In our daily news e-mail here at the U of R, the lead story this morning was titled, “Physician Calls for More Rigorous Standards for Drugs Up for FDA Approval.” In it, the U of R Med School Dr. O’Connor calls for:
… more stringent Federal guidelines governing the approval of potential new drugs
Among the guidelines he suggests are (just after noting how overworked the FDA is already):
the agency should be empowered to demand more detailed data when a new drug is being considered for approval.
How can that be bad? More data is good, how could it not be. After all, in economics don’t we argue that choices expand freedom, so more data must expand our possibilities, no? However, information is not free to obtain. Information is not free to analyze. And information is not guaranteed to be used properly by those obtaining it. Even if I do not wish to impugn the motives and abilities of the folks at FDA, by simply arguing that we should have more detailed data – even if that data is very useful – is not necessarily a good idea. Why? Because information is not free. Just think of the dating/marriage market. If each of us were able to collect “more detailed data” on the likes, dislikes, tendencies, etc. of potential partners, I am sure we would all make terrific matches for each other. But assembling this information, and ultimately acting on it, is prohibitively costly. All the time we would spend analyzing information could have been spent … enjoying relationships. And no matter how good your relationship, it is a statistical certainty that there is always a better one to be had out there – but the existence of a better outcome does not mean it makes romantic or economic sense to try to attain it. By the way, the FDA regulates over 11,000 existing drugs, evaluates over 100 drugs per year and regulates products from food to vaccines to medical devices making up near 25% of all consumer spending.
There is good reason for some change at the FDA, or so it seems:
“When new medications have been compared only to placebo, not to drugs already on the market, it’s very difficult to know whether and how we should prescribe them to patients,”
What Dr. O’Connor is referring to is the fact that the FDA has two mandates – one to assure drug safety (i.e. that it will not kill you or otherwise seriously harm you) and the other to assure drug efficacy (i.e. that the drug actually does what it is supposed to do). Proponents of the FDA can probably make a decent case for the first of these two roles (though for reasons you will see later, I am unpersuaded). When it comes to the second, not so much. Certainly among economists who study these things, the efficacy mandate for the FDA is among the most unpopular economic policies imposed by the U.S. government – and there is certainly no shortage of bad policies.
Dr. O’Connor then adds a very sensible thought:
The current system makes it very difficult to evaluate a new, more expensive treatment against an established, less expensive option,” added O’Connor. “We should consider a different standard for approval – that a drug not just outperform placebo, but that it also undergo direct comparison with at least one treatment considered the standard of care. In this way we can ensure that new, less effective treatments do not replace older, more effective and cheaper ones.
Which should lead readers to question, “why is this not happening already?” In a world where we rely on a central, uniform regulatory body, problems like this are to be expected. Why? Because if someone, for some reason, gets the rules or procedures wrong, there is no other agency or company trying something different. You see, the problem with one-size-fits-all regulation is not that the good things will work for everyone the same way, it is that when there are oversights or mistakes, well, it results in a problem for everybody. Without a centralized regulatory authority, consumers and regulators would have incentives to make sure a variety of methods of testing were used, and consumers and their doctors would be able to choose drugs based on the multiple sources of evidence that are out there. If you wish to empower the FDA with anything, how about simply making it an information repository? A second major problem when you have an FDA to evaluate drug efficacy is that people mistakenly believe that since the FDA is checking into things, then “everything must be alright.” We become far less vigilant ourselves (if you don’t believe this is the case, then ask yourself if you have ever investigated the soundness of your bank’s balance sheet before deciding to deposit thousands of dollars there … and ask yourself if you would have done so in the absence of bank regulation like deposit insurance?), and we think that simply by having an FDA all our problems are solved. Furthermore, having a single body like the FDA do these things makes it completely uneconomical for third parties to evaluate drug efficacy. And did we ever consider the possibility that just maybe the FDA itself is influenced by the major drug companies? Is it not at all possible that this oversight has not really been an oversight? Nah, that would never happen.
