Apologies for dual postings.  - Richard

A student sent me a link to a video of a lecture on epidemiology and its abuse: http://www.ted.com/talks/ben_goldacre_battling_bad_science.html.

.

The student noted the relevance of the video to our discussion in class of whether FDA approval should be viewed as a complete assurance that a drug is safe and should thus preempt tort liability.  The video also sheds further light on the highly questionable assumptions by Judge Kozinski in Daubert on remand that published empirical research done in labs (generally industry labs) is more trustworthy than empirical research done for purposes of litigation because (1) such work in labs supposedly is not motivated or affected by regulatory and liability issues and (2) thus is the best evidence, per se (independent of investigation of its methodology), of its being done through proper scientific methods.