Prevalence of Overweight and Obesity in the United States, 1999-2004 FREE

Obesity continues to be a leading public health concern in the United States.^1- 2 Between 1980 and 2002, obesity prevalence doubled in adults aged 20 years or older and overweight prevalence tripled in children and adolescents aged 6 to 19 years.^3- 5 This article provides the most recent prevalence estimates of overweight and obesity based on national measurements of weight and height in 2003-2004 and compares these estimates with estimates from 1999-2000 and 2001-2002 to determine if the trend is continuing.

Prevalence estimates of overweight and obesity were calculated using data from the National Health and Nutrition Examination Survey (NHANES), a complex, multistage probability sample of the US civilian, noninstitutionalized population.⁶ Race/ethnicity was reported by survey participants. During a physical examination in a mobile examination center, height and weight were measured using standardized protocols and calibrated equipment. Body mass index (BMI) was calculated as weight in kilograms divided by the square of height in meters and was rounded to the nearest tenth.

The NHANES 2003-2004 overall response rate (of those originally selected for participation) was 68.6% (4742/6916) for adults aged 20 years or older and 83.2% (4105/4932) for children and adolescents aged 2 to 19 years. Less than 3% of examined children and adolescents and 7% of adults had missing data for BMI. Data for analyses were available for 3958 children and adolescents and 4431 adults in 2003-2004 (Table 1), in addition to previously published data for 1999-2000 and 2001-2002.^3- 5

Among children and adolescents aged 2 to 19 years, overweight and at risk for overweight were defined based on the sex-specific BMI for age growth charts from the Centers for Disease Control and Prevention.⁷ The CDC growth charts were created from national data in the United States collected from 1963 through 1994. Data for the BMI for age charts were collected between 1971 and 1994 for children 2 to 6 years of age and between 1963 and 1980 for children 6 through 19 years of age. The “at risk for overweight” category was defined as a BMI for age at or above the sex-specific 85th percentile but less than the 95th percentile.⁸ Overweight was defined as a BMI for age at or above the sex-specific 95th percentile. For adults aged 20 years or older, overweight was defined as a BMI of 25.0 to 29.9; obesity, 30.0 or higher; and extreme obesity, 40.0 or higher.⁹

Statistical analyses were performed using SAS (version 9, SAS Institute Inc, Cary, NC) and SUDAAN (version 9, Research Triangle Institute, Research Triangle Park, NC) software. All analyses excluded pregnant females. Sample weights were used to account for differential nonresponse, noncoverage, and to adjust for planned oversampling of some groups. Estimates for adults were age-standardized to the projected estimates of the 2000 US Census using the age groups of 20 to 39 years, 40 to 59 years, and 60 years or older. SEs for all survey years were estimated with SUDAAN software using Taylor series linearization.

Trends in prevalence of overweight for children and adolescents and in obesity for adults over 1999-2000, 2001-2002, and 2003-2004 were modeled separately using logistic regression. To test for trends, the NHANES survey years were included as an ordinal variable in logistic regression models that included age group and race/ethnicity. Trends were similarly tested using 6 separate years of data for 1999, 2000, 2001, 2002, 2003, and 2004.

An interaction was found between sex and the NHANES survey years so the models were run separately for males and females. All 2-way interaction terms were dropped from the sex-specific models because they were not statistically significant. No 3-way interactions were tested. Because no interaction was found between race/ethnicity and the NHANES survey years, tests for differences between racial/ethnic groups were performed in these models with the 3 NHANES survey periods (1999-2000, 2001-2002, 2003-2004). In contrast, overall differences by sex were tested in 2003-2004 using t tests at the .05 significance level for adults and for children and adolescents separately.

Overall, among children and adolescents aged 2 to 19 years, 17.1% were overweight in 2003-2004 and 32.2% of adults aged 20 years or older were obese. The prevalence of extreme obesity among adults was 4.8%. Detailed prevalence estimates of overweight in children and adolescents (Table 2 and Table 3) and obesity in adults (Table 4 and Table 5) are shown by sex, age group, and race/ethnicity for 2003-2004.