The FDA has its own statisticians who analyze data and arrive at their own conclusions regarding studies,” said O’Connor. “They have no financial conflict of interest, and there is no pressure to produce favorable results for a drug. Such an independent analysis is a gold mine for researchers and physicians seeking information about the safety and efficacy of a medication that has recently been approved.
Dr. O’Connor only sees conflicts of interest working in one direction. Even if we take seriously the claim that FDA researchers and policymakers have no pressure to produce favorable results for a drug, I don’t see any reason for the converse to be true. It is quite possible, even expected, that the FDA and its researchers would be overly cautious about the results for drugs. This has nothing to do with the quality of character of the folks at FDA, but rather with the incentive structures that the FDA places. If the FDA is empowered with keeping drugs safe and showing they are effective, then their only goal is to release drugs that are safe and effective. In fact, releasing a drug that hurts people, or that is widely understood to be ineffective would represent massively bad publicity for the FDA – and hence they would have an interest in being too conservative in their drug releases. Below I show you evidence that this is exactly how they are operating today.
I have a huge amount of respect for FDA. The amount of time and effort the agency puts into each new drug review is incredible. They do an incredible task with a finite amount of time and resources. The FDA is an ally in maintaining public health and protecting our patients,” he added.
I never cease to be amazed about the faith people place in government. Would Dr. O’Connor say the same thing about the drug companies? After all they put hundreds of times more effort and resources into developing drugs to ease our pain, extend our lives and cure us from all manner of ailments. They operate with finite time and resources. They also operate with an enormous target on their back from the political class and the consumers. I love that they want to make a profit by selling me drugs. Would people rather that they tried NOT to make a profit by selling drugs. Who is the real ally in maintaining public health and protecting patients? From my experience with my doctors, it is certainly not them. If doctors cared about patients and public health they would recognize that the largest correlate with negative health outcomes is poverty. If they really cared about patients they would eliminate their own cartels and allow something resembling a market in medicine – where medical costs would be much, much lower than they are today. If doctors cared about patients, medical school enrollments would be exploding, nursing schools would be expanding, and retail medicine would be exploding with the support of doctors. Whatever growth those things are enjoying is coming despite the cartel that the medical profession has arranged for itself. And if the FDA was truly a partner in public health, wouldn’t we want to see evidence that in fact, it has improved public health? As you will see below, the evidence is far from clear that they are even a small net benefit. In fact, the invention of the automobile probably did more for public health than all of the work of the FDA in its 60 year history.
What follows then, is a very brief primer on the Economics of the FDA. In it we shall ask what the FDA has done for new drug development, for drug safety, for drug costs, for competition in the pharmaceutical industry, and for private incentives to maintain quality. Many thanks go to economist Dan Klein, whose work on the FDA has been an invaluable resource to my understanding of the institution.
Brief History of the FDA
The FDA as we know it first got any meaningful power as a result of the Food, Drug and Cosmetics Act of 1938. Some of the impetus for this Act was likely due to the release of a sulfa drug in 1937 which killed over 100 Americans (mostly children). In Europe in the 1950s, the Thalidomide episode which led to the deformity of 10,000 children also added to the groundswell support for a regulator to “do something” to prevent such horrible injustices.
Its powers were substantially enhanced by the 1962 Kefauver-Harris Amendments which added a “proof of efficacy” mission to the FDA in addition to its original safety mission.
The FDA has monopoly authority to not only determine whether a company can manufacture a drug or medical device, but also what the company can even say about it (what was that uproar about corporate free speech? we are not even talking here about political speech, but rather what companies are allowed to say about themselves).
For years, men at risk for heart troubles have known that taking half an aspirin per day would reduce their risk of myocardial occlusion. By 1988, research had well established this fact. But for years, the FDA forbid aspirin makers from advertising this fact. By this simply restriction alone, tens of thousands of Americans have been killed. And this is just an advertising restriction on ONE simple drug.
The Tufts Center for the Study of Drug Development lists dozens of drugs where FDA delays have stifled development and cost thousands of lives. These include, but are not restricted to, Practolol, Interleukin-2, Taxotere, Vasoseal, Ancrod, Glucophage, Navelbine, Lamictal, Ethyol, Photofrin, Rilutek, Citicoline, Panorex, Femara, Prostar, Omnicath, and Transform. Each of these has been subject to long delays allowing many thousands of deaths and countless suffering to continue.
