyer X, Lewis CE, Grunfeld C, Heshka S, Heymsfield SB. WC correlates with metabolic syndrome indicators better than percentage fat. Obesity (Silver Spring). 2006;14(4):727–36. [DOI] [PMC free article] [PubMed] [Google Scholar]
34. Wildman RP, Muntner P, Reynolds K, McGinn AP, Rajpathak S, Wylie-Rosett J, Sowers MR. The obese without cardiometabolic risk factor clustering and the normal weight with cardiometabolic risk factor clustering: prevalence and correlates of 2 phenotypes among the US population (NHANES 1999–2004). Arch Intern Med. 2008;168(15):1617–24. [DOI] [PubMed] [Google Scholar]
35. Conus F, Rabasa-Lhoret R, Peronnet F. Characteristics of metabolically obese normal-weight (MONW) subjects. Appl Physiol Nutr Metab. 2007;32(1):4–12. [DOI] [PubMed] [Google Scholar]
36. Karelis AD, St-Pierre DH, Conus F, Rabasa-Lhoret R, Poehlman ET. Metabolic and body composition factors in subgroups of obesity: what do we know?. J Clin Endocrinol Metab. 2004;89(6):2569–75. [DOI] [PubMed] [Google Scholar]
37. Srinivasan SR, Wang R, Chen W, Wei CY, Xu J, Berenson GS. Utility of waist-to-height ratio in detecting central obesity and related adverse cardiovascular risk profile among normal weight younger adults (from the Bogalusa Heart Study). Am J Cardiol. 2009;104(5):721–4. [DOI] [PubMed] [Google Scholar]
38. Teixeira TF, Alves RD, Moreira AP, Peluzio Mdo C. Main characteristics of metabolically obese normal weight and metabolically healthy obese phenotypes. Nutr Rev. 2015;73(3):175–90. [DOI] [PubMed] [Google Scholar]
39. Liu PJ, Ma F, Lou HP, Zhu YN. Normal-weight central obesity is associated with metabolic disorders in Chinese postmenopausal women. Asia Pac J Clin Nutr. 2017;26(4):692–7. [DOI] [PubMed] [Google Scholar]
40. Page JH, Rexrode KM, Hu F, Albert CM, Chae CU, Manson JE. Waist-height ratio as a predictor of coronary heart disease among women. Epidemiology. 2009;20(3):361–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
41. Park YS, Kim JS. Association between waist-to-height ratio and metabolic risk factors in Korean adults with normal body mass index and WC. Tohoku J Exp Med. 2012;228(1):1–8. [DOI] [PubMed] [Google Scholar]
42. Sardinha LB, Santos DA, Silva AM, Grontved A, Andersen LB, Ekelund U. A comparison between BMI, WC, and waist-to-height ratio for identifying cardio-metabolic risk in children and adolescents. PLoS One. 2016;11(2):e0149351. [DOI] [PMC free article] [PubMed] [Google Scholar]
43. Thomas EL, Frost G, Taylor-Robinson SD, Bell JD. Excess body fat in obese and normal-weight subjects. Nutr Res Rev. 2012;25(1):150–61. [DOI] [PubMed] [Google Scholar]
44. Rask-Andersen M, Karlsson T, Ek WE, Johansson A. Genome-wide association study of body fat distribution identifies adiposity loci and sex-specific genetic effects. Nat Commun. 2019;10(1):339. [DOI] [PMC free article] [PubMed] [Google Scholar]
45. de Koning L, Merchant AT, Pogue J, Anand SS. WC and waist-to-hip ratio as predictors of cardiovascular events: meta-regression analysis of prospective studies. Eur Heart J. 2007;28(7):850–6. [DOI] [PubMed] [Google Scholar]
46. Han TS, van Leer EM, Seidell JC, Lean ME. WC action levels in the identification of cardiovascular risk factors: prevalence study in a random sample. BMJ. 1995;311(7017):1401–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
47. Janssen I, Katzmarzyk PT, Ross R. WC and not body mass index explains obesity-related health risk. Am J Clin Nutr. 2004;79(3):379–84. [DOI] [PubMed] [Google Scholar]
48. Siparsky PN, Kirkendall DT, Garrett WE Jr. Muscle changes in aging: understanding sarcopenia. Sports Health. 2014;6(1):36–40. [DOI] [PMC free article] [PubMed] [Google Scholar]
49. Silver JJ, Einhorn TA. Osteoporosis and aging. Current update. Clin Orthop Relat Res. 1995;316:10–20. [PubMed] [Google Scholar]
50. Fani Marvasti F, Stafford RS. From sick care to health care – reengineering prevention into the U.S. system. N Engl J Med. 2012;367(10):889–91. [DOI] [PMC free article] [PubMed] [Google Scholar]
51. Agarwal SK, Misra A, Aggarwal P, Bardia A, Goel R, Vikram NK, Wasir JS, Hussain N, Ramachandran K, Pandey RM. WC measurement by site, posture, respiratory phase, and meal time: implications for methodology. Obesity (Silver Spring). 2009;17(5):1056–61. [DOI] [PubMed] [Google Scholar]
52. Wong MC, Ng BK, Kennedy SF, Hwaung P, Liu EY, Kelly NN, Pagano IS, Garber AK, Chow DC, Heymsfield SB et al. Children and adolescents’ anthropometrics body composition from 3-D optical surface scans. Obesity (Silver Spring). 2019;27(11):1738–49. [DOI] [PMC free article] [PubMed] [Google Scholar]
53. Anothaisintawee T, Sansanayudh N, Thamakaison S, Lertrattananon D, Thakkinstian A. Neck circumference as an anthropometric indicator of central obesity in patients with prediabetes: a cross-sectional study. Biomed Res Int. 2019;2019:4808541. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

