Wednesday, May 16, 2012

mildly mathematical: a familiar face, and the maths of obesιty.

when i was an undergrad, my academic advisor (i.e. the one who signed my registration forms) advised me to go into applied maths .. and in particular, into mathematιcal biology. suffice it to say that i didn't listen to him.

one thing that i remember about him was that he was always a direct, no-nonsense kind of guy. as it happens, the NYT recently interviewed him about the mathematιcs of obesιty; it seems that his manner is much the same.

he also discusses some interesting conclusions.
-- ✂ -- --
..
Why would mathematιcs have the answer?

Because to do this experimentally would take years. You could find out much more quickly if you did the math.

Now, prior to my coming on staff, the institute had hired a mathematical physiologist, Kevιn Hall. Kevιn developed a model that could predict how your body composition changed in response to what you ate. He created a math model of a human being and then plugged in all the variables — height, weight, food intake, exercise. The model could predict what a person will weigh, given their body size and what they take in.

However, the model was complicated: hundreds of equations. Kevιn and I began working together to boil it down to one simple equation. That’s what applied mathematicιans do. We make things simple. Once we had it, the slimmed-down equation proved to be a useful platform for answering a host of questions.

What new information did your equation render?

That the conventional wisdom of 3,500 calories less is what it takes to lose a pound of weight is wrong. The body changes as you lose. Interestingly, we also found that the fatter you get, the easier it is to gain weight. An extra 10 calories a day puts more weight onto an obese person than on a thinner one.

Also, there’s a time constant that’s an important factor in weight loss. That’s because if you reduce your caloric intake, after a while, your body reaches equilibrιum. It actually takes about three years for a dieter to reach their new “steady state.” Our model predicts that if you eat 100 calories fewer a day, in three years you will, on average, lose 10 pounds — if you don’t cheat.
-- ✂ -- --
the article continues here.  for those interested, it also links to a "body weight simulator" java applet that realises their model, in online form.

No comments: