This week’s econtalk, featuring Richard Epstein talking about patent issues with respect to pharmaceuticals, I found to be really taxing mentally. I have thought a great deal about these issues, yet Epstein was talking faster than I could absorb. Also, just at the end, Russ Roberts raised some really interesting issues.
First, do big drug companies benefit from FDA regulation, because the compliance costs are so high that small competitors cannot enter? Epstein’s answer was negative, because big companies end up bidding for the research results of small companies.
I found this answer too glib. I think that industry structure would be very different without the FDA. (I believe the same about the car industry–my joke is that I bet it would take more lawyers than engineers to start a new automobile company.)
Another issue that got raised near the very end is “unauthorized uses” of drugs–trying a drug that has been approved for one kind of cancer on another kind of cancer. One would think that if authorized use needs to be heavily regulated, then unauthorized use would be, also. But unauthorized use tends to be not only unregulated, but unstudied. So the results of unauthorized use, rather than being subject to statistical analysis and widely disseminated, are passed along by word of mouth from physician to physician.
In fact, one defense I would make of the FDA is that it does promote a scientific process, with controlled experiments. Contrast this with education, where new “treatments” are constantly introduced without any results from controlled experiments. In medicine, I think we make real progress, in part due to the use of the scientific method. In education, new methods of teaching and curriculum “reforms” are often a step backward, because nobody bothered to run a controlled experiment before introducing the new approach. On the whole, I think that progress has been much slower in education than in medicine, in part because education lacks the equivalent of the FDA.
READER COMMENTS
Bruce G Charlton
Feb 19 2007 at 11:00am
There may be a misconception about ‘unauthorized’ uses of drugs – the ‘license’ from the FDA is just a license to market the drug. It does not restrict usage. So unauthorized, off-label drugs uses simply refer to usages which are not able to be used in marketing.
A lot of drug use in children and small niches such as rare diseases is necessarily off label, for no reason more sinister than that it would cost far more to get a license than could be recouped from sales.
I would say that the FDA style regulation is just about as scientific as scientific socialism by central planning – which is to say that there is indeed a feedback loop, but the lag is measured in decades – therefore way too slow for medical science.
Medical progress is currently, objectively, being strangled by regulation and the costs of licensing. Pretty soon we will have to decide whether or not we want new treatments – if we do will will have to reduce regulation, and thereby re-introduce competition-by-results.
I think the license should merely certify a level of safety, and the usefulness of drugs will be discovered later – as has always happened. It took 100 years to discover that aspirin prevented heart attacks and a couple of decades to discover the many uses of ACE inhibitors in cardiac disease.
Was it Bryan Caplan that suggested that it should be possible to sell unlicensed drugs – so long as this was made clear? Sounds good to me.
Chris R
Feb 19 2007 at 11:12am
But education does have an equivalent of the FDA. It’s the Board of Ed of a given state or district.
In fact, what you’re seeing is more of a movement toward controlled education experiments. Caroline Hoxby’s work and a similar ongoing study of charter schools in San Diego have exactly taken an experimental approach (and come to different conclusions so far). The movement is glacial because a subset of education people don’t like to think of their work as scientifically falsifiable and aren’t trained in any statistics. Also, incentives aren’t correctly aligned. More charter schools, for instance = less union control. This is a problem not so prevalent with the FDA.
dWj
Feb 19 2007 at 12:57pm
What Chris R said.
Also, education is almost entirely government-provided. Pharmaceutical development may be heavily regulated, but it still features a significant amount of private-sector competition. Having a government agency ensuring that consumers actually know what works and what doesn’t (as opposed to sort of thinking they know) to a greater extent than might otherwise be the case is probably a benefit, perhaps worth its costs, but ultimately the drug company that tries new things that work out will be rewarded on a level that a school that tries new things that work out will not be.
Brad Hutchings
Feb 19 2007 at 2:57pm
Isn’t the unapproved usage scenario a “good” example of the way that markets disseminate information? Obviously, we should study it. Obviously, experts in statistics should add information. But when I read Chris R’s in application to education that “a subset” resists falsifiable experiments and isn’t good at stats, I cringe. (1) Teaching is not physics. (2) Education is not a scientocracy, nor should it be. No Child Left Behind is a pretty good federal approximation of the FDA for education. The complaint that it makes teachers teach to the test is vastly understated. Now, you may have all your data and statistics to scientifically manage the education sector in as Tayloristic a manner as an Ed bureaucrat could fantasize over, but what if the measurements are rigged or don’t correlate to learning? It may be tough for some to understand, but education success depends to some extent on parent, teacher, and student buy-in. Teaching to the tests is setting off a lot of bullshit detectors ;-).
J
Feb 19 2007 at 3:09pm
I would recommend you have a look at:
Richard E. Doblin. Regulation of the medical use of psychedelics and marijuana. PhD Thesis, Harvard, 2000.
This thesis is an in-depth study of the FDA regulatory process and the hurdles it imposes on a small, fringy outfit like Dr Doblin’s MAPS. The main question in Doblin’s thesis is obviously the feasibility of MAPS (the non-profit pharmaceutical company he now runs, which tries to develop psychedelic drugs, also known as hallucinogens, into approved medicines). His thesis should also be interesting to regulation wonks because it can be interpreted as a case study of the barriers to entry imposed by the regulatory process. It exposes many details that usually don’t come out because drug companies are (understandably) quite protective of their practical knowledge of how to navigate the FDA.
Also, Epstein does talk too fast. It seems to me that 90% of the time, he makes perfect sense, and although his speaking style makes it hard to pick out the other 10%, I would still ascribe his phenomenal speed to enthusiasm rather than malice.
Cyrus
Feb 19 2007 at 8:52pm
A comparison to heavily regulated pharmaceutical medicine might be the somewhat less regulated market in surgical procedures. New procedures are taught by one practitioner to another, insofar as they seem useful. On the other hand, there are a number of procedures performed where research suggests they are no more beneficial than less costly and less invasive treatments: what seems useful is not a standard of statistical scrutiny.
Dave
Feb 20 2007 at 9:17pm
There are a few smaller car companies starting up on the west coast, primarily making very high end electric cars. I’m not sure about their lawyer to engineer ratio, but it has to be low enough that more than one of them exists. When they came scouting for talent in the Detroit area, it wasn’t for experts on automotive law but for engineers.
Comments are closed.