I know that BMI is imperfect, but it seems like a reasonable enough proxy measure when you're looking at whole-population trends for the USA. I don't think a mo

I know that BMI is imperfect, but it seems like a reasonable enough proxy measure when you're looking at whole-population trends for the USA. I don't think a more reasonable measure is really needed for this purpose. The BMI of athletes, or the limited options of people who live in food deserts, really seem like missing-the-forest-for-the-trees tangents in these sorts of discussions. According to the "rise of obesity in America" chart in the article, overweight-or-more among American women went from about 40% in 1960 to about 80% in 2014, with the highest percentage increase in the "extremely obese" subgroup. It's interesting to speculate about what drives it. But I don't think we can explain the trend with "they got shredded" or "they moved to food deserts" when we're talking about 40% of all American women. Maybe my intuition is wrong -- maybe there are enough people with high muscle mass that it's skewing entire charts like the Vox article shows. If you have citations to that effect, I'll read them with gratitude.