ArticlePDF Available
Letters to the Editor
Indian Journal of Endocrinology and Metabolism / Sep-Oct 2012 / Vol 16 | Issue 5 857
Table 2: Pearson’s Correlations (r values) between lipid
prole and anthropometric measurements.
Lipid Prole WC#SFT WHR BMI
Cholesterol 0.747 (**) 0.671 (**) 0.610 (**) 0.593 (**)
LDL 0.614 (**) 0.583 (**) 0.537 (**) 0.510 (**)
TGL 0.526 (**) 0.503 (**) 0.506 (**) 0.338 (**)
HDL -0.283 (**) -0.172 -0.212 (*) -0.187
LDL/HDL 0.593 (**) 0.524 (**) 0.484 (**) 0.481 (**)
**Correlation significant at 0.01 level, *Correlation significant at 0.05 level,
# WC - Highest correlation coefficient (r values)
Table 1: General guidelines for using total sum
(in millimeters) of the seven main skinfold sites
Excellent Good Average Below average Poor
Normal
Male 60–80 81–90 91– 110 111–150 150+
Female 70–90 91–100 101–120 121–150 150+
Relation between
anthropometric
measurements and serum
lipid prole among cardio-
metabolically healthy
subjects: A pilot study
Sir,
Excess body fatness is a risk factor associated with
premature mortality, type 2 diabetes and cardiovascular
disease. Serum lipid levels as cardio-metabolic risk factor
has been well known.[1] This study was done to correlate and
understand the association between measures of adiposity
like waist circumference (WC), waist/hip ratio (WHR),
skinfold thickness (SFT) and body mass index (BMI) with
serum lipid levels and to determine the best predictor of
serum lipid prole among them.
One-hundred subjects between 20 and 60 years of age
attending the outpatient department of RL Jalappa
Hospital, Kolar, were enrolled after obtaining clearance
from Institutional Ethical committee and informed consent
from them. Detailed history was taken and subjects with
H/O diabetes mellitus, cardiovascular disease, carcinoma,
liver disease and on lipid-lowering agents suggestive of
cardio-metabolic abnormality were excluded from the study.
Body weight was measured in kg by a mechanical scale to
the nearest kg. Height was measured to the nearest one
cm. BMI (kg/m2) was calculated using Quitelet’s index.
WC was measured midway between the lowest rib and the
iliac crest and hip circumference at the level of the greater
trochanters with legs close together, using a non-stretchable
measuring tape by average of three measurements nearest to
0.5 cm. The WHR equals WC divided by hip circumference.
SFT in mm was assessed at seven sites: biceps, triceps,
abdomen, subscapular, suprailiac, thigh and calf, using digital
skin caliper whereby a pinch of skin is precisely measured
by caliper at these sites to determine the subcutaneous fat
layer thickness. General guidelines for using total sum (in
millimeters) of the seven main skin fold sites [Table 1]:[2]
Lipid prole of the subjects was done in biochemistry
laboratory of RL Jalappa Hospital, Kolar, using Vitrow’s
250 Autoanalyser (Johnson & Johnson, Rochester, New
York, USA).
WC, BMI, WHR and SFT were correlated individually
with lipid prole using Pearson’s correlation analysis using
SPSS version 14.
The mean age in the study group was 39.3 ± 10.5 years
and the proportion of males and females was 67% and
37%, respectively.
All parameters of lipid prole namely serum cholesterol,
low density lipoprotein (LDL), triglyceride and LDL/
HDL ratio except high density lipoprotein (HDL) had
signicant positive correlation with all parameters of
anthropometric measures as shown in Table 2. HDL had
negative correlation with all anthropometric measures,
which was statistically signicant only with WC and
WHR and not with SFT and BMI. Correlations were
stronger for WC compared with other anthropometric
measurements.
Pearson’s Correlations (r values) between lipid prole and
anthropometric measurements [Table 2].
All the anthropometric variables had signicant positive
correlation with each other. According to the present study,
WC best correlates with all the parameters of lipid prole.
