[ExI] FDA Tells Google-Backed 23andMe to Halt DNA Test Service

[Brian Manning Delaney]

El 2013-11-27 21:03, Dave Sill escribió:

> Well, gosh, yes, if a genetic test shows that I'm more likely to get a
> disease, I'll probably change my behavior to avoid that disease. How is
> that harmful, even if the test turns out to have been wrong?

Well, here's an example.

Everyone "knows" that, in general, light to moderate daily consumption 
of alcohol reduces mortality more than drinking more, or drinking less. 
(Turns out that's a bit controversial, but let's leave that aside.) The 
link to dementia is particularly strong. Let's say you get your 23andMe 
results back, and discover you're an APOE-ε4 carrier. 23andMe tells you 
the odds of your developing Alzheimer's is much greater than avg. You do 
some digging and learn that, for ε4-carriers, the "little is better than 
none" rule about drinking doesn't apply: the more you drink, starting 
from zero, the greater the risk.

So you become a teetotaler.

But say it turns out you aren't an ε4-carrier. Well, you changed your 
behavior, and it was harmful (assuming the J-curve idea about alcohol 
consumption -- at least for non-ε4 carriers -- is correct).

Brian