Preventing Childhood Obesity: What Are We Doing Right?

Abstract

After decades of increases, the prevalence of childhood obesity has declined in the past decade in New York City, as measured in children participating in the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) and public school students, with the greatest reductions occurring in the youngest children.

Possible explanations were changes in demographics; WIC, day care, and school food policies; citywide obesity prevention policies, media messages; and family and community food consumption.

Although the decreases cannot be attributed to any one cause, the most plausible explanation is changes in food consumption at home, prompted by media messages and reinforced by school and child care center policy changes. Continued media messages and policy changes are needed to sustain these improvements and extend them to other age groups.

Recent reports have suggested that the prevalence of childhood obesity in the United States may have leveled off or declined after decades of increases since the late 1970s.¹ In New York City (NYC), significant decreases in childhood obesity have been found during the past decade among low-income children aged two to four years participating in the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC; Figure 1) and among NYC public school students in kindergarten through eighth grade.²

FIGURE 1—

Obesity prevalence among New York City children aged two to four years enrolled in Women, Infants, and Children and children aged five to 14 years enrolled in public school with timeline of selected policy changes.

Note. Obesity is defined as body mass index at or above the 95th percentile. Data for two- to four-year-old children are from New York State Pediatric Nutrition Surveillance Data (Personal communication, Lynn Edmunds, New York State Department of Health), and data for five- to 14-year-old children are from the New York City Department of Education.²

The reasons for this decline are unclear but very important. Understanding what has caused the decline in some jurisdictions may inform local efforts to accelerate decreases and promote decreases in other jurisdictions. Several hypotheses may explain these trends in childhood obesity during the past decade in NYC.

TRENDS IN CHILDHOOD OBESITY IN NEW YORK CITY

New York State measures yearly the height and weight of approximately 230 000 low-income two- to four-year-old children who are enrolled in WIC, including approximately 130 000 children in NYC. Among these NYC children, obesity prevalence—defined as body mass index (BMI) for age and sex at or above the 2000 Centers for Disease Control and Prevention growth chart 95th percentile—decreased from 18.0% in 2002 to 13.9% in 2010 (4.1 percentage points, or 22.8%; Figure 1). In comparison, among two- to four-year-old WIC participants from 30 US states and the District of Columbia consistently reporting data from 1998 through 2010, obesity prevalence increased through 2003 to 15.2% and then decreased to 14.9% in 2010.³ A study of WIC recipients aged three to four years in NYC found that from 2003 to 2011, obesity prevalence decreased among Blacks, Whites, and Hispanics but increased slightly among Asians.⁴ Hispanic children accounted for the largest proportion of WIC enrollees, had the greatest prevalence of obesity each year, and had the largest decrease in obesity prevalence over this time period.⁴

From 2006–2007 to 2010–2011, obesity prevalence among NYC public school students in kindergarten through eighth grade (students aged five to 14 years) decreased from 21.9% to 20.7% (a decline of 1.2 percentage points, or 5.5%); the decline was more marked in this group among children aged five to six years than among older children (Figure 1).² Obesity prevalence decreased in all age and racial/ethnic groups and in children of all school neighborhood poverty levels. Decreases in obesity prevalence were greater among White (12.5%) and Asian/Pacific Islander (7.6%) children than among Hispanic (3.4%) or Black (1.9%) children and were greater in schools in neighborhoods in which fewer than 10% of residents were below the federal poverty level (7.8%) than in schools in neighborhoods in which at least 30% of residents lived in poverty (2.9%).²

Across these two data sets, the annual decline in obesity prevalence was greatest in the youngest children (0.6% per year from 2003 to 2009 for three-year-old children⁴) and diminished as age increased (0.2% per year from 2006–2007 through 2010–2011 for 11- to 14-year-old children²). Taken together, these data suggest that obesity prevalence has been decreasing among children of all racial/ethnic and socioeconomic groups in NYC since 2002, with the greatest declines among the youngest children.

POSSIBLE EXPLANATIONS

We considered several changes in the population of children and their environment in NYC as possible explanations of these trends. Most changes did not fit the observed data well.

Demographic Changes

Overall enrollment in WIC increased in NYC between 2002 and 2011. It is possible that increased representation of subgroups with lower obesity prevalence contributed to decreases in obesity prevalence. An analysis of information on three- and four-year-old children in WIC in NYC from 2003 to 2011 showed that the proportions of Hispanic, White, and Asian children enrolled increased between 2003 and 2011, and the proportion of Black children decreased.⁴ What effects these changes would have on obesity prevalence among WIC recipients overall is unclear. However, obesity prevalence declined within groups stratified by race/ethnicity, with the exception of Asian children, suggesting that these demographic changes were not responsible for overall declines in obesity.

Increasing numbers of NYC public school students in kindergarten through eighth grade had BMI measurements recorded over time, from 61% in 2006–2007 to 93% in 2010–2011.² Schools participating earlier may have had students who differed from students in schools initiating participation later. However, we saw the same pattern of declines in obesity among schools with more than 90% participation in 2006–2007 and continuing to record BMI measurements for five years.

Women, Infants, and Children Food Package Changes

New York State implemented changes to the WIC food package in January 2009 that reflected Institute of Medicine recommendations to include a wider variety of healthy foods⁵ and were consistent with the US Department of Agriculture’s revision to the WIC Food Package.⁶ Changes included cash-value vouchers for fruits and vegetables, additional whole-grain foods, milk limited to low fat (1%) or nonfat for children older than two years, and restrictions on juice.

This revision influenced families’ food purchases for children.⁷ However, given that these changes were implemented in 2009, they cannot explain the decreases in obesity prevalence among WIC recipients between 2002 and 2009 and would be unlikely to explain much of the change in obesity prevalence among five- to 14-year-old public school children between 2006 and 2011.

Day Care Food and Physical Activity Changes

An amendment to NYC’s health code, effective in January 2007, required all group day care centers with health department permits to limit beverages to 100% juice (maximum six oz a day), 1% milk (for those older than two years), and water, with water available and easily accessible to children throughout the day. In addition, the amendment limited television and video viewing and required 60 minutes of physical activity daily. New York State’s “Eat Well Play Hard” program promoted physical activity and improved diet for children in nutrition programs, beginning in 2008 in NYC.

Limiting calorie intake from beverages and promoting physical activity might have had an impact on obesity prevalence among preschool-aged children in WIC in NYC and possibly in later years among school-aged children after day care attendees aged into this group. However, the timing of these changes was too late to entirely explain decreases in obesity prevalence that began after 2002 among preschool-aged children. They could have reinforced changes that may have occurred earlier, though, such as parents’ altering the beverages they offer young children at home.

School Food and Physical Activity Changes

New York City has undertaken several interventions to improve nutrition in public schools in the past decade, including elimination of sugary drinks from beverage vending machines in 2003–2004, shifting from whole milk to 1% fat and skim milk in school meals in 2005, increasing fiber in school meals and introducing water jets in 2008–2009, and limiting calories in beverage vending to less than 10 calories per eight ounces in elementary and middle schools in 2010–2011. In addition, since 2009, the Move to Improve program has trained more than 6000 elementary school teachers to integrate physical activity into classroom academics. Beginning in 2005, personalized BMI reports were sent to parents of public school students.

Some of these changes may have had an impact on obesity among students attending public elementary schools. For example, switching from a cup of whole milk to 1% fat milk could save approximately 40 kilocalories daily,⁸ or approximately 8000 kilocalories over a 200-day school year. It is also possible that BMI reports may have raised concerns among parents and persuaded them to alter food patterns at home. BMI reporting has a mixed record: Among Arkansas students in kindergarten through 12th grade, a halt in progression of obesity rates was seen after schools began sending BMI reports to parents,⁹ although this program was implemented concurrent with other school-based initiatives, whereas in California BMI reporting was not followed by changing trends in BMI scores.¹⁰

Citywide Obesity Prevention Efforts

In 2008, the city established nutrition standards for the 270 million meals and snacks various government agencies serve each year, including restricting trans fat, sodium, and sugar; requiring that fruit and vegetables be served at lunch and dinner; requiring that water be served at every meal; and prohibiting sugary drinks. Also in 2008, NYC mandated calorie labeling on menus and menu boards in chain restaurants.

Beginning in 2009, the city ran high-profile paid advertisements on subway trains and television warning about the obesogenic effects of sugary drinks. In addition, the city advocated for a sugary drink tax in 2009 and 2010, but it did not pass the New York State legislature. The city also proposed a restriction on sugary drinks in the Supplemental Nutrition Assistance Program in 2010 that was rejected by the US Department of Agriculture in 2011, and the Board of Health passed a 16-ounce sugary drink restaurant portion cap in 2012 that was challenged by the beverage industry and overturned in court.

These initiatives may have raised awareness of obesity and of the importance of a healthier diet, with a particular emphasis on the value of reducing sugary drink consumption. In addition, both adults and adolescents in NYC reported decreasing sugary drink consumption, with the percentage of adults reporting consumption of at least one sugary drink a day decreasing from 32.6% in 2008 to 28.2% in 2012.¹¹ None of these initiatives occurred early enough to explain the decreasing obesity prevalence among NYC children beginning in the early 2000s, but as with changes in school and day care food, they could have reinforced change toward healthier diets in children that began earlier.

Changes in Breastfeeding

Breastfeeding has been found to be associated with decreased risk of childhood obesity,^12,13 although this decrease may be the result of unmeasured confounders (e.g., mothers choosing to breastfeed may also choose other healthier behaviors for themselves and their children). The percentage of new mothers breastfeeding in NYC was likely increasing during the early 2000s, based on data from different surveys.^14,15

However, breastfeeding was also increasing from 1980 to 2000 in NYC ¹⁵ while childhood obesity rates were increasing. Nevertheless, even if these increases in breastfeeding are not causally related to decreases in obesity, they may demonstrate that during this period of time mothers were adopting behaviors with their children to improve their children’s health.

