If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Requests for reprints should be addressed to Uri Ladabaum, MD, MS, Division of Gastroenterology and Hepatology, Stanford University School of Medicine, 300 Pasteur Drive, M211, Stanford, CA 94305.
Division of Gastroenterology and Hepatology, Stanford University School of Medicine, Stanford, CalifDepartment of Medicine, Stanford University School of Medicine, Stanford, Calif
Division of Gastroenterology and Hepatology, Stanford University School of Medicine, Stanford, CalifDepartment of Medicine, Stanford University School of Medicine, Stanford, Calif
Division of Gastroenterology and Hepatology, Stanford University School of Medicine, Stanford, CalifDepartment of Medicine, Stanford University School of Medicine, Stanford, Calif
Division of Gastroenterology and Hepatology, Stanford University School of Medicine, Stanford, CalifDepartment of Medicine, Stanford University School of Medicine, Stanford, Calif
Obesity and abdominal obesity are associated independently with morbidity and mortality. Physical activity attenuates these risks. We examined trends in obesity, abdominal obesity, physical activity, and caloric intake in US adults from 1988 to 2010.
Methods
Univariate and multivariate analyses were performed using National Health and Nutrition Examination Survey data.
Results
Average body mass index (BMI) increased by 0.37% (95% confidence interval [CI], 0.30-0.44) per year in both women and men. Average waist circumference increased by 0.37% (95% CI, 0.30-0.43) and 0.27% (95% CI, 0.22-0.32) per year in women and men, respectively. The prevalence of obesity and abdominal obesity increased substantially, as did the prevalence of abdominal obesity among overweight adults. Younger women experienced the greatest increases. The proportion of adults who reported no leisure-time physical activity increased from 19.1% (95% CI, 17.3-21.0) to 51.7% (95% CI, 48.9-54.5) in women, and from 11.4% (95% CI, 10.0-12.8) to 43.5% (95% CI, 40.7-46.3) in men. Average daily caloric intake did not change significantly. BMI and waist circumference trends were associated with physical activity level but not caloric intake. The associated changes in adjusted BMIs were 8.3% (95% CI, 6.9-9.6) higher among women and 1.7% (95% CI, 0.68-2.8) higher among men with no leisure-time physical activity compared with those with an ideal level of leisure-time physical activity.
Conclusions
Our analyses highlight important dimensions of the public health problem of obesity, including trends in younger women and in abdominal obesity, and lend support to the emphasis placed on physical activity by the Institute of Medicine.
Average body mass index and waist circumference, obesity and abdominal obesity prevalence, and the population fraction reporting no leisure-time physical activity increased substantially in US adults from 1988 to 2010.
•
Body mass index and waist circumference trends were associated with physical activity level but not daily caloric intake.
•
Although US obesity rates may be stabilizing, our results lend support to the emphasis placed on physical activity in the Institute of Medicine report on obesity.
The Institute of Medicine (IOM) identifies obesity as a health issue of monumental importance to the nation.
Obesity, defined as a body mass index (BMI) of ≥30 kg/m2, is associated with multiple comorbidities, including cardiovascular disease and cancer, and a higher risk of all-cause mortality.
Clinical Guidelines on the Identification, Evaluation, and Treatment of Overweight and Obesity in Adults: The Evidence Report. National Heart, Lung and Blood Institutes, National Institutes of Health,
Washington, DC1998
The National Institutes of Health recommends measuring waist circumference in addition to weight and height in primary care practice to help guide weight management.
The Practical Guide Identification, Evaluation, and Treatment of Overweight and Obesity in Adults. National Heart, Lung and Blood Institutes, National Institutes of Health,
Washington, DC2000
The Practical Guide Identification, Evaluation, and Treatment of Overweight and Obesity in Adults. National Heart, Lung and Blood Institutes, National Institutes of Health,
Washington, DC2000
Our aims were to characterize trends in and associations among overweight and obesity, abdominal obesity, physical activity, and caloric intake in US adults in the last 2 decades. We explored trends in the prevalence of abdominal obesity within strata defined by BMI, given the health risks of increased waist circumference independent of BMI, and we explored the relationships between measures of obesity and levels of physical activity and caloric intake.
Methods
Data Source
We used data from the National Health and Nutrition Examination Survey (NHANES), beginning with NHANES-III (1988-1994) and including the continuous NHANES 2-year survey cycles from 1999-2000 to 2009-2010.
Centers for Disease Control and Prevention (CDC). National Center for Health Statistics (NCHS). National Health and Nutrition Examination Survey Data. Hyattsville, MD. Available at: http://www.cdc.gov/nchs/nhanes/nhanes_questionnaires.htm. Accessed April 17, 2013.
National Center for Health Statistics. Analytical and Reporting Guidelines. The Third National Health and Nutrition Examination Survey, 1988–1994, Hyattsville, MD. 1996. Available at: http://www.cdc.gov/nchs/data/nhanes/nhanes3/nh3gui.pdf. Accessed April 17, 2013.
All examined participants aged ≥18 years, except pregnant women, were included. For the caloric intake analyses, only eligible participants who had reliable and complete 24-hour dietary recall data were included. Subgroup analyses were performed for subgroups of self-identified race/ethnicity with sufficiently large samples: non-Hispanic white, non-Hispanic black, and Mexican-American.
Centers for Disease Control and Prevention (CDC). National Center for Health Statistics (NCHS). National Health and Nutrition Examination Survey Data. Hyattsville, MD. Available at: http://www.cdc.gov/nchs/nhanes/nhanes_questionnaires.htm. Accessed April 17, 2013.
During the examination component of NHANES, height, weight, and waist circumference were measured. On the basis of BMI, individuals were categorized as normal weight (18.5-24.99 kg/m2), overweight (25.0-29.99 kg/m2), or obese (≥30 kg/m2).
Clinical Guidelines on the Identification, Evaluation, and Treatment of Overweight and Obesity in Adults: The Evidence Report. National Heart, Lung and Blood Institutes, National Institutes of Health,
Washington, DC1998
Clinical Guidelines on the Identification, Evaluation, and Treatment of Overweight and Obesity in Adults: The Evidence Report. National Heart, Lung and Blood Institutes, National Institutes of Health,
Washington, DC1998
Definition of Leisure-time Physical Activity Levels
Participants were grouped into 3 levels of leisure-time physical activity: ideal, intermediate, and none. This was accomplished by following the methods and definitions of the specific NHANES surveys (Appendix, online).
Definition of Dietary Measures
Total caloric intake per day was estimated on the basis of the total energy intake (kcal, or calories) reported in the first 24-hour recall data set. These data were obtained during the examination component of NHANES by trained dietary interviewers using an automated dietary interview system.
National Center for Health Statistics. Analytical and Reporting Guidelines. The Third National Health and Nutrition Examination Survey, 1988–1994, Hyattsville, MD. 1996. Available at: http://www.cdc.gov/nchs/data/nhanes/nhanes3/nh3gui.pdf. Accessed April 17, 2013.
Analyses were performed for all adults and after stratification by gender; predefined age strata of 18 to 39 years, 40 to 64 years, and ≥65 years; and race/ethnicity. Data from each cycle were considered to be from the midpoint of each period.
To examine trends in BMI, waist circumference, and daily energy intake, we calculated national estimates of age-adjusted averages and 95% confidence intervals (CIs) for each survey cycle. We used linear regression to assess trends in log-transformed BMI, waist circumference, and daily energy intake by including the midpoints of each survey cycle as a continuous variable. To address possible changes in BMI trends over time, we used joinpoint analysis (Appendix, online).
The results are expressed as annual percentage change.
To examine trends in the prevalence of obesity and abdominal obesity, and the levels of physical activity, we calculated national estimates of age-adjusted proportions and 95% CIs for each survey cycle. We assessed changes over time using logistic regression analysis by including the midpoints of each survey cycle as a continuous variable. We present odds ratios and 95% CIs reflecting the odds of fulfilling criteria for a specific category in a given year versus the previous year. We performed separate analyses to examine trends in abdominal obesity for overweight and obese adults, defined by BMI.
Analyses stratified by 3 prespecified levels of leisure-time physical activity were performed for trends in the measures of obesity and daily energy intake. To examine the independent effects of physical activity and daily energy intake, we performed multivariate regression analysis with log-transformed BMI and waist circumference as response variables, and year, physical activity level, and daily energy intake as explanatory variables, adjusted for age, age squared, and gender. For sensitivity analyses, we first used only the continuous NHANES data, in which total minutes spent on physical activity can be estimated; and second, we used only data from 1999-2000 to 2005-2006, including activity related to transportation and domestic work, because the physical activity questionnaire was consistent in these cycles (Appendix).
All analyses were performed with appropriate survey procedures in SAS v9.3 (SAS Institute Inc, Cary, NC). Survey design variables and weight variables were used to account for the complexity of the NHANES survey design with stratification and clustering. Averages and prevalence rates were age-adjusted to the year 2000 US standard population.
Population Projections of the United States by Age, Sex, Race, and Hispanic Origin: 1995 to 2050, U.S. Bureau of the Census, Current Population Reports, P25-1130.
U.S. Government Printing Office,
Washington DC1996
From 1988-1994 to 2009-2010, the average BMI in the United States increased by 0.37% (95% CI, 0.30-0.44) per year in both women and men. Average BMI increased in all age and racial/ethnic groups (Figure 1, Table 1). The age subgroup with the greatest annual increase was 18- to 39-year-old women (Table 1). Joinpoint analyses identified a change in the trend for average BMI only in women, with annual increases of 0.60% (95% CI, 0.50-0.69) from 1991 to 2002, and smaller annual increases of 0.22% (95% CI, 0.11-0.32) from 2002 to 2010. Among women, average BMI was highest among non-Hispanic blacks and lowest among non-Hispanic whites; among men, it was similar across racial/ethnic groups (Figure 1, Table 1, Appendix Table 1 [online]).
Figure 1Trends in average BMI in US adults, 1988-2010. (A) Trends in women by age group. (B) Trends in men by age group. (C) Trends in women by racial/ethnic group. (D) Trends in men by racial/ethnic group. BMI = body mass index.
The prevalence of obesity increased from 24.9% (95% CI, 23.1-26.8) to 35.4% (95% CI, 33.6-37.3) in women, and from 19.9% (95% CI, 18.4-21.3) to 34.6% (95% CI, 31.0-38.1) in men. These increases were associated with decreases in the proportions of normal-weight women and men, and no change in the proportions of overweight women and men (Table 1). Compared with the previous year, the odds ratio for being categorized as obese in any given year was 1.026 (95% CI, 1.019-1.033) for women and 1.042 (95% CI, 1.033-1.050) for men. Similar trends were observed in all age groups (Table 1). Sex-specific differences between racial/ethnic groups in obesity prevalence paralleled the BMI results (Table 1, Appendix Figure 1 [online], Appendix Table 1 [online]).
Waist Circumference and Abdominal Obesity
The average waist circumference increased by 0.37% (95% CI, 0.30-0.44) per year in women and 0.27% (95% CI, 0.22-0.32) per year in men. Increases in average waist circumference were observed in all age and racial/ethnic groups (Figure 2, Table 1). As with BMI, the subgroup with the greatest annual increase in average waist circumference was 18- to 39-year-old women (Table 1). Among women, average waist circumference was largest among non-Hispanic blacks and smallest among non-Hispanic whites; in contrast, among men, it was largest among non-Hispanic whites and smallest among non-Hispanic blacks (Figure 2, Table 1, Appendix Table 1 [online]).
Figure 2Trends in average waist circumference in US adults, 1988-2010. (A) Trends in women by age group. (B) Trends in men by age group. (C) Trends in women by racial/ethnic group. (D) Trends in men by racial/ethnic group.
The changes in average waist circumference were associated with increases in the prevalence of abdominal obesity (Appendix Figure 2, Table 1), from 46.0% (95% CI, 43.9-48.0) to 61.5% (95% CI, 58.8-64.2) in women, and from 29.1% (95% CI, 27.4-30.7) to 42.0% (95% CI, 38.6-45.3) in men. The annual increases were most pronounced in younger persons, but the prevalence of abdominal obesity increased with age (Table 1). Sex-specific differences between racial/ethnic groups in abdominal obesity prevalence paralleled the waist circumference results (Table 1, Appendix Figure 2 [online], Appendix Table 1 [online]).
Abdominal Obesity by Body Mass Index Category
Among adults categorized as normal weight on the basis of BMI, the prevalence of abdominal obesity increased significantly only in women, from 9.6% (95% CI, 8.1-11.0) to 13.8% (95% CI, 10.1-17.5) (Appendix Table 2 [online]).
Among adults categorized as overweight on the basis of BMI, the prevalence of abdominal obesity increased from 66.1% (95% CI, 62.7-69.4) to 80.8% (95% CI, 78.1-83.5) in women, and from 24.5% (95% CI, 22.1-26.8) to 27.8% (95% CI, 24.3-31.3) in men (Appendix Table 2, online). Compared with the previous year, the odds ratio for fulfilling criteria for abdominal obesity in a given year was 1.045 (95% CI, 1.032-1.057) for overweight women and 1.017 (95% CI, 1.007-1.027) for overweight men. The majority (>90%) of US adults categorized as obese on the basis of BMI also were categorized as having abdominal obesity (Appendix Table 2, online). Appendix Table 2 (online) displays the results by age group.
Physical Activity
The proportion of adults who reported no leisure-time physical activity increased from 19.1% (95% CI, 17.3-21.0) to 51.7% (95% CI, 48.9-54.5) in women, and from 11.4% (95% CI, 10.0-12.8) to 43.5% (95% CI, 40.7-46.3) in men. These changes were accompanied by decreases in the proportions of adults who reported intermediate or ideal levels of leisure-time physical activity (Table 2). Compared with the previous year, the odds ratio for reporting no leisure-time physical activity in a given year was 1.074 (95% CI, 1.064-1.084) for women and 1.083 (95% CI, 1.071-1.094) for men (Table 2). Among both women and men, the proportions reporting no leisure-time physical activity were higher among non-Hispanic blacks and Mexican Americans than among non-Hispanic whites (Table 2). Table 2 and Appendix Table 3 (online) display the results by age and race/ethnicity.
Table 2Trends in Reported Levels of Leisure-time Physical Activity and Daily Caloric Intake, US Adults, 1988-2010
Average daily caloric intake did not change significantly over time for women or men as a whole or when stratified by race/ethnicity (Table 2). The results differed by age, gender, and leisure-time physical activity level (Table 3, Appendix Table 4 [online]). Average daily caloric intake increased significantly for women who reported no leisure-time physical activity and decreased significantly for men who reported an ideal level of leisure-time physical activity (Table 3).
Table 3Trends in Prevalence of Obesity, Abdominal Obesity, and Daily Caloric Intake Stratified by Reported Level of Leisure-time Physical Activity, US Adults, 1988-2010
Category Based on BMI, OR (95% CI) of Being in Category in Any Given Year Compared with Previous Year
Category Based on Waist Circumference, OR (95% CI) of Being in Category in Any Given Year Compared with Previous Year
Mean Energy Intake, APC (95% CI)
Obese
Overweight
Normal Weight
Abdominal Obesity
Women
No reported leisure-time physical activity
1.020 (1.010-1.030)
0.997 (0.985-1.009)
0.987 (0.978-0.995)
1.026 (1.015-1.038)
0.27 (0.02-0.52)
Intermediate level of reported leisure-time physical activity
1.029 (1.018-1.040)
1.003 (0.993-1.014)
0.974 (0.963-0.985)
1.037 (1.027-1.048)
0.09 (−0.11-0.28)
Ideal level of reported leisure-time physical activity
1.020 (1.008-1.032)
1.013 (1.002-1.025)
0.977 (0.966-0.989)
1.037 (1.025-1.048)
0.08 (−0.13-0.30)
Men
No reported leisure-time physical activity
1.053 (1.037-1.070)
1.003 (0.991-1.016)
0.947 (0.933-0.962)
1.048 (1.032-1.065)
0.19 (−0.14-0.53)
Intermediate level of reported leisure-time physical activity
1.029 (1.016-1.041)
1.004 (0.993-1.014)
0.971 (0.959-0.983)
1.029 (1.018-1.039)
−0.13 (−0.34-0.09)
Ideal level of reported leisure-time physical activity
1.045 (1.030-1.060)
1.000 (0.992-1.008)
0.967 (0.955-0.978)
1.036 (1.024-1.048)
−0.22 (−0.40 to −0.04)
APC = annual percentage change; BMI = body mass index; CI = confidence interval; OR = odds ratio.
Body Mass Index and Obesity by Physical Activity Strata
Average BMI increased in women and men in all 3 strata of leisure-time physical activity (Appendix Table 4 [online]). Increases in average BMI were observed in nearly all age and gender subgroups stratified by leisure-time physical activity (Appendix Table 4 [online]). The prevalence of obesity increased in women and men in all 3 strata of leisure-time physical activity (Table 3). Similar patterns were seen across racial/ethnic groups (data not shown).
Waist Circumference and Abdominal Obesity by Physical Activity Strata
Average waist circumference increased over time in women and men in all 3 strata of leisure-time physical activity (Appendix Table 4 [online]). Increases in average waist circumference were observed in nearly all age and gender subgroups stratified by leisure-time physical activity (Appendix Table 4 [online]). The prevalence of abdominal obesity increased in women and men within all strata of leisure-time physical activity (Table 3). Similar patterns were seen across racial/ethnic groups (data not shown).
Body Mass Index and Waist Circumference: Associations with Leisure-time Physical Activity and Caloric Intake
BMI and waist circumference both increased significantly in women and men after adjusting for leisure-time physical activity, caloric intake, and age. The magnitude of the increases was associated with level of leisure-time physical activity, but not with caloric intake.
The associated changes in BMI were 5.5% (95% CI, 4.3-6.7) higher among women with an intermediate level of leisure-time physical activity and 8.3% (95% CI, 6.9-9.6) higher among women with no leisure-time physical activity than among women with an ideal level of leisure-time physical activity. For men, the respective measures were 1.2% (95% CI, 0.15-2.2) higher and 1.7% (95% CI, 0.68-2.8) higher.
The associated changes in waist circumferences were 4.2% (95% CI, 3.3-5.0) higher among women with an intermediate level of leisure-time physical activity and 6.4% (95% CI, 5.4-7.4) higher among women with no leisure-time physical activity than among women with an ideal level of leisure-time physical activity. For men, the respective measures were 2.1% (95% CI, 1.4-2.9) higher and 2.5% (95% CI, 1.7-3.3) higher.
In analyses stratified by race/ethnicity, the associations between increases in BMI and waist circumference and level of leisure-time physical activity were statistically significant for non-Hispanic white women and men, and Mexican American women (Appendix Table 5 [online]). Sensitivity analyses showed similar results (Appendix [online]).
Discussion
Our analyses highlight important dimensions of the public health problem of obesity in the United States. The average BMI, average waist circumference, prevalence of obesity, and prevalence of abdominal obesity increased substantially over the past 2 decades in both women and men. The annual increases in BMI have been most pronounced in younger women, the greatest increases in the prevalence of abdominal obesity have been in younger women and men, and both obesity and abdominal obesity preferentially affected non-Hispanic black women. Furthermore, the prevalence of abdominal obesity has increased among normal-weight women and overweight women and men. These changes have occurred in the context of substantial increases in the proportion of adults reporting no leisure-time physical activity, but in the absence of any significant population-level changes in average daily caloric intake. At the population level, we found a significant association between the level of leisure-time physical activity, but not daily caloric intake, and the increases in both BMI and waist circumference.
Although the overall trends in obesity in the United States are well appreciated and obesity prevalence may be stabilizing,
our analyses highlight troublesome trends in younger adults, in women, and in abdominal obesity prevalence, as well as persistent racial/ethnic disparities. Overall, normal-weight women and overweight women and men have become more “abdominally obese.” Given that obesity prevalence increased significantly between 1999 and 2010 in male subjects aged 2 to 19 years, but not in female subjects,
it is possible that young adult women may be particularly vulnerable to weight gain. It remains controversial whether overweight alone increases mortality risk,
but the trends in abdominal obesity among the overweight are concerning in light of the risks associated with increased waist circumference independent of BMI.
Clinical Guidelines on the Identification, Evaluation, and Treatment of Overweight and Obesity in Adults: The Evidence Report. National Heart, Lung and Blood Institutes, National Institutes of Health,
Washington, DC1998
Our findings do not support the popular notion that the increase in obesity in the United States can be attributed primarily to sustained increases over time in the average daily caloric intake of Americans.
We found no evidence that average daily caloric intake has increased over the last 2 decades. Previous analyses have noted increases over longer time periods,
In contrast, our analyses identified significant associations between the reported level of leisure-time physical activity in the population and all measures of obesity. Other investigators have examined this relationship.
However, an ideal level of physical activity does not by itself ensure a normal weight (Table 3).
Our results are based on a series of cross-sectional surveys, and not a prospective cohort study. They do not suggest that caloric intake is not an important determinant of obesity at the level of an individual. A previous analysis of NHANES from 1971-1975 to 1999-2002 reported an increase in food energy density approximately parallel to US obesity prevalence rates.
Diet composition may be relevant at the population level, but we found no substantial changes in the daily fat, carbohydrate, and protein intake during our study period (Appendix Table 6 [online]). However, analyses of the major macronutrients do not address specific sources of calories, such as simple sugars, and interactions with genetic risk.
Our results should be viewed in the context of existing studies. Studies restricted to more recent time periods suggest that the trends in obesity prevalence in US adults may be stabilizing.
Our joinpoint analyses of BMI in women are consistent with these results. Although this is encouraging, the prevalence of obesity remains unacceptably high.
Pleis JR, Lucas JW. Summary Health Statistics for U.S. Adults: National Health Interview Survey, 2007. Vital and Health Statistics Series 10, Data from the National Health Survey. Hyattsville MD: U.S. Department of Health and Human Services; 2009:1-159.
The heterogeneity in results may be due in part to diversity in surveys and study methods.
Study Limitations
The NHANES surveys do not follow a cohort of individuals prospectively, so our reported associations with physical activity must be interpreted with caution. Although height, weight, and waist circumference are measured, caloric intake and physical activity are self-reported, and we analyzed caloric intake on 24-hour recall. Caloric intake may be under-reported and physical activity overestimated,
but this cannot explain the high rates of no leisure-time physical activity in our study. Changes in survey methodology over time could have affected physical activity estimates. We used cycle midpoints as a continuous variable.
Societal considerations merit discussion. Obesity is a public health problem across racial, ethnic, and socioeconomic groups,
Societal realities, including divergent views about the roles of individuals and government, compound the challenges encountered in the medical setting.
The IOM also recommends transforming messages about physical activity and nutrition, stressing the roles of health care providers, insurers, employers, and schools in promoting health, with physical activity as a major focus.
Average BMI and waist circumference both increased among US adults from 1988 to 2010. Abdominal obesity, an independent predictor of mortality, has become more prevalent among normal-weight women and overweight women and men. At the population level, we found associations between the reported level of leisure-time physical activity and obesity, but even among adults reporting an ideal level of leisure-time physical activity, obesity prevalence has increased. Although it is encouraging that obesity rates may be stabilizing, our results lend support to the emphasis placed on physical activity in the IOM report on obesity.
Appendix. Definition of Leisure-time Physical Activity Levels
During the home interview component of NHANES, a physical activity questionnaire was administered. In NHANES-III, participants were asked to record the frequency of leisure-time activities in the past month in terms of times per month. Each activity was given an intensity rating, reflecting the ratio of energy expenditure relative to that of the resting metabolic rate in metabolic equivalent of task (MET). In NHANES 1999-2000 to 2005-2006, participants were asked to report the frequency, duration, and level of exertion for each of the reported leisure-time activities in the past month. MET scores were assigned for each activity. From NHANES 2007-2008 onward, participants were asked to report the number of days per week that they engaged in moderate or vigorous leisure-time activities and the duration of these activities per day. The MET score thresholds used to define moderate and vigorous activity were 4 and 8, respectively.
To define the physical activity levels of participants, different approaches were followed for NHANES-III and the continuous NHANES waves, because NHANES-III did not consider duration of the activities. For NHANES-III, based on the MET scores, the intensity of activities was defined as moderate (3-6 METS) or vigorous (>6 METS). An ideal level of physical activity was defined as engaging in moderate activities ≥5 times per week or in vigorous activities ≥3 times per week.
An intermediate level of physical activity was defined as engaging in at least some leisure-time physical activities without meeting the criteria for the ideal group.
For continuous NHANES surveys, the number of minutes of moderate activity per week and the number of minutes of vigorous activity per week were calculated on the basis of the reported frequency, intensity, and duration of activities. An ideal level of physical activity was defined as ≥150 minutes per week of moderate activities, ≥75 minutes per week of vigorous activities, or ≥150 minutes per week of a combination of moderate and vigorous activities.
An intermediate level of physical activity was defined as 1 to 149 minutes per week of moderate activities, 1 to 74 minutes per week of vigorous activities, or 1 to 149 minutes per week of a combination of moderate and vigorous activities.
Trends in Body Mass Index: Joinpoint Analyses
Trends in age-adjusted mean BMI over time were analyzed using Joinpoint Regression Program version 4.0.1 from the National Cancer Institute,
JoinPoint Regression Program, Version 4.0.1-January 2013; Statistical Methodology and Applications Branch and Data Modeling Branch, Surveillance Research Program National Cancer Institute. Available at: http://surveillance.cancer.gov/joinpoint/. Accessed April 17, 2013.
Joinpoint regression methodology is widely used to identify statistically significant changes in trends. The joinpoint program determines the number of joinpoints (a time point at which the direction or magnitude of trend changes), tests whether the change in the trend is significant by performing permutation tests, and fits a log-linear model to each segment. The specifications set in advance were as follows: The maximum number of joinpoints was set to 1, and the minimum number of observations for a joinpoint from the beginning or end of data series was set to 3. The slope of a line segment is described as the Annual Percentage Change (APC).
Sensitivity Analyses: Effect of Physical Activity on Body Mass Index and Waist Circumference
Methods
To further tease out the effect of physical activity on BMI and waist circumference, we performed a sensitivity analysis using only the continuous NHANES data, in which the number of minutes of participation in physical activity are available. To estimate total minutes spent on physical activity, we calculated intensity-adjusted leisure-time physical activity minutes per week on the basis of the recommendations given in “Physical Activity Guidelines for Americans: Appendix 1” (http://www.health.gov/paguidelines/guidelines/appendix1.aspx), which states that “1 minute of vigorous-intensity activity counts the same as 2 minutes of moderate-intensity activity.” The intensity-adjusted total minutes were top-truncated to 1680 minutes/week. We performed the multivariate regression analysis with log-transformed BMI and waist circumference as the response variables, and year, intensity-adjusted leisure-time physical activity minutes per week, and calorie intake as the explanatory variables, and adjusted for age, age squared, and gender. Analyses were performed for all adults and after stratification by gender.