Prevalence estimates of overweight and at risk for overweight in children and adolescents and overweight, obesity, and extreme obesity among adults for 1999-2000 and 2001-2002 also appear in Table 2, Table 3, Table 4, and Table 5. Tests of trend using logistic regression adjusted for age and race/ethnicity showed a significant increase in the prevalence of overweight in children and adolescents over 1999-2000, 2001-2002, and 2003-2004 (P = .0396 for males and P = .0463 for females). Among men, the increase in obesity was not significant (P = .0503). Tests for trends using 6 single years of data showed significant increases in prevalence of obesity for men (P = .02) and overweight for male children and adolescents (P = .01), but no significant increase in overweight for female children and adolescents (P = .10). Among women, no significant trends were found using either test of trends.

During 1999-2004, significant differences between racial/ethnic groups existed among children and adolescents (Table 6). After adjusting for age and the NHANES survey years in a multiple logistic regression model using data from 1999-2000, 2001-2002, and 2003-2004, the prevalence of overweight in Mexican American male children and adolescents was significantly greater than in non-Hispanic white male children and adolescents. However, prevalence of overweight among non-Hispanic white male children and adolescents did not differ significantly from non-Hispanic black male children and adolescents. Mexican American and non-Hispanic black female children and adolescents were significantly more likely to be overweight compared with non-Hispanic white female children and adolescents. Adolescents were more likely to be overweight compared with children. Among children and adolescents aged 2 to 19 years in 2003-2004, no significant difference in prevalence of overweight was found between the sexes.

Among adult men, no differences were found between racial/ethnic groups (Table 6). However, Mexican American and non-Hispanic black women were significantly more likely to be obese compared with non-Hispanic white women. Older adults were more likely to be obese than their younger counterparts. The only exception was adults aged 80 years or older who were not significantly different from adults aged 20 to 39 years (for both sexes). In 2003-2004, there was no significant difference in the prevalence of obesity in adults between the sexes.

These estimates from the NHANES conducted in 2003-2004 provide the most recent prevalence estimates of overweight and obesity in the United States. Among children and adolescents aged 2 to 19 years, 17.1% were overweight and 32.2% of adults aged 20 years or older were obese. The prevalence of overweight among children and adolescents and obesity among men increased significantly between 1999 and 2004. No increase in overall obesity prevalence among women was observed over the 6-year period.

Multivariable model results indicate that the only statistically significant trends were those noted overall for men and for children and adolescents. Trends did not differ significantly by age or racial/ethnic group. Although differences in point estimates may appear large in some subgroups, the differences are not statistically significant. Subgroup estimates by sex, age, and race/ethnicity are less precise than the overall estimates. This is due to smaller sample sizes and fewer degrees of freedom than for the overall estimates. Future data from NHANES 2005-2006 should provide more information about changes in subgroups.

Increases in prevalence of overweight and obesity among both adults and children have been observed in many countries throughout the world.^10- 11 For example, the prevalence of obesity among adults in Great Britain almost tripled between 1980 and 2002.¹² Among preschool children in urban areas of China, the prevalence of obesity increased from 1.5% in 1989 to 12.6% in 1997.¹³

Similar to previous analyses,^4- 5 there continues to be differences in overweight and obesity prevalence by racial/ethnic group among women and among children and adolescents. Among women, almost 58% of non-Hispanic black women aged 40 to 59 years were obese in 2003-2004 compared with about 38% of non-Hispanic white women of the same age. Among men, however, the prevalence of obesity did not differ by racial/ethnic group. Among children and adolescents, differences by racial/ethnic group exist for both sexes. These racial/ethnic group differences in female children and adolescents were observed in 1988-1994 but these differences were not observed among male adolescents in 1988-1994.⁴

The prevalence of overweight and obesity in the United States remains high. The prevalence of obesity has continued to increase in men and the prevalence of overweight has continued to increase in children and adolescents between 1999 and 2004. However, no significant increases were observed among women. There is little indication that the prevalence is decreasing in any subgroup of the population. These prevalence estimates, based on a 6-year period (1999-2004), suggest that the increases in body weight may be leveling off in women.