Boston Scientific Taxus Liberte study: statistical smoking gun or who cares?

I was surprised, both by content and by tone, of an article printed in today's Wall Street Journal:

http://online.wsj.com/article/SB121867148093738861.html

It questioned the statistical integrity of the study that was submitted to the FDA for approval of the Taxus Liberte (by the way, does anyone know where I can find the publication of this study?). In almost breathless tone, the article described how the Wall Street Journal conducted an independent analysis to determine that the Wald Interval is the inappropriate test for statistical significance in this kind of study and criticized that the study was not randomized (comparison was made to historical controls of patients that received Taxus Express) or non-blinded. But is there any real "gotcha" here?

First, is it really necessary to have a randomized trial to approve an incremental design change from Taxus Express to Taxus Liberte? Second, is it even possible to blind it? Third, who cares if it only meets statistical significance with the Wald only? It's commonly used and every other p value in the WSJ analysis had p=0.05.

It is interesting to note that no cardiologists (besides Baim of Boston Scientific) were interviewed for the story, and the expert criticism comes from statisticians, a mathematician and an internist.

I think that most interventionalists would still say, "Give me Liberte!"

The most fascinating aspect of this article, I think, is that it is another example of the new phenomenon of a post-publication "peer review" by the lay press.

Any thoughts? (BTW, I have no relevant disclosures, but if you really want to nitpick, I do have a Boston Scientific pen lying around my home somewhere).

Brian Choi, Asst Moderator