To compare the trends with any physical activity or leisure-time physical activity as an explanatory variable, we performed a subanalysis using only NHANES data from 1999-2000 to 2005-2006 because the physical activity questionnaires were consistent in these survey cycles. Any physical activity includes walking or bicycling for transport, activities around the home or yard information, and leisure-time physical activity.
Results
The results of these sensitivity analyses were similar to the results of the primary analyses. BMI increased significantly over the years in both women (APC, 0.21; 95% CI, 0.03-0.38) and men (APC, 0.35; 95% CI, 0.17-0.53) after adjusting for physical activity, caloric intake, and age. The level of physical activity was associated with the degree of increase in BMI. In contrast, the effect of caloric intake was not significant.
In women, BMI decreased by 0.94% (95% CI, 0.76-1.1) for every 100 minutes/week increase in physical activity. In men, BMI decreased by 0.24% (95% CI, 0.11-0.36) for every 100 minutes/week increase in physical activity.
Waist circumference increased significantly over the years in both women (APC, 0.31; 95% CI, 0.15-0.46) and men (APC, 0.19; 95% CI, 0.06-0.32) after adjusting for physical activity, caloric intake, and age. The level of physical activity was associated with the degree of increase in waist circumference. In contrast, the effect of caloric intake was not significant.
In women, waist circumference decreased by 0.79% (95% CI, 0.66-0.91) for every 100 minutes/week increase in physical activity. In men, waist circumference decreased by 0.32% (95% CI, 0.23-0.42) for every 100 minutes/week increase in physical activity.
In the analyses restricted to NHANES data from 1999-2000 to 2005-2006, average BMI, average waist circumference, prevalence of obesity, and prevalence of abdominal obesity all increased significantly only in the subgroup with no physical activity. The trends were comparable with any physical activity or leisure-time physical activity as an explanatory variable.
Appendix Table 1Trends in Average Body Mass Index, Waist Circumference, and Prevalence of Obesity and Abdominal Obesity, US Adults Stratified by Race/Ethnicity, 1988-2010
BMI and Waist Circumference
Category Based on BMI
Category Based on Waist Circumference
Mean BMI in kg/m2 (95% CI)
Mean waist Circumference in cm (95% CI)
Obese, % of Population (95% CI)
Overweight, % of Population (95% CI)
Normal Weight, % of Population (95% CI)
Abdominal Obesity, % of Population (95% CI)
Women
18-39 y, non-Hispanic white
1988-1994
24.6 (24.1-25.0)
82.0 (80.9-83.2)
17.7 (14.9-20.5)
16.4 (14.2-18.7)
60.5 (57.2-63.8)
25.9 (21.8-30.0)
2009-2010
27.1 (26.4-27.8)
90.4 (88.3-92.5)
26.6 (22.6-30.7)
23.4 (16.6-30.3)
46.4 (40.5-52.4)
46.3 (39.7-52.9)
Trend
APC, 0.48 (0.32-0.64)
APC, 0.52 (0.39-0.65)
OR, 1.031 (1.015-1.048)
OR, 1.019 (1.004-1.035)
OR, 0.970 (0.958-0.983)
OR, 1.048 (1.032-1.065)
40-64 y, non-Hispanic white
1988-1994
27.3 (26.8-27.8)
91.5 (90.3-92.6)
27.5 (24.0-31.1)
28.8 (26.1-31.5)
41.0 (37.4-44.7)
52.7 (49.6-55.8)
2009-2010
28.5 (27.9-29.1)
96.4 (95.1-97.7)
35.6 (30.7-40.5)
28.1 (23.0-33.2)
33.0 (29.0-36.9)
66.0 (61.7-70.4)
Trend
APC, 0.25 (0.12-0.38)
APC, 0.30 (0.20-0.40)
OR, 1.021 (1.007-1.035)
OR, 0.998 (0.985-1.012)
OR, 0.981 (0.970-0.993)
OR, 1.032 (1.021-1.044)
≥65 y, non-Hispanic white
1988-1994
26.3 (26.1-26.6)
92.9 (91.8-93.9)
22.4 (20.0-24.8)
32.5 (30.3-34.7)
41.5 (38.3-44.6)
63.1 (59.1-67.1)
2009-2010
29.0 (28.2-29.8)
98.4 (97.2-99.6)
37.6 (32.4-42.8)
32.0 (26.5-37.5)
28.3 (24.6-32.0)
73.6 (70.3-76.9)
Trend
APC, 0.38 (0.27-0.50)
APC, 0.26 (0.17-0.35)
OR, 1.028 (1.015-1.041)
OR, 1.006 (0.994-1.017)
OR, 0.975 (0.964-0.986)
OR, 1.026 (1.011-1.041)
18-39 y, non-Hispanic black
1988-1994
27.4 (26.8-27.9)
87.6 (86.3-88.9)
28.7 (25.4-31.9)
25.9 (23.4-28.4)
41.8 (37.7-45.9)
42.1 (38.1-46.2)
2009-2010
31.1 (29.5-32.7)
96.5 (92.2-100.7)
54.1 (44.3-63.8)
18.8 (13.5-24.0)
23.6 (17.2-30.0)
65.7 (54.3-77.1)
Trend
APC, 0.64 (0.44-0.84)
APC, 0.51 (0.35-0.66)
OR, 1.051 (1.034-1.068)
OR, 0.990 (0.977-1.004)
OR, 0.952 (0.937-0.968)
OR, 1.049 (1.030-1.068)
40-64 y, non-Hispanic black
1988-1994
30.3 (29.7-30.9)
98.0 (96.5-99.5)
46.8 (42.1-51.4)
30.0 (26.5-33.6)
20.9 (16.8-24.9)
72.3 (67.8-76.9)
2009-2010
33.8 (32.3-35.2)
104.6 (102.8-106.3)
61.4 (55.0-67.7)
27.3 (22.0-32.5)
10.4 (6.8-14.0)
83.9 (76.7-91.1)
Trend
APC, 0.39 (0.21-0.57)
APC, 0.29 (0.17-0.41)
OR, 1.024 (1.009-1.038)
OR, 0.991 (0.977-1.005)
OR, 0.974 (0.956-0.993)
OR, 1.024 (1.002-1.047)
≥65 y, non-Hispanic black
1988-1994
29.1 (28.4-29.8)
97.9 (96.1-99.7)
36.3 (31.6-40.9)
36.1 (31.3-40.8)
25.1 (21.5-28.8)
75.7 (70.8-80.7)
2009-2010
31.4 (29.9-33.0)
100.7 (98.1-103.4)
55.6 (44.4-66.7)
25.8 (16.1-35.5)
17.1 (9.8-24.5)
81.9 (72.0-91.8)
Trend
APC, 0.35 (0.12-0.57)
APC, 0.18 (0.03-0.32)
OR, 1.042 (1.020-1.065)
OR, 0.968 (0.947-0.990)
OR, 0.983 (0.961-1.006)