nqaa194_Supplemental_File

Click here for additional data file.^{(304KB, docx)}

[bib1] 1. Heymsfield SB, Cefalu WT. Does body mass index adequately convey a patient's mortality risk?. JAMA. 2013;309(1):87–8. [DOI] [PubMed] [Google Scholar]

[bib2] 2. Shah NR, Braverman ER. Measuring adiposity in patients: the utility of body mass index (BMI), percent body fat, and leptin. PLoS One. 2012;7(4):e33308. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib3] 3. Aune D, Sen A, Prasad M, Norat T, Janszky I, Tonstad S, Romundstad P, Vatten LJ. BMI and all cause mortality: systematic review and non-linear dose-response meta-analysis of 230 cohort studies with 3.74 million deaths among 30.3 million participants. BMJ. 2016;353:i2156. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib4] 4. Jensen MD, Ryan DH, Apovian CM, Ard JD, Comuzzie AG, Donato KA, Hu FB, Hubbard VS, Jakicic JM, Kushner RF et al. 2013 AHA/ACC/TOS guideline for the management of overweight and obesity in adults: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines and The Obesity Society. Circulation. 2014;129(25 Suppl 2):S102–38. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib5] 5. Ross R, Neeland IJ, Yamashita S, Shai I, Seidell J, Magni P, Santos RD, Arsenault B, Cuevas A, Hu FB et al. WC as a vital sign in clinical practice: a Consensus Statement from the IAS and ICCR Working Group on Visceral Obesity. Nat Rev Endocrinol. 2020;16(3):177–89. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib6] 6. Eckel RH, Grundy SM, Zimmet PZ. The metabolic syndrome. Lancet. 2005;365(9468):1415–28. [DOI] [PubMed] [Google Scholar]

[bib7] 7. Hwaung P, Heo M, Kennedy S, Hong S, Thomas DM, Shepherd J, Heymsfield SB. Optimum WC-height indices for evaluating adult adiposity: an analytic review. Obes Rev. 2020;21(1):e12947. [DOI] [PubMed] [Google Scholar]

[bib8] 8. Heymsfield SB, Peterson CM, Thomas DM, Heo M, Schuna JM Jr. Why are there race/ethnic differences in adult body mass index-adiposity relationships? A quantitative critical review. Obes Rev. 2016;17(3):262–75. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib9] 9. Heymsfield SB, Peterson CM, Thomas DM, Heo M, Schuna JM Jr., Hong S, Choi W. Scaling of adult body weight to height across sex and race/ethnic groups: relevance to BMI. Am J Clin Nutr. 2014;100(6):1455–61. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib10] 10. Centers for Disease Control and Prevention (CDC). Version current January 2015. [Internet] Available from: https://wwwn.cdc.gov/nchs/nhanes/2011-2012/DEMO_G.htm#DMDEDUC2 (accessed 15 May, 2020).

