Freakonomics » The History of Obesity Revisited

(Photo: davidd)

We’ve blogged before about the obesity epidemic, and whether or not it is a recent phenomenon; John Komlos and Marek Brabec have argued that obesity rates actually began rising in the early 20th century. A new study (abstract; PDF) by Paul von Hippel and Ramzi Nahhas looks at 60 years of data on child obesity and finds that the increase in obesity rates started with children born in the 1970s and 1980s. Von Hippel wrote to us in an email:

Intrigued by the conflicting extrapolation results, Ramzi Nahhas and I decided to look at measurements that were actually taken before 1960. We analyzed the heights and weights of children in the Fels Longitudinal Study, an ongoing study that since the 1930s has measured children from shortly after birth until age 18. Most of the children come from the area near Dayton, Ohio, which is not a mirror of the nation but has an obesity rate that is close to the national average.

Our results lined up pretty well with the conventional wisdom, suggesting that the obesity epidemic is not particularly old but took off in the 1980s. We found that child obesity rates were low and stable among children born in the 1930s, 40s, 50s, 60s, and part of the 70s, and then rose rapidly through the 1980s and 1990s. Unlike Komlos and Brabec, we did not find evidence that the obesity epidemic was underway earlier. We did see some evidence that girls (but not boys) were getting a bit heavier before 1960, but significant numbers of girls didn’t break into the obese category until after 1980. In fact, much of the increase in girls’ weight before 1960 consisted of girls moving out of the underweight category and into a normal weight range.

Another source of historical evidence comes from measurements on young men drafted into the military, who on average grew about 1.5 BMI points heavier in the hundred years between the Civil War and the first national surveys in the 1960s. That’s a meaningful change—about 10 pounds for a man of average modern height—but it’s not as large as the 2.3 BMI point (16 pound) increase that national surveys observed among young men over just the 20-year period between 1980 and 2000. In other words, after 1980 the average BMIs of young men increased at a rate more than seven times faster than the rate observed over the previous century.

While it may be true that BMIs have been increasing slowly for a long time, the increases observed in recent decades are much faster and have pushed many adults and children over the obesity threshold in a remarkably short time. The trend is distressing, but to reverse it we only need to turn the clock back to 1980. We don’t need to go back to 1900.