OR, 1.024 (0.995-1.054)
18-39 y, Mexican American
1988-1994
26.9 (26.5-27.3)
86.7 (85.8-87.7)
26.6 (23.4-29.8)
30.2 (25.8-34.6)
40.5 (37.2-43.7)
42.5 (39.9-45.2)
2009-2010
28.8 (27.9-29.6)
94.3 (91.8-96.8)
36.8 (32.1-41.6)
29.3 (24.2-34.5)
33.2 (26.8-39.6)
60.5 (51.3-69.7)
Trend
APC, 0.40 (0.24-0.57)
APC, 0.51 (0.38-0.65)
OR, 1.031 (1.014-1.049)
OR, 0.998 (0.983-1.014)
OR, 0.976 (0.961-0.990)
OR, 1.051 (1.031-1.071)
40-64 y, Mexican American
1988-1994
29.7 (29.0-30.3)
96.4 (95.2-97.6)
43.6 (39.3-48.0)
33.7 (30.2-37.3)
22.2 (17.3-27.2)
74.4 (70.6-78.2)
2009-2010
31.0 (30.1-31.9)
100.0 (97.5-102.6)
54.2 (45.7-62.7)
36.0 (27.4-44.6)
9.4 (5.5-13.2)
84.6 (78.6-90.6)
Trend
APC, 0.21 (0.02-0.40)
APC, 0.19 (0.04-0.35)
OR, 1.022 (1.001-1.044)
OR, 1.000 (0.980-1.021)
OR, 0.964 (0.943-0.986)
OR, 1.029 (1.004-1.054)
≥65 y, Mexican American
1988-1994
28.0 (27.1-28.9)
97.0 (94.5-99.6)
29.7 (23.1-36.4)
38.8 (31.1-46.5)
30.3 (23.0-37.6)
72.7 (64.5-80.9)
2009-2010
29.3 (28.6-29.9)
98.8 (97.5-100.1)
35.7 (27.0-44.4)
41.0 (31.0-51.0)
21.6 (15.8-27.3)
82.2 (73.2-91.2)
Trend
APC, 0.26 (0.04-0.48)
APC, 0.18 (−0.01-0.37)
OR, 1.017 (0.991-1.043)
OR, 1.006 (0.981-1.031)
OR, 0.975 (0.951-1.001)
OR, 1.039 (1.001-1.078)
Men
18-39 y, non-Hispanic white
1988-1994
25.7 (25.3-26.1)
91.0 (90.2-91.9)
14.4 (12.1-16.7)
34.7 (31.3-38.1)
49.1 (45.8-52.4)
15.2 (12.8-17.5)
2009-2010
27.8 (27.0-28.5)
96.2 (94.1-98.2)
33.1 (25.7-40.4)
30.4 (26.7-34.0)
35.4 (28.3-42.5)
30.6 (24.0-37.1)
Trend
APC, 0.35 (0.23-0.47)
APC, 0.26 (0.18-0.35)
OR, 1.051 (1.034-1.069)
OR, 0.994 (0.983-1.006)
OR, 0.972 (0.960-0.984)
OR, 1.045 (1.030-1.060)
40-64 y, non-Hispanic white
1988-1994
27.6 (27.2-27.9)
99.8 (99.1-100.5)
25.5 (22.8-28.2)
44.1 (40.8-47.3)
29.8 (26.3-33.3)
38.7 (35.2-42.2)
2009-2010
29.3 (28.7-29.9)
104.2 (102.7-105.7)
36.6 (31.7-41.5)
44.5 (38.3-50.8)
18.1 (12.9-23.3)
51.3 (45.9-56.7)
Trend
APC, 0.33 (0.22-0.43)
APC, 0.25 (0.18-0.33)
OR, 1.030 (1.018-1.043)
OR, 1.000 (0.987-1.012)
OR, 0.963 (0.948-0.978)
OR, 1.033 (1.020-1.047)
≥65 y, non-Hispanic white
1988-1994
26.7 (26.4-27.1)
101.0 (100.2-101.9)
20.5 (17.3-23.8)
45.3 (41.5-49.0)
32.5 (28.9-36.1)
46.7 (43.2-50.2)
2009-2010
28.9 (28.4-29.4)
106.7 (105.6-107.8)
37.3 (31.3-43.3)
39.3 (33.9-44.8)
22.3 (18.8-25.8)
62.1 (57.6-66.6)
Trend
APC, 0.40 (0.30-0.49)
APC, 0.30 (0.24-0.37)
OR, 1.043 (1.027-1.059)
OR, 0.992 (0.979-1.004)
OR, 0.970 (0.958-0.982)
OR, 1.040 (1.030-1.051)
18-39 y, non-Hispanic black
1988-1994
26.1 (25.7-26.4)
87.8 (87.0-88.6)
18.8 (16.7-21.0)
31.1 (28.1-34.0)
48.3 (44.8-51.8)
13.9 (12.2-15.5)
2009-2010
28.9 (27.6-30.2)
92.8 (90.2-95.4)
35.2 (26.6-43.9)
27.1 (20.3-33.8)
35.7 (30.2-41.1)
28.0 (19.8-36.1)
Trend
APC, 0.52 (0.36-0.68)
APC, 0.36 (0.24-0.48)
OR, 1.052 (1.035-1.071)
OR, 0.995 (0.981-1.008)
OR, 0.967 (0.955-0.979)
OR, 1.052 (1.033-1.072)
40-64 y, non-Hispanic black
1988-1994
27.0 (26.7-27.4)
95.3 (94.2-96.5)
22.4 (18.8-25.9)
41.6 (37.7-45.6)
34.4 (30.9-38.0)
27.9 (24.2-31.6)
2009-2010
30.0 (29.0-30.9)
102.3 (100.1-104.4)
42.9 (36.8-49.0)
31.7 (26.1-37.2)
24.8 (19.9-29.7)
41.1 (35.5-46.6)
Trend
APC, 0.49 (0.33-0.65)
APC, 0.33 (0.21-0.44)
OR, 1.053 (1.034-1.072)
OR, 0.981 (0.967-0.994)
OR, 0.970 (0.956-0.985)
OR, 1.035 (1.020-1.051)
≥65 y, non-Hispanic black
1988-1994
26.0 (25.4-26.6)
96.9 (95.4-98.3)
20.7 (16.5-25.0)
35.9 (31.5-40.2)
38.4 (33.1-43.8)
34.2 (29.2-39.3)
2009-2010
28.0 (26.8-29.1)
102.6 (99.6-105.6)
32.9 (23.6-42.2)
34.7 (28.5-41.0)
30.4 (23.6-37.2)
54.8 (46.8-62.9)
Trend
APC, 0.48 (0.27-0.70)
APC, 0.37 (0.21-0.52)
OR, 1.045 (1.020-1.071)
OR, 0.995 (0.979-1.012)
OR, 0.971 (0.950-0.992)
OR, 1.052 (1.030-1.074)
18-39 y, Mexican American
1988-1994
26.0 (25.6-26.5)
89.8 (88.6-90.9)
15.0 (11.8-18.2)
40.6 (37.1-44.2)
43.1 (38.9-47.3)
13.5 (9.9-17.0)
2009-2010
28.4 (27.3-29.6)
96.6 (93.8-99.4)
30.5 (23.5-37.4)
44.5 (42.0-46.9)
24.4 (17.6-31.2)
31.8 (23.0-40.6)
Trend
APC, 0.50 (0.33-0.68)
APC, 0.40 (0.27-0.54)
OR, 1.048 (1.024-1.072)
OR, 1.010 (0.997-1.024)
OR, 0.954 (0.937-0.971)
OR, 1.051 (1.026-1.077)
40-64 y, Mexican American
1988-1994
28.5 (28.1-28.9)
100.0 (98.9-101.1)
33.4 (28.9-37.8)
45.0 (40.7-49.2)
21.1 (17.9-24.4)
40.5 (35.9-45.1)
2009-2010
29.5 (29.0-30.0)
102.0 (100.4-103.5)
38.7 (32.8-44.7)
47.8 (41.0-54.6)
13.5 (10.6-16.4)
44.6 (37.2-52.1)
Trend
APC, 0.22 (0.08-0.36)
APC, 0.16 (0.05-0.27)
OR, 1.019 (1.000-1.040)
OR, 1.004 (0.988-1.020)
OR, 0.968 (0.949-0.987)
OR, 1.017 (0.996-1.038)
≥65 y, Mexican American
1988-1994
26.7 (26.1-27.3)
99.1 (97.6-100.5)
19.8 (14.0-25.6)
47.7 (40.0-55.4)
31.0 (25.5-36.5)
40.0 (31.6-48.4)
2009-2010
29.0 (27.2-30.8)
104.8 (102.1-107.4)
36.1 (20.7-51.5)
47.1 (35.9-58.2)
16.9 (6.5-27.3)
58.7 (49.7-67.7)
Trend
APC, 0.44 (0.11-0.77)
APC, 0.31 (0.16-0.46)
OR, 1.057 (1.010-1.106)
OR, 0.993 (0.966-1.021)
OR, 0.960 (0.930-0.992)
OR, 1.040 (1.013-1.069)
APC = annual percentage change; BMI = body mass index; CI = confidence interval; OR = odds ratio.
