Obesity in K–8 Students — New York City, 2006–07 to 2010–11 School Years

Obesity in K–8 Students — New York City, 2006–07 to 2010–11 School YearsWeekly
December 16, 2011 / 60(49);1673-1678

Overweight and obese children are more likely to develop risk factors that can lead to respiratory, metabolic, and cardiovascular illness (1–3). The increase in prevalence of childhood overweight and obesity in the United States since the 1960s has been well documented (4). In New York City, in 1996, an estimated 19.7% of third grade children and 21.2% of sixth grade children in public and private schools were found to be overweight (5); in 2003, an estimated 43% of the city's public elementary school students were found to be overweight, and 24% of these students were obese (6). To update city data on childhood obesity and evaluate public health interventions, the New York City Department of Health and Mental Hygiene analyzed body mass index (BMI) data for public schoolchildren in kindergarten through eighth grade (K–8), using data from the 2006–07 to 2010–11 school years. This report summarizes the results of that analysis, which found that, overall, the prevalence of obesity in grades K–8 decreased 5.5%, from 21.9% in 2006–07 to 20.7% in 2010–11. Obesity decreased significantly among children in all age groups and in all socioeconomic and racial/ethnic populations; however, the decrease was smaller among black (1.9%) and Hispanic (3.4%) children than among Asian/Pacific Islander (7.6%) and white (12.5%) children. Despite the decreases in obesity, continued public health interventions are needed to further reduce the prevalence of obesity and to eliminate disparities among schoolchildren in New York City.

According to the American Community Survey,* approximately 900,000 children attend elementary and middle schools in New York City, and 78% of those attend a public school. In 2005, during physical education classes, the New York City Department of Education (DOE) began annually measuring the BMI (weight [kg] / height [m]2) of public school students in grades K–12 and the fitness of students in grades 4–12 as part of an overall fitness program. Physical education teachers were trained in taking height and weight measurements using standard protocols developed by DOE. Using these measurements, DOE now provides students and their parents with an annual assessment of the child's BMI and fitness status. The findings in this report are based on analysis by the New York City Department of Health and Mental Hygiene of BMI data obtained from DOE records, including information on student height, weight, race/ethnicity,† date of birth, sex, grade, place of birth, language spoken at home, school postal code, and free lunch status (a proxy measure of poverty).

Data were limited to children in grades K–8 who were aged 5–14 years and enrolled in non–alternative and non–special education public schools. During the 5 school years studied, approximately 2 million BMI measurements were completed for 947,765 K–8 students. Among individual students, the number of annual measurements ranged from one to five. Biologically implausible measurements (2%–3% of all measurements), as defined by CDC's BMI percentile-for-sex and age criteria, were excluded from analysis. Children with BMI at or above the 95th percentile were categorized as obese. The percentage of enrolled K-8 students measured as part of the New York City fitness program was 61% in 2006–07, 76% in 2007–08, 86% in 2008–09, 92% in 2009–10, and 93% in 2010–11.

For each school year, observations were weighted to ensure that data were representative of the enrollment population for that year. Weights were calculated using a raking process, with race/ethnicity, a combination of borough and district public health office (DPHO) neighborhoods (neighborhoods defined by low income and disproportionate rates of morbidity and mortality), free lunch status (free versus not free), grade, sex, age, and school type (elementary versus middle) as population marginal control totals.§ To test for obesity prevalence trends from 2006–07 to 2010–11, a multivariate model was built that included a linear term for time, along with sex, age, race/ethnicity, school borough, free lunch status, DPHO, place of birth, language spoken at home, and an interaction of age, sex, and race/ethnicity, as covariates. School and student codes were used as cluster variables, and statistical procedures that account for intercluster correlation were used to ensure that variance estimates were calculated correctly. Separate multivariate models were built to test trends for age group, race/ethnicity, and socioeconomic status. The significance level for all analyses was set at p<0.05. For presentation of prevalence estimates by school neighborhood poverty, school postal codes were characterized by the percentage of residents living below the federal poverty level (as defined by the 2000 U.S. Census). The percentage of residents living below the poverty level in the school postal code area was categorized as low (<10% of residents), medium (10% to <20%), high (20% to <30%), and very high (≥30%).

