[Skip to Content]
Sign In
Individual Sign In
Create an Account
Institutional Sign In
OpenAthens Shibboleth
Purchase Options:
[Skip to Content Landing]
Figure.
Body Mass Index Distribution Among US Men and Women by Urbanization Level, 2001-2004 and 2013-2016
Body Mass Index Distribution Among US Men and Women by Urbanization Level, 2001-2004 and 2013-2016

Data are from the National Health and Nutrition Examination Survey. The estimates are weighted and adjusted for age to the projected 2000 US census population using the age groups of 20 to 39 years, 40 to 59 years, and 60 years or older. The body mass index distributions were calculated using local polynomial smoothing. The proportion of the population with a body mass index within a given range is estimated by the area under the density curve within that range. The sample sizes are unweighted.

Table 1.  
Unweighted Sample Sizes for US Adults Aged 20 Years or Older With Body Mass Index Data by Sex, Age Group, Race and Hispanic Origin, Education Level, Smoking Status, and Urbanization Level From the National Health and Nutrition Examination Survey, 2013-2016
Unweighted Sample Sizes for US Adults Aged 20 Years or Older With Body Mass Index Data by Sex, Age Group, Race and Hispanic Origin, Education Level, Smoking Status, and Urbanization Level From the National Health and Nutrition Examination Survey, 2013-2016
Table 2.  
Unadjusted and Age-Adjusted Prevalence of Obesity and Severe Obesity Among US Adults Aged 20 Years or Older, 2013-2016
Unadjusted and Age-Adjusted Prevalence of Obesity and Severe Obesity Among US Adults Aged 20 Years or Older, 2013-2016
Table 3.  
Trends in Prevalence of Obesity and Severe Obesity Among US Adults Aged 20 Years or Older, 2001-2004 to 2013-2016a
Trends in Prevalence of Obesity and Severe Obesity Among US Adults Aged 20 Years or Older, 2001-2004 to 2013-2016a
1.
Ogden  CL, Carroll  MD, Fryar  CD, Flegal  KM.  Prevalence of obesity among adults and youth: United States, 2011-2014.  NCHS Data Brief. 2015;(219):1-8.PubMedGoogle Scholar
2.
Flegal  KM, Kruszon-Moran  D, Carroll  MD, Fryar  CD, Ogden  CL.  Trends in obesity among adults in the United States, 2005 to 2014.  JAMA. 2016;315(21):2284-2291. doi:10.1001/jama.2016.6458PubMedGoogle ScholarCrossref
3.
Befort  CA, Nazir  N, Perri  MG.  Prevalence of obesity among adults from rural and urban areas of the United States: findings from NHANES (2005-2008).  J Rural Health. 2012;28(4):392-397. doi:10.1111/j.1748-0361.2012.00411.xPubMedGoogle ScholarCrossref
4.
Cohen  SA, Cook  SK, Kelley  L, Foutz  JD, Sando  TA.  A closer look at rural-urban health disparities: associations between obesity and rurality vary by geospatial and sociodemographic factors.  J Rural Health. 2017;33(2):167-179. doi:10.1111/jrh.12207PubMedGoogle ScholarCrossref
5.
Voss  JD, Masuoka  P, Webber  BJ, Scher  AI, Atkinson  RL.  Association of elevation, urbanization and ambient temperature with obesity prevalence in the United States.  Int J Obes (Lond). 2013;37(10):1407-1412. doi:10.1038/ijo.2013.5PubMedGoogle ScholarCrossref
6.
Trivedi  T, Liu  J, Probst  J, Merchant  A, Jhones  S, Martin  AB.  Obesity and obesity-related behaviors among rural and urban adults in the USA.  Rural Remote Health. 2015;15(4):3267.PubMedGoogle Scholar
7.
Li  TY, Rana  JS, Manson  JE,  et al.  Obesity as compared with physical activity in predicting risk of coronary heart disease in women.  Circulation. 2006;113(4):499-506. doi:10.1161/circulationaha.105.574087PubMedGoogle ScholarCrossref
8.
Hsu  CY, McCulloch  CE, Iribarren  C, Darbinian  J, Go  AS.  Body mass index and risk for end-stage renal disease.  Ann Intern Med. 2006;144(1):21-28. doi:10.7326/0003-4819-144-1-200601030-00006PubMedGoogle ScholarCrossref
9.
Ogden  CL, Carroll  MD, Kit  BK, Flegal  KM.  Prevalence of childhood and adult obesity in the United States, 2011-2012.  JAMA. 2014;311(8):806-814. doi:10.1001/jama.2014.732PubMedGoogle ScholarCrossref
10.
Ogden  CL, Fakhouri  TH, Carroll  MD,  et al.  Prevalence of obesity among adults, by household income and education—United States, 2011-2014.  MMWR Morb Mortal Wkly Rep. 2017;66(50):1369-1373. doi:10.15585/mmwr.mm6650a1PubMedGoogle ScholarCrossref
11.
Johnson  CL, Dohrmann  SM, Burt  VL, Mohadjer  LK.  National Health and Nutrition Examination Survey: sample design, 2011-2014.  Vital Health Stat 2. 2014;(162):1-33.PubMedGoogle Scholar
12.
National Center for Health Statistics. NHANES response rates. https://wwwn.cdc.gov/nchs/nhanes/ResponseRates.aspx. Accessed April 13, 2018.
13.
National Center for Health Statistics. National Health and Nutrition Examination Survey (NHANES): examination manuals. https://www.cdc.gov/nchs/nhanes/search/nhanes_continuous.aspx. Accessed April 13, 2018.
14.
Jensen  MD, Ryan  DH, Apovian  CM,  et al; American College of Cardiology/American Heart Association Task Force on Practice Guidelines; Obesity Society.  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-S138. doi:10.1161/01.cir.0000437739.71477.eePubMedGoogle ScholarCrossref
15.
National Center for Health Statistics. NCHS urban-rural classification scheme for counties. https://www.cdc.gov/nchs/data_access/urban_rural.htm. Accessed September 26, 2017.
16.
Hales  CM, Carroll  MD, Fryar  CD, Ogden  CL.  Prevalence of obesity among adults and youth: United States, 2015-2016.  NCHS Data Brief. 2017;(288):1-8.PubMedGoogle Scholar
17.
