I was all set to write a different article on this subject when I ran across a study – and that study went and upset my convictions. Yes, I can change my mind when given contradicting evidence.
Let me do bottom line first. BMI sucks as a mortality risk indicator, but bodymass and bodyfat proportions are GOOD indicators. BMI is both, and itself has no way to tell the respective proportions. On with the show.
Now, I’m still going to say that doctor’s reliance on the BMI is excessive. Here’s the deal (and the start point of the original). Arguments that reduced fat increases longevity runs into a crapload of studies that show mortality is u-shaped on the BMI charts. That is, when age of death is tracked against BMI, the people in the middle live longer than those with high or low numbers. Oddly, the peak is on the edge of – and even slightly above – the first danger line for obesity. Anyway, I started doing some research to pull this together. I was going to note (among other things) that while longevity to fasting studies have shown a correlation in some animals it’s not been all, and further that “WHY” is still unknown. Hypotheses keep getting kicked out, and then demonstrated wrong. (Don’t you love science when it is done right? We know it happens, we don’t know why. Let’s find out by trying and failing, and if we find something that’s sorta right let’s test it some more.)
Yeah, all that. Then I ran across this study. You don’t have to read it if you don’t want – I’m going to summarize it here.
A bunch of 60 year old Swedish males, all born in 1913, had a lot of statistical information taken that happened to include bodyfat. For several years after, concluding 22 years later, there were follow ups. For the most part, during their lives they kept approximately the same health standards – smokers still smoked, bodyfats were about the same, etc. The study excluded individuals who’d died from accidents (car wrecks, that sort of thing) and did some correlations. As shown by study after study, BMI to mortality was still U-shaped. Enter the kicker.
Fatfree mass to height^2 is an ascending curve. Bodyfat (alone) mass to height^2 is a descending curve. In this case, higher correlates to decreased mortality.
The less bodyfat you have, the more likely you are to live a long time. The more muscle mass you have, the more likely you are to live a long time. They are curves, not lines. Let’s take muscle mass for an example. There’s a ‘sharp break’ (my paraphrase) in this curve at around 16 kg/m2. That is, below that number and the chances of dying increased A LOT. Above that line wasn’t flat, mind you, but its increase was a lot flatter.
For what it’s worth, the bodyfat curve is a lot smoother; in the study it’s said to approach linearity (but that doesn’t square with the shape of the curve for increasing BMI, so I’m suspicious).
Oh – and while this study was only about mortality, I’m aware of others that show other health problems accruing to people who let their bodyfat get too low. In other words I’m suspicious of the “linearity rate” of the study in regard to risk:bodyfat.
Still: don’t use BMI for your fitness determinations. Use actual bodyfat measures. And strive to keep that bodyfat down.