1 Boring Old Man

At the risk of overworking my Rip Van Winkle analogy, before I went to sleep in the early 1980s, I recall that conflict of interest standards in medicine and science were similar to those for our judicial counterparts – even the possibility of a conflict of interest was grounds for recusal. Had someone announced in a conference, "I am a Consultant for by SmithKline Beecham, but that will not cloud my judgements about Paxil," the room would’ve dissolved into laughter and cat-calls. Likewise, scientific experiments built-in procedures to eliminate the possibility of bias – randomization, double-blinding, etc are there for a reason and sacrosanct.

So  when I got really into reading the volumes of material we had about Paxil Study 329 and read in the Clinical Study Report [CSR] that the Outcome Parameters had been changed in the final months of the trial, my eyes crossed. And they came out of their sockets when I realized that the only significant outcome variables were those very ones added in the 11th hour. No argument can convince me that somebody didn’t "peek" at the data before the study was unblinded and made that change, though I could never prove such an assertion.

Reply: In response to Dr Garland, our combined analysis was defined a priori, well before the last participant was entered into the study and before the study was unblinded. The decision to present the combined analysis as a primary analysis and study report was made based on …

Being asked to accept statements like that is particularly annoying from people who otherwise preach the gospel of evidence-based medicine from any available pulpit on any given Sunday. As a matter of fact, the whole point of doing RCTs in the first place is because people wouldn’t accept, "Newbery’s Brain Salt offers ‘A POSITIVE RELIEF AND CURE FOR Brain Troubles, Headaches, Sea Sickness,’ etc," particularly if the person saying it worked for F. Newbery & Sons.