Appendix Table 3Trends in Reported Levels of Leisure-time Physical Activity and Daily Caloric Intake, US Adults, Stratified by Race/Ethnicity, 1988-2010
Reported Leisure-time Physical Activity
Mean Energy Intake in kcal/d (95% CI)
None, % of Population (95% CI)
Intermediate Level, % of Population (95% CI)
Ideal level, % of Population (95% CI)
Women
18-39 y, non-Hispanic white
1988-1994
10.4 (8.2-12.6)
51.3 (46.9-55.7)
38.3 (34.2-42.4)
1939 (1874-2005)
2009-2010
41.3 (34.9-47.7)
21.9 (17.5-26.4)
36.8 (31.5-42.1)
1861 (1820-1903)
Trend
OR, 1.080 (1.061-1.100)
OR, 0.933 (0.920-0.945)
OR, 1.010 (0.996-1.024)
APC, −0.16 (−0.38-0.06)
40-64 y, non-Hispanic white
1988-1994
15.2 (12.7-17.8)
48.6 (45.8-51.4)
36.2 (32.1-40.2)
1715 (1685-1746)
2009-2010
43.4 (38.4-48.4)
21.3 (18.0-24.7)
35.2 (31.5-39.0)
1795 (1708-1882)
Trend
OR, 1.077 (1.056-1.099)
OR, 0.936 (0.925-0.948)
OR, 0.995 (0.980-1.011)
APC, 0.17 (−0.08-0.42)
≥65 y, non-Hispanic white
1988-1994
29.3 (25.8-32.7)
30.9 (28.0-33.9)
39.8 (35.3-44.3)
1491 (1459-1523)
2009-2010
64.2 (58.0-70.4)
12.1 (7.9-16.3)
23.7 (18.0-29.4)
1607 (1523-1690)
Trend
OR, 1.073 (1.055-1.090)
OR, 0.957 (0.945-0.970)
OR, 0.957 (0.940-0.974)
APC, 0.40 (0.18-0.62)
18-39 y, non-Hispanic black
1988-1994
19.5 (16.2-22.8)
47.4 (43.8-51.0)
33.0 (29.4-36.7)
1970 (1915-2025)
2009-2010
59.9 (54.1-65.8)
19.4 (13.7-25.1)
20.7 (15.8-25.5)
1957 (1790-2125)
Trend
OR, 1.070 (1.055-1.086)
OR, 0.936 (0.922-0.949)
OR, 0.989 (0.975-1.004)
APC, −0.14 (−0.52-0.24)
40-64 y, non-Hispanic black
1988-1994
33.0 (28.7-37.3)
39.6 (35.5-43.8)
27.4 (23.1-31.6)
1645 (1587-1703)
2009-2010
71.1 (64.0-78.2)
14.3 (10.2-18.3)
14.6 (8.4-20.8)
1728 (1613-1843)
Trend
OR, 1.064 (1.046-1.083)
OR, 0.948 (0.933-0.964)
OR, 0.972 (0.954-0.990)
APC, 0.41 (0.02-0.79)
≥65 y, non-Hispanic black
1988-1994
51.3 (44.7-57.9)
30.1 (25.2-35.0)
18.6 (13.5-23.7)
1340 (1261-1419)
2009-2010
69.7 (61.4-77.9)
18.0 (11.7-24.3)
12.4 (3.7-21.1)
1509 (1298-1720)
Trend
OR, 1.040 (1.015-1.066)
OR, 0.962 (0.938-0.985)
OR, 0.980 (0.949-1.011)
APC, 0.56 (−0.03-1.15)
18-39 y, Mexican American
1988-1994
29.9 (26.6-33.3)
44.0 (40.8-47.2)
26.0 (22.2-29.9)
1873 (1833-1913)
2009-2010
57.2 (49.4-65.0)
16.7 (12.2-21.2)
26.1 (18.5-33.7)
1844 (1698-1990)
Trend
OR, 1.056 (1.034-1.079)
OR, 0.934 (0.917-0.952)
OR, 1.002 (0.981-1.024)
APC, −0.10 (−0.51-0.30)
40-64 y, Mexican American
1988-1994
30.8 (27.7-33.9)
42.8 (39.2-46.4)
26.4 (22.7-30.0)
1697 (1621-1772)
2009-2010
70.3 (60.3-80.2)
9.6 (3.9-15.3)
20.1 (13.2-27.1)
1733 (1595-1872)
Trend
OR, 1.088 (1.063-1.114)
OR, 0.917 (0.897-0.937)
OR, 0.975 (0.954-0.997)
APC, 0.18 (−0.37-0.72)
≥65 y, Mexican American
1988-1994
47.2 (41.5-53.0)
28.9 (23.0-34.8)
23.8 (16.9-30.7)
1271 (1198-1344)
2009-2010
80.2 (74.1-86.3)
6.8 (3.6-10.0)
13.0 (5.9-20.0)
1455 (1381-1528)
Trend
OR, 1.081 (1.058-1.105)
OR, 0.934 (0.909-0.959)
OR, 0.950 (0.918-0.984)
APC, 0.77 (0.20-1.35)
Men
18-39 y, non-Hispanic white
1988-1994
7.3 (5.5-9.1)
46.0 (42.2-49.8)
46.7 (42.6-50.7)
3050 (2965-3134)
2009-2010
28.6 (21.6-35.6)
16.7 (12.0-21.4)
54.7 (45.7-63.8)
2773 (2632-2913)
Trend
OR, 1.064 (1.042-1.085)
OR, 0.936 (0.922-0.949)
OR, 1.016 (1.001-1.032)
APC, −0.38 (−0.63 to −0.12)
40-64 y, non-Hispanic white
1988-1994
9.1 (6.8-11.3)
48.0 (44.0-52.0)
42.9 (39.5-46.3)
2547 (2472-2621)
2009-2010
43.9 (40.7-47.1)
17.3 (13.5-21.0)
38.8 (33.7-44.0)
2590 (2497-2682)
Trend
OR, 1.107 (1.085-1.130)
OR, 0.928 (0.915-0.942)
OR, 0.983 (0.970-0.997)
APC, 0.17 (−0.05-0.40)
≥65 y, non-Hispanic white
1988-1994
16.3 (13.9-18.7)
35.3 (32.6-38.1)
48.4 (44.9-51.8)
2013 (1957-2069)
2009-2010
54.9 (50.2-59.5)
12.8 (9.7-15.9)
32.3 (28.5-36.2)
2018 (1951-2085)
Trend
OR, 1.094 (1.079-1.109)
OR, 0.942 (0.931-0.954)
OR, 0.962 (0.951-0.972)
APC, −0.03 (−0.28-0.22)
18-39 y, non-Hispanic black
1988-1994
6.7 (4.6-8.7)
35.9 (32.9-38.8)
57.4 (53.9-61.0)
2917 (2800-3034)
2009-2010
37.6 (28.9-46.3)
5.3 (1.4-9.3)
57.0 (46.3-67.7)
2510 (2388-2632)
Trend
OR, 1.070 (1.047-1.093)
OR, 0.922 (0.908-0.937)
OR, 1.004 (0.987-1.022)
APC, −0.56 (−0.85 to −0.26)
40-64 y, non-Hispanic black
1988-1994
19.6 (15.8-23.4)
42.3 (37.9-46.7)
38.1 (34.2-42.0)
2314 (2226-2403)
2009-2010
57.4 (48.3-66.5)
11.8 (7.2-16.4)
30.8 (23.5-38.0)
2593 (2285-2901)
Trend
OR, 1.070 (1.050-1.091)
OR, 0.928 (0.913-0.944)
OR, 0.987 (0.970-1.003)
APC, 0.58 (0.08-1.08)
≥65 y, non-Hispanic black
1988-1994
29.6 (22.5-36.7)
35.0 (28.9-41.2)
35.4 (29.0-41.7)
1651 (1559-1744)
2009-2010
71.9 (63.5-80.3)
9.9 (1.4-18.4)
18.2 (10.3-26.1)
1912 (1803-2021)
Trend
OR, 1.080 (1.056-1.103)
OR, 0.939 (0.913-0.965)
OR, 0.957 (0.934-0.980)
APC, 1.04 (0.43-1.65)
18-39 y, Mexican American
1988-1994
19.5 (15.2-23.8)
42.0 (37.0-46.9)
38.6 (34.3-42.8)
2792 (2697-2887)
2009-2010
52.8 (42.6-63.1)
10.7 (4.6-16.9)
36.4 (29.9-43.0)
2613 (2442-2784)
Trend
OR, 1.052 (1.032-1.074)
OR, 0.914 (0.895-0.934)
OR, 1.018 (1.001-1.036)
APC, −0.35 (−0.72-0.03)
40-64 y, Mexican American
1988-1994
24.2 (20.0-28.5)
44.3 (40.5-48.0)
31.5 (27.7-35.3)
2424 (2364-2484)
2009-2010
63.2 (53.6-72.9)
13.5 (8.5-18.5)
23.3 (16.2-30.3)
2409 (2240-2578)
Trend
OR, 1.080 (1.055-1.106)
OR, 0.919 (0.900-0.938)
OR, 0.976 (0.955-0.997)
APC, 0.21 (−0.30-0.71)
≥65 y, Mexican American
1988-1994
32.1 (25.6-38.6)
28.8 (22.5-35.1)
39.1 (32.4-45.8)
1777 (1657-1897)
2009-2010
59.4 (48.2-70.6)
12.0 (5.4-18.6)
28.6 (22.0-35.3)
1919 (1693-2144)
Trend
OR, 1.061 (1.029-1.095)
OR, 0.947 (0.919-0.976)
OR, 0.968 (0.941-0.997)
APC, 0.45 (−0.21-1.11)
APC = annual percentage change; CI = confidence interval; OR = odds ratio.
