To the Editor
Obesity and extreme obesity in childhood, which are more prevalent among minority and low-income families, have been associated with other cardiovascular risk factors, increased health care costs, and premature death.1,2 Obesity and extreme obesity during early childhood are likely to continue into adulthood.3 Understanding trends in extreme obesity is important because the prevalence of cardiovascular risk factors increases with severity of childhood obesity.2 However, national trends in extreme obesity among young children living in low-income families are unknown.
Methods
The Pediatric Nutrition Surveillance System (PedNSS) includes almost 50% of children eligible for federally funded maternal and child health and nutrition programs. The study population included 27.5 million children aged 2 through 4 years from 30 states and the District of Columbia that consistently reported data to PedNSS from 1998 through 2010. We excluded those with missing (n=297 999; 1.1%), miscoded (n=106 844; 0.4%), or biologically implausible height, weight, or body mass index (BMI) (n=427 051; 1.6%), leaving 26 708 517 children. The study was exempt from ethics review by the US Centers for Disease Control and Prevention (CDC).
One routine clinic visit with demographic information and measured height and weight was randomly selected for each child.4 Obesity (BMI ≥95th percentile for age and sex) and extreme obesity (BMI ≥120% of the 95th percentile) were defined according to the 2000 CDC growth charts.5 We examined trends from 1998 through 2010 in mean BMI and the prevalence of obesity and extreme obesity. Significant changes in overall trends were identified by the Joinpoint regression program version 3.5.3 (National Cancer Institute). Piecewise logistic regression adjusting for age, sex, and race/ethnicity was performed to examine trends using SAS version 9.2 (SAS Institute Inc). Using the transition year for overall trend detected by Joinpoint, separate line segments prior to and after that year were fitted.
Results
The 2010 study population was slightly younger and had proportionally more Hispanics and fewer non-Hispanic whites and blacks compared with the 1998 population (Table 1). Joinpoint regression found significant changes in trends of obesity and extreme obesity in 2003. The prevalence of obesity increased from 13.05% (95% CI, 13.00%–13.09%) in 1998 to 15.21% (95% CI, 15.16%–15.26%) in 2003. The prevalence of extreme obesity increased from 1.75% (95% CI, 1.73%–1.77%) in 1998 to 2.22% (95% CI, 2.20%–2.24%) in 2003. However, the prevalence of obesity decreased slightly to 14.94% (95% CI, 14.89%–14.98%) in 2010. Similarly, the prevalence of extreme obesity decreased to 2.07% (95% CI, 2.05%–2.09%) in 2010 (Table 1).
Table 1.
Distribution of the Study Population and Prevalence of Obesity and Extreme Obesity Among US Children Aged 2 Through 4 Years Living in Low-Income Families, 1998–2010
| Year | Overall, No. |
Age, Mean, mo |
Race/Ethnicity, %a
|
BMI, Mean (SD)b |
Prevalence, % (95% CI)
|
|||||
|---|---|---|---|---|---|---|---|---|---|---|
| Non-Hispanic White |
Non-Hispanic Black |
Hispanic | American Indian/ Alaska Native |
Asian/ Pacific Islander |
Obesityb,c | Extreme Obesityb,d |
||||
| 1998 | 1 945 115 | 40.9 | 43.7 | 23.5 | 28.6 | 0.8 | 3.4 | 16.52 (1.88) | 13.05 (13.00–13.09) | 1.75 (1.73–1.77) |