Bieler  GS, Brown  GG, Williams  RL, Brogan  DJ.  Estimating model-adjusted risks, risk differences, and risk ratios from complex survey data.  Am J Epidemiol. 2010;171(5):618-623. doi:10.1093/aje/kwp440PubMedGoogle ScholarCrossref
18.
Kleinbaum  DG, Kupper  LL, Muller  KE, Nizam  A. Polynomial regression. In:  Applied Regression Analysis and Other Multivariable Methods. 3rd ed. Pacific Grove, CA: Duxbury Press; 1998:281-316.
19.
American Lung Association. Cutting tobacco’s rural roots: tobacco use in rural communities. http://www.lung.org/assets/documents/research/cutting-tobaccos-rural-roots.pdf. Accessed December 15. 2018.
20.
Klein  RJ, Schoenborn  CA.  Age adjustment using the 2000 projected U.S. population.  Healthy People 2000 Stat Notes. 2001;20(20):1-9.PubMedGoogle Scholar
21.
Korn  EL, Graubard  BI.  Confidence intervals for proportions with small expected number of positive counts estimated from survey data.  Surv Methodol. 1998;23:193-201.Google Scholar
22.
National Center for Health Statistics. National Center for Health Statistics data presentation standards for proportions. https://www.cdc.gov/nchs/data/series/sr_02/sr02_175.pdf. Accessed May 15, 2018.
23.
R package survey: analysis of complex survey samples [computer program]. Version 3.32. Vienna, Austria: R Foundation for Statistical Computing; 2017.
24.
Sobal  J, Troiano  R, Frongillo  E.  Rural-urban differences in obesity.  Rural Sociol. 1996;61(2):289-305. doi:10.1111/j.1549-0831.1996.tb00621.xGoogle ScholarCrossref
25.
Zhang  X, Holt  JB, Lu  H,  et al.  Neighborhood commuting environment and obesity in the United States: an urban-rural stratified multilevel analysis.  Prev Med. 2014;59:31-36. doi:10.1016/j.ypmed.2013.11.004PubMedGoogle ScholarCrossref
26.
Connor Gorber  S, Tremblay  M, Moher  D, Gorber  B.  A comparison of direct vs self-report measures for assessing height, weight and body mass index: a systematic review.  Obes Rev. 2007;8(4):307-326. doi:10.1111/j.1467-789X.2007.00347.xPubMedGoogle ScholarCrossref
27.
O’Connor  A, Wellenius  G.  Rural-urban disparities in the prevalence of diabetes and coronary heart disease.  Public Health. 2012;126(10):813-820. doi:10.1016/j.puhe.2012.05.029PubMedGoogle ScholarCrossref
28.
Boring  MA, Hootman  JM, Liu  Y,  et al.  Prevalence of arthritis and arthritis-attributable activity limitation by urban-rural county classification—United States, 2015.  MMWR Morb Mortal Wkly Rep. 2017;66(20):527-532. doi:10.15585/mmwr.mm6620a2PubMedGoogle ScholarCrossref
29.
Peen  J, Schoevers  RA, Beekman  AT, Dekker  J.  The current status of urban-rural differences in psychiatric disorders.  Acta Psychiatr Scand. 2010;121(2):84-93. doi:10.1111/j.1600-0447.2009.01438.xPubMedGoogle ScholarCrossref
30.
Goeres  LM, Gille  A, Furuno  JP,  et al.  Rural-urban differences in chronic disease and drug utilization in older Oregonians.  J Rural Health. 2016;32(3):269-279. doi:10.1111/jrh.12153PubMedGoogle ScholarCrossref
31.
Moy  E, Garcia  MC, Bastian  B,  et al.  Leading causes of death in nonmetropolitan and metropolitan areas—United States, 1999-2014.  MMWR Surveill Summ. 2017;66(1):1-8. doi:10.15585/mmwr.ss6601a1PubMedGoogle ScholarCrossref
32.
Ivey-Stephenson  AZCA, Crosby  AE, Jack  SPD, Haileyesus  T, Kresnow-Sedacca  MJ.  Suicide trends among and within urbanization levels by sex, race/ethnicity, age group, and mechanism of death—United States, 2001-2015.  MMWR Surveill Summ. 2017;66(18):1-16. doi:10.15585/mmwr.ss6618a1PubMedGoogle ScholarCrossref
33.
Singh  GK, Siahpush  M.  Widening rural-urban disparities in all-cause mortality and mortality from major causes of death in the USA, 1969-2009.  J Urban Health. 2014;91(2):272-292. doi:10.1007/s11524-013-9847-2PubMedGoogle ScholarCrossref
34.
Singh  GK, Siahpush  M.  Widening rural-urban disparities in life expectancy, US, 1969-2009.  Am J Prev Med. 2014;46(2):e19-e29. doi:10.1016/j.amepre.2013.10.017PubMedGoogle ScholarCrossref
35.
Douthit  N, Kiv  S, Dwolatzky  T, Biswas  S.  Exposing some important barriers to health care access in the rural USA.  Public Health. 2015;129(6):611-620. doi:10.1016/j.puhe.2015.04.001PubMedGoogle ScholarCrossref
36.
Doogan  NJ, Roberts  ME, Wewers  ME,  et al.  A growing geographic disparity: rural and urban cigarette smoking trends in the United States.  Prev Med. 2017;104(104):79-85. doi:10.1016/j.ypmed.2017.03.011PubMedGoogle ScholarCrossref
37.
Heo  M, Faith  MS, Pietrobelli  A, Heymsfield  SB.  Percentage of body fat cutoffs by sex, age, and race-ethnicity in the US adult population from NHANES 1999-2004.  Am J Clin Nutr. 2012;95(3):594-602. doi:10.3945/ajcn.111.025171PubMedGoogle ScholarCrossref
38.
Zheng  W, McLerran  DF, Rolland  B,  et al.  Association between body-mass index and risk of death in more than 1 million Asians.  N Engl J Med. 2011;364(8):719-729.PubMedGoogle ScholarCrossref
Original Investigation
June 19, 2018

Differences in Obesity Prevalence by Demographic Characteristics and Urbanization Level Among Adults in the United States, 2013-2016

Author Affiliations
  • 1National Center for Health Statistics, US Centers for Disease Control and Prevention, Hyattsville, Maryland
  • 2US Public Health Service, Rockville, Maryland
  • 3National Center for Chronic Disease Prevention and Health Promotion, US Centers for Disease Control and Prevention, Atlanta, Georgia
JAMA. 2018;319(23):2419-2429. doi:10.1001/jama.2018.7270
Key Points

Question  During 2013-2016, were there differences in the prevalence of obesity and severe obesity by demographics and urbanization level among US adults?

