Friday, November 15, 2013

22 hours

Latest news flash from the scientists with more time on their hands than most.

Men who walk 22 hours a week reduce the chances of a stroke by 2/3rds.  What a great statistic.  That's 3 hours a day walking.  For one hour a day the risk only drops by 1/3rd.

My gut feel says that there is a lot more behind this stat than meets the eye.  In summary, exercise is better than no exercise.  Some exercise is better than others.

Personally, I wish I had 3 hours a day to go walking.

These types of statistic fascinate me.  I am very interested in the philosophy of numbers, probability, causality and this stat strikes at the heart of that.  What is this telling us.

On the surface we might judge that it is telling us that if we walk 3 hours a day we are less likely to have a stroke by 2/3rds.  But if that were true it could only be true if I were going to have a stroke in the first place.  If I was never going to have a stroke then the walk will have made no difference.

What is more likely is that when compared to a set of men who did not walk against a set of men who did walk then the ones who walked had less strokes than the ones who did.  One assumes though that the pool of people observed was very large to avoid other factors of culture, age, gender, location etc that might make this truth be caused by other things than the walk.

It may be the case that men who take their health seriously and walk also do other things that mean they are less likely to stroke.  I also suspect that men who take their health seriously at this stage of their life have probably done a good job earlier.

There is another theory that the sample pool is already biased.  The system does not count the numbers of men who did not make it to that age due to earlier deaths for other reasons.

Interpreting statistics with such specificity is a strange thing to do.  Where evidence is based upon some difficult and complex studies with lots of caveats the safest thing to do is to stick with the general advice.  Exercise is better than no exercise, some exercise is better than others.  Take advice for your particular circumstances.

The other factor in this little story is the issue of the conditional.  If you walk for 3 hours then your chances of a stroke reduces by 2/3rds.

But that does not logically mean that if you want to drop your stroke probability by 2/3rds you should walk 3 hours.  A counterfactual is true only when the if then statement occurs.

If a then b is true when a and b are true.  All other combinations of a and b are false.  What this says is that while we can tell truth from a counterfactual condition we cannot interpret any other state from this.  Only that if a is true then b will be true.  We cannot say that if a is not true then b is not true.  That is important.


Post a Comment

Subscribe to Post Comments [Atom]

<< Home