Does the FDA Save Lives?

0:00:00.0 Peter Van Doren: There’s nothing wrong with people trying out those kinds of products and finding out they don’t work very well, because the consequences are not forever, it’s just, I got a bad game.

0:00:10.8 Aaron Ross Powell: You wouldn’t say that if you were the eight-​year-​old who’d been saving for some box to…

0:00:14.8 Peter Van Doren: Well…

0:00:15.2 Aaron Ross Powell: Those consequences are pretty dramatic.

[music]

0:00:19.2 Trevor Burrus: Welcome to Free Thoughts, I’m Trevor Burrus.

0:00:21.2 Aaron Ross Powell: And I’m Aaron Powell.

0:00:22.2 Trevor Burrus: Joining us today is Peter Van Doren. Senior Fellow at the Cato Institute, and editor of Regulation magazine. Welcome back to the show, Peter.

0:00:32.7 Peter Van Doren: Thanks for having me.

0:00:33.8 Trevor Burrus: Yeah, I think that since I now have a bank of synthesizers, that I should maybe re-​compose your theme music, as our only guest with his own theme music. I could make a real acid banger out of it, maybe put some robot voices in there, but we can put that on the table ’cause today we’re talking about the FDA. Why do we have an FDA?

0:00:54.9 Peter Van Doren: There’s an intro. Well, every time in history when there has been a scandal or a muckraking incident regarding, first mostly food, and later of drugs, the Congress responds by passing statutes that say, “We need to do something about this because there was an incident and people died,” or populist press said, “Consumers were not being well-​served by markets as we understood them at the time, and there was too much fraud, too much deceit, too much… Fill in the blank, and thus, consumers were not well-​served by what happened.”

0:01:48.4 Trevor Burrus: Well, it’s 1906 for food, right?

0:01:50.8 Peter Van Doren: 1906.

0:01:51.6 Trevor Burrus: But that was just food and labelling.

0:01:52.9 Peter Van Doren: And this goes back to stuff we learned in high school about Upton Sinclair and the meat-​packing industry in Chicago, and the 60 Minutes of its time showing what stuff was being put into what we now call “industrialized food” and how different that was, and blah blah, blah. So in 1906, a statute was passed… The first Food and Drug Act, which basically had a lot to do with food, and as always, even now, was labeling, which is what… So something called “poisonous products” could not be sold in interstate commerce. And then we had our first labeling requirement, because remember back then, in terms of medicines, we really didn’t have what we now think of as pharmaceuticals, instead we had, at the time were called “elixirs”, which were mostly opiate-​based and alcohol-​based stuff; Coca-​Cola had cocaine in it when it was invented; and what we now call patent medicines which, Lydia Pinkham’s cure for everything, and there were just thousands of these things.

0:03:09.2 Peter Van Doren: And the 1906 Act said, “Well, there’s stuff in there, and you have to tell people that there’s stuff in there,” so if there’s alcohol or morphine or opium or cannabis, those were all required additives that had to be disclosed to consumers, and you could be fined, the manufacturer could be fined if the label was not accurate, if it had those things in it. You could call it an elixir, for example, and if it did not have alcohol or morphine or opium in it, you could not label it as what was at the time a common term called an “elixir”.

0:03:50.3 Aaron Ross Powell: A question about how this plays out though, because there are lots of crappy products all the time, and we all have the experience of buying a product and then learning that it’s not very good. It was terrible as a kid to save up your money, the 60 bucks for a Nintendo game, or 50 bucks I guess at the time, and then come home and it’s just a crappy game. And we’ve all had that experience, but we don’t have the government hop in and say… We’ll call on the government to say, “It should certify that all video games aren’t crappy and buggy before they can be released.” There’s lots of bad products. Here it seems like so many of these products, there’s a health angle, but I’m not gonna make a product that poisons people because I’m either gonna get in trouble for it, or I’m going to turn off my customers. You can only sell so many products that poison people before people stop buying them, but on the other hand, things like elixirs and whatever else, the way that bad products get filtered out is someone buys that, takes it, sees it doesn’t actually help them, it doesn’t accomplish much, and they tell their friends and it gets a reputation for not working very well, just like different pizza chains get a reputation for having bad pizza. So, why do we need to inject the government in here, specifically?

