Like President George W. Bush, First Lady Michelle Obama is declaring “Mission Accomplished” in her war on childhood obesity. Appearing on ABC News’ Good Morning America today, she told Robin Roberts, co-anchor, “Our kids are eating better at school. They’re moving more. And we’re starting…to see a change in trends. We’re starting to see rates of obesity coming down like never before.”
Really? Time for a little fact checking. But first, a note of caution. It seems whenever someone in the political spotlight, as First Lady Michelle Obama is, claims victory, they are taking a risk. You might recall that in 2003, a dynamic governor of Arkansas signed the nation’s first comprehensive childhood obesity legislation. The law garnered much national attention for its provision to sending parents a BMI report card, as well as limiting access to vending machines. On August 17, 2006, Governor Mike Huckabee declared, “Arkansas has halted the progression of childhood obesity. We have stopped the locomotive train of childhood obesity in its tracks, now it’s time to turn the train around and move full speed ahead to healthier living.” Medical News Today
How’s that going? Reports by the Arkansas Center for Health Improvement show little difference in obesity rates in children from 2005-6 to 2011-12. (Note: there was a change in categorizing childhood obesity during this period but it does not appear to be significant.)
Now, the statement, “We’re starting to see rates of obesity coming down like never before,” just isn’t accurate. What’s the source of the confusion? Well, it is like playing ‘Telephone.’
In January 2012 Cynthia Ogden and colleagues from the National Center for Health Statistics, part of the CDC, published a paper in the Journal of the American Medical Association which indicated a leveling of obesity rates between 2007-8 and 2009-10. At the time, many pointed out that this was a very small time period – 4 years – to use in reaching an important conclusion. The paper did report a significantly greater prevalence in obesity in boys aged 2-19 over a 12 year period but not in females. There was also a significant increase in BMI among adolescent males 12-19 but not among any other age group or among females.
In May of 2012, Department of Health and Human Secretary Kathleen Sebelius told the attendees at the Weight of the Nation conference, “The good news is that in the last decade, we’ve seen the rise in obesity rates begin to stabilize in some areas. A recent study out of Massachusetts even showed a decline in obesity among children under 6 in parts of the state. That’s progress.”
In December of 2012, the Robert Wood Johnson Foundation issued a brief, “Declining childhood obesity rates-where are we seeing the most progress.” The report was widely covered in the media.
The Robert Wood Johnson Foundation “Declining childhood” report cited Philadelphia, New York City, Mississippi and California as places reporting declines in childhood obesity rates. (Massachusetts appears to have dropped off the list of success stories. The study Secretary Sebelius referred to was a census in a multi-center pediatric practice.)
California’s reported decline was based on a Robert Wood Johnson Foundation funded study. I critiqued it at the time for the wide variation in reports of increases and decreases by county in the state. See Is California’s Childhood Obesity Picture Really Improving? I noted that the number of overweight/obese students had actually gone up even though the prevalence rate had gone down 1.1%, due to an increase in enrollment.
Mississippi’s prominence in the RWJF report may surprise many since CDC rates it as one of the states with the highest rates of obesity, 34%. Regardless, the RWJF report referred us to another RWJF funded project, the Center for Mississippi Health Policy. The “Declining Childhood” report states, “In 2006, the Mississippi State Board of Education set nutritional standards for food and beverages sold in school vending machines. The Healthy Students Act of 2007 required the state’s public schools to provide more physical activity time, offer healthier foods and beverages, and develop health education programs…Starting in 2008, Mississippi’s Childcare Licensure Division helped facilitate the Color Me Healthy program and training classes to create a healthier environment in childcare centers. Mississippi’s Fruits and Veggies-More Matters program reached more than 15,000 residents in 2009 through worksite wellness programs, health fairs and school events…Mississippi reported a significant drop in overweight and obesity only among White students. However, rates for Black students appear to be leveling off, which is a major shift after years of steady increases.”
Actually, it’s a bit more complicated than that. The report from the Center for Mississippi Health Policy states, “In 2011, the combined prevalence of overweight and obesity for all public school students was 40.9%…Statistically significant declines in overweight and obesity were found among white students and elementary students from 2005 to 2011…In 2011, the prevalence of overweight and obesity was significantly lower among white students than black students and significantly lower among white female students than black female students in all three grade categories.”
But, rates for black males and females are actually increasing from their low point in 2007. So they are ‘leveling off’ to the extent that they are back to earlier, higher levels.
Philadelphia was cited in the “Declining Childhood” report for its efforts to help corner stores offer fresh foods, connect schools with local farms, bring supermarkets to underserved areas and ensure that farmers’ markets accept food stamps, as well as improving the nutritional quality of foods and beverages in the schools. Data from Philadelphia is strong. A report showed that, between 2006-2007 and 2009-2010, the prevalence of obesity in Philadelphia schools decreased from 21.5% to 20.5%. Decreases were generally smaller in groups with the highest prevalence, including high school students, Hispanic males and African American females. But he report noted, “The inconsistency of findings between subgroups and the small increases in obesity, severe obesity or both in some groups in the most recent year of data indicate that it is not yet certain that the epidemic increases in child obesity are over.”
New York City has by far the best combination of enacting comprehensive approaches to childhood with pretty solid data showing improvements. A CDC study showed the prevalence of obesity declined in grades K-8 in public schools from 2006-7 to 2010-11 by 5.5%. Obesity declined significantly in all age groups and in all socioeconomic and racial/ethnic populations. However, the decrease was smaller among black and Hispanic children than among Asian/Pacific Islanders and white children. By age, the largest decrease was seen among children age 5-6 years and, within this age group, the largest decline was among white children. Among children age 5-6, large differences were observed in obesity reduction by neighborhoods with high income white neighborhoods having greater decreases than in poorer black neighborhoods. The editorial note suggested that the large changes in ages 5-6 indicated that the home environment or pre-school was particularly important.
So, what are the questions?
First, are national trends in childhood obesity reversing? We do not have good long term data yet. New York and Philadelphia are showing some results but even though these are big cities they are still a small part of the country. Even in those cities, there is a great diversity in patterns among subgroups.
Second, are rates of childhood obesity at least leveling off? Well, this is tougher to assess. There is some positive indication as in the JAMA study in January 2012. In addition Rokholm and colleagues reported on leveling of the epidemic in children and adolescent in Australia, Europe, and Japan with some decreases along with increases. Two additional considerations. Most studies use the Body Mass Index which is a crude instrument for determining body weight. Its use for children and adolescents in the United States is of very recent vintage. Because height is twice the value of weight in the BMI equation, and as children grow in height so rapidly, sudden changes in height can skew the picture. In addition, BMI cannot measure excess visceral adipose tissue in normal BMI individuals and so it may miss an important part of the population.
One’s time frame is also important. Some of the comparison periods in the studies above are very short, five years or so. Others may be 10 or more. So what is your frame of reference? If you look at the long term trends in the United States, the prospect of leveling much less reversing this trend would be unprecedented in our history.
Of course, one can go back some 300 years to observe powerful trends in ever greater human height and weight, but not without periods of regression. See The Changing Body.
Third, but at least we have some models of what works, correct? Not exactly. The Robert Wood Johnson Foundation report and the First Lady’s Let’s Move campaign each are taking some credit for the changes we are seeing. But when you add in the diverse activities of New York, Philadelphia, Mississippi, and California the picture gets more cloudy of what works and what doesn’t. More focused examinations including one by Jeffrey Mechanick,another which I co-authored and as well as other in depth Cochrane Review on interventions for obesity in children found limited evidence of the effectiveness of any specific policies. It should be noted that many communities have undertaken programs like those reported in California, Mississippi, New York City and Philadelphia but we do not have reports from them.
Finally, in obesity discussions, cause and effect are often mixed up. And well intentioned participants may omit some factors out of either ignorance or lack of complete understanding of how an external factor may affect a conclusion. Two such externalities come to mind here. The first is that almost all these studies of a leveling effect span, at least in part, the recent recession, starting around 2007. The obesity literature has not assessed what impact this might have had. On the one hand, economic insecurity may have caused individuals or families to reduce food expenditures, such as meals away from home, which might cause a reduction in weight. On the other hand, an increase in stress due to economic uncertainty might have driven up the consumption of comfort foods. By the same token, individuals looking for employment may be more sensitive to their appearance. Perhaps closer to home, during this period, the Federal Trade Commission in a December 2012 report indicates a reduction in food marketing to children and adolescents during the same timeframe. If food marketing to children is as influential as many advocates believe, this may be a significant, if overlooked contributor to the putative leveling.
Bottom Line: It’s complicated. Keep On Moving but put the champagne on ice.