The Story of Influenza Antivirals: Part 13

Untrustworthy and unreliable

When we started the 2009 update of the review, we received a comment from a Japanese paediatrician, Keiji Hayashi. He pointed out that the data in a Roche-sponsored meta-analysis published by Kaiser and colleagues in a prestigious journal came from ten placebo-controlled trials on various populations. But only two had ever been published; the remaining eight (including what looked like the largest trial ever undertaken – M76001) remained hidden. 

Hayashi compelled us to do something about this. So we asked Kaiser, one of his co-authors, and the first authors of the two published trials for the “raw” data from their papers. None had access to the dataset; they referred us back to Roche. The situation deteriorated when the BMJ/C4 investigation disclosed evidence of ghostwriting and probably guest authorship. 

The results of M76001, the largest-ever trial of Tamiflu, had briefly been presented as an abstract at a congress in New Orleans in 2000. So naturally, it was much quoted, but when t…