Fitness vs. Fatness

This study weighs in on the question of whether fitness or fatness is a more important marker of health (abstract here, press release here). This is a nice, clear one. The researchers followed 14,345 men (with an average age of 44 at the start of the study). They did two detailed physical exams just over six years apart (on average), then followed the men for another 11 years to see which ones died, and why.
Basically, there are two variables of interest:

1. Did the men lose weight, gain weight, or stay stable between the two exams?
2. Did the men lose fitness (as measured by a maximal treadmill test), gain fitness, or stay stable?

The result: those who maintained or improved fitness were less likely to die by ~30-40% compared to those who lost fitness — even when you control for factors like BMI. Obvious and expected.
But what about weight? As you’d guess, those who gained weight were more likely (by 35-39%) to die of heart disease than those who lost weight or stayed stable. BUT if you take into account changes in fitness, then the effect of changes in weight almost disappeared. So this is further support for Stephen Blair’s argument that it’s fitness that matters, not fatness, when it comes to predicting health (and Blair is, indeed, one of the authors of this paper). As the press release puts it:

“This is good news for people who are physically active but can’t seem to lose weight,” said Duck-chul Lee, Ph.D. [...] “You can worry less about your weight as long as you continue to maintain or increase your fitness levels.”

But let me add one clarification. Whenever this topic comes up, I often hear from people who say something along the lines of “See, BMI is meaningless! This proves that doctors should never even measure weight, because it doesn’t predict health.” Not quite. Let me reiterate: those who gained weight in this study were 39% more likely to die of a heart attack than those who lost weight. The reason weight gain doesn’t stay as an independent predictor of death is that those who gained weight also (on average) lost fitness, and those who lost weight also (on average) gained or maintained fitness.
So the very important message that this study reinforces is that it’s fitness that matters most. Keep exercising even if you don’t see changes in your weight, and you’ll be gaining extremely important benefits. But don’t interpret it to mean “it doesn’t matter if I gain weight, because weight is meaningless as a health marker” — because there’s a decent chance (though it’s certainly not guaranteed) that if you’re gaining weight, you’re also losing fitness.
(One final caveat: as the press release notes, 90 percent of the men were either “normal weight” or “overweight” — i.e. BMI under 30. So you can’t assume that the same lack of problems would hold true for the “obese” category with BMI over 30.)