Jan

23

Drug Investing, from Andy Aiken

January 23, 2017 |

For those interested in pharma/biotech investing, or simply interested in innovation or public health, here is an interesting, easy-to-read study recently published by the FDA: "22 Case Studies Where Phase 2 and Phase 3 Trials Had Divergent Results".

To paraphrase the political scientist cum statistician Andrew Gelman, the problem with the Neyman-Pearson statistical decision framework (set a null hypothesis, perform a statistical test to accept or reject it) is that the difference between statistical significance and non-significance is itself non-significant. Computer models are often inaccurate, and even animal surrogates may not be predictive either. We often don't understand how a drug really works until it is put into clinical use.


Comments

Name

Email

Website

Speak your mind

Archives

Resources & Links

Search