Letter to the Editor: Tufts' Joseph A. DiMasi debunks the Andrew Witty R&D cost meme

Dear Editor,

Joseph A. DiMasi

I read with interest your article in FiercePharmaMarketing on pharma pricing debates (March 16, 2016). I have been the principal investigator for the Tufts Center for the Study of Drug Development studies on pharmaceutical R&D costs (including the latest one which is available now in the Journal of Health Economics). It is your reference to that work that I wish to address.

In particular, I would like to correct what I believe is a misinterpretation of something that GSK CEO Andrew Witty stated at a healthcare conference in 2013. Some industry critics have picked up on a comment he was reported to have made, taken it out of context, used it to try to discredit our work and repeated it so often that it has become a meme. The quote that the critics use, without context, is "one of the great myths of the industry," referring to a $1 billion R&D cost figure.

The original reporting on Mr. Witty's comments is from Reuters. An explanation by the reporter of what Mr. Witty meant immediately follows the quote in question ("since it was an average figure that includes money spent on drugs that ultimately fail"). That, of course, is what we measured, and our estimates rightfully consider the costs of failures.

Mr. Witty, I believe, was not taking issue with the accuracy of our estimate of industry R&D costs for the period covered. He was apparently saying, in effect, that it does not have to cost that much. As the article indicates, he recognized that failures have costs when noting that GSK revamped the way it did research and had fewer drugs failing recently in late stage testing. He is quoted in the article as having said that, "If you stop failing so often you massively reduce the cost of drug development."

There is nothing in the original reporting that indicates that Mr. Witty was commenting on the accuracy of our work, or that he even had in mind a comprehensive analysis of GSK costs. Rather, what he appears to have been saying was that companies can do much better than $1 billion per success (accounting for all R&D costs) if they can manage to substantially improve their regulatory approval success rates.


Joseph A. DiMasi, Ph.D.
Director of Economics Analysis
Tufts Center for the Study of Drug Development
Tufts University