Monday, January 11, 2010

Number Crunching

My doctor recently mentioned his concerns about the swine flu vaccine and the lack of extensive testing. When you test on 600 subjects, and then give that drug to a hundred thousand, what is going to happen?

Well, that's an easy math problem. Let's assume that there's a one in a thousand chance that the vaccine has some serious side effects. That's a .1% chance. Not a likely event for any given individual, but still something that should never be approved by the Food and Drug Administration.

Testing measures differ from one drug company to the other, I'm sure, but the two brands of the H1N1 vaccine I Googled both mentioned using 600 subjects in their clinical trials. So, what are the odds that they'd miss that .1% chance? If the odds of an individual coming through okay is .999, then the odds of six hundred people not showing any symptoms is .999 to the power of 600. This comes out to .5486, so there is roughly a 55% chance that a potentially fatal side effect could be missed.

Not that a few more test subjects will make me feel warm and fuzzy. Doubling the number of subjects only gives a 70% chance of finding out your serum is poisonous. Tripling it gives an 83% chance.

This sums up in a nutshell my biggest concern about vaccinations. Before injecting something into millions of people, make sure the testing process is more vigorous than for cough drops and foot powder.

No comments: