29 November 2017

Yay! Miners

In keeping with today's theme that data isn't always real enough for some people, we get an admission that pharma isn't above lowering the level.
Basically, whenever a trial fails, there's almost always a group of patients — left-handed Peruvian men who hate the taste of basil, say — in whom it works. Drawing conclusions from such analyses is statistically fraught, to say the least, so why do scientists so often do so?

"One answer — unpleasant but real — is that pharmaceutical companies want to put a positive spin on their drugs, even when the trials fail to show benefit," writes Dr. Siddhartha Mukherjee, who has personal experience in the subject.

No comments: