Is the Obesity Epidemic Leveling Off? Don’t be Too Sure.

Recently, CDC epidemiologists published an article in the Journal of the American Medical Association to the effect that there was not a significant change over the past 12 years in the nation’s obesity rate of 35.5% for adult men and 35.8% for adult women. Many, but not all, media reports interpreted this as meaning the obesity epidemic was leveling off. But is that really the case? Let’s look behind the headlines.

First, let’s keep in mind that, even if BMI is highly correlated with body fatness, it still misses a lot of groups for whom the BMI is a crude indicator of body fatness or excess adipose tissue.

Second, obesity prevalence should not be judged by making the BMI of 30 the sole criteria. This is better explained if you look at the “etables” accompanying the article. These breakdown the categories by age and gender and graphically plot the changes in BMI from 1988 to 2010. While each graph is slightly different, they basically all show three trends: the population at normal weight is declining, the population with a BMI of 30 is getting even heavier and the BMI level for 90% of each subgroup is progressively increasing. Bottom line: fewer Americans are at a normal weight and the overweight and obese subgroups are gaining weight.

    eFigure 4: Smoothed distribution of body mass index for women 20-39 years of age

Third, during this period when supposedly the obesity epidemic was leveling off, there were roughly 7 million new cases of obesity. How did this occur? Well, the population increased from about 281 million in 2000 census to 308 million in the 2010 census. That’s an increase of 28 million. About ¾ are adults or 21 million. Multiply that by the 38% prevalence rate and you get the 7 million figure. I would argue this figure is more relevant that the prevalence rate. These are the persons dealing with the health and social issues related to obesity. Further, the increases in the number of cases reflects the strain on the health care system, health care utilization and health care costs. So that figure is much more relevant to policy makers than the prevalence figure.

Fourth, being normal weight may not mean what it used to. Recent evidence indicates that many individuals at normal BMIs can have elevated levels of adipose tissue, meaning that they are at normal weight for height but are carrying extra fat tissue, putting them at risk for the metabolic syndrome.  

The Atlantic recently reported on a small study which showed that our dietary intake may be responsible for this phenomenon. 

Fifth, the CDC studies only look back over 12 years. When you consider that we are talking about changes in the fundamental parameters of the human body, this is a pretty short period of time. Taking a longer view, say 300 years, one sees a pretty dramatic upward march in both height and weight. This is supported by more recent studies going back to the 1948 Framingham Study.

Sixth, going forward, the “etables” show a very strong increase in BMI for women  and men ages 20-39 who are in their child-bearing years. This carries serious implications, especially for women, as obesity is associated with more difficult pregnancies as well as increased risk for obesity in the child.

The percentage of the population with a BMI over 30 is a relevant figure but it is not the whole story. Looking at the trends of obesity in the population indicate the picture is getting worse, not leveling off.

Morgan Downey, Chief Health Policy Officer at Obesity PPM, Policy Advisor for the STOP Obesity Alliance, and Publisher & Editor, The Downey Obesity Report