[bib11] 11. Centers for Disease Control and Prevention (CDC). Version current September 2017. [Internet] Available from: https://wwwn.cdc.gov/Nchs/Nhanes/2015-2016/SMQ_I.htm (accessed 15 May, 2020).

[bib12] 12. Piercy KL, Troiano RP, Ballard RM, Carlson SA, Fulton JE, Galuska DA, George SM, Olson RD. The physical activity guidelines for Americans. JAMA. 2018;320(19):2020–8. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib13] 13. Centers for Disease Control and Prevention (CDC). Version current January 2007. [Internet] Available from: https://www.cdc.gov/nchs/data/nhanes/nhanes_07_08/manual_an.pdf (accessed 15 May, 2020).

[bib14] 14. Centers for Disease Control and Prevention (CDC), National Center for Health Statistics (NCHS). Version current 21 February, 2020. [Internet] Available from: https://wwwn.cdc.gov/nchs/nhanes/nhanes3/AnthropometricVideos.aspx (accessed 15 May, 2020).

[bib15] 15. Centers for Disease Control and Prevention (CDC), National Center for Health Statistics (NCHS). Version current 21 February, 2020. [Internet] Available from: https://wwwn.cdc.gov/nchs/nhanes/dxa/dxa.aspx (accessed 15 May, 2020).

[bib16] 16. Kim J, Wang Z, Heymsfield SB, Baumgartner RN, Gallagher D. Total-body skeletal muscle mass: estimation by a new dual-energy X-ray absorptiometry method. Am J Clin Nutr. 2002;76(2):378–83. [DOI] [PubMed] [Google Scholar]

[bib17] 17. Ng BK, Sommer MJ, Wong MC, Pagano I, Nie Y, Fan B, Kennedy S, Bourgeois B, Kelly N, Liu YE et al. Detailed 3-dimensional body shape features predict body composition, blood metabolites, and functional strength: the Shape Up! studies. Am J Clin Nutr. 2019;110(6):1316–26. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib18] 18. Centers for Disease Control and Prevention (CDC), National Center for Health Statistics (NCHS). Version current 21 February, 2020. [Internet] Available from: https://wwwn.cdc.gov/nchs/nhanes/continuousnhanes/labmethods.aspx?BeginYear = 2011 (accessed 15 May, 2020).

[bib19] 19. Centers for Disease Control and Prevention (CDC). Version current September 2017. [Internet] Available from: https://wwwn.cdc.gov/Nchs/Nhanes/2015-2016/BPX_I.htm (accessed 15 May, 2020).

[bib20] 20. Centers for Disease Control and Prevention (CDC), National Center for Health Statistics (NCHS). Version current 21 February, 2020. [Internet] Available from: https://wwwn.cdc.gov/nchs/nhanes/tutorials/module2.aspx (accessed 15 May, 2020).

[bib21] 21. Centers for Disease Control and Prevention (CDC), National Center for Health Statistics (NCHS). Version current 21 February, 2020. [Internet] Available from: https://wwwn.cdc.gov/nchs/nhanes/tutorials/module3.aspx (accessed 15 May, 2020).

[bib22] 22. Lumley T. Analysis of complex survey samples. J Stat Soft. 2004;9(8):1–19. [Google Scholar]

[bib23] 23. Lumley T. Complex Surveys: A Guide to Analysis Using R. Hoboken, NJ: John Wiley; 2010. [Google Scholar]

[bib24] 24. Barnard J, Rubin D. Small-sample degrees of freedom with multiple imputation. Biometrika. 1999;86(4):948–55. [Google Scholar]

[bib25] 25. Snijder MB, Visser M, Dekker JM, Goodpaster BH, Harris TB, Kritchevsky SB, De Rekeneire N, Kanaya AM, Newman AB, Tylavsky FA et al. Low subcutaneous thigh fat is a risk factor for unfavourable glucose and lipid levels, independently of high abdominal fat. The Health ABC Study. Diabetologia. 2005;48(2):301–8. [DOI] [PubMed] [Google Scholar]