Changes in Consumption Prompted by Media Messages

One finding that is evident from both WIC and public school data is that the prevalence of obesity decreased impressively among children born within a few years after 2002. In national surveys, after increasing for decades, reported calorie intake among two- to six-year-old children decreased from an average of 1735 kilocalories per day in 2003–2004 to an average of 1614 kilocalories per day in 2005–2006 and 1584 kilocalories per day in 2007–2008.¹⁶ A decrease in calories from sugary drinks and fruit juice accounted for 53 kilocalories per capita per day between 2003 and 2008, or one third of the 151-kilocalorie per capita per day decline during this time period. Other contributors to the decrease were reductions in calories from milk, ready-to-eat cereals, and poultry, each consumed at 18 fewer kilocalories per capita per day.¹⁶

Increasing public awareness of the risk of obesity in childhood may have prompted a change in what parents feed their children and could partially explain this consumption pattern. Media reports in the United States on obesity and childhood obesity increased sharply in the early 2000s, with news stories mentioning childhood obesity nearly doubling between 2001 and 2003 and continuing at 2003 levels through 2007 (Figure 2).¹⁷ Media reports on obesity in general (not only in children) also rose dramatically in the same period, but parents may be more motivated to make changes they perceive will improve their children’s health than to adopt changes for themselves. Our data show that decreases in obesity prevalence have been greatest in the youngest children, which might reflect that their diets are more under parental control than those of older children. Although we do not have local data for NYC children on calorie intake or food consumption trends, NYC parents were likely to be aware of media messages and may have changed what they fed their children. In addition, local media attention to obesity and sugary drinks spiked in the years in which NYC proposed the sugary drink tax, Supplemental Nutrition Assistance Program, and portion policies, which may have had a reinforcing effect.

FIGURE 2—

Counts of news stories that mentioned search terms for obesity (n = 30 014) and childhood obesity (n = 16 102): United States, 2000–2009.

Source. Reproduced from Barry et al.¹⁷ Reproduced with permission from Pediatrics, Vol. 128, pp. 132–145. Copyright © 2011 by the American Academy of Pediatrics.

CONCLUSIONS

The good news is that the prevalence of childhood obesity in NYC is declining after decades of increases nationally. We cannot be certain what factors are most directly responsible for this improvement. The pattern of a greater reduction in obesity in preschool children and the timing for the initial reversal, occurring after the surge in media reports about childhood obesity but before most policy changes, suggests that the initial mechanism was the increased media attention prompting changes in food served to children by parents at home, which led to reductions in calorie intake, especially from sugary drinks and fruit juices. This explanation would be consistent with the finding of greater declines in obesity in younger children, who are more influenced by parental actions at home than are older children. The fact that obesity rates in school-aged children fell more quickly in higher income neighborhoods may indicate that parents with higher income and more education were more responsive to the media messages in changing their children’s diets. These changes may have later been reinforced by changes in day care and school food and physical activity policies as well as NYC-specific media messages and media coverage of policy proposals.

There is a dynamic relationship between individual health-related behavior change and policy changes that support that behavior change. Policies that make healthier behaviors easier are typically followed by changes in those behaviors. At the same time, individuals trying to alter their behavior to be healthier are more supportive of policies that make that behavior change easier, which increases the likelihood that policymaking bodies will approve them. Individual behavior change sometimes both precedes and follows population-level interventions that facilitate risk reduction. For example, per capita cigarette consumption in the United States began to decline in 1964,¹⁸ the year that the first surgeon general’s report on smoking and health¹⁹ was published, stimulating considerable media coverage that made the report one of the top news stories in 1964.²⁰ This change helped make possible the enactment of excise taxes for cigarettes in the 1960s. Smoking prevalence subsequently declined from more than 40% in 1965 to less than 25% in 2000²¹ before there was sufficient political and social will to implement smoke-free air policies in many localities and states.²² However, further declines in smoking rates seemed to require additional policy interventions. For example, smoking prevalence in NYC remained stable at about 22% from 1993 to 2002 but decreased to 14% by 2010 after sharp increases in city and state excise taxes and passage of legislation making all NYC workplaces smoke-free.²³

The experience in NYC suggests, without providing definitive guidance, ways to accelerate the declines in childhood obesity seen so far. Media and communications interventions, including reducing advertising of unhealthy food to children as well as continued messages about obesity and the risks of unhealthy products, are likely to help promote and sustain changes in food consumption and decreases in obesity. In addition, policies that promote access to healthier food and limit access to less healthy food would reinforce these media messages and individual shifts in diet; these policies can be promoted through media advocacy. For example, healthier food policies could be implemented at all venues with children, including after-school programs and summer camps. Healthier food could be made available and promoted at grocery and convenience stores. Healthier Supplemental Nutrition Assistance Program policies could encourage purchase of more nutritious foods that are less likely to promote obesity.

Despite some recent signs of progress among children, obesity in both children and adults is still a substantial problem in NYC, in the United States, and globally. Reductions in obesity rates in young children do not necessarily predict reductions in older children and adults; in fact, one cohort study demonstrated that children born during the obesity epidemic had lower preschool BMI values but higher school-aged BMI values than children born earlier.²⁴ Sustained improvement in obesity rates across the entire age spectrum will require continued media attention and policy change to support healthier choices for everyone.