0:05:18.7 Peter Van Doren: Well, let me get… Now we’re into the metaphysics of knowledge and knowledge acquisition, and then, how do people make decisions based on what knowledge exists? And there’s certainly… The market-​oriented argument is, as you stated, which is that there’s learning, there’s updating, there’s etcetera, etcetera, and then we put a little asterisk to say, “Well, well, the health… Ooh… ” If you… Sometimes there’s learning and then there’s bad effects from that kind of learning through disease or death, and so the question… And then people vary as to how much knowledge they want to have about something before they ingest it or use it in a medical context. The bad news for our side, I think, is what I call the nutraceutical industry, the nutritional supplement industry, which is not regulated in the way that pharmaceuticals are. And so, there’s… People everyday put in lots and lots of stuff into their bodies based on claims they see in advertising that, this improves memory or this improves the cold, or this does this, right, natural alternatives to what people call pharmaceutical of, chemically-​created and scientifically-​studied drugs.

0:06:45.2 Peter Van Doren: And the evidence is overwhelming from studies that have been done about these kinds of supplements, that there is no learning that people adjust these things, even though the effects are zero. As best we can tell, the negative effects of these things are not that large either, so basically lots of Americans continually waste their money over hundreds of years, starting with elixirs and things like that, things that really didn’t have any effects on health, other than alcohol made you sleepy and when you were sick, that in fact was the most useful thing we could do for you back then. So, the evidence is not striking, that there’s lots of learning in the market we have, which is relatively unregulated for what I call nutritional supplements, and…

0:07:35.9 Trevor Burrus: Is that necessarily a bad thing though? You said that…

0:07:39.1 Peter Van Doren: No, I’m not… I’m just saying the… Aaron makes… Market-​oriented people and economics says, “If you keep making stupid decisions, eventually you run out of money,” that’s usually… There’s a budget-​constraint, and people, they learn and they [0:07:54.4] ____ and update, we have all sorts of technical terms for this kind of process, but there’s a whole industry out there where there’s not much updating. Every night on television, I see people saying, “I take Prevagen and my memory has improved.” Well, [chuckle] I suspect that that actually isn’t true. So, what Aaron says, they talk to their friends. Well, that’s how these kinds of over-​the-​counter things are in fact sold, which is people make inferences, even doctors, right, there’s observational studies in medical journals all the time, which is, “I studied these patients, I gave them some stuff, and then ooh, they got better or something followed and nothing bad seemed to happen, therefore, this is a cure, or this is something.”

0:08:39.7 Peter Van Doren: And then, we do trials and we find out almost all these claims, or many, many, many of the claims that people make based on observations or their personal experience as well as the experience of others, turns out, in a random assignment placebo-​controlled environment, very, very few of these knowledge understandings that people have are actually true. And part of the struggle over the FDA is the different epistemological concepts people have about knowledge, knowledge acquisition, how they learn about things and what they think of something called science and scientific experiments and what they learn from them.

0:09:27.9 Trevor Burrus: It doesn’t seem like the worst thing ever, people do a lot of things to make themselves feel better, they believe in horoscopes, and they believe in Tarot cards, you can list them on and on and on and on, and generally we say it’s like, “Hey, it’s harmless, maybe you spend a lot of money on Tarot cards and you think that people… These guide your life in some way, but ultimately it’s your money and it’s kind of harmless.” And it seems to me that if any of these supplements was significantly harmful, that would be indicated pretty quickly, so it just indicates that they don’t really do anything, but if it makes you feel better then go for it, kind of thing. That would be one way an economist would look at it, right?

0:10:13.4 Peter Van Doren: Well, and actually the key, the key is how rare are the negative effects or how common, that is the ability of people to see negative things happening from observational data is much, much easier if in fact the harm is immediate, obvious, and large. Small negative effects that are detected at the rate of 1% or 2% or 5% of the population actually are even difficult to find in trials with normal number of subjects in them, because trying to differentiate actual harm from random error is actually difficult, the more rare the negative event is, and that… So that plays into this… Yes, the nutritional supplements does… In fact, the trials that have been of them don’t seem to find any even rare harms, they just don’t have any positive results, it’s just a waste of money. The issue of finding real harm, we need trials to find those things if in fact the harm is real but small. And then it’s difficult even then because the trials aren’t big enough, and that’s an issue over time in the FDA when… The difference between doing a trial with 8000 subjects or something, which is, as big as a trial as the FDA would ever require, versus having it out there in the marketplace where 300 million people use it. Same thing with, right now, the Johnson & Johnson vaccine that we’ve got 224 rare blood clots in Europe, out of what? 200 million, something like that.

