Jack Scannell (the person who coined the phrase “Eroom’s law” for the rise of drug discovery expenses) has a long piece up at Forbes on drug pricing, and it’s terrific. He goes over the various rationales for high drug prices in turn, with extensive comment. First off is the “recouping costs” argument, which he doesn’t have much time for:
Cost covering is a palatable spiel to make if one is trying to sell an expensive drug. In fact, the cost of production story has been repeated so many times for so long, that it has become plausible to lots of people who should know better. I still read in health policy papers that drug companies need to “recoup their costs.” This is nonsense. Sunk costs are sunk. If companies are going to spend on R&D they need to believe there are decent odds that they will make a good return on investment, but this is a different thing to recouping anyone’s historic R&D costs.
Next is pricing based on value – what this drug can do for you – which gets a more sympathetic hearing, but is also rejected in the end. One of the biggest problems is that it’s very hard to assign the real value to a drug, especially at first. This is especially tricky for the national heath services that are trying to do cost/benefit analysis:
Nobody knows for sure at the point of launch which drugs will transform medicine, which will turn out to be duds, and which will poison people. Drug companies and Wall Street analysts are hopeless at forecasting drug sales, even when most of the clinical trial data have been collected. I would be surprised if national health economic agencies have clairvoyant powers that the companies and the analyst lack. Therefore, rushing to do technical health economic analyses to come up with what will often be the wrong value-based price, which may be ignored by the drug company if it is lower than other countries’ wrong numbers, seems somewhat futile.
Arguing drug costs based on value, Scannell says, is also an invitation to exaggeration and salesmanship, which the industry does not need any more invitations to. As he notes, the UK is backing off the whole value-based pricing model, which is something to think about.
After that comes pricing power, since patents after all grant someone a temporary monopoly. Emphasis on “temporary”, because this does collapse when the patent expires and the drug goes generic. (At least it should – the recent furor over T*ring Pharmaceuticals and their ilk is largely about keeping this from happening and putting ancient generic drugs into the as-high-as-it-will-go category). But when the system works, it’s something to see. As Scannell says: “I would guess that one can buy today, at rock bottom generic prices, a set of small-molecule drugs that has greater medical utility than the entire set available to anyone, anywhere, at any price in 1995.” This is, as he says, a tough sell, but it may be one of the best drug-price arguments out there.
And finally comes the prize-winning lottery-playing aspect of drug research, and if anyone would like to deny that as a factor, I will gladly meet them in person to hit them over the head. After 26 years of drug research, that’s one thing I’m sure of: money, hard work, and brains are essential to success in this business, but are in themselves still insufficient. Time and chance happeneth to them all, folks, and equally smart and hard-working teams will come up with different results, sometimes dramatically jackpot-winning or job-losing different, and there is (so far) not a whole lot we can do about it. Here’s Scannell:
Only someone lacking ambition would play the lottery hoping to “recoup” their costs. . .the R&D lottery is expensive to play, most games are a bust, and the rare wins take a long time to pay out. Investors and drug companies choose their games (cancer, Alzheimer’s, obesity, etc.) by guessing at the value of the monopoly, the cost of the R&D, and the odds of success.
. . .It takes luck. It’s not fair. Roughly nine out of ten drug candidates that enter clinical trials in man are never launched. Even for the few drugs that are ultimately approved, the winnings are skewed. The most successful 10% of approved drugs, only 1% of those that entered clinical trials, maybe 3 new drugs each year, generate half of the profits of the entire drug industry.
He does not wrap up with a big list of suggestions that will fix everything – he has some ideas, which I’ll send people over to his post to read, but the big take-home is that this is a hard problem and there is almost certainly no way to make everyone happy about it. Along the way, he has a lot of excellent points about the various estimates of the costs of drug discovery and development, about the relationships of US prices to European prices, and many more topics. You really should read the whole thing. A lot of us have been making many of these points for a long time, but it’s really great to see everything wrapped up in one big package – instead of sending people to my my last dozen years of blog posts on drug pricing, I’m going to start sending them to this piece instead. Things like this prompt me to do so:
A mistake that many critics of the drug industry make is to imagine that because new drugs are very expensive, the industry is systematically cheating, there must be a lot of slack in the system, and if only the slack were removed, we would have a flow of good cheap new drugs. I think that the truth is much worse: The industry is trying hard to discover and develop things to sell; it is charging as much as the market will bear; even First World health systems are baulking; access to the newest drugs is problematic; but outside of a select few disease areas the financial returns on R&D investment are poor; and without the private sector investors, there would be vanishingly few new drugs.
Sounds easy to fix, doesn’t it? Gosh, if only all those folks running for office could just get their hands on the levers; they’d have things straightened out in no time. Despite all the political fireworks, Scannell says that “a likely outcome is that things grind slowly along their current trajectories”. I’d take that over some of the other possibilities, believe me.