|
| ||||||||||
| 1999 | 1 887 622 | 40.8 | 42.5 | 23.9 | 29.6 | 0.7 | 3.2 | 16.56 (1.90) | 13.58 (13.53–13.63) | 1.85 (1.83–1.87) |
|
| ||||||||||
| 2000 | 1 869 593 | 40.6 | 42.6 | 22.7 | 30.5 | 0.8 | 3.4 | 16.62 (1.93) | 14.37 (14.32–14.42) | 2.02 (2.00–2.04) |
|
| ||||||||||
| 2001 | 1 823 377 | 40.5 | 43.0 | 22.7 | 30.2 | 0.7 | 3.4 | 16.61 (1.93) | 14.32 (14.26–14.37) | 2.03 (2.01–2.05) |
|
| ||||||||||
| 2002 | 1 887 032 | 40.5 | 42.9 | 22.4 | 30.7 | 0.7 | 3.4 | 16.65 (1.96) | 14.89 (14.84–14.94) | 2.15 (2.13–2.17) |
|
| ||||||||||
| 2003e | 1 958 480 | 40.4 | 42.3 | 21.5 | 31.6 | 0.8 | 3.8 | 16.68 (1.97) | 15.21 (15.16–15.26) | 2.22 (2.20–2.24) |
|
| ||||||||||
| 2004f | 2 014 085 | 40.5 | 42.4 | 21.7 | 31.5 | 0.8 | 3.6 | 16.70 (1.96) | 15.36 (15.31–15.41) | 2.22 (2.20–2.24) |
|
| ||||||||||
| 2005 | 2 078 203 | 40.5 | 40.8 | 21.1 | 33.9 | 0.7 | 3.4 | 16.69 (1.95) | 15.13 (15.09–15.18) | 2.16 (2.14–2.18) |
|
| ||||||||||
| 2006 | 2 042 050 | 40.4 | 38.9 | 20.5 | 36.5 | 0.8 | 3.3 | 16.70 (1.94) | 15.17 (15.12–15.22) | 2.15 (2.13–2.17) |
|
| ||||||||||
| 2007 | 2 095 442 | 40.4 | 38.9 | 20.5 | 36.4 | 0.8 | 3.3 | 16.70 (1.94) | 15.23 (15.18–15.28) | 2.14 (2.12–2.16) |
|
| ||||||||||
| 2008 | 2 249 621 | 40.4 | 38.9 | 20.6 | 36.5 | 0.8 | 3.2 | 16.71 (1.92) | 15.24 (15.19–15.29) | 2.10 (2.08–2.12) |
|
| ||||||||||
| 2009 | 2 453 464 | 40.3 | 38.5 | 20.0 | 37.6 | 0.7 | 3.1 | 16.71 (1.92) | 15.12 (15.07–15.16) | 2.09 (2.07–2.11) |
|
| ||||||||||
| 2010 | 2 404 433 | 40.5 | 38.9 | 20.5 | 36.7 | 0.7 | 3.2 | 16.68 (1.92) | 14.94 (14.89–14.98) | 2.07 (2.05–2.09) |
Abbreviation: BMI, body mass index (calculated as weight in kilograms divided by height in meters squared).
Reported by parent or caregiver. The percentages describe the distribution of the study population.
Excluded missing, miscoded, or biologically implausible height, weight, and BMI. Because the distribution of weight, height, and BMI in the study population was skewed, we expanded the upper cutoff points for biological implausible z scores recommended by the World Health Organization to the following: height for age, −5.0 or less or greater than 5.0; weight for age, −5.0 or less or greater than 8.0; and BMI for age, −4.0 or less or greater than 8.0.
Defined as having a BMI for age and sex in the 95th percentile or greater according to 2000 Centers for Disease Control and Prevention growth charts.
Defined as having a BMI for age and sex in the 120% or greater of the 95th percentile according to 2000 Centers for Disease Control and Prevention growth charts.
Results of Joinpoint regression showed that the upward trends in the prevalence of obesity and extreme obesity ended in 2003.
Results of Joinpoint regression showed that the upward trend in mean BMI ended in 2004.
From 1998 through 2003, the prevalence of extreme obesity significantly increased overall (adjusted odds ratio [AOR], 1.047; 95% CI, 1.045–1.049) and in all groups except Asians/Pacific Islanders; the greatest average annual increases were among those aged 4 years and non-Hispanic whites (Table 2). From 2003 through 2010, extreme obesity significantly decreased overall (AOR, 0.983; 95% CI, 0.981–0.984) and in all groups except American Indians/Alaska Natives; the greatest decreases were among those aged 2 years and Asians/Pacific Islanders (Table 2).
Table 2.
