America's Obesity Crisis Is Ending - As Long As You're Not Poor

After years of warnings that Americans are gaining more and more weight, the obesity rate in the United States is finally leveling off, some studies suggest.

Time to celebrate? Not so fast, caution researchers in a Jan. 13 report in Proceedings of the National Academy of Sciences (PNAS).

A closer look at the data indicates that in this case the average doesn't tell the whole story.

The researchers looked at adolescent obesity (an important predictor of adult obesity) by analyzing two large health surveys. While they confirmed that the obesity rate among adolescents is no longer climbing - after it "more than doubled over the past three decades" - this did not hold true across different socioeconomic groups.

Obesity is decreasing among adolescents who come from well-educated families, but it has continued to increase in poor teens. Looking at the obesity rate overall, this reads as a plateau.

"There's a danger in aggregating trends in groups that are so different," study author Kaisa Snellman, a sociologist at INSEAD and the Saguaro Seminar at Harvard Kennedy School of Government, tells Business Insider. "The picture becomes very murky."

Take a look at these charts (NHANES and NSCH are the two health surveys):

For adolescents whose parents have at least a college degree (black dots), obesity is on the decline. But for those whose parents have no more than a high school diploma (red dots), obesity rates are still rising - and the gap seems to be getting larger. (The short lines above and below the dots show the possible margin of error.)

Why are the rates of obesity so different among members of different socioeconomic groups, to the point where they are actually trending in opposite directions? The answer is more complicated than it might seem.

Recent research has questioned the popular idea that a lack of access to healthy food is a major contributor to obesity (for example, that poorer people live in "food deserts"). The new results of the PNAS study seems to fuel the finding that the lack of healthy food causes obesity, since the main difference between the two groups is parental education.

When the researchers looked deeper into the survey data to examine eating habits, they found that the downward trend in terms of daily calories consumed was steeper among kids of college-educated parents than among kids of high-school-educated parents. But both groups were eating fewer calories per day in 2009-2010 than they were twenty years earlier, and the kids of high-school-educated parents were actually eating fewer calories overall than the kids of college-educated parents.

Of course, nutrition is about more than total calories consumed. "Someone could get scurvy, lose all their teeth, and still not gain weight," Snellman points out. Since the survey data doesn't tell us exactly what the teens were eating, it's hard to tease out all of the possible health differences between the groups. But "when it comes to obesity," Snellman says, "it's really the total calories that matter more."

The PNAS study indicates that a difference in exercise habits, not eating habits, may be what's really driving the adolescent obesity gap. Take a look at these physical activity charts, which seem to mirror the obesity rate charts above:

The NSCH counts children as "active" if they had at least 20 minutes of physical activity in the past seven days; the NHANES counts children as "active" if they had at least 10 minutes of physical activity in the past 30 days, an alarmingly low bar. (Only a quarter of U.S. adolescents get the recommended 60 minutes of activity every day.)

The disparity in rates of physical activity increased sharply during the most recent round of surveys, just like the disparity in obesity rates.

While the proportion of high socioeconomic-status adolescents getting at least some physical activity in a 30-day period has stayed about the same in recent years (most recently 94.7%), the latest NHANES counted just 82.1% of low socioeconomic-status adolescents as active - down from 90% in 2003.

The researchers can't say for sure what's causing the different levels of physical activity. Snellman notes that adolescents from poorer families are less likely to participate in school sports, which can cost money, and are also more likely to attend schools that have had to reduce or eliminate their sports programs.

Some poor neighborhoods have fewer opportunities for exercise (though this doesn't seem to be the case everywhere), and the safety of a neighborhood can also affect whether kids are able or allowed to run around outside or use parks and other recreational facilities.

Rich and middle-class adolescents are also four times as likely as poor adolescents to have a regular doctor, so they may be more exposed to messages about the importance of regular exercise.

More contact with doctors, safer neighborhoods, more green spaces, and better school programs may be part of the solution. And parents, Snellman notes, "can do a lot to encourage kids to move" - even if it's just suggesting an active video game or a walk to the grocery store.