Findings  In this cross-sectional analysis that included 10 792 adults aged 20 years or older, differences were found in the prevalence of obesity and severe obesity by age group, race and Hispanic origin, and education level. The prevalence of obesity was significantly greater among women living in nonmetropolitan statistical areas (non-MSAs; 47.2%) compared with women living in large MSAs (38.1%), and the prevalence of severe obesity in non-MSAs was higher than in large MSAs among men (9.9% vs 4.1%, respectively) and women (13.5% vs 8.1%, respectively).

Meaning  Differences in age group, race and Hispanic origin, education level, or smoking status were not related to the differences in the prevalence of obesity and severe obesity by urbanization level.

Abstract

Importance  Differences in obesity by sex, age group, race and Hispanic origin among US adults have been reported, but differences by urbanization level have been less studied.

Objectives  To provide estimates of obesity by demographic characteristics and urbanization level and to examine trends in obesity prevalence by urbanization level.

Design, Setting, and Participants  Serial cross-sectional analysis of measured height and weight among adults aged 20 years or older in the 2001-2016 National Health and Nutrition Examination Survey, a nationally representative survey of the civilian, noninstitutionalized US population.

Exposures  Sex, age group, race and Hispanic origin, education level, smoking status, and urbanization level as assessed by metropolitan statistical areas (MSAs; large: ≥1 million population).

Main Outcomes and Measures  Prevalence of obesity (body mass index [BMI] ≥30) and severe obesity (BMI ≥40) by subgroups in 2013-2016 and trends by urbanization level between 2001-2004 and 2013-2016.

Results  Complete data on weight, height, and urbanization level were available for 10 792 adults (mean age, 48 years; 51% female [weighted]). During 2013-2016, 38.9% (95% CI, 37.0% to 40.7%) of US adults had obesity and 7.6% (95% CI, 6.8% to 8.6%) had severe obesity. Men living in medium or small MSAs had a higher age-adjusted prevalence of obesity compared with men living in large MSAs (42.4% vs 31.8%, respectively; adjusted difference, 9.8 percentage points [95% CI, 5.1 to 14.5 percentage points]); however, the age-adjusted prevalence among men living in non-MSAs was not significantly different compared with men living in large MSAs (38.9% vs 31.8%, respectively; adjusted difference, 4.8 percentage points [95% CI, −2.9 to 12.6 percentage points]). The age-adjusted prevalence of obesity was higher among women living in medium or small MSAs compared with women living in large MSAs (42.5% vs 38.1%, respectively; adjusted difference, 4.3 percentage points [95% CI, 0.2 to 8.5 percentage points]) and among women living in non-MSAs compared with women living in large MSAs (47.2% vs 38.1%, respectively; adjusted difference, 4.7 percentage points [95% CI, 0.2 to 9.3 percentage points]). Similar patterns were seen for severe obesity except that the difference between men living in large MSAs compared with non-MSAs was significant. The age-adjusted prevalence of obesity and severe obesity also varied significantly by age group, race and Hispanic origin, and education level, and these patterns of variation were often different by sex. Between 2001-2004 and 2013-2016, the age-adjusted prevalence of obesity and severe obesity significantly increased among all adults at all urbanization levels.

Conclusions and Relevance  In this nationally representative survey of adults in the United States, the age-adjusted prevalence of obesity and severe obesity in 2013-2016 varied by level of urbanization, with significantly greater prevalence of obesity and severe obesity among adults living in nonmetropolitan statistical areas compared with adults living in large metropolitan statistical areas.

Introduction

Previous reports have shown differences in the prevalence of obesity among adults in the United States by sex, age group, and race and Hispanic origin. In 2011-2014, non-Hispanic black women had a higher prevalence of obesity than non-Hispanic white and Hispanic women, whereas Hispanic men had a higher prevalence than non-Hispanic white men, but the prevalence among non-Hispanic black men was not significantly different from non-Hispanic white men. Non-Hispanic Asian adults had the lowest prevalence of obesity among both men and women.1

Differences in obesity trends by other demographic characteristics also have been reported. From 2005 to 2014, the prevalence of obesity among women increased from 35% to 40% and severe obesity increased from 7% to 10%, whereas among men, the prevalence of obesity increased from 33% to 35% and severe obesity increased from 4.2% to 5.5%, but the changes were not statistically significant.2

Differences in the prevalence of chronic disease risk factors have been reported among adults living in rural areas compared with those living in urban areas.3-6 These studies, using both measured and self-reported weight and height, have found a higher prevalence of obesity among adults living in rural areas compared with urban areas of the United States. It is unclear whether these differences have changed over time.

The objectives of this analysis were to provide the latest 4-year estimates (2013-2016) for the prevalence of obesity among US adults by sex, age group, race and Hispanic origin, education level, smoking status, and urbanization level. Among adults with obesity, those with severe obesity have even higher risks of obesity-related adverse health outcomes such as coronary heart disease7 and end-stage renal disease8; therefore, prevalence estimates for severe obesity also were examined. Trends in the prevalence of obesity among adults have been previously reported by sex, age group, race and Hispanic origin, and education level2,9,10; using measured height and weight, this report adds trends by urbanization level from 2001 to 2016.

Methods

The National Health and Nutrition Examination Survey (NHANES) is a cross-sectional survey designed to monitor the health and nutritional status of the US population and has been conducted continuously since 1999. Data are released in 2-year cycles, with the most recent release for 2015-2016. Each NHANES cycle uses a stratified, multistage probability sampling design to produce a nationally representative sample of the US civilian, noninstitutionalized population.11

NHANES was approved by the National Center for Health Statistics (NCHS) research ethics review board. Adult participants provided written consent. The NHANES examination response rate for adults aged 20 years or older was 64% in 2013-2014 and 55% in 2015-2016; response rates for adults between 2001-2002 and 2011-2012 ranged between 65% and 73%.12

NHANES consisted of interviews conducted in participants’ homes and physical examinations conducted in mobile examination centers, in which height and weight were measured using standardized techniques.13 Body mass index (BMI; calculated as weight in kilograms divided by height in meters squared) was rounded to 1 decimal place. Obesity was defined as a BMI of 30 or greater and severe obesity was defined as a BMI of 40 or greater according to clinical guidelines.14

Demographic variables included sex, age group, race and Hispanic origin, and education level because previous analyses have shown differences in the prevalence of obesity by these factors.2,9,10 Participant age was categorized in the following age groups: 20 to 39 years, 40 to 59 years, and 60 years or older.

Participants’ self-reported race and Hispanic origin was categorized as non-Hispanic white, non-Hispanic black, non-Hispanic Asian, Hispanic, or other (including multiple races). Data for the “other” category were included in the analyses of the entire population but the results for this category are not reported. Adjusted trend analyses between 2001-2002 and 2015-2016 included race and Hispanic origin categorized as non-Hispanic white, non-Hispanic black, Mexican American, or other due to NHANES sample design changes and for consistency across time.11 The term race and Hispanic origin reflects the specific designation in NHANES.

Participants’ smoking status was categorized as never smoker (<100 cigarettes during lifetime), former smoker (≥100 cigarettes during lifetime but not currently smoking), or current smoker (currently smoke cigarettes every day or on most days). Education level was categorized as high school diploma or less, some college, or college graduate.

Level of urbanization was assigned based on the participant’s county of residence (2001-2014) or the county where the physical examinations occurred (2015-2016). Unweighted agreement between urbanization level based on county of residence and county of the examination was 99.9% for NHANES 2007-2014. County of residence and county of examination can be obtained through the research data center (https://www.cdc.gov/rdc/).

Urbanization level was based on 1990 (for NHANES 2001-2004), 2006 (for NHANES 2005-2012), and 2013 (for NHANES 2013-2016) NCHS classification schemes for counties.15 The NCHS scheme was based on 4 metropolitan and 2 nonmetropolitan categories. The metropolitan categories included: (1) large central metropolitan or counties in metropolitan statistical areas (MSAs) with a population of 1 million or more that contain all or part of the area’s principal city; (2) large fringe metropolitan or counties in MSAs with a population of 1 million or more that surrounded the large central metropolitan counties; (3) medium metropolitan or counties in MSAs with a population of 250 000 to 999 999; and (4) small metropolitan or counties in MSAs with a population of less than 250 000.

Nonmetropolitan, or the most rural areas, included: (1) micropolitan or counties in micropolitan statistical areas, which contained an urban cluster with 2500 to 49 999 inhabitants and (2) noncore or nonmetropolitan counties that did not qualify as micropolitan. The NCHS metropolitan categories were collapsed into the following 3 categories to increase the sample size and the degrees of freedom for the subgroup analysis: large MSAs (combined large central and fringe metropolitan counties); medium or small MSAs (combined medium and small metropolitan counties); and non-MSAs (combined micropolitan and noncore counties).

Statistical Analysis

Prevalence estimates of obesity and severe obesity and 95% CIs were calculated by sex, age group, race and Hispanic origin, education level, smoking status, and urbanization level for 2013-2016. Estimates were calculated for men and women separately because of known differences in patterns of obesity prevalence among adults by sex.2,9,16

Adjusted prevalence ratios, adjusted differences, and Wald 95% CIs were calculated for men and women separately using logistic regression models that included age group, race and Hispanic origin, education level, smoking status, and urbanization level.17 Tests of trend over the categories of age group, education level, smoking status, and urbanization level (and their associated P values) were calculated by including these as continuous variables in the logistic regression models.

The analyses of trends in obesity and severe obesity from 2001-2002 to 2015-2016 by urbanization level were evaluated overall and for men and women separately. Obesity prevalence estimates by urbanization level from 2001 to 2016 were calculated using 4-year periods (2001-2004, 2005-2008, 2009-2012, and 2013-2016) to increase the stability and statistical reliability of the estimates. Linear and quadratic trends for age-adjusted obesity prevalence by urbanization level were tested using linear regression. The 2-year survey cycle was modeled as an orthogonal polynomial.18

Linear time trends by urbanization level were adjusted by age group, race and Hispanic origin, education level, and smoking status because these variables were associated with both obesity and urbanization level.2,6,19 The linear regression estimate (β coefficient) for the continuous 2-year survey cycle can be interpreted as the average percentage point change every 2 years; the 95% CI and P value for this estimate also were calculated. The differences in the mean percentage point change in the prevalence of obesity and severe obesity were evaluated by testing the urbanization level × survey cycle interaction in the fully adjusted models.

The differences in obesity trends by urbanization level were further evaluated by examining changes in the distribution of BMI over time. Weighted and age-adjusted BMI distributions by sex, urbanization level, and period (2001-2004 vs 2013-2016) were calculated using local polynomial smoothing; statistical tests were not performed. The BMI distribution is presented as the probability density within the range of BMI in the actual data. The proportion of the population with a BMI within a given range is estimated by the area under the density curve within that range.

All analyses used sample weights for the NHANES physical examination that adjust for nonresponse, noncoverage, and unequal probabilities of selection. Four-year estimates were adjusted by age group using the direct method to the projected 2000 US census population.20

All variance estimates accounted for the complex sample design, which includes stratification and clustering in addition to weighting. The standard errors for prevalence estimates were calculated using Taylor series linearization and 95% CIs, and were constructed using the method by Korn and Graubard.21 Effective sample size, absolute and relative 95% CI width, and degrees of freedom were evaluated to determine the reliability of prevalence estimates.22

A 2-sided P value of .05 was used to assess statistical significance. No adjustments were made for multiple comparisons. The analyses were conducted using R version 3.4.1 (R Foundation for Statistical Computing),23 SAS version 9.4 (SAS Institute Inc), SUDAAN version 11.0 (RTI International), and Stata version 13 (StataCorp).

Results

The analyses of the 2013-2016 data excluded 135 pregnant women and 135 adults with missing height or weight data, leaving a total examined sample size of 10 792 adults (mean age, 48 years; 51% female [weighted]; unweighted percentage excluded, 270/11 062 = 2.4%). An additional 16 adults missing education level or smoking status were excluded from the analyses that included these variables.

Trend analyses between 2001-2002 and 2013-2016 excluded 1157 pregnant women, 883 adults missing height or weight data, and 17 adults missing urbanization level, leaving an examined sample size of 40 632 adults (unweighted percentage excluded, 2057/42 689 = 4.8%). An additional 77 adults with missing education level or smoking status were excluded from the trend analyses that adjusted for these variables. The sample sizes for men and women by age group, race and Hispanic origin, education level, smoking status, and urbanization level appear in Table 1.

Prevalence estimates for obesity and severe obesity in 2013-2016 by demographic characteristics and urbanization level appear in Table 2 and the unweighted numbers appear in eTable 1 in the Supplement. Among all adults, 38.9% (95% CI, 37.0% to 40.7%) had obesity and 7.6% (95% CI, 6.8% to 8.6%) had severe obesity (eTable 2 in the Supplement).

The age-adjusted prevalence of obesity was 36.5% (95% CI, 34.0% to 39.0%) among men and 40.8% (95% CI, 38.7% to 42.9%) among women. The age-adjusted prevalence of severe obesity was 5.5% (95% CI, 4.6% to 6.6%) among men and 9.8% (95% CI, 8.7% to 11.0%) among women. The prevalence estimates and unweighted numbers for obesity and severe obesity stratified by sex, race and Hispanic origin, and age group appear in eTables 2 and 3 in the Supplement.

The adjusted differences and the adjusted prevalence ratios for obesity and severe obesity by age group, race and Hispanic origin, education level, smoking status, and urbanization level appear in Table 2. Adults aged 40 to 59 years had a significantly higher prevalence of obesity compared with adults aged 20 to 39 years among both men (39.0% vs 33.2%, respectively; adjusted difference, 6.1 percentage points [95% CI, 2.1 to 10.0 percentage points]) and women (44.7% vs 36.8%, respectively; adjusted difference, 8.7 percentage points [95% CI, 5.2 to 12.2 percentage points]).

Hispanic men had a significantly higher age-adjusted prevalence of obesity compared with non-Hispanic white men (40.6% vs 36.2%, respectively; adjusted difference, 7.0 percentage points [95% CI, 1.7 to 12.3 percentage points]). Non-Hispanic Asian men had a significantly lower age-adjusted prevalence of obesity compared with non-Hispanic white men (11.2% vs 36.2%, respectively; adjusted difference, −22.0 percentage points [95% CI, −25.1 to −18.9 percentage points]). However, the prevalence of obesity among non-Hispanic black and non-Hispanic white men was not significantly different.

The age-adjusted prevalence of obesity was significantly higher among non-Hispanic black women compared with non-Hispanic white women (55.9% vs 38.1%, respectively; adjusted difference, 16.9 percentage points [95% CI, 13.7 to 20.1 percentage points]). The age-adjusted prevalence of obesity was significantly higher among Hispanic women compared with non-Hispanic white women (48.9% vs 38.1%, respectively; adjusted difference, 8.3 percentage points [95% CI, 3.7 to 12.9 percentage points]). Non-Hispanic Asian women had a significantly lower age-adjusted prevalence of obesity compared with non-Hispanic white women (13.6% vs 38.1%, respectively; adjusted difference, −22.9 percentage points [95% CI, −26.5 to −19.3 percentage points]).

Men with some college education had a significantly higher age-adjusted prevalence of obesity than male college graduates (44.0% vs 30.0%, respectively; adjusted difference, 11.3 percentage points [95% CI, 6.0 to 16.6 percentage points]). However, the age-adjusted prevalence of obesity among men with a high school diploma or less was not statistically different from male college graduates.

There was a significant increasing linear trend in the prevalence of obesity with decreasing education level among women (P < .001). The age-adjusted prevalence of obesity was higher among women with some college compared with female college graduates (45.1% vs 29.3%, respectively; adjusted difference, 13.2 percentage points [95% CI, 9.7 to 16.7 percentage points]) and among women with a high school diploma or less compared with female college graduates (47.3% vs 29.3%, respectively; adjusted difference, 14.7 percentage points [95% CI, 10.7 to 18.7 percentage points]).

Men who currently smoked had a significantly lower age-adjusted prevalence of obesity compared with men who never smoked (31.1% vs 35.5%, respectively; adjusted difference, −6.8 percentage points [95% CI, −11.8 to −1.8 percentage points]). Men who were former smokers had a significantly higher age-adjusted prevalence of obesity compared with men who never smoked (40.2% vs 35.5%, respectively; adjusted difference, 4.1 percentage points [95% CI, 0.3 to 8.0 percentage points]). Among women, the prevalence of obesity did not differ significantly by smoking status.

There was a significantly increasing linear trend in obesity prevalence across the 3 categories of urbanization level from large MSAs to non-MSAs among both men and women (P = .03 for both). Men living in medium or small MSAs had a higher age-adjusted prevalence of obesity compared with men living in large MSAs (42.4% vs 31.8%, respectively; adjusted difference, 9.8 percentage points [95% CI, 5.1 to 14.5 percentage points]). However, the prevalence of obesity among men living in non-MSAs compared with men living in large MSAs was not significantly different (38.9% vs 31.8%, respectively; adjusted difference, 4.8 percentage points [95% CI, −2.9 to 12.6 percentage points]).

The age-adjusted prevalence of obesity was higher among women living in medium or small MSAs compared with women living in large MSAs (42.5% vs 38.1%, respectively; adjusted difference, 4.3 percentage points [95% CI, 0.2 to 8.5 percentage points]). The age-adjusted prevalence of obesity was higher among women living in non-MSAs compared with women living in large MSAs (47.2% vs 38.1%, respectively; adjusted difference, 4.7 percentage points [95% CI, 0.2 to 9.3 percentage points]).

The prevalence of severe obesity did not differ significantly by age group among men. However, women aged 60 years or older had a significantly lower prevalence of severe obesity than women aged 20 to 39 years (7.2% vs 9.8%, respectively; adjusted difference, −2.8 percentage points [95% CI, −5.1 to −0.6 percentage points]).