In 1992, the CDC recommended that women of childbearing age take folic acid supplements. Studies showed that taking folic acid reduced risks of babies suffering neural-tube birth defects such as anencephaly and spina bifida. The FDA immediately announced, however, that it would prosecute any food or vitamin manufacturer that placed the CDC recommendation in its advertising or product labeling (much like they are threatening Cheerios today). However. since 1998, the agency has required manufacturers to fortify a variety of grain products with folic acid—that which is not prohibited is mandatory!
Scope of the FDA
20 cents of every dollar consumers spend in the economy are on products that fall under the purview of the FDA. Oft repeated canards like “safety,” “consumer protection,” and “information asymmetries,” have given the government a virtual strangehold on the drug industry through the FDA. And despite the obvious economic implications of its existence the FDA has never sought to gain expertise in the economics of drug regulation although many economists have studied it for a living. Certainly, the medical profession seems painfully unaware of the economics of drug regulation (or of much of the problem in the medical field in general).
Economic Rationale for the Existence of the FDA (see here for a more complete illustration)
The only plausible economic rationale for the existence of the FDA is that there is a serious information asymmetry problem which has the result of driving all good medicines from the market. In order to justify the existence of the FDA, one would have to demonstrate that FDA veto power over the bringing of a drug to market will correct some systematic error made in the face of grave uncertainties. So, one must not only demonstrate a grave failure, but one must also demonstrate that FDA power overcomes this grave failure.
Remember the hoops you must jump through to prove this. You would have to argue for the FDA to approve drugs in the first place, that doctors when prescribing medicines, make dramatic errors in a predictable and serious way. Do the good doctors of the AMA want to tell their customers whom they presumably care so much about, that this, is in fact, a true statement? If not, it begs the question of why the good doctors feel like the FDA needs to be involved in screening drugs for them. Is not the threat of litigation and an otherwise well functioning feedback mechanism between good doctoring and acquiring patients enough?
Of course, in today’s world, there is not exactly a great feedback loop between patients and doctors. For example, given the cartelization of medicine by doctors, many doctors are no longer accepting new patients. If this is the case, then there are few choices for dissatisfied customers to go. My wife and I have had this exact experience in Rochester. We are unhappy with our primary care physicians, yet are having trouble finding a new one that is accepting new patients (indeed, ask yourself how much of the health reform being jammed down our throats will address this problem – itself created by regulation?). Our choice – patronize the “bad” doctor, or not have a doctor at all. If our county allowed retail clinics to be established, we would be first in line.
In any event, even if you could prove that there was an intolerable information problem in the market for drugs, at best that is justification for the FDA to provide information to solve the problem – not justification for the FDA banning drugs entirely.
How would the market failure happen? In short, drug companies are presumed to know more about the quality of their drugs than customers. Customers, not knowing what drugs are good or bad (doctors remember would be no help, nor would websites, journals, third-party assessors, etc), would look at a given drug and give it some probability of being good and some probability of being bad.
For the sake of concreteness, suppose there are only two drugs – the Red pill which works but cost $100 per dose to produce and the Blue pill which does not work as well but cost only $20 per dose to produce. Consumers would be willing to pay $120 to be healed really well, and only $40 to be healed a little bit. Since consumers do not know if the red or blue pill is better for them, they (assuming 50% of the potential pills are red and 50% blue) they would at most be willing to pay $80 for a pill that has a 50% chance of making them better.
Knowing this, drug companies would no longer produce the good Red pills. Why? Because the Red pill cost $100 to make and consumers are not willing to pay more than $80 for it. The market therefore “unravels” because of this information asymmetry and we see only bad pills selling for low prices. Of course, we don’t actually see this happening in drug markets, but it does make for a nice exam question. In fact, aren’t we all up in a tizzy that drugs are too expensive? If that is true, and I tend to agree with the data, then it does not support the market failure story at all.
