I actually covered much of the ground that Light and Warburton traverse back in HOOKED, but since they provide additional details and calculations, and we have not discussed the issue for a while, I figured it was time for a revisit.
In 2004 Merrill Goozner, formerly economics reporter for the Chicago Tribune (and now blooger at GoozNews), published his expose, The $800 Million Pill. The title came from the research by DiMasi and associates at the Tufts Center for the Study of Drug Development, which is primarily funded by industry. That group published their estimates of what it cost industry to discover one new drug (that is, a novel molecule never before used) given all the dead ends that have to be navigated before one good drug emerges. They came up with the figure of $802M. Light and Warburton follow in the footsteps of Goozner and Marcia Angell (The Truth about the Drug Companies) in deconstructing this myth.
The brief highlights of their analysis:
- The data DiMasi et al. used for their study was secret industry data that has never been revealed, so we have no way to check its accuracy. There are a number of sources of independent information that would call its accuracy and representativeness into question.
- The costs of actual drug discovery are essentially unknown and highly variable. What did it cost Alexander Fleming to discover penicillin? Sometimes you just trip over a useful new drug by chance, sometimes a new drug emerges after 30 years of arduous research. Light and Warburton note in passing other findings that some 84% of new drugs are not discovered in house by the drug companies but by university and other noncommercial labs.
- The drug industry rakes in considerable tax savings that represent as much as 50% of the actual costs of their R&D. Light and Warburton go into a long economic talk about this that goes way over my head, but seems to amount to noting that drug companies regard R&D costs as operating expenses and not capital investments for purposes of getting their tax breaks. Then when somebody asks why they don't deduct the tax savings from their reported R&D costs, they answer as if the R&D expenses were capital investments, so that tax savings if any would accrue only slowly over many years. They can't have it both ways.
- As I noted in HOOKED, the industry then pads its R&D costs by adding in the profits that that same money would have earned had they invested it in the stock market for the same number of years, and saying that they deserve to be repaid for the foregone profits as well as for the actual costs. Light and Warburton note that no other research-intensive industry makes any such claim. If you purport to be in the business of discovering new drugs, of course you invest your money in drug R&D, and of course you don't put that same money in the stock market instead. Claiming that the company should effectively be reimbursed for profits forgone as well as for actual costs nearly doubles the estimated cost of discovering a new drug.
- DiMasi et al. appear to have used unrealistically high estimates both for how long research trials go on and how many subjects are enrolled. Finally, instead of using median figures they use means, which can be inflated by a few high-end outliers.
- Light and Warburton are so impolite to add that only a small proportion of the new drugs marketed by the industry are genuinely new molecular entities; and an even smaller percentage actually represent significant therapeutic over existing drugs.
So what does it actually cost to discover one new drug? I believe that in HOOKED I settled on the figure of maybe $100-200M. Light and Warburton suggest that the best answer is"we don't know" but did some calculations to show that the true number could even be as low as $43M.
I think the real take-home message is this: even if we don't know the exact right answer, there has been a lot of reason to regard the $800M as spurious practically from the moment it was released. It is surely at least twice the real figure and more likely, at least four times. Yet PhRMA and the media continue to bruit about that figure as if it were gospel. That just shows when you have a lot of money at your back, you can specify the "truth" to be pretty much whatever you wish.
So--why am I not worried about Don Light being on a Pharma hit list? The only example I know of, of the drug industry murdering people who they don't like, is John Le Carre's novel The Constant Gardener (and the film version). Le Carre was careful to document that all the industry skullduggery he used for the plot of his novel was based on real events, except for the murder part. When I met with a group of students recently who had seen the film, I asked them why they thought Pharma didn't actually go around killing people. The students immediately got the right answer--Pharma doesn't have to. They get what they want just fine, thanks, without the need for any such crude tactics.
Light DW, Warburton R. Demythologizing the high costs of pharmaceutical research. Biosocieties 6:34-50, 2011.
Light DW, ed. The Risk of Prescription Drugs. New York: Columbia University Press, 2010.
DiMasi JA, Hansen RW, Grabowski H. The price of innovation: new estimates of drug development costs. Journal of Health Economics 22:151-185, 2003.
Goozner M. The $800 Million Pill: The Truth Behind the Cost of New Drugs. Berkeley, CA: University of California Press, 2004.
Angell M. The Truth About the Drug Companies: How They Deceive Us and What To Do About It. New York: Random House, 2004.