Non-Hispanic black men had a significantly higher age-adjusted prevalence of severe obesity compared with non-Hispanic white men (7.3% vs 5.6%, respectively; adjusted difference, 2.3 percentage points [95% CI, 0.3 to 4.4 percentage points]). Non-Hispanic black women had a significantly higher age-adjusted prevalence of severe obesity compared with non-Hispanic white women (15.5% vs 9.4%, respectively; adjusted difference, 6.3 percentage points [95% CI, 3.5 to 9.0 percentage points]). Non-Hispanic Asian men had a significantly lower age-adjusted prevalence of severe obesity compared with non-Hispanic white men (0.4% vs 5.6%, respectively; adjusted difference, −4.9 percentage points [95% CI, −6.4 to −3.4 percentage points]). Non-Hispanic Asian women had a significantly lower age-adjusted prevalence of severe obesity compared with non-Hispanic white women (0.3% vs 9.4%, respectively; adjusted difference, −8.8 percentage points [95% CI, −10.5 to −7.1 percentage points]). However, among men and women there was no significant difference in the prevalence of obesity between non-Hispanic and Hispanic adults.

Men with some college had a higher age-adjusted prevalence of severe obesity compared with men who were college graduates (7.7% vs 3.8%, respectively; adjusted difference, 3.0 percentage points [95% CI, 0.8 to 5.2 percentage points]). Women with some college had a higher age-adjusted prevalence of severe obesity compared with women who were college graduates (12.2% vs 6.6%, respectively; adjusted difference, 4.2 percentage points [95% CI, 0.9 to 7.6 percentage points]). However, among men and women, there was no significant difference in the prevalence of severe obesity between those with a high school diploma or less and those who were college graduates.

The prevalence of severe obesity did not vary by smoking status among men or women. The prevalence of severe obesity showed a significant linear increase among both men and women across the 3 urbanization level categories from large MSAs to non-MSAs (P < .001 for men and P = .004 for women).

The age-adjusted prevalence of severe obesity was significantly higher among men living in medium or small MSAs compared with men living in large MSAs (6.1% vs 4.1%, respectively; adjusted difference, 1.8 percentage points [95% CI, 0 to 3.6 percentage points]). The age-adjusted prevalence of severe obesity was significantly higher among women living in medium or small MSAs compared with women living in large MSAs (11.1% vs 8.1%, respectively; adjusted difference, 2.8 percentage points [95% CI, 0.3 to 5.4 percentage points]).

The age-adjusted prevalence of severe obesity was significantly higher among men living in non-MSAs compared with men living in large MSAs (9.9% vs 4.1%, respectively; adjusted difference, 5.0 percentage points [95% CI, 1.8 to 8.2 percentage points]). The age-adjusted prevalence of severe obesity was significantly higher among women living in non-MSAs compared with women living in large MSAs (13.5% vs 8.1%, respectively; adjusted difference, 3.7 percentage points [95% CI, 0.6 to 6.7 percentage points]).

At all levels of urbanization, there were significantly increasing trends in the age-adjusted prevalence of obesity and severe obesity from 2001-2004 to 2013-2016 (P < .001 to P = .03 for linear trend), except that the increase in severe obesity among men living in medium or small MSAs was not statistically significant (P = .20 for linear trend) (Table 3 and eFigures 1-4 in the Supplement). Trends were very similar in both the age-adjusted and fully adjusted models. There were no significant quadratic trends in any age-adjusted or fully adjusted model.

The prevalence of obesity increased at similar rates for both men and women at all urbanization levels (range, 0.8-1.6 percentage points every 2 years). The prevalence of severe obesity increased among adults living in large MSAs from 4.1% to 6.2% (β = 0.3 percentage points every 2 years [95% CI, 0.2 to 0.5 percentage points]). The prevalence of severe obesity increased among adults living in non-MSAs from 4.6% to 11.8% (β = 0.9 percentage points every 2 years [95% CI, 0.4 to 1.3 percentage points]).

Among men, the mean percentage point change every 2 years varied significantly by urbanization level (P = .02 for interaction). The prevalence of severe obesity increased among men living in large urban MSAs from 2.5% to 4.1% (β = 0.2 percentage points every 2 years [95% CI, 0 to 0.4 percentage points]) and among men living in non-MSAs from 2.8% to 9.9% (β = 1.0 percentage point every 2 years [95% CI, 0.4 to 1.5 percentage points]). Among women, the average percentage point change every 2 years did not vary significantly by level of urbanization.

The Figure shows weighted and age-adjusted BMI distributions in 2001-2004 and 2013-2016 by sex and urbanization level. The BMI distributions in 2013-2016 among adults living in non-MSAs appear shifted and skewed to the right of the 2001-2004 distributions, indicating that BMI has increased overall and that the proportion of adults with a BMI at the upper end of the distribution has increased. These shifts are more apparent among adults living in non-MSAs compared with those living in large MSAs among both men and women.

Discussion

The prevalence of obesity and severe obesity increased with decreasing level of urbanization from large MSAs to medium or small MSAs and non-MSAs. This pattern was seen among both men and women. The prevalence of obesity and severe obesity also varied by age group, race and Hispanic origin, education level, and smoking status, often with different patterns by sex.

Shifts in BMI distribution between 2001-2004 and 2013-2016 toward the upper end are most evident among adults living in non-MSAs. The differences in the prevalence of obesity and severe obesity by urbanization level were not explained by demographic characteristics and smoking status.

The differences in the prevalence of obesity by urbanization level in the United States have been examined in previous studies.3-6,24,25 However, the results are difficult to compare because of differing classification schemes for urbanization level and differences in obesity assessment. Many studies have reported differences using self-reported weight and height,4,5,25 which are known to underestimate obesity prevalence,26 and others have used measured height and weight.3,6

Previous reports3,6 based on NHANES data from 2005-2008 reported that differences in the prevalence of obesity between urban and rural areas were independent of daily energy intake, percentage of calorie intake from fat, physical activity, and marital status. However, a study24 based on earlier NHANES data (1976-1980) concluded that urban-rural differences in the prevalence of obesity were largely explained by age group, education level, and parity among women, and age group and marital status among men.

Urban-rural health disparities also have been reported for various health conditions such as diabetes and cardiovascular disease,27 arthritis,28 mental health disorders,29 and prescription medication use.30 Furthermore, compared with adults living in urban areas, those living in rural areas have higher mortality from certain chronic diseases,31 suicide,32 and all causes.33 Life expectancy in rural areas in 2005-2009 was 2 years shorter than in metropolitan areas.34

Certain differences in health status may be related to differences in access to health services35 and health-related behaviors such as physical activity, intake of sweetened beverages,6 and cigarette smoking.36 Differences by urbanization level in life expectancy34 and smoking status36 also have increased over time.