If you wish to appreciate the fact that third-parties are already paying for most of American medicine, that raises substantially the amount that Americans are willing to pay for medicine. In other words, the semi-socialized medical system we have now already overcomes the information problem (if there ever was one) by encouraging consumers to be willing to pay more than $80 for medicine! Remember, if you only pay 15% of the cost of medicine out of pocket, and if your willingness to pay for good medicine is $80, then you would support society spending $533 for your medicine – well more than drug companies need to be persuaded to produce the safe and effective medicine!
The market failure story also conveniently ignores the possibility that drug companies and consumers of drugs can do anything at all to overcome this information problem. But it is simply implausible to argue that corporations are greedy all the time, except when it comes to not being able to sell costly, high quality drugs. Do you mean to say that companies would not try to find all kinds of ways to persuade their customers that these high quality and costly drugs are worth paying for? And do you also mean to argue that consumers, who spend hours agonizing over which model cell phone to buy, would not try, at least a little bit, to find out what drugs are good for them and which might not be? Call me unpersuaded. And as you will see below, combine this lack of theory supporting market failures in drug delivery with the empirical record of the FDA, and it is hard to find legitimate economic justification for the existence of the FDA.
The FDA, Cartelization and the Crushing of Small Business
As with much regulation, it is entirely conceivable that big drug companies LIKE being regulated by the FDA. See the history of the US tobacco settlement for a really vivid illustration of this. When the FDA regulates, it imposes huge costs on doing drug research. It may take ten years and very expensive approval processes to get your drug through the FDA Rube Goldberg device. Small drug companies simply cannot get the resources to navigate this process. Often drug formulae will be developed and then sold to Big Pharma, or small companies will outsource the testing and approval work to their larger competitors. In other words, the high fixed costs of getting drugs and devices to markets creates a monopoly-like position for the Big Pharma companies. Even though FDA regulations reduce the profits that could be made from selling more drugs more quickly, the stifling of drug company competitors more than makes up for these losses. And so it goes with all kinds of regulations.
Have you ever wondered why there are not more small and innovative car companies? Have you ever wondered why there aren’t myriad more ways to generate power in the U.S.? Understanding how regulation creates monopolies and cartels will get you a long way toward answering those questions. Thus, calling for an expansion of FDA powers, asking companies to provide more data, to run more tests, to prove efficacy for more treatments – is as much of a cost to large firms as small. And this added cost provides a relative advantage to large and incumbent firms. If you want less medical innovation, there is hardly a better way to get it.
As an aside, think about what such expensive testing and delays do to the ability and interest of drug companies to find cures and ameliorative medicines for rare diseases or diseases that do not strike many Americans.
Quality, Safety and Efficacy Without the FDA
Thoughtful people can understand that high costs and delays in delivery can lead to unnecessary deaths. But they may also suspect that this is an unfortunate cost of keeping us safe. In other words, isn’t it possible that the FDA screens out drugs that would otherwise hurt people?
But remember that there is no discrete concept called “safety.” Even things which are not drugs are unsafe when taken in too large a quantity, or in combination with something else. And remember that some medicine is, in fact, poison. Understanding this means that even a good deal of FDA approved drugs can and do kill people every year. Is that reason to ban the FDA? No. But it should lead one to be suspicious of the ability of any agency to prevent some amount of deaths. For example, in 1994, adverse reactions to FDA approved drugs killed 106,000 hospital patients.
No doubt, virtually all drugs approved by the FDA when properly administered are not very unsafe. But does that mean that absent the FDA that drug safety would be worse? Alas few Americans can even imagine this counterfactual world, and no amount of evidence we present would seem to encourage them to give it a try. But why would one believe that absent the FDA people would be dropping like flies in the street from dangerous, tainted and ineffective drugs?
Remember that private companies are greedy – and the best way to capitalize on that greed is by not killing current or prospective customers. I know that is hard to believe – but ask yourselves whether Wegmans is likely to make more money by pursuing short term profits selling low-quality and tainted food, or by actually promising and delivering high quality, healthy and safe food to its customers. Does it need a big brother looking over its should to ensure it doesn’t feed me rotten pizza or spoiled milk? Let us explore this idea a little further.