[bib26] 26. Ruderman NB, Berchtold P, Schneider S. Obesity-associated disorders in normal-weight individuals: some speculations. Int J Obes. 1982;6(Suppl 1):151–7. [PubMed] [Google Scholar]

[bib27] 27. Ruderman NB, Schneider SH, Berchtold P. The “metabolically-obese,” normal-weight individual. Am J Clin Nutr. 1981;34(8):1617–21. [DOI] [PubMed] [Google Scholar]

[bib28] 28. Ruderman N, Chisholm D, Pi-Sunyer X, Schneider S. The metabolically obese, normal-weight individual revisited. Diabetes. 1998;47(5):699–713. [DOI] [PubMed] [Google Scholar]

[bib29] 29. Vague J. La Differentiacion Sexuelle Humaine: Ses Incidences en Pathologie. Paris: Masson Editeur; 1953. [Google Scholar]

[bib30] 30. Krotkiewski M, Bjorntorp P, Sjostrom L, Smith U. Impact of obesity on metabolism in men and women. Importance of regional adipose tissue distribution. J Clin Invest. 1983;72(3):1150–62. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib31] 31. Kelishadi R, Cook SR, Motlagh ME, Gouya MM, Ardalan G, Motaghian M, Majdzadeh R, Ramezani MA. Metabolically obese normal weight and phenotypically obese metabolically normal youths: the CASPIAN Study. J Am Diet Assoc. 2008;108(1):82–90. [DOI] [PubMed] [Google Scholar]

[bib32] 32. Mathew H, Farr OM, Mantzoros CS. Metabolic health and weight: understanding metabolically unhealthy normal weight or metabolically healthy obese patients. Metabolism. 2016;65(1):73–80. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib33] 33. Shen W, Punyanitya M, Chen J, Gallagher D, Albu J, Pi-Sunyer X, Lewis CE, Grunfeld C, Heshka S, Heymsfield SB. WC correlates with metabolic syndrome indicators better than percentage fat. Obesity (Silver Spring). 2006;14(4):727–36. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib34] 34. Wildman RP, Muntner P, Reynolds K, McGinn AP, Rajpathak S, Wylie-Rosett J, Sowers MR. The obese without cardiometabolic risk factor clustering and the normal weight with cardiometabolic risk factor clustering: prevalence and correlates of 2 phenotypes among the US population (NHANES 1999–2004). Arch Intern Med. 2008;168(15):1617–24. [DOI] [PubMed] [Google Scholar]

[bib35] 35. Conus F, Rabasa-Lhoret R, Peronnet F. Characteristics of metabolically obese normal-weight (MONW) subjects. Appl Physiol Nutr Metab. 2007;32(1):4–12. [DOI] [PubMed] [Google Scholar]

[bib36] 36. Karelis AD, St-Pierre DH, Conus F, Rabasa-Lhoret R, Poehlman ET. Metabolic and body composition factors in subgroups of obesity: what do we know?. J Clin Endocrinol Metab. 2004;89(6):2569–75. [DOI] [PubMed] [Google Scholar]

[bib37] 37. Srinivasan SR, Wang R, Chen W, Wei CY, Xu J, Berenson GS. Utility of waist-to-height ratio in detecting central obesity and related adverse cardiovascular risk profile among normal weight younger adults (from the Bogalusa Heart Study). Am J Cardiol. 2009;104(5):721–4. [DOI] [PubMed] [Google Scholar]

[bib38] 38. Teixeira TF, Alves RD, Moreira AP, Peluzio Mdo C. Main characteristics of metabolically obese normal weight and metabolically healthy obese phenotypes. Nutr Rev. 2015;73(3):175–90. [DOI] [PubMed] [Google Scholar]

[bib39] 39. Liu PJ, Ma F, Lou HP, Zhu YN. Normal-weight central obesity is associated with metabolic disorders in Chinese postmenopausal women. Asia Pac J Clin Nutr. 2017;26(4):692–7. [DOI] [PubMed] [Google Scholar]