An increased WHR may reect both a relative abundance
of abdominal fat (increased WC) or a relative lack of
gluteal muscle (decreased hip circumference), questioning
its reliability.[3]
SFT has limitations like skin fold calipers cannot open wide
enough to measure total fat thickness, thus grossly under
Presentation at a Meeting:
Organization: PSI-SAAP, Conference Place: Bangalore, Date:
15th Dec 2010
[Downloaded free from http://www.ijem.in on Wednesday, September 28, 2016, IP: 78.189.111.93]
Letters to the Editor
Indian Journal of Endocrinology and Metabolism / Sep-Oct 2012 / Vol 16 | Issue 5
858
estimates body fat percentage in the obese population, and
has wide observer bias, requiring proper skills in skin fold
measurements.[4]
BMI does not account for factors such as body fat
distribution, specically abdominal obesity, and cannot
distinguish between lean and fat body mass.[1]
WC reects abdominal fat, which contains higher amounts
of visceral fat. Visceral fat is made by liver, turned into
cholesterol, and released into the bloodstream where it
forms plaque on the artery walls, resulting in high blood
pressure and cardiovascular disease.[5]
The results of this study are consistent with previous
studies that report a stronger association between
anthropometric measures accounting for abdominal
adiposity like WC and cardiovascular disease risk
factors.
[1,3,5]
This study revealed that all the anthropometric measures
were signicantly correlated with lipid prole. However,
WC was the best predictor of lipid prole and hence the
most important risk factor for cardio-metabolic diseases.
It is a very simple, economic and less time-consuming
procedure, which can be used as a screening test to
predict the cardio-metabolic risk of an individual. Further
studies with larger population are needed to quantify
the results for application to community health lifestyle
modications.
acknowledgment
The authors acknowledge the support of the biochemistry
department for providing lab Reports of lipid prole of
study subjects.
Garg Sumit, Shankar Vinutha, Kutty Karthiyanee,
Annamalai Nachal
Department of Physiology, Sri Devaraj Urs Medical College,
Kolar, Karnataka, India
Corresponding Author: Dr. Sumit Garg,
Department of Physiology, Sri Devaraj Urs Medical College,
Tamaka, Kolar – 563 101, Karnataka, India.
E-mail: sumitgargdr@gmail.com
RefeRences
1. Brenner DR, Tepylo K, Eny KM, Cahill LE, Sohemy AE. Comparison
of body mass index and waist circumference as predictors of
cardiometabolic health in a population of young Canadian adults.
Diabetol Metab Syndr 2010;2:28.
2. Skinfold Measurement [Internet] 2010. Available from: http://www.
topendsports.com/testing/tests/skinfolds.htm. [Updated 25 Jun 2010;
Cited 2010 Oct 8].
3. Wannamethee SG, Shaper AG, Morris RW, Whincup PH. Measures
of adiposity in the identification of metabolic abnormalities in elderly
men. Am J Clin Nutr 2005;81:1313-21.
4. Doyle AJ. The Exercise and Physical Fitness [homepage on the
Internet]. Georgia: Georgia State University; Body Composition. [About
5 screens]. Available from: http://www2.gsu.edu/~wwwfit/bodycomp.
html#Skinfold. [Updated 1998 Mar 18; cited 2011 oct 18].
5. Menke A, Muntner P, Wildman RP, Reynolds K, He J. Measures of
adiposity and cardiovascular disease risk factors. Obesity (Silver
Spring) 2007;15:785-95.
Access this article online
Quick Response Code:
Website:
www.ijem.in
DOI:
10.4103/2230-8210.100686
It is short-but so what!
Sir,
A 34-year-old male presented with gradually worsening
right upper-quadrant pain. There was no history of any
signicant past illness or abdominal surgery. His laboratory
test results revealed hemoglobin 12.9 g/dl; total leukocyte
count 13 200 cells/mm3 (neutrophils-63%); normal liver and
kidney function tests. Ultrasound of the abdomen revealed
features suggesting acute-on-chronic cholecystitis following
which the patient was subjected to contrast-enhanced CT.
Apart from calculus cholecystitis, an important incidental
nding was detected. CT sections through the pancreas
revealed normal size head, neck and the uncinate process of
the pancreas with absent pancreatic body and tail. The distal
pancreatic bed was seen lled by stomach and intestine
[Figure 1]. The patient was subsequently evaluated and
was found to have mildly elevated serum glucose (fasting
blood glucose – 105 mg/ dl).
Figure 1: CT sections through the abdomen reveal normal size head,
neck and the uncinate process of the pancreas with absent pancreatic
body and tail
[Downloaded free from http://www.ijem.in on Wednesday, September 28, 2016, IP: 78.189.111.93]
... This finding was similar to observation reported by Manjareeka et al. and Shamai et al. 23,25 On the contrary, a study by Garg et al. found a significant correlation of BMI with TC and LDL-C. 26 BMI showed a non-significant inverse correlation with HDL-C. Conversely, Shamai et al. reported that BMI showed a significant inverse correlation with HDL-C. ...
... 25 There was a significant correlation between BMI and TG. This finding was similar to the observation reported by Garg et al. 26 However, a study done in India reported no correlation between BMI and TG. 23 WC in the present study showed a statistically nonsignificant correlation with TC, HDL-C, and LDL-C. ...
... On the other hand, a study conducted by Garg et al. reported a significant correlation with TC, HDL-C, and LDL-C. 26 WC in the present study showed a significant correlation with TG. This finding was similar to observation reported by Manjareeka et al. and Garg et al. 23,26 In the present study, WHpR showed a significant correlation with TC. ...
Article
Full-text available
Background Anthropometric parameters of individuals are good at predicting functional impairment, mortality, and future cardiometabolic diseases. The relationship between anthropometric parameters and lipid profiles have been studied in different parts of the world. But to date, no such studies have been conducted in Nepal. Objective To investigate the association between anthropometric parameters and lipid profile in the adult population of Kaski district, Nepal. Method This study was carried out at Manipal Teaching Hospital, Pokhara, Kaski, Nepal. The fasting lipid profiles were analyzed in a total of 400 subjects aged > 18 years with an automated OCD Vitros 350 dry chemistry analyzer. The Kolmogorov-Smirnov test was used to test the normality of the data. The mean values of fasting lipids were compared within the subjects with different body mass index groups using ANOVA and waist circumference, waist-hip ratios, waist-height ratios, and neck circumference using independent samples t-test. The anthropometric indices evaluated were body mass index, waist circumference, waist-hip ratio, waist-height ratio, head circumference, neck circumference, and mid-upper arm circumference. Pearson’s correlation coefficients and multiple regression analysis were performed to identify the association between the lipid profile and anthropometric parameters. The difference was considered statistically significant when p values (two-tailed) were < 0.050. Result The mean values of the serum lipid parameters other than high-density lipoprotein cholesterol were found to be higher in the subjects with an above than normal BMI, waist circumference, waist-hip ratio, waist-height ratio, and neck circumference. Pearson’s correlation coefficient and multiple regression analysis showed that waist height ratio best predicts serum triglycerides (β=0.622, p < 0.001) and high-density lipoprotein cholesterol (β=-0.711, p < 0.001) among all measured anthropometric parameters. Conclusion Among all the studied anthropometric parameters, the WHtR was found to be the most powerful predictor of serum triglycerides and high-density lipoprotein cholesterol.
... This finding was similar to observation reported by Manjareeka et al. and Shamai et al. 23,25 On the contrary, a study by Garg et al. found a significant correlation of BMI with TC and LDL-C. 26 BMI showed a non-significant inverse correlation with HDL-C. Conversely, Shamai et al. reported that BMI showed a significant inverse correlation with HDL-C. ...
... 25 There was a significant correlation between BMI and TG. This finding was similar to the observation reported by Garg et al. 26 However, a study done in India reported no correlation between BMI and TG. 23 WC in the present study showed a statistically nonsignificant correlation with TC, HDL-C, and LDL-C. ...
... On the other hand, a study conducted by Garg et al. reported a significant correlation with TC, HDL-C, and LDL-C. 26 WC in the present study showed a significant correlation with TG. This finding was similar to observation reported by Manjareeka et al. and Garg et al. 23,26 In the present study, WHpR showed a significant correlation with TC. ...
... WC is being increasingly accepted as the best anthropometric indicator of abdominal adiposity and metabolic risk. [9] It is less known, however, which one of these anthropometric variables (BMI or WHR or WC) is a better link to lipid profile. ...
... [6,13] Many Indian studies have been reported relating anthropometric parameters with lipid profile in type 2 diabetes [14] and also in hypothyroid patients. [15] Not many studies have been reported in this regard involving healthy subjects in India, [9] especially in Eastern India. The present study attempts to correlate some anthropometric variables with lipid parameters in apparently healthy subjects, as also to assess the anthropometric variable which best reflects the altered lipid profile. ...
... While a South Indian pilot study reports a strong correlation between anthropometric parameters and lipid profile in healthy adults, [9] the present study shows no correlation between BMI and lipid profile; in the BMI ≥25 group, BMI showed significant negative correlation with HDL. Results of a few other studies support our findings. ...
Article
Full-text available
Background: Cardiovascular diseases constitute one class of common contributors to morbidity and mortality worldwide. Prevalence of overweight and obesity has dramatically increased in developing countries and is related to cardiovascular risk factors. Anthropometric parameters have the advantages in daily clinical practice of being a simple to measure tool with good reproducibility, especially in a developing country like India. Aim of this study is to correlate some anthropometric variables with lipid parameters in healthy subjects and to assess the anthropometric variable which best reflects the altered lipid profile. Materials and methods: A hospital based cross-sectional study was conducted after the Institutional Ethical Committee Clearance. Included participants (1187) were subjected to anthropometric measurements such as height, weight, waist circumference (WC), and hip circumference using standard procedures on the same morning of the day, as the blood sample was collected after overnight fast and estimated for fasting blood sugar and lipid profile. Results: There is a weak correlation between body mass index (BMI) and lipid parameters. Among all the anthropometric variables studied, WC is best correlated to lipid parameters. The mean values of lipid parameters were not significantly different in BMI <25 and BMI ≥25 groups. Conclusions: WC remains one of the simple and reliable variables which best reflects the lipid profile. In a developing country like India, where measurement of cardiovascular risk factors such as body fat saturation and lipid profile remains difficult in the rural population, WC may be used as an effective tool, without being used as a substitute.
... Lingkar pinggang merupakan prediktor yang paling baik untuk melihat abnormalitas profil lipid dan paling mudah diaplikasikan. 13,14 Tujuan penelitian ini untuk mengobservasi korelasi pengukuran antropometri dengan abnormalitas profil lipid di daerah pedesaan. Hasil penelitian diharapkan dapat memberikan aplikasi terkait pengukuran anthropometri sebagai parameter abnormalitas profil lipid. ...
... 22 Penelitian di Turki menyatakan bahwa LP memiliki peran besar yang memengaruhi risiko penyakit kardiovaskular. 13,14,21 Pada penelitian ini terdapat perbedaan yang signifikan (p<0,05) antara kadar HDL dan trigliserida antara kelompok normal dan obesitas berdasarkan parameter LP dan RLPP. Penelitian di Cina juga menunjukkan hasil serupa di mana hanya terdapat perbedaan signifikan di profil HDL dan trigliserida terhadap parameter LP. 23 Hasil penelitian di Nigeria justru menunjukkan bahwa terdapat perbedaan proporsi hanya dijumpai antara RLPP dan trigliserida. ...
... In a study done by X Zhang, XO Shu, Y-T Gao, G yang et al., "Anthropometric predictors of coronary heart disease in Chinese women", it was concluded that, Waist Hip Ratio was positively associated with the risk of Coronary Artery Disease in both younger and older women, while other anthropometric indices, including Body Mass Index, were related to Cardiovascular disease risk primarily among younger women. 17 In another study done by A Esmaillzadeh, P. Mirmiran & F. Azizi, it was concluded that Waist Hip Ratio, as compared to Body Mass Index, Waist Circumference & Waist Hip Ratio, may be a better indicator of cardiovascular risk factors. 18 In 20 In a study done by Fu-Ling Chu, ChungHuei Hsu &Chi Jeng on premenopausal taiwanese women it was concluded that Waist Hip Ratio had the best performance in predicting hypertension and Diabetes Mellitus. ...
... For example, BMI indicates concise information about muscle mass and fat percentage indirectly, while WC and WHR are more reliable indices to detect fat distribution, especially the abdominal region [15]. Considering the importance of anthropometric indices, to identify disease-related risk factors, there are a few reports on the relationship between these parameters and lipid profiles in healthy subjects [16]. ...
... Cholesterol is the source of most of the steroids found in increased amounts in the circulation of pregnant women, which has a significant role of lipid metabolism in pregnancy. The lipid change during pregnancy may be due to formation of zygote in the uterine wall in the first trimester in response to the maternal switch from carbohydrate to fat metabolism, which is an alternative pathway for energy generation due to high energy demand in second trimester and development of fetal organ in the third trimester (29). ...
Article
Full-text available
Background Pregnancy is a natural physiological statement with hormonal and metabolic changes that helps the growth and survival of the fetus. However, biochemical profiles derangement may lead to pregnancy complications. Therefore, there is a need for determining biochemical profiles among pregnant women. Methods A comparative cross-sectional study was conducted among pregnant and non-pregnant women at the University of Gondar Hospital, from February to April, 2015. Fasting blood sample was collected from 139 pregnant and 139 age matched non-pregnant women using systematic random sampling technique. Interviewer-administered questionnaire was used to collect socio-demographic and clinical data. Fasting blood glucose and lipid profile were measured by A25 Biosytemchemistry analyzer using enzymatic calorimetric methods. Data analysis was done using SPSS version 20. Level of significance between groups was analyzed using independent student t-test and Mann-Whitney U test. A p-value of <0.05 was considered as statistically significant. Result Pregnant women as compared to non-pregnant had significantly increased glucose (96.35±14.45 and 81.12±9.86 mg/dl), total cholesterol (211.9±40.88 and 172.40±29.64 mg/dl) [p<0.05], respectively. It had also significantly high triglycerides (190.81±81.04 and 107.43±45.80 mg/dl) and low-density lipoprotein cholesterol (116.03±37.26 and 86.12±27.29mg/dl) [p<05] in pregnant as compared to non-pregnant women. The level of high-density lipoprotein cholesterol was significantly lower in pregnant women (59.58±14.26) than control (63.63±11.4, P <0.05). Conclusion There were statistically significant increment in glucose, total cholesterol, triglycerides, low-density lipoprotein cholesterol and decrement in high-density lipoprote in cholesterol levels among pregnant women compared with non-pregnant women. Therefore, pregnant women have to be monitored closely for their biochemical profiles to avoid adverse pregnancy outcomes.
... Cholesterol is the source of most of the steroids found in increased amounts in the circulation of pregnant women, which has a significant role of lipid metabolism in pregnancy. The lipid change during pregnancy may be due to formation of zygote in the uterine wall in the first trimester in response to the maternal switch from carbohydrate to fat metabolism, which is an alternative pathway for energy generation due to high energy demand in second trimester and development of fetal organ in the third trimester (29). ...
Article
Full-text available
BACKGROUND: Pregnancy is a natural physiological statement with hormonal and metabolic changes that helps the growth and survival of the fetus. However, biochemical profiles derangement may lead to pregnancy complications. Therefore, there is a need for determining biochemical profiles among pregnant women. METHODS: A comparative cross-sectional study was conducted among pregnant and non-pregnant women at the University of Gondar Hospital, from February to April, 2015. Fasting blood sample was collected from 139 pregnant and 139 age matched non-pregnant women using systematic random sampling technique. Interviewer-administered questionnaire was used to collect socio-demographic and clinical data. Fasting blood glucose and lipid profile were measured by A25 Biosytemchemistry analyzer using enzymatic calorimetric methods. Data analysis was done using SPSS version 20. Level of significance between groups was analyzed using independent student t-test and Mann-Whitney U test. A p-value of <0.05 was considered as statistically significant.
... The WHR equals WC divided by hip circumference. [6] Statistical Analysis ...
... Obesity is related to altered reproductive hormone status and is now considered a growing disease in developed and underdeveloped countries [13]. BMI, a measure of general obesity cannot distinguish between lean and fat body mass [32]. Changes in BMI can be attributed to skeletal muscle rather than body fat [33]. ...
Article
Full-text available
This study aimed to investigate whether waist circumference (WC) or body mass index (BMI) is a better predictor of blood lipid concentrations among young men and women from different ethnocultural groups. Participants were 1181 healthy men (n = 358) and women (n = 823) aged 20-29 years taken from the cross-sectional Toronto Nutrigenomics and Health Study. Analyses were conducted separately for men and women, and for Caucasian and East Asian ethnocultural groups. Serum triglycerides (TG) and total to HDL cholesterol ratio (TC:HDL cholesterol) were used as outcomes. Associations between the adiposity and blood lipid measures were examined using partial correlations and odds ratios derived from logistic regression models. WC had a stronger association with serum lipid concentrations than BMI. WC was significantly related to TG and TC:HDL cholesterol after adjusting for BMI and covariates among men and women (P </= 0.01). However, after adjusting for WC and covariates, BMI was not significantly associated with the two serum lipid measures. WC was a better predictor of TG and TC:HDL among all sex and ethnocultural subgroups except among East Asian women where little difference between the two measures was observed. WC is a stronger predictor of cardiometabolic health when compared with BMI among young adults, especially among men.
Article
Background: Body mass index (BMI; in kg/m²) is considered a poor indicator of overall and abdominal obesity in the elderly. Objectives: Our goal was to determine which simple anthropometric measurements [BMI, waist-to-hip ratio (WHR), waist circumference (WC), percentage body fat (%BF), or fat mass (FM)] are most closely associated with metabolic risk factors and insulin resistance in elderly men. Design: This was a cross-sectional study of 2924 men aged 60–79 y with no history of coronary heart disease, stroke, or diabetes who were drawn from general practices in 24 British towns. Results: BMI and WC were the measures most strongly associated with the metabolic syndrome (≥3 of the following: hypertension, low HDL cholesterol, high triacylglycerols, or high blood glucose) and insulin resistance. For a 1-SD increase in BMI, WC, WHR, %BF, and FM, the odds ratios (95% CIs) of having the metabolic syndrome after adjustment for age, socioeconomic status, smoking status, and physical activity were as follows: BMI, 1.61 (1.44, 1.79); WC, 1.65 (1.48, 1.81); WHR, 1.49 (1.34, 1.66); %BF, 1.41 (1.25, 1.59); and FM, 1.53 (1.38, 1.70). For insulin resistance, the odds ratios (95% CIs) were as follows: 2.48 (2.22, 2.77), 2.46 (2.19, 2.65), 1.75 (1.59, 1.93), 1.79 (1.60, 2.00), and 2.10 (1.88, 2.34), respectively. In normal-weight (BMI < 25) and overweight (BMI 25–29.9) men, the presence of the metabolic syndrome and insulin resistance increased with increasing WC; this did not occur in obese men. Conclusions: BMI and WC are the simple measures of adiposity most strongly associated with metabolic abnormalities in elderly men. Our findings suggest that WC can be used as a complementary measurement to identify health risks in normal-weight and overweight elderly persons.
Article
Body mass index (BMI; in kg/m(2)) is considered a poor indicator of overall and abdominal obesity in the elderly. Our goal was to determine which simple anthropometric measurements [BMI, waist-to-hip ratio (WHR), waist circumference (WC), percentage body fat (%BF), or fat mass (FM)] are most closely associated with metabolic risk factors and insulin resistance in elderly men. This was a cross-sectional study of 2924 men aged 60-79 y with no history of coronary heart disease, stroke, or diabetes who were drawn from general practices in 24 British towns. BMI and WC were the measures most strongly associated with the metabolic syndrome (>/=3 of the following: hypertension, low HDL cholesterol, high triacylglycerols, or high blood glucose) and insulin resistance. For a 1-SD increase in BMI, WC, WHR, %BF, and FM, the odds ratios (95% CIs) of having the metabolic syndrome after adjustment for age, socioeconomic status, smoking status, and physical activity were as follows: BMI, 1.61 (1.44, 1.79); WC, 1.65 (1.48, 1.81); WHR, 1.49 (1.34, 1.66); %BF, 1.41 (1.25, 1.59); and FM, 1.53 (1.38, 1.70). For insulin resistance, the odds ratios (95% CIs) were as follows: 2.48 (2.22, 2.77), 2.46 (2.19, 2.65), 1.75 (1.59, 1.93), 1.79 (1.60, 2.00), and 2.10 (1.88, 2.34), respectively. In normal-weight (BMI < 25) and overweight (BMI 25-29.9) men, the presence of the metabolic syndrome and insulin resistance increased with increasing WC; this did not occur in obese men. BMI and WC are the simple measures of adiposity most strongly associated with metabolic abnormalities in elderly men. Our findings suggest that WC can be used as a complementary measurement to identify health risks in normal-weight and overweight elderly persons.
Article
To determine which of five measures of adiposity maintains the strongest association with cardiovascular disease risk factors. A nationally representative sample of 12,608 adult participants of the third National Health and Nutrition Examination Survey were examined. Waist circumference, total body fat, percent body fat, BMI, and skinfold thickness were measured following a standardized protocol. In multivariable adjusted models including waist circumference and BMI as independent variables, waist circumference was a significantly better predictor. The odds ratios (95% confidence intervals) for each standard deviation higher waist circumference and BMI for men were as follows: 1.88 (1.43, 2.48) and 0.99 (0.76, 1.29), respectively, for hypertension; 1.51 (0.87, 2.59) and 1.23 (0.76, 1.99), respectively, for diabetes; and 1.85 (1.48, 2.32) and 1.00 (0.80, 1.24), respectively, for low high-density lipoprotein-cholesterol. The analogous odds ratios (95% confidence intervals) for women were as follows: 2.28 (1.74, 3.00) and 0.91 (0.69, 1.19), respectively, for hypertension; 2.72 (1.85, 4.00) and 0.82 (0.55, 1.23), respectively, for diabetes; and 1.90 (1.47, 2.47) and 1.07 (0.83, 1.38), respectively, for low high-density lipoprotein-cholesterol. Results were markedly similar for waist circumference in models adjusting for total body fat, percent body fat, and skinfold thickness separately. In contrast, waist circumference was not a significantly better predictor of elevated C-reactive protein than the other measures of adiposity. Waist circumference maintains a stronger association with cardiovascular disease risk factors than other measures of adiposity.
The Exercise and Physical Fitness [homepage on the Internet]. Georgia: Georgia State University; Body Composition. [About 5 screens] Available from
  • Aj Doyle
Doyle AJ. The Exercise and Physical Fitness [homepage on the Internet]. Georgia: Georgia State University; Body Composition. [About 5 screens]. Available from: http://www2.gsu.edu/~wwwfit/bodycomp. html#Skinfold. [Updated 1998 Mar 18; cited 2011 oct 18].
Georgia: Georgia State University; Body Composition
  • A J Doyle
Doyle AJ. The Exercise and Physical Fitness [homepage on the Internet]. Georgia: Georgia State University; Body Composition. [About 5 screens]. Available from: http://www2.gsu.edu/~wwwfit/bodycomp. html#Skinfold. [Updated 1998 Mar 18; cited 2011 oct 18].