I was listening to the radio the other day and heard a doctor being interviewed about yet another study "showing" that the slightly pudgy (1 to 30 pounds overweight) have a lower risk of dying from various diseases than do people who are not overweight at all. I didn't write down the author or anything, but it's not the first study of its type that I've heard of.

Does this mean that those of you who can see your abs should stop off at Krispy Kreme on your way home from the gym if you want to stay alive? I'm sure some people might think so, and I'm even more sure that plenty of people who have a small but growing potbelly will use studies like that one to justify continuing their lazy and self destructive lifestyle. Why go to the gym and get really lean when it just makes us more likely to die?

The truth is that there are a lot of problems with studies like this one. Let's look at a few.

Basically the way you do one of these is to study epidemiological data. You look at records of lots of people, figure out their weight category (overweight, obese, whatever) by either calculating BMI or comparing their weight to a height/weight chart, then do some sophisticated tallying of who dies of what. You find that obese people are more likely to die of heart disease, which is no surprise, and of cancer, which might be a surprise to some of us. Sometimes you find things that surprise you - like slightly overweight people die less often than non-overweight people.

Why do the studies show this? There are a few reasons.

Imagine twins, let's call them Adolf and Ben (A & B, okay?), who are both slightly overweight, say 15 pounds heavier than they should be, and it's fat. Let's suppose both are equally active, both have comparable diets, and so forth. Now let's suppose that for whatever reason Adolf gets inoperable testicular cancer - maybe he got X-rayed one too many times, maybe he lives over an old toxic waste dump, maybe he's just unlucky.

What's going to happen to Adolf? Well, he's going to die. On his way to dying he'll probably lose weight. Why? Because he has cancer. By the time he dies he might be normal weight or even underweight. Then, next year, when scientists do their studies, they count Ben as an overweight guy who lived and Adolf as a non-overweight guy who died, and it looks a little bit like being overweight is protective, when that just isn't the case.

Another problem is linked to but different from the Adolf-Ben issue. The studies can analyze who is over- or underweight, but not how they are overweight. They use height and weight to categorize people. The studies can't differentiate between a healthy, fit lean person and someone who is low in weight because they have lost the will to live or can't feed themselves. It also can't distinguish the sedentary, low weight people who are what Art De Vany calls "skinny-fat." That is, people of normal weight because of low muscle mass but still holding significant amount of fat, possibly in their organs where it isn't easily seen. These studies also can't distinguish the overweight but lean and very muscular from the overweight but just fat.

Why do these studies go on, as flawed as they are? Because every time you visit your doctor they take your height and weight. They don't measure your bodyfat or fitness levels - that would be a lot more expensive and require a lot more time. BMI or your weight relative to a height weight chart do correlate somewhat with actually being overfat or obese, just imperfectly. These studies do show how unhealthy it is to be obese, but may not be doing a good job of representing the health benefits or costs of being lean.

Lastly, another set of problems arise from epidemiological studies in general. The point of this study is to inform people's decisions to lose or gain weight, I suppose. That is, if it showed that being obese was unhealthy, presumably that might motivate obese people to lose weight. But it might be that obesity is caused by some genetic factor that also happens to cause heart disease or cancer or whatever. In other words, the fact that the obese are more likely to die of anything is not by itself proof that the obesity caused the extra deaths or that an obese person who lost weight would be less likely to die. To prove that, we'd have to take a bunch of obese people and have some lose weight and some not lose weight and see what happens.

Now it may be true that having some extra bodyfat is actually healthy, except that a lot of other more careful studies show just the opposite. How would we test this for sure? We'd have to take a bunch of lean people and force half of them to put on 15 pounds of excess fat. Or find a bunch of 15 lbs. overweight people and force half of them to lean out. This is expensive and difficult to do - we'd have to consider only subjects who could maintain their new weights for years on end, and how many people do you know who manage this?

The morale of this story is to be as lean as you can be. If you can't see your abs clearly then you're too fat (unless you can't see them because of body hair, like me, in which case you should shave your belly and repeat the test).

Plus, reducing your weight will improve your karate performance, so you're in for a win-win situation.

Osu.