Trends in the Prevalence of Extreme Obesity From 1998–2003 and From 2003–2010 Among US Children Aged 2 Through 4 Years Living in Low-Income Familiesa
| Characteristic | Prevalence, %
|
OR (95% CI) Per Year Change
|
|||||
|---|---|---|---|---|---|---|---|
| Unadjustede
|
Adjustedf
|
||||||
| 1998b | 2003c | 2010d | 1998–2003 | 2003–2010 | 1998–2003 | 2003–2010 | |
| Overall | 1.75 | 2.22 | 2.07 | 1.048 (1.046–1.050) | 0.988 (0.987–0.990) | 1.047 (1.045–1.049) | 0.983 (0.981–0.984) |
|
| |||||||
| Age, y | |||||||
| 2 | 1.18 | 1.39 | 1.14 | 1.028 (1.023–1.032) | 0.971 (0.968–0.973) | 1.021 (1.017–1.026) | 0.965 (0.962–0.967) |
|
| |||||||
| 3 | 1.76 | 2.28 | 2.08 | 1.052 (1.049–1.056) | 0.988 (0.986–0.990) | 1.048 (1.044–1.051) | 0.982 (0.979–0.984) |
|
| |||||||
| 4 | 2.40 | 3.21 | 3.24 | 1.062 (1.058–1.065) | 0.998 (0.996–1.000) | 1.059 (1.056–1.063) | 0.993 (0.991–0.995) |
|
| |||||||
| Sex | |||||||
| Boys | 1.69 | 2.17 | 2.01 | 1.053 (1.050–1.057) | 0.986 (0.985–0.988) | 1.052 (1.049–1.055) | 0.980 (0.979–0.982) |
|
| |||||||
| Girls | 1.81 | 2.27 | 2.14 | 1.042 (1.039–1.045) | 0.991 (0.989–0.992) | 1.042 (1.039–1.045) | 0.985 (0.983–0.987) |
|
| |||||||
| Race/ethnicityg | |||||||
| Non-Hispanic white | 1.17 | 1.64 | 1.60 | 1.071 (1.067–1.075) | 0.987 (0.985–0.990) | 1.072 (1.068–1.076) | 0.987 (0.985–0.989) |
|
| |||||||
| Non-Hispanic black | 1.40 | 1.73 | 1.50 | 1.045 (1.040–1.050) | 0.979 (0.976–0.983) | 1.047 (1.042–1.052) | 0.980 (0.976–0.983) |
|
| |||||||
| Hispanic | 2.85 | 3.30 | 2.92 | 1.028 (1.025–1.031) | 0.982 (0.980–0.984) | 1.032 (1.029–1.036) | 0.982 (0.980–0.984) |
|
| |||||||
| American Indian/Alaska Native | 2.24 | 2.95 | 2.98 | 1.046 (1.025–1.068) | 1.003 (0.989–1.018) | 1.048 (1.027–1.070) | 1.003 (0.989–1.018) |
|
| |||||||
| Asian/Pacific Islander | 1.99 | 1.97 | 1.58 | 0.996 (0.985–1.007) | 0.968 (0.963–0.973) | 1.002 (0.991–1.013) | 0.966 (0.960–0.971) |
Abbreviation: OR, odds ratio.
Extreme obesity defined as body mass index for age and sex in the 120% or greater of the 95th percentile according to 2000 Centers for Disease Control and Prevention growth charts.
Bonferroni adjustments were used during the multiple subgroup comparisons with t tests. Results of all subgroup comparisons are all statistically significant at P < .005, except for the prevalence difference between American Indian/Alaska Native and Asian/Pacific Islander children.
Bonferroni adjustments were used during the multiple subgroup comparisons with t tests. Results of all subgroup comparisons are all statistically significant at P < .005, except the prevalence difference between Hispanic and American Indian/Alaska Native children.
Bonferroni adjustments were used during the multiple subgroup comparisons with t tests. Results of all subgroup comparisons are all statistically significant at P < .005, except the prevalence differences between Asian/Pacific Islander and non-Hispanic white or black children and between Hispanic and American Indian/Alaska Native children.
Unadjusted odds of being extremely obese for 1 year’s increase in time, calculated from piecewise logistic regression.