From 2006–07 to 2010–11, the overall prevalence of obesity in grades K–8 decreased 5.5%, from 21.9% to 20.7% (Table). The prevalence of obesity decreased significantly among children in all age groups, neighborhood poverty levels, and racial/ethnic populations. By age group, the largest decrease was observed among children aged 5–6 years (9.9%, from 20.2% to 18.2%) (Figure 1). Among children in this age group, the largest decrease was among white children (23.6%, from 16.1% to 12.3%), followed by a decrease of 13.5% (from 15.5% to 13.4%) among Asian/Pacific Islanders, 7.0% (from 18.5% to 17.2%) among blacks, and 6.0% (from 24.9% to 23.4%) among Hispanics (Table).

Among children aged 5– 6 years, large differences also were observed in obesity reduction by school neighborhood poverty level, with a decrease of 16.7% (from 16.8% to 14.0%) in low poverty areas, compared with a nonsignificant decrease of 2.7% (from 22.2% to 21.6%) in very high poverty areas. Among children in all age groups, the greatest decreases were observed among white children (12.5%, from 17.6% to 15.4%) and Asian/Pacific Islander children (7.6%, from 14.5% to 13.4%) (Figure 2). After further stratification by age group, race/ethnicity, and neighborhood poverty level, decreases in the prevalence of obesity were not consistently significant among all children attending school in neighborhoods with high poverty levels (Table).


Reported by
Magdalena Berger, MPH, Kevin Konty, MS, Sophia Day, Lynn D. Silver, MD, Cathy Nonas, MS, Bonnie D. Kerker, PhD, Carolyn Greene, MD, Thomas Farley, MD, New York City Dept of Health and Mental Hygiene; Lindsey Harr, New York City Dept of Education. Corresponding contributor: Magdalena Berger, mberger@health.nyc.gov, 347-396-4134.

Editorial Note
The findings in this report indicate that, from 2006–7 to 2010–11, the prevalence of obesity among New York City public elementary and middle school students decreased overall and across all demographic groups. Decreases in obesity prevalence were most notable among children aged 5–6 years and were greater among white and Asian/Pacific Islander children than among Hispanic and black children.
In the last decade, the prevalence of obesity appears to have stabilized nationally among preschool and school-aged children (7,8). Although studies in New York and California have shown recent declines in pediatric obesity (9,10), this report describes the largest documented decline to date in a large city in the United States, using comprehensive K–8 public school data.

During 2003–2009, New York City implemented multiple interventions to address the increase in childhood obesity. These measures included establishment of regulations to require improved nutrition, increased physical activity time and limited screen time (e.g., video game, television, or computer) in group child care, provision of extensive nutrition education training and physical activity equipment to 80% of group child care centers, and provision of on-site nutrition education workers at 300 centers. School nurses were trained to identify and monitor children at high risk for obesity and to know when to notify parents that a problem exists and when to refer children for additional medical care. Nurses also were given information about obesity prevention programs offered at schools and in the community. In schools, substantial improvements in cafeteria food were made, including a shift from whole milk to 1% fat and skim milk in 2005. The number of middle schools in a before-school and after-school physical activity program was expanded from 40 to 225, and nearly 4,000 elementary classroom teachers were trained to provide in-class physical activity breaks.
Additionally, individualized BMI and fitness reports were sent to all parents of K–8 public school students beginning in 2005, with guidance on how to help their children maintain a healthy weight.

The findings in this report are subject to at least two limitations. First, although this study uses objectively measured height and weight data collected by trained physical education teachers, which is likely an improved method compared with surveys using self-reported data, some measurement error is possible.
Measurement equipment was not standardized across schools, but obvious measurement errors (i.e., implausible height or weight values, as determined by CDC's BMI percentile-for-age and sex criteria) were excluded from analysis. Second, although DOE sought to assess all eligible children, certain schools that began participating in early, rather than later years might differ in some unmeasured way. However, no evidence indicates that trends were caused by changes in socioeconomic or demographic characteristics of the public school population over time. Additionally, in each year, BMI values for those participating were weighted to be representative of the entire enrollment for that year, thus minimizing selection bias.

The objectives of this study were to create obesity prevalence estimates that are representative of the New York City public school population and to examine trends. Because of the nature of this analysis, a causal relationship cannot be inferred between the BMI and fitness interventions implemented by New York City in schools and the decrease in prevalence of child obesity described in this report. Nevertheless, the trend toward reduced prevalence of obesity is encouraging. The larger decreases in obesity prevalence among children aged 5–6 years suggest that changes in the preschool or home environment might have been particularly important. The smaller reductions among older children might indicate that changes in school-based nutrition and physical activity programs also helped reduce the prevalence of obesity. Nonetheless, the uneven gains among minorities and those with lower incomes highlight the need for further targeted measures to reduce childhood obesity.


Acknowledgments
Lori Rose Benson, Michael Larkin, Subir Saha, New York City Dept of Education; Andrew Goodman, Karen Lee, Roger Platt, New York City Dept of Health and Mental Hygiene.


References
1.Duncan GE, Li SM, Zhou XH. Prevalence and trends of a metabolic syndrome phenotype among U.S. adolescents, 1999–2000. Diabetes Care 2004;27:2438–43.
2.Okosun IS, Boltri JM, Lyn R, Davis-Smith M. Continuous metabolic syndrome risk score, body mass index percentile, and leisure time physical activity in American children. J Clin Hypertens 2010;12:636–44.
3.Visness CM, London SJ, Daniels JL, et al. Association of childhood obesity with atopic and nonatopic asthma: results from the National Health and Nutrition Examination Survey 1999–2006. J Asthma 2010;47:822–9.
4.Ogden CL, Fryar CD, Carroll MD, Flegal KM. Mean body weight, height, and body mass index, United States 1960–2002. Adv Data 2004;27:1–17.
5.CDC. Prevalence of overweight among third- and sixth-grade children—New York City, 1996. MMWR 1998;47:980–4.
6.Thorpe LE, List DL, Marx T, May L, Helgerson SD, Frieden TR. Childhood obesity in New York City elementary school students. Am J Pub Health 2004;94:1496–500.
7.CDC. Obesity prevalence among low-income, preschool-aged children—United States, 1998–2008. MMWR 2009;58:769–73.
8.Ogden CL, Carroll MD, Curtin LR, Lamb MM, Flegal KM. Prevalence of high body mass index in US children and adolescents, 2007–2008. JAMA 2010;303:242–9.
9.Madesen KA, Weeden AE, Crawford PB. Disparities in peaks, plateaus, and declines in prevalence of high BMI among adolescents. Pediatrics 2010;126:434–42.
10.Sekhobo JP, Edmunds LS, Reynolds DK, Dalenius K, Sharma A. Trends in prevalence of obesity and overweight among children enrolled in the New York State WIC program, 2002–2007. Public Health Rep 2010;125:218–24.

* Available at http://www.census.gov/acs/www.
† Students were categorized as Asian/Pacific Islander, black, white, or Hispanic. Those categorized as Asian/Pacific Islander, black, or white all were non-Hispanic. Those categorized as Hispanic might be of any race.
§ The weighting of the data follows procedures similar to those for nonresponse adjustments (or post-stratification) in surveys. In particular, the weighting is similar to that used by the Youth Risk Behavior Survey in New York City. However, the large sample size allowed for adjustment over some additional variables. The control totals are tabulated directly from the DOE enrollment file. An iterative proportional fitting procedure (generally referred to as raking) was used to adjust the observations in the dataset to match the following marginal control totals: EMH (elementary or middle school) by school borough by DPHO status by race/ethnicity, EMH by school borough by DPHO status by grade by sex, EMH by school borough by DPHO status by meal code (free, full, or reduced), and EMH by single year of age (truncated depending on EMH status). Race/ethnicity was a five-level variable (Asian/Pacific Islander, Hispanic, black, white, and other) with groups contributing less than 5% of control total population collapsed into the largest group.


Obesity in K–8 Students — New York City, 2006–07 to 2010–11 School Years