The latest word ( but likely not the last) on the use of the neuraminidase inhibitors (Tamiflu and Relenza) comes from the Evidence Based Medicine epicenter, The Cochrane review) and can be found here.
The article in the BMJ did not go unnoticed by Dr. Howard Brody who correctly recognized that there was more there than simply a lack of definitive data about the role of Tamiflu. See here for his blog comments regarding that series of events in the broader context of an integrity crisis in medical research which I believe is part of an even broader context of "How the heck do we know what is correct in medical research publications".
An ingrained as Tamiflu is in the current medical thinking about management of flu (and anything that looks like flu) I doubt the BMJ article will have any significant impact on the sales of the drug. I bet that sometime in the near future we will see a meta-analysis to counter the analysis from Cochrane.