Limitations

This study has several limitations. First, BMI is an indirect measure of adiposity and health risk. The level of adiposity differs among age groups and race and Hispanic origin groups at the same BMI. For example, non-Hispanic black adults have lower adiposity compared with non-Hispanic white adults with the same BMI, and older adults have higher adiposity compared with younger adults with the same BMI.37 Health risk begins at a lower BMI among Asian adults than among non-Hispanic white adults.38

Second, NHANES was not designed to provide estimates for individual urbanization levels. As a result, estimates for medium or small MSAs or non-MSAs are based on smaller sample sizes and fewer degrees of freedom, leading to imprecise estimates of standard errors.22 The analysis of population subgroups within medium or small MSAs or non-MSAs also is limited.

Third, although sampling weights adjust for nonresponse, residual bias may remain due to incomplete nonresponse adjustment and may vary with changing response rates.

Conclusions

In this nationally representative survey of adults in the United States, the age-adjusted prevalence of obesity and severe obesity in 2013-2016 varied by level of urbanization, with significantly greater prevalence of obesity and severe obesity among adults living in nonmetropolitan statistical areas compared with adults living in large metropolitan statistical areas.

Back to top
Article Information

Corresponding Author: Craig M. Hales, MD, National Center for Health Statistics, US Centers for Disease Control and Prevention, 3311 Toledo Rd, Hyattsville, MD 20782 (chales@cdc.gov).

Accepted for Publication: May 14, 2018.

Author Contributions: Dr Hales had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Hales, Ogden.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Hales.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Hales, Fryar, Carroll, Freedman, Aoki.

Supervision: Ogden.

Conflict of Interest Disclosures: The authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest and none were reported.

Funding/Support: The National Center for Health Statistics and the US Centers for Disease Control and Prevention funded this study.

Role of the Funder/Sponsor: The National Center for Health Statistics and the US Centers for Disease Control and Prevention had a role in the design and conduct of the National Health and Nutrition Examination Survey, in the collection and management of the data, and in the review and approval of the manuscript; however, the National Center for Health Statistics and the US Centers for Disease Control and Prevention had no role in the analysis and interpretation of the data, in the preparation of the manuscript, or in the decision to submit the manuscript for publication.

Disclaimer: The findings and conclusions in this report are those of the authors and do not necessarily reflect the official position of the National Center for Health Statistics and the US Centers for Disease Control and Prevention.

