[sghodas]

The positive predictive value for the prostate-specific antigen test for prostate cancer is only 30%. This means that only 30% of people who test positive for elevated levels of prostate-specific antigen actually have prostate cancer. This is an intrinsic problem with testing for prostate-specific antigen not with Theranos' testing.

[danieltillett]

The issue is not how accurate the PSA test is for detecting prostate cancer, but that Theranos’ tests give wildly different results to everyone else’s FDA approved tests.

[srunni]

A great book was published on this topic 2 months ago, called Risky Medicine: http://www.amazon.com/Risky-Medicine-Quest-Cure-Uncertainty-...

Podcast interview of the author, released yesterday: http://www.econtalk.org/archives/2015/11/robert_aronowit.htm...

[sk5t]

This would suggest that one cannot draw a definite conclusion from the test result. Not that the test result itself should fluctuate wildly.