[bib40] 40. Page JH, Rexrode KM, Hu F, Albert CM, Chae CU, Manson JE. Waist-height ratio as a predictor of coronary heart disease among women. Epidemiology. 2009;20(3):361–6. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib41] 41. Park YS, Kim JS. Association between waist-to-height ratio and metabolic risk factors in Korean adults with normal body mass index and WC. Tohoku J Exp Med. 2012;228(1):1–8. [DOI] [PubMed] [Google Scholar]

[bib42] 42. Sardinha LB, Santos DA, Silva AM, Grontved A, Andersen LB, Ekelund U. A comparison between BMI, WC, and waist-to-height ratio for identifying cardio-metabolic risk in children and adolescents. PLoS One. 2016;11(2):e0149351. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib43] 43. Thomas EL, Frost G, Taylor-Robinson SD, Bell JD. Excess body fat in obese and normal-weight subjects. Nutr Res Rev. 2012;25(1):150–61. [DOI] [PubMed] [Google Scholar]

[bib44] 44. Rask-Andersen M, Karlsson T, Ek WE, Johansson A. Genome-wide association study of body fat distribution identifies adiposity loci and sex-specific genetic effects. Nat Commun. 2019;10(1):339. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib45] 45. de Koning L, Merchant AT, Pogue J, Anand SS. WC and waist-to-hip ratio as predictors of cardiovascular events: meta-regression analysis of prospective studies. Eur Heart J. 2007;28(7):850–6. [DOI] [PubMed] [Google Scholar]

[bib46] 46. Han TS, van Leer EM, Seidell JC, Lean ME. WC action levels in the identification of cardiovascular risk factors: prevalence study in a random sample. BMJ. 1995;311(7017):1401–5. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib47] 47. Janssen I, Katzmarzyk PT, Ross R. WC and not body mass index explains obesity-related health risk. Am J Clin Nutr. 2004;79(3):379–84. [DOI] [PubMed] [Google Scholar]

[bib48] 48. Siparsky PN, Kirkendall DT, Garrett WE Jr. Muscle changes in aging: understanding sarcopenia. Sports Health. 2014;6(1):36–40. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib49] 49. Silver JJ, Einhorn TA. Osteoporosis and aging. Current update. Clin Orthop Relat Res. 1995;316:10–20. [PubMed] [Google Scholar]

[bib50] 50. Fani Marvasti F, Stafford RS. From sick care to health care – reengineering prevention into the U.S. system. N Engl J Med. 2012;367(10):889–91. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib51] 51. Agarwal SK, Misra A, Aggarwal P, Bardia A, Goel R, Vikram NK, Wasir JS, Hussain N, Ramachandran K, Pandey RM. WC measurement by site, posture, respiratory phase, and meal time: implications for methodology. Obesity (Silver Spring). 2009;17(5):1056–61. [DOI] [PubMed] [Google Scholar]

[bib52] 52. Wong MC, Ng BK, Kennedy SF, Hwaung P, Liu EY, Kelly NN, Pagano IS, Garber AK, Chow DC, Heymsfield SB et al. Children and adolescents’ anthropometrics body composition from 3-D optical surface scans. Obesity (Silver Spring). 2019;27(11):1738–49. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib53] 53. Anothaisintawee T, Sansanayudh N, Thamakaison S, Lertrattananon D, Thakkinstian A. Neck circumference as an anthropometric indicator of central obesity in patients with prediabetes: a cross-sectional study. Biomed Res Int. 2019;2019:4808541. [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

Distinct phenotypic characteristics of normal-weight adults at risk of developing cardiovascular and metabolic diseases

Abishek Stanley

John Schuna

Shengping Yang

Samantha Kennedy

Moonseong Heo

Michael Wong

John Shepherd

Steven B Heymsfield

ABSTRACT

Background

Objective

Methods

Results

Conclusions

Introduction

Methods

Study design

FIGURE 1.

TABLE 1.

Data sources/study population

TABLE 2.

Measurements

Anthropometry

Body composition

Avatar image development

Health risk factors

Statistical analyses

Results

WCI prediction models

Characteristics of the study population

Demographic

Body composition

FIGURE 2.

Health risk biomarkers

TABLE 3.

Discussion

FIGURE 3.

Limitations

Conclusions

Supplementary Material

Acknowledgments

Notes

Contributor Information

References

Associated Data

Supplementary Materials

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases