One of the interesting bits is how to hack experimental science to promote a substance. Pick a substance, pick a small test group and watch as many variables as

One of the interesting bits is how to hack experimental science to promote a substance. Pick a substance, pick a small test group and watch as many variables as you can during a limited test span. The interesting bit is not only that the media got took for a spin - this hacking is often accidental and real problem in contemporary science publication as it creates an ocean of pure noise that gives the first impression of good data. Until this situation clears out I'm not holding high hopes for pure data mining based medical research.