This is the Cato Daily Podcast for Wednesday, January 23rd, 2008. I'm Caleb Brown.

The Centers for Disease Control calculate how many people die each year from smoking related illnesses.

But the manner in which that number is calculated remains shrouded in secrecy.

Brad Rodu, a professor of medicine at the University of Louisville, questions the CDC's numbers

and has developed an alternative method for calculating it. His article, Calculating The Big Kill,

is in the new issue of Regulation magazine.

If the CDC model is shrouded in secrecy, why do people believe the numbers that that model produces?

Well, in effect, the CDC has been the sole source for estimates of smoking related deaths for the past 20 years and they've literally

never been challenged. You know the CDC not only estimates smoking related

deaths but estimates the number of smokers.

And while the number of smokers is easily accessible and easily verified from federal sources. The widely quoted 400,000 number is totally inaccessible except

for investigators within the CDC itself and their

collaborators at the American Cancer Society and we thank a number that is so

important to the establishment of tobacco policy,

especially tobacco regulation, should be a number that is more easily accessible by investigators outside those agencies.

What is the argument from the CDC favoring keeping this model secretive?

Well, they've developed a computer algorithm that's accessible by the Internet, but there's really

no argument for this number not being more widely available and its calculation not being

more easily explained.

What are some of the differences between your method of calculating smoking related deaths from the CDC's?

We used an entirely open and fully disclosed model where we look at the number of current smokers, we look at the

number of former smokers, we develop algorithms that are clearly accessible by anyone and we believe as a result

...