Appendix Table 4Annual Changes in Mean Body Mass Index, Mean Waist Circumference, and Daily Caloric Intake, Stratified by Reported Level of Leisure-time Physical Activity, US Adults, 1988-2010
Mean BMI in kg/m2 (95% CI)
Mean Waist Circumference in cm (95% CI)
Mean Energy Intake in kcal/d (95% CI)
1988-1994
2009-2010
APC (95% CI)
1988-1994
2009-2010
APC (95% CI)
1988-1994
2009-2010
APC (95% CI)
Women
No reported leisure-time physical activity
≥18 y
27.9 (27.3-28.6)
29.7 (29.2-30.2)
0.28 (0.16-0.40)
92.9 (91.4-94.4)
97.3 (95.6-99.1)
0.25 (0.14-0.37)
1714 (1664-1764)
1791 (1753-1828)
0.27 (0.02-0.52)
18-39 y
26.7 (25.6-27.8)
28.5 (27.7-29.3)
0.26 (0.03-0.49)
87.4 (84.6-90.2)
92.9 (90.4-95.4)
0.31 (0.11-0.51)
1875 (1793-1958)
1914 (1827-2001)
−0.04 (−0.52-0.44)
40-64 y
29.5 (28.5-30.6)
30.7 (29.9-31.5)
0.16 (−0.04-0.35)
97.4 (95.3-99.6)
100.4 (98.5-102.3)
0.18 (0.03-0.33)
1681 (1600-1762)
1762 (1668-1855)
0.29 (−0.14-0.72)
≥65 y
27.2 (26.6-27.9)
30.2 (29.2-31.2)
0.50 (0.33-0.67)
95.6 (93.7-97.5)
100.9 (99.1-102.8)
0.30 (0.17-0.43)
1395 (1336-1454)
1556 (1497-1616)
0.63 (0.29-0.96)
Intermediate level of reported leisure-time physical activity
≥18 y
26.6 (26.2-27.0)
28.4 (27.9-28.8)
0.39 (0.28-0.51)
89.1 (88.2-90.0)
94.7 (93.7-95.7)
0.38 (0.30-0.46)
1776 (1738-1814)
1783 (1714-1851)
0.09 (−0.11-0.28)
18-39 y
25.4 (24.9-25.9)
27.9 (26.9-29.0)
0.50 (0.33-0.68)
83.8 (82.6-84.9)
92.2 (89.9-94.5)
0.53 (0.41-0.65)
1969 (1889-2050)
1757 (1663-1852)
−0.27 (−0.60-0.06)
40-64 y
27.7 (27.2-28.2)
28.8 (27.7-29.9)
0.32 (0.15-0.49)
92.4 (91.1-93.7)
96.5 (94.1-98.9)
0.28 (0.16-0.41)
1702 (1665-1739)
1874 (1776-1972)
0.38 (0.13-0.64)
≥65 y
26.9 (26.4-27.4)
28.3 (26.3-30.3)
0.26 (0.06-0.47)
94.3 (93.0-95.6)
96.5 (91.5-101.6)
0.14 (−0.02-0.29)
1478 (1427-1530)
1627 (1392-1861)
0.41 (0.00-0.82)
Ideal level of reported leisure-time physical activity
≥18 y
25.5 (25.2-25.7)
26.7 (26.1-27.2)
0.28 (0.16-0.39)
86.0 (85.2-86.7)
90.5 (89.2-91.8)
0.33 (0.25-0.41)
1765 (1722-1808)
1774 (1705-1843)
0.08 (−0.13-0.30)
18-39 y
24.3 (23.9-24.7)
26.6 (25.4-27.8)
0.48 (0.32-0.65)
80.9 (79.9-81.8)
88.5 (85.9-91.2)
0.52 (0.39-0.64)
1926 (1856-1996)
1892 (1795-1988)
−0.12 (−0.42-0.18)
40-64 y
26.5 (26.1-26.9)
26.8 (25.9-27.6)
0.10 (−0.08-0.28)
89.2 (87.8-90.5)
91.6 (89.6-93.6)
0.20 (0.06-0.33)
1697 (1616-1779)
1719 (1603-1835)
0.18 (−0.23-0.59)
≥65 y
25.9 (25.4-26.3)
26.6 (25.2-27.9)
0.12 (−0.06-0.31)
90.8 (89.7-91.9)
92.8 (90.6-95.0)
0.13 (0.01-0.24)
1530 (1472-1588)
1616 (1508-1723)
0.41 (0.07-0.75)
Men
No reported leisure-time physical activity
≥18 y
26.6 (26.1-27.1)
29.2 (28.6-29.8)
0.52 (0.37-0.66)
96.1 (94.4-97.8)
102.6 (100.9-104.2)
0.35 (0.23-0.47)
2438 (2343-2533)
2487 (2384-2589)
0.19 (−0.14-0.53)
18-39 y
26.0 (25.0-26.9)
28.4 (27.6-29.3)
0.44 (0.20-0.67)
90.7 (88.1-93.2)
97.8 (95.3-100.2)
0.32 (0.13-0.51)
2714 (2536-2891)
2663 (2492-2834)
−0.06 (−0.63-0.50)
40-64 y
27.6 (26.8-28.4)
29.8 (29.0-30.6)
0.52 (0.32-0.71)
100.1 (97.5-102.7)
105.2 (103.0-107.4)
0.34 (0.18-0.51)
2471 (2322-2620)
2548 (2428-2669)
0.34 (−0.15-0.83)
≥65 y
25.9 (25.1-26.7)
29.4 (28.9-29.9)
0.66 (0.51-0.81)
99.7 (97.7-101.8)
108.1 (107.1-109.2)
0.43 (0.32-0.53)
1679 (1595-1762)
1903 (1820-1986)
0.38 (−0.02-0.77)
Intermediate level of reported leisure-time physical activity
≥18 y
27.0 (26.7-27.4)
28.3 (27.6-28.9)
0.28 (0.17-0.38)
97.1 (96.3-97.9)
100.0 (98.3-101.7)
0.21 (0.14-0.29)
2636 (2576-2695)
2532 (2408-2655)
−0.13 (−0.34-0.09)
18-39 y
25.8 (25.3-26.2)
28.0 (26.6-29.5)
0.37 (0.21-0.52)
91.2 (90.0-92.3)
96.3 (93.0-99.7)
0.29 (0.17-0.41)
2972 (2864-3080)
2710 (2483-2937)
−0.36 (−0.68 to −0.03)
40-64 y
28.2 (27.7-28.7)
28.9 (28.0-29.8)
0.21 (0.05-0.37)
101.2 (100.1-102.3)
102.9 (100.8-105.1)
0.15 (0.04-0.26)
2548 (2466-2631)
2537 (2337-2738)
0.10 (−0.21-0.40)
≥65 y
27.3 (26.8-27.9)
27.4 (25.9-28.9)
0.14 (−0.05-0.33)
102.1 (100.5-103.7)
101.9 (98.1-105.8)
0.12 (−0.02-0.27)
2015 (1934-2096)
2077 (1818-2336)
−0.03 (−0.44-0.39)
Ideal level of reported leisure-time physical activity
≥18 y
26.2 (25.9-26.4)
28.0 (27.5-28.6)
0.35 (0.26-0.44)
93.9 (93.2-94.5)
98.2 (96.7-99.7)
0.25 (0.18-0.32)
2639 (2579-2699)
2536 (2452-2620)
−0.22 (−0.40 to −0.04)
18-39 y
25.5 (25.1-25.9)
27.5 (26.6-28.3)
0.37 (0.24-0.50)
89.0 (87.9-90.0)
93.6 (91.6-95.6)
0.28 (0.17-0.38)
3023 (2912-3134)
2678 (2564-2792)
−0.60 (−0.87 to −0.32)
40-64 y
26.7 (26.4-27.1)
28.7 (27.9-29.6)
0.35 (0.21-0.49)
96.6 (95.7-97.4)
100.9 (98.9-102.9)
0.24 (0.13-0.34)
2491 (2395-2587)
2585 (2447-2722)
0.23 (−0.09-0.55)
≥65 y
26.4 (26.0-26.7)
27.8 (27.0-28.6)
0.28 (0.15-0.42)
99.5 (98.4-100.5)
103.1 (100.4-105.8)
0.22 (0.11-0.34)
2047 (1981-2113)
2066 (1942-2191)
0.06 (−0.29-0.41)
APC = annual percentage change; CI = confidence interval.
