Monday, December 17, 2018

New Obesity Study From the Annals of Idiotic Goalposts

As has been my tradition, in December I repost old favourites from years gone by. This year am looking back to 2015.
If I looked at 279,000 men and women for a decade and studied whether or not they qualified for the Boston Marathon, but I didn't actually look to see if they were runners, and if they were runners I didn't bother exploring what their training plans and distances were like, but instead simply looked at how many people from that 279,000 qualified for Boston, I'm guessing I'd be left with an incredibly small number.

And yet, that's pretty much exactly what the latest depressing weight loss study did. They followed 279,000 men and women for ten years to see what was the probability of those with obesity losing back down to "normal" weight (a BMI less than 25). They didn't exclude people who weren't trying to lose weight or who might not have wanted to lose weight. They also didn't pay any attention to the means with which those who did lose weight only to regain it lost it in the first place.

The odds weren't good. Over the course of a decade, only 1 in 210 men with obesity, and 1 in 124 women managed to bring their weights down to a place where a table would define them as "normal".

It's not particularly surprising. Putting aside the surprising fact that this study didn't exclude people who weren't trying to lose weight, it remains that the vast majority of folks trying to lose weight these days do so by undertaking ridiculous diets. Go figure people don't sustain the results of ridiculous diets; weight lost through suffering comes back when you get sick of suffering. This study of course misses all of that.

And is getting down to a "normal" weight really the right yardstick to measure success? I mean getting down to a BMI under 25 is to weight loss what qualifying for the Boston Marathon is to running. Most runners will never qualify, and consequently qualifying would be a very poor way to measure whether or not people were runners.

But what if you change the goal posts?

If for instance, you set out to study the number of runners who continue to enjoy running as often and as much as they're able to enjoy, rather than simply the number of runners who qualified for Boston, well suddenly the number of runners will be much much higher, though of course not all of those who take up running, keep up with it either.

Extended that to weight, if the goal posts become your "best weight" which is whatever weight you reach when you're living the healthiest life that you can enjoy, suddenly the numbers change.

How much do they change?

Looking at, for instance, the 8 year data from the LOOK AHEAD trial, where lifestyle changes were thoughtful and the goal wasn't qualifying for Boston, 8 years out and 1 in 2 of the participants were maintaining losses of greater than 5 percent of their presenting weights, and more than 1 in 4 were maintaining losses of greater than 10 percent.

So did the publication of this depressing study add to obesity's literature? Quantifying the number of people who don't qualify for the Boston Marathon of weight loss, without quantifying how many of them were actually runners, and what sort of training program they were adopting doesn't strike me as a helpful addition.

I also can't help but wonder what the impact the publication and coverage of studies like these have on individuals who might be considering lifestyle change - both in terms of reinforcing idiotic goalposts, and in terms of them even starting out of the gates.