Drug Safety already can be, and certainly would be, certified and assured by a large number of institutions were it not for the FDA. These institutions would mostly consist of private-voluntary organizations backed up by the tort system. Let us ask what happens to Wegmans, or to a drug company that harms its customers with unsafe products? Not only would they suffer devastating losses by paying damages to victims, the loss in reputation and future business can very well put the entire firm out of business. Market processes provide feedback loops between customers and firms to make sure customers get the products that they want and to ensure that firms are not behaving badly. The process is ugly, unsmooth and imperfect – but less imperfect than a one-size-fits all regulatory policy where the feedback loops are muted or severed.
There are literally dozens of ways consumers of drugs can verify or seek assurance about drug quality. Here I quote from Dan Klein:
In other words, the very essence of what doctors and drug companies do is about safety and efficacy. All parties try to work hard to burnish their reputations without the FDA having to be a part of it. And since confidence is such an important part of a consumer’s purchasing decision, third-parties have strong incentives (as well as first parties) to provide means to assure customers of quality.
A world of voluntary association can and has done quite well in protecting customers in the face of the supposed “information” problems which need to be solved by regulators.
Public Choice Theory and the FDA
I am tired. But remember that all of the theory above is assuming that government actors are not self-interested, that government actors are not influenced in any way by lobbyists, interest groups and other third parties, and that government actors do not make as many mistakes as private actors do. Regular readers will not need me to detail how likely these assumptions are to be true. Simply put, it is a Herculean wish to think that public choice issues not only do not exist, but do not swamp any potential good that is done by the FDA. The analysis set forth already and following does not require this public choice idea to be incorporated to demonstrate that the FDA is not a good idea. If you want to throw gas on the fire, then by all means, feel free to add your own little public choice story to the mix.
A relatively benign illustration of the public choice problems was actually laid out by Dr. O’Connor in his article:
he was able to unearth a great deal of information from unpublished studies available deep within the FDA web site. Oftentimes these studies contained negative data about new medications – information that was largely unknown in the medical community.
In other words, the FDA who is tasked with getting this information out there, was not able to do it, although it collected the information that was important. And the good doctor’s solution is to give the FDA more power, to ask it to run more tests and ask it to (hopefully) get that information out to people. Really.
Some Empirical Evidence
In short, the FDA has done extreme harm to the public health of Americans by raising the cost of pharmaceutical research and reducing the supply of new and effective drugs and by delaying the approval of the drugs that ultimately make their way to the markets. In numerous surveys of economic research on drug regulation, virtually none has found that the existing FDA should have more powers or even defending the FDA as it exists with its current powers.
But the findings and advice of economists here have been ignored. When economists offer advice that expand the power of regulators and government (see the debate on financial systemic risk or on raising prices on carbon emissions) their proscriptions are heartily endorsed, yet when the very same batch of economists do research suggesting deregulation and less government involvement in other areas like drug approval and public education – well, then, they are not only ignored, but that are called just about every name in the book. Odd indeed.
Three bodies of evidence show that the costs of the FDA well exceed the benefits of having it. First, we study the safety and availability of drugs in the US and abroad. Second we study drug approval times and rates of drug introduction pre- and post-expansion of the FDA’s powers in 1962. Third, we analyze the use of drugs for off-label treatments (see above). And for a treat, we also study the impact of the FDA’s advertising restrictions (also see above).
Comparing the US to Europe
In a 20 year old study we can see that the drug delays we can list, are just the tip of the iceberg. That study catalogued 192 generic and 1,535 brand-name tested drugs available abroad but not approved in the United States. Of the drugs approved by the FDA between 1987 and 1993, fully 73 percent had already been approved abroad. And because the FDA process is so expensive, so protracted, and so uncertain, thousands of untold drugs are never discovered or developed. It is impossible to estimate the suffering and death caused, but surely it greatly exceeds 50,000 premature deaths annually. This is more than the number of people who perish in automobile accidents ever year. In other words, if we banned driving altogether, we would not save nearly as many lives as the FDA kills each year. In recent years, however, the lag times between Europe and the US have narrowed – but this does not change the fact that the length of time for the approval process represents many thousands of lives that could otherwise have been saved.
For example, the slow adoption by the FDA of beta-blockers that had been widely used in Europe was likely responsible for tens of thousands of deaths between 1968 and 1981. In general, research has found that the number of lives saved from withholding dangerous drugs is anywhere from 1/2 to one-twenty-fourth as big as the number of lives lost due to drug lags.
Maybe the longer wait times for drugs to appear in the US brought us more safety? Maybe, but it appears to have not. For example, from 1970 to 1993 comparing safety in the US with Great Britain and Spain – each of whom approved more drugs than the US over that time period, the rates of drug withdrawals for safety reasons were 3% in Spain, 3% in the US and 4% in the UK. And this extra bit of recall in the UK seems to have bought them more gains than it has cost.
Comparing pre-1962 to post-1962
In work begun by Sam Peltzman in 1973 and continued by him and others to this day, they have tracked drug introduction pre- and post-1962, and have associated this with lives saved per drug. The average number of new drugs introduced pre-1962 was 40 per year. Since then, it fell to 16 drugs per year in the decade after the law passed. In total, the delay in the total number of drugs coming to market has had terrible consequences. Estimates of the loss of life due simply to drug delay are in the hundreds of thousands (ignoring the pain and suffering of those who did not die too).
Briefly in Closing
The article reported on by the U of R could very well have been titled, “Rochester Physician Calls for More Deaths and for the Elimination of Small Drug Companies”? How would that have gone over in the community?
Just because that is not the doctor’s intention, does not mean it would not be the result. The ugly history of the 20th century and now into the 21st century demonstrates the vast chasm between the intentions of society’s aspiring philosopher kings for improving the human condition, and the actual results achieved. One needs to look no further than our own backyard in Rochester for proof of this.
I suspect Dr. O’Connor has very good intentions – we both wish for consumers to have myriad safe and effective drugs available. And I guess that when he points to suggested reforms, he has in mind a fairly narrow range of examples for drugs that are not as effective, and perhaps just as costly, as existing drugs. No doubt about it. Every market is littered with these sorts of things – expensive beer that is no tastier and has just as much alcohol as less expensive kinds, cars, electronic gadgets, etc. Ought we demand a regulator step in and evaluate with piles of data the efficacy of new shirts and sweaters as compared to what is out there already? How about new iPhone-type gadgets?
The beauty of the marketplace is that consumers get to decide, and the trial and error process is vital for sorting this out. Of course, consumers can never have full information about virtually any product they buy – but taking this to its logical conclusion means that every single thing we do ought to be evaluated for efficacy, and than we should never be permitted to make choices for ourselves – because we can never know how to make even the simplest item in the world – and in a world where literally millions of people had a hand in making even the simplest products, consumers ought to be hopelessly terrified that no product will work and that no product will be safe for them. So why not? And why is a pill any different than a pie or a scooter or a razor blade or a power tool or a car for that matter? It is not — there is certainly no economic reason to treat these differently.
Which reminds me, how come the good doctor can do nothing but think of the guys with the guns to solve any and all problems? Are human beings that uncivilized that we need to revert to Barbarian methods in order to solve our current problems? It would be nice to see someone, some elite, think outside the government box – just for once. Why couldn’t the good doctor have said he intends to establish a private research institute here at UR that is tasked with doing exactly what he wishes FDA would do?
So, what do we have the FDA for? Power. Period. And so long as the medical professionals feel like they are part of the elites who get to exercise power over us, don’t hope for a minute that you will find many of them asking hard questions about whether the FDA should or should not exist.
UPDATE: Just after I finished this, David Henderson penned this. A highlight:
Seventeen years of human data in millions of patients showing that gabapentin doesn’t cause pancreatic cancer in humans were trumped by a two-year study in 200 rats given extremely high doses. The real danger to our health comes from such myopic views of safety and a hyperconservative, opaque, and capricious FDA keeping useful medicines out of the hands of American doctors and patients. The FDA needs to be reminded that it is reviewing drugs for humans, not for rodents.