Appendix Table 5Relative Increase in Body Mass Index and Waist Circumference in “Intermediate” and “No Reported” Groups Compared with “Ideal” Reported Leisure-time Physical Activity in US Adults, 1988-2010
Relative Increase in BMI Compared with Ideal Level of Reported Leisure-time Physical Activity
Relative Increase in Waist Circumference Compared with Ideal Level of Reported Leisure-time Physical Activity
Appendix Table 6Annual Changes in Daily Intake of Fat, Carbohydrate, and Protein, Stratified by Reported Level of Leisure-time Physical Activity, US Adults, 1988-2010
Reported Level of Leisure-time Physical Activity
None
Intermediate
Ideal
Mean Fat Intake in g/d (95% CI)
Mean Carbohydrate Intake in g/d (95% CI)
Mean Protein Intake in g/d (95% CI)
Mean Fat Intake in g/d (95% CI)
Mean Carbohydrate Intake in g/d (95% CI)
Mean Protein Intake in g/d (95% CI)
Mean Fat Intake in g/d (95% CI)
Mean Carbohydrate Intake in g/d (95% CI)
Mean Protein Intake in g/d (95% CI)
Women
Aged ≥18 y
1988-1994
66 (64-68)
216 (209-224)
64 (62-67)
69 (67-71)
219 (213-225)
66 (64-68)
65 (63-68)
225 (220-230)
67 (65-69)
2009-2010
67 (65-68)
228 (223-233)
67 (66-69)
67 (63-72)
224 (217-231)
68 (64-72)
65 (62-68)
222 (212-232)
69 (67-71)
Trend
APC, 0.26 (−0.07-0.59)
APC, 0.21 (−0.07-0.49)
APC, 0.50 (0.18-0.81)
APC, −0.02 (−0.32-0.28)
APC, 0.18 (−0.03-0.39)
APC, 0.22 (−0.03-0.48)
APC, 0.45 (0.11-0.78)
APC, −0.19 (−0.44-0.07)
APC, 0.26 (0.02-0.49)
18-39 y
1988-1994
72 (67-76)
238 (228-248)
68 (64-73)
77 (73-81)
240 (229-251)
71 (67-74)
71 (68-75)
244 (234-254)
70 (67-72)
2009-2010
71 (67-75)
245 (235-256)
68 (65-72)
64 (58-69)
227 (214-240)
66 (61-71)
69 (64-73)
239 (227-250)
72 (70-75)
Trend
APC, −0.15 (−0.79-0.49)
APC, −0.03 (−0.57-0.51)
APC, 0.37 (−0.26-1.00)
APC, −0.55 (−1.03 to −0.07)
APC, −0.04 (−0.41-0.33)
APC, −0.10 (−0.51-0.30)
APC, 0.18 (−0.26-0.62)
APC, −0.36 (−0.71-0.00)
APC, 0.22 (−0.16-0.59)
40-64 y
1988-1994
65 (61-69)
208 (198-219)
64 (59-68)
67 (64-69)
208 (202-214)
65 (63-67)
64 (59-68)
214 (204-224)
66 (63-69)
2009-2010
65 (61-69)
223 (211-235)
69 (65-73)
71 (66-76)
233 (222-244)
72 (66-77)
63 (59-67)
212 (192-232)
68 (64-71)
Trend
APC, 0.11 (−0.48-0.70)
APC, 0.31 (−0.14-0.75)
APC, 0.52 (0.01-1.03)
APC, 0.26 (−0.12-0.63)
APC, 0.50 (0.20-0.79)
APC, 0.53 (0.21-0.84)
APC, 0.53 (−0.03-1.09)
APC, −0.16 (−0.66-0.34)
APC, 0.30 (−0.10-0.71)
≥65 y
1988-1994
52 (49-54)
182 (174-190)
55 (52-58)
54 (51-57)
193 (185-202)
58 (56-60)
54 (50-57)
203 (195-211)
62 (59-65)
2009-2010
59 (56-63)
197 (188-205)
61 (58-64)
67 (55-79)
194 (165-224)
64 (56-71)
60 (55-64)
205 (192-218)
64 (58-70)
Trend
APC, 1.02 (0.43-1.61)
APC, 0.37 (0.03-0.71)
APC, 0.62 (0.19-1.05)
APC, 1.01 (0.47-1.54)
APC, −0.01 (−0.47-0.45)
APC, 0.44 (−0.01-0.90)
APC, 0.97 (0.48-1.46)
APC, 0.20 (−0.14-0.54)
APC, 0.26 (−0.16-0.68)
Men
≥18 y
1988-1994
95 (90-100)
293 (278-308)
92 (89-96)
102 (99-106)
309 (301-317)
100 (97-103)
100 (97-104)
320 (311-329)
99 (97-102)
2009-2010
93 (88-98)
296 (284-308)
97 (92-102)
96 (91-101)
305 (288-321)
96 (91-101)
93 (88-97)
297 (287-308)
102 (99-105)
Trend
APC, 0.22 (−0.23-0.67)
APC, 0.04 (−0.35-0.43)
APC, 0.37 (0.00-0.75)
APC, −0.13 (−0.41-0.15)
APC, −0.12 (−0.36-0.13)
APC, 0.07 (−0.19-0.32)
APC, −0.31 (−0.59 to −0.02)
APC, −0.41 (−0.62 to −0.20)
APC, 0.17 (−0.01-0.34)
18-39 y
1988-1994
105 (96-115)
323 (299-348)
101 (94-107)
115 (110-119)
346 (335-358)
110 (105-114)
116 (111-122)
363 (348-378)
111 (106-116)
2009-2010
95 (87-103)
326 (307-345)
102 (93-110)
101 (89-113)
327 (298-356)
100 (90-110)
94 (89-99)
324 (308-340)
106 (102-110)
Trend
APC, −0.30 (−1.09-0.50)
APC, −0.23 (−0.79-0.34)
APC, 0.30 (−0.39-1.00)
APC, −0.43 (−0.85-0.00)
APC, −0.27 (−0.60-0.05)
APC, −0.13 (−0.54-0.28)
APC, −0.93 (−1.34 to −0.53)
APC, −0.64 (−0.94-−0.34)
APC, −0.17 (−0.45-0.11)
40-64 y
1988-1994
97 (89-105)
297 (273-321)
95 (89-101)
100 (95-105)
297 (285-310)
99 (95-103)
94 (88-99)
302 (287-316)
95 (91-98)
2009-2010
99 (94-104)
293 (275-311)
100 (94-106)
97 (88-105)
304 (278-329)
98 (88-108)
97 (89-105)
292 (276-308)
106 (101-112)
Trend
APC, 0.56 (−0.10-1.21)
APC, 0.19 (−0.43-0.81)
APC, 0.45 (−0.05-0.95)
APC, 0.16 (−0.24-0.57)
APC, 0.01 (−0.38-0.40)
APC, 0.28 (−0.09-0.65)
APC, 0.37 (−0.08-0.83)
APC, −0.13 (−0.50-0.23)
APC, 0.68 (0.35-1.01)
≥65 y
1988-1994
62 (57-67)
209 (194-225)
67 (62-71)
78 (73-82)
245 (234-256)
78 (75-82)
76 (72-79)
255 (245-264)
81 (77-84)
2009-2010
72 (68-76)
228 (216-240)
76 (73-80)
82 (68-95)
250 (216-285)
80 (70-90)
80 (73-86)
243 (225-262)
83 (77-89)
Trend
APC, 0.48 (−0.03-1.00)
APC, 0.24 (−0.28-0.76)
APC, 0.35 (−0.11-0.81)
APC, −0.02 (−0.59-0.55)
APC, 0.05 (−0.45-0.56)
APC, 0.08 (−0.33-0.49)
APC, 0.34 (−0.11-0.78)
APC, −0.23 (−0.62-0.17)
APC, 0.13 (−0.26-0.52)
APC = annual percentage change; CI = confidence interval.
Clinical Guidelines on the Identification, Evaluation, and Treatment of Overweight and Obesity in Adults: The Evidence Report. National Heart, Lung and Blood Institutes, National Institutes of Health,
Washington, DC1998
The Practical Guide Identification, Evaluation, and Treatment of Overweight and Obesity in Adults. National Heart, Lung and Blood Institutes, National Institutes of Health,
Washington, DC2000
Centers for Disease Control and Prevention (CDC). National Center for Health Statistics (NCHS). National Health and Nutrition Examination Survey Data. Hyattsville, MD. Available at: http://www.cdc.gov/nchs/nhanes/nhanes_questionnaires.htm. Accessed April 17, 2013.
National Center for Health Statistics. Analytical and Reporting Guidelines. The Third National Health and Nutrition Examination Survey, 1988–1994, Hyattsville, MD. 1996. Available at: http://www.cdc.gov/nchs/data/nhanes/nhanes3/nh3gui.pdf. Accessed April 17, 2013.
Population Projections of the United States by Age, Sex, Race, and Hispanic Origin: 1995 to 2050, U.S. Bureau of the Census, Current Population Reports, P25-1130.
U.S. Government Printing Office,
Washington DC1996 (Available at:)
Pleis JR, Lucas JW. Summary Health Statistics for U.S. Adults: National Health Interview Survey, 2007. Vital and Health Statistics Series 10, Data from the National Health Survey. Hyattsville MD: U.S. Department of Health and Human Services; 2009:1-159.
Funding: National Institutes of Health T32 Training Grant T32DK007056 Awarded to PAM. The sponsor had no design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Conflict of Interest: UL has served as a consultant for Endosphere.
Authorship: All authors had access to the data and played a role in writing this manuscript.
Although we agree with Ladabaum et al1 that obesity is a major health problem, we do not believe that any conclusion can be justified from their data as to whether the increase in body weight over the past 2 decades is primarily due to an increase in energy intake or a decrease in energy expenditure. They base their conclusion on their analysis of National Health and Nutrition Examination Survey (NHANES) dataset, indicating that energy intake did not increase during this period while self-reported leisure-time activity decreased.
A recent article by Ladabaum et al1 raised an important question regarding the leading drivers for the obesity epidemic in US adults. We appreciate that the authors called attention to physical inactivity as an important public health problem and attempted to relate the rise in obesity to trends in leisure-time physical activity and energy intake. We agree that the prevalence of adult obesity has increased during this period. This is consistent with previous studies using the National Health and Nutrition Examination Survey (NHANES), which defined obesity according to waist circumference2 or body mass index.
00191-0&id=gr1.jpg){kind=link}
00191-0&id=gr2.jpg){kind=link}
00191-0&id=fx1.jpg){kind=link}
00191-0&id=fx2.jpg){kind=link}