References
1.
Ogden  CL, Carroll  MD, Fryar  CD, Flegal  KM.  Prevalence of obesity among adults and youth: United States, 2011-2014.  NCHS Data Brief. 2015;(219):1-8.PubMedGoogle Scholar
2.
Flegal  KM, Kruszon-Moran  D, Carroll  MD, Fryar  CD, Ogden  CL.  Trends in obesity among adults in the United States, 2005 to 2014.  JAMA. 2016;315(21):2284-2291. doi:10.1001/jama.2016.6458PubMedGoogle ScholarCrossref
3.
Befort  CA, Nazir  N, Perri  MG.  Prevalence of obesity among adults from rural and urban areas of the United States: findings from NHANES (2005-2008).  J Rural Health. 2012;28(4):392-397. doi:10.1111/j.1748-0361.2012.00411.xPubMedGoogle ScholarCrossref
4.
Cohen  SA, Cook  SK, Kelley  L, Foutz  JD, Sando  TA.  A closer look at rural-urban health disparities: associations between obesity and rurality vary by geospatial and sociodemographic factors.  J Rural Health. 2017;33(2):167-179. doi:10.1111/jrh.12207PubMedGoogle ScholarCrossref
5.
Voss  JD, Masuoka  P, Webber  BJ, Scher  AI, Atkinson  RL.  Association of elevation, urbanization and ambient temperature with obesity prevalence in the United States.  Int J Obes (Lond). 2013;37(10):1407-1412. doi:10.1038/ijo.2013.5PubMedGoogle ScholarCrossref
6.
Trivedi  T, Liu  J, Probst  J, Merchant  A, Jhones  S, Martin  AB.  Obesity and obesity-related behaviors among rural and urban adults in the USA.  Rural Remote Health. 2015;15(4):3267.PubMedGoogle Scholar
7.
Li  TY, Rana  JS, Manson  JE,  et al.  Obesity as compared with physical activity in predicting risk of coronary heart disease in women.  Circulation. 2006;113(4):499-506. doi:10.1161/circulationaha.105.574087PubMedGoogle ScholarCrossref
8.
Hsu  CY, McCulloch  CE, Iribarren  C, Darbinian  J, Go  AS.  Body mass index and risk for end-stage renal disease.  Ann Intern Med. 2006;144(1):21-28. doi:10.7326/0003-4819-144-1-200601030-00006PubMedGoogle ScholarCrossref
9.
Ogden  CL, Carroll  MD, Kit  BK, Flegal  KM.  Prevalence of childhood and adult obesity in the United States, 2011-2012.  JAMA. 2014;311(8):806-814. doi:10.1001/jama.2014.732PubMedGoogle ScholarCrossref
10.
Ogden  CL, Fakhouri  TH, Carroll  MD,  et al.  Prevalence of obesity among adults, by household income and education—United States, 2011-2014.  MMWR Morb Mortal Wkly Rep. 2017;66(50):1369-1373. doi:10.15585/mmwr.mm6650a1PubMedGoogle ScholarCrossref
11.
Johnson  CL, Dohrmann  SM, Burt  VL, Mohadjer  LK.  National Health and Nutrition Examination Survey: sample design, 2011-2014.  Vital Health Stat 2. 2014;(162):1-33.PubMedGoogle Scholar
12.
National Center for Health Statistics. NHANES response rates. https://wwwn.cdc.gov/nchs/nhanes/ResponseRates.aspx. Accessed April 13, 2018.
13.
National Center for Health Statistics. National Health and Nutrition Examination Survey (NHANES): examination manuals. https://www.cdc.gov/nchs/nhanes/search/nhanes_continuous.aspx. Accessed April 13, 2018.
14.
Jensen  MD, Ryan  DH, Apovian  CM,  et al; American College of Cardiology/American Heart Association Task Force on Practice Guidelines; Obesity Society.  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-S138. doi:10.1161/01.cir.0000437739.71477.eePubMedGoogle ScholarCrossref
15.
National Center for Health Statistics. NCHS urban-rural classification scheme for counties. https://www.cdc.gov/nchs/data_access/urban_rural.htm. Accessed September 26, 2017.
16.
Hales  CM, Carroll  MD, Fryar  CD, Ogden  CL.  Prevalence of obesity among adults and youth: United States, 2015-2016.  NCHS Data Brief. 2017;(288):1-8.PubMedGoogle Scholar
17.
Bieler  GS, Brown  GG, Williams  RL, Brogan  DJ.  Estimating model-adjusted risks, risk differences, and risk ratios from complex survey data.  Am J Epidemiol. 2010;171(5):618-623. doi:10.1093/aje/kwp440PubMedGoogle ScholarCrossref
18.
Kleinbaum  DG, Kupper  LL, Muller  KE, Nizam  A. Polynomial regression. In:  Applied Regression Analysis and Other Multivariable Methods. 3rd ed. Pacific Grove, CA: Duxbury Press; 1998:281-316.
19.
American Lung Association. Cutting tobacco’s rural roots: tobacco use in rural communities. http://www.lung.org/assets/documents/research/cutting-tobaccos-rural-roots.pdf. Accessed December 15. 2018.
20.
Klein  RJ, Schoenborn  CA.  Age adjustment using the 2000 projected U.S. population.  Healthy People 2000 Stat Notes. 2001;20(20):1-9.PubMedGoogle Scholar
21.
Korn  EL, Graubard  BI.  Confidence intervals for proportions with small expected number of positive counts estimated from survey data.  Surv Methodol. 1998;23:193-201.Google Scholar
22.
National Center for Health Statistics. National Center for Health Statistics data presentation standards for proportions. https://www.cdc.gov/nchs/data/series/sr_02/sr02_175.pdf. Accessed May 15, 2018.
23.
R package survey: analysis of complex survey samples [computer program]. Version 3.32. Vienna, Austria: R Foundation for Statistical Computing; 2017.
24.
Sobal  J, Troiano  R, Frongillo  E.  Rural-urban differences in obesity.  Rural Sociol. 1996;61(2):289-305. doi:10.1111/j.1549-0831.1996.tb00621.xGoogle ScholarCrossref
25.
Zhang  X, Holt  JB, Lu  H,  et al.  Neighborhood commuting environment and obesity in the United States: an urban-rural stratified multilevel analysis.  Prev Med. 2014;59:31-36. doi:10.1016/j.ypmed.2013.11.004PubMedGoogle ScholarCrossref
26.
Connor Gorber  S, Tremblay  M, Moher  D, Gorber  B.  A comparison of direct vs self-report measures for assessing height, weight and body mass index: a systematic review.  Obes Rev. 2007;8(4):307-326. doi:10.1111/j.1467-789X.2007.00347.xPubMedGoogle ScholarCrossref
27.
O’Connor  A, Wellenius  G.  Rural-urban disparities in the prevalence of diabetes and coronary heart disease.  Public Health. 2012;126(10):813-820. doi:10.1016/j.puhe.2012.05.029PubMedGoogle ScholarCrossref
28.
Boring  MA, Hootman  JM, Liu  Y,  et al.  Prevalence of arthritis and arthritis-attributable activity limitation by urban-rural county classification—United States, 2015.  MMWR Morb Mortal Wkly Rep. 2017;66(20):527-532. doi:10.15585/mmwr.mm6620a2PubMedGoogle ScholarCrossref
29.
Peen  J, Schoevers  RA, Beekman  AT, Dekker  J.  The current status of urban-rural differences in psychiatric disorders.  Acta Psychiatr Scand. 2010;121(2):84-93. doi:10.1111/j.1600-0447.2009.01438.xPubMedGoogle ScholarCrossref
30.
Goeres  LM, Gille  A, Furuno  JP,  et al.  Rural-urban differences in chronic disease and drug utilization in older Oregonians.  J Rural Health. 2016;32(3):269-279. doi:10.1111/jrh.12153PubMedGoogle ScholarCrossref
31.
Moy  E, Garcia  MC, Bastian  B,  et al.  Leading causes of death in nonmetropolitan and metropolitan areas—United States, 1999-2014.  MMWR Surveill Summ. 2017;66(1):1-8. doi:10.15585/mmwr.ss6601a1PubMedGoogle ScholarCrossref
32.
Ivey-Stephenson  AZCA, Crosby  AE, Jack  SPD, Haileyesus  T, Kresnow-Sedacca  MJ.  Suicide trends among and within urbanization levels by sex, race/ethnicity, age group, and mechanism of death—United States, 2001-2015.  MMWR Surveill Summ. 2017;66(18):1-16. doi:10.15585/mmwr.ss6618a1PubMedGoogle ScholarCrossref
33.
Singh  GK, Siahpush  M.  Widening rural-urban disparities in all-cause mortality and mortality from major causes of death in the USA, 1969-2009.  J Urban Health. 2014;91(2):272-292. doi:10.1007/s11524-013-9847-2PubMedGoogle ScholarCrossref
34.
Singh  GK, Siahpush  M.  Widening rural-urban disparities in life expectancy, US, 1969-2009.  Am J Prev Med. 2014;46(2):e19-e29. doi:10.1016/j.amepre.2013.10.017PubMedGoogle ScholarCrossref
35.
Douthit  N, Kiv  S, Dwolatzky  T, Biswas  S.  Exposing some important barriers to health care access in the rural USA.  Public Health. 2015;129(6):611-620. doi:10.1016/j.puhe.2015.04.001PubMedGoogle ScholarCrossref
36.
Doogan  NJ, Roberts  ME, Wewers  ME,  et al.  A growing geographic disparity: rural and urban cigarette smoking trends in the United States.  Prev Med. 2017;104(104):79-85. doi:10.1016/j.ypmed.2017.03.011PubMedGoogle ScholarCrossref
37.
Heo  M, Faith  MS, Pietrobelli  A, Heymsfield  SB.  Percentage of body fat cutoffs by sex, age, and race-ethnicity in the US adult population from NHANES 1999-2004.  Am J Clin Nutr. 2012;95(3):594-602. doi:10.3945/ajcn.111.025171PubMedGoogle ScholarCrossref
38.
Zheng  W, McLerran  DF, Rolland  B,  et al.  Association between body-mass index and risk of death in more than 1 million Asians.  N Engl J Med. 2011;364(8):719-729.PubMedGoogle ScholarCrossref
×