Adjusted odds of being extremely obese for 1 year’s increase in time, calculated from piecewise logistic regression controlling for age, sex, and race/ethnicity.
Trends are presented because they are important for identifying health disparities.
Comment
Results of a previous study6 based on a broader sample of children aged 2 through 4 years in PedNSS indicated that the prevalence of obesity increased from 12.4% in 1998 to 14.5% in 2003, but remained essentially unchanged until 2008. Few studies have focused on extreme obesity because of its relatively low prevalence in national data. With data through 2010, we found that the upward trends in obesity and extreme obesity turned downward slightly in 2003 among preschool-aged children living in low-income families. To our knowledge, this is the first national study to show that the prevalence of obesity and extreme obesity among young US children may have begun to decline.
Major strengths of the study were that weight and height were measured and the sample size was large. Limitations included that we only included children from 30 states and the District of Columbia and our findings may have been slightly affected by the cut points for the biologically implausible values.
The results of this study indicate modest recent progress of obesity prevention among young children. These findings may have important health implications because of the lifelong health risks of obesity and extreme obesity in early childhood.
Footnotes
Conflict of Interest Disclosures: The authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest and none were reported.
Disclaimer: The findings and conclusions of this report are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention.
Additional Contributions: The data used in this study were collected by the state Special Supplemental Nutrition Program for Women, Infants, and Children grantees and the California Child Health Disability Prevention Program. We gratefully acknowledge David S. Freedman, PhD, Deborah A. Galuska, PhD, Ashleigh L. May, PhD, and Barbara Polhamus, PhD, MPH, RD (all with the Centers for Disease Control and Prevention), for assisting in the preparation and revision of this research letter. None received compensation for their contributions.
Author Contributions: Dr Pan 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.
Study concept and design: Pan, Sherry, Dalenius, Grummer-Strawn.
Acquisition of data: Dalenius, Grummer-Strawn.
Analysis and interpretation of data: Pan, Blanck, Dalenius, Grummer-Strawn.
Drafting of the manuscript: Pan.
Critical revision of the manuscript for important intellectual content: Pan, Blanck, Sherry, Dalenius, Grummer-Strawn.
Statistical analysis: Pan, Grummer-Strawn.
Administrative, technical, or material support: Grummer-Strawn.
Study supervision: Blanck, Sherry.
References
- 1.Franks PW, Hanson RL, Knowler WC, Sievers ML, Bennett PH, Looker HC. Childhood obesity, other cardiovascular risk factors, and premature death. N Engl J Med. 2010;362(6):485–493. doi: 10.1056/NEJMoa0904130. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Sen Y, Kandemir N, Alikasifoglu A, Gonc N, Ozon A. Prevalence and risk factors of metabolic syndrome in obese children and adolescents: the role of the severity of obesity. Eur J Pediatr. 2008;167(10):1183–1189. doi: 10.1007/s00431-007-0658-x. [DOI] [PubMed] [Google Scholar]
- 3.Whitaker RC, Wright JA, Pepe MS, Seidel KD, Dietz WH. Predicting obesity in young adulthood from childhood and parental obesity. N Engl J Med. 1997;337(13):869–873. doi: 10.1056/NEJM199709253371301. [DOI] [PubMed] [Google Scholar]
- 4.Polhamus B, Dalenius K, Mackentosh H, Smith B, Grummer-Strawn L. [Accessibility verified November 1, 2012];Pediatric nutrition surveillance 2009 report. http://www.cdc.gov/pednss/pdfs/PedNSS_2009.pdf.
- 5.Flegal KM, Wei R, Ogden CL, Freedman DS, Johnson CL, Curtin LR. Characterizing extreme values of body mass index-for-age by using the 2000 Centers for Disease Control and Prevention growth charts. Am J Clin Nutr. 2009;90(5):1314–1320. doi: 10.3945/ajcn.2009.28335. [DOI] [PubMed] [Google Scholar]
- 6.Centers for Disease Control and Prevention (CDC). . Obesity prevalence among low-income, preschool-aged children—United States, 1998–2008. MMWR Morb Mortal Wkly Rep. 2009;58(28):769–773. [PubMed] [Google Scholar]