View source for How Bad?

<center>''[[Blistiki]]'' --> here</center>
----

Lots of things are bad for you.  This is not news, but every new instance is reported as if it were.  Recently people are getting a little more sophisticated about statistics, with messages like, "Substance ''X'' has been found to increase your chances of contracting cancer of the ''Y'' with 95% confidence."  This tells you absolutely nothing useful until you have some more information, like the number of subject in the study, the amounts of ''X'' to which they were exposed, and so on.  Leaving out all the minutiae and assuming (charitably) that the statistical analysis has been done competently, there is one vital piece of information that every such message must include to be meaningful:

<CENTER> '''How''' bad is it for you? </CENTER>

If you are exposed to a specified amount of ''X'', '''how much''' does your (already finite) probability of contracting cancer of the ''Y'' increase by?  If you don't wear a helmet while riding a bicycle, '''how much''' larger is your (already finite) probability of suffering a fatal head injury?  This must be well established, or surely the law requiring helmets could not have passed.  But how many people know the numbers?

Why do we tolerate such nonsense?  Because we are mentally lazy and don't want to have to deal with the complex balancing of priorities implied by a quantitative comparison.   How much easier to say, "The only acceptable amount of ''X'' in the environment is zero," or, "The only acceptable risk of a core meltdown in a reactor is zero," (see ''[[Zero Tolerance]]'') than to ask what we get in return for accepting a finite amount of ''X'' exposure or a reactor with a mean time between meltdowns of 1,000 years.  

"If even one person might die as a result of ''Z'', then ''Z'' must be forbidden, no matter what the cost."  (See ''[[The Price of Life]]''.)  How many times have you heard this comfortable rationale expressed?  What an absurd notion!  Lowering the bus fare by 10% in a city of a million people will inevitably result in some people dying "prematurely" who would otherwise have lived longer.  (See ''[[Premature Death]]''.)  The question is, would ''more'' people be ''saved'' by this action?  Such policy decisions must always involve ''quantitative'' comparisons of costs and benefits, and human lives will ''always'' be part of the equation.