0:12:15.9 Peter Van Doren: So, finding that kind of error can’t occur even in a trial setting, all the trials that Moderna and Pfizer and J&J ran, you couldn’t detect that kind of error. Same thing with, sort of I guess, the so-​called Vioxx scandal that happened with the drug Vioxx and the FDA. The heart attack rate from Vioxx appears to be in the kind of 1% land, and that even in a trial of 8000 people, that’s the kind of error that can be easily confused with a noise in which there’s nothing really going on. So, the more risk-​averse you are and the more zealous you are, the more you want a larger and greater observational data or word of mouth, if you will, you wanna hear from thousands of people that they have done this and nothing bad has happened to any one of them, etcetera, etcetera, etcetera. And people vary as to how worried they are about this. And it depends on the kind of thing we’re worried about, heart attack and death you don’t get to redo; whereas, the kind of learning that Aaron talked about, about video games, there’s nothing wrong with people trying out those kinds of products and finding out they don’t work very well because the consequences are not forever, it’s just, I got a bad game.

0:13:45.4 Aaron Ross Powell: You wouldn’t say that if you were the eight-​year-​old who’d been saving for some box to…

0:13:49.7 Peter Van Doren: Well…

0:13:50.2 Aaron Ross Powell: Those consequences are pretty dramatic. But we started this conversation with, in the early 1900s we have a reaction to the Upton Sinclair stuff, and then you sent us… One of the reasons that we keep bringing you back on the show is because you always send us these really comprehensive outlines to work off of ahead of time, so Trevor and I… You do the research for us. But in that you have this long list of events that led to more and more and more stuff, either falling under the purview or arguments that should be regulated and so on. And I’m curious how often, in retrospect it turns out to have been wrong, because we can say, “We should have stopped this medication, or, “We should have… ” They were putting these additives in the food or whatever, but so much of this is like it’s small risks, it’s large studies. How often do we say, we react to something, there was something bad that we thought was happening, we react to it, we give the FDA more power, we control for it, we ban it, and then later on it turns out like it wasn’t that big of a deal really, or we got the science wrong, or those additives weren’t necessarily a problem. Does this run in the other direction as far as updating our knowledge?

0:15:08.2 Peter Van Doren: Well, sort… The honest answer, I think, is we don’t know. One way to think about this is, if we think of the denominator of a ratio as the number of new drugs, investigational new drugs, that where some manufacturer says, “I’m gonna study the properties of this ’cause I think it might have this beneficial effect for people, but I need to do trials because of the current FDA regime, and I’m gonna pursue this first for safety in a small trial, phase one, and then two and three are safety, and two, in efficacy and dosage, and then three is mostly efficacy and not much safety.” And the data are overwhelming about how few drugs ever come out of this process, it’s only about 8%. So you’re saying, “How about those other 92% of things that in laissez-​faire world, some manufacturer might have had the cajones?” if I could use that technical term to say, “I’m just gonna send this out there and say, ‘I think we’ve identified this and it might be good and it might be bad, we don’t know very much, but here it is, alright, and here’s what’s in it, and here’s what some smart people think it might do, and here’s what some other smart people think it might not do.’ ”

0:16:46.2 Peter Van Doren: But in the current world, there isn’t a re-​investigation of those things. Right? Now and then there is, there’s a repurposing of products, mostly that have already been approved for other purposes and are shown to have been safe and efficacious. For example, Viagra was a blood pressure drug, it turned out to have other attributes, and… But it had already been approved, right, for this, and then it was served another purpose. The question you ask is, “What would our world look like if all those 92% of things had been tried out? How many bad events would the media and whatever ever be able to find?” The answer is, I think probably a lot more than we now hear about, because in fact, none of those things get to market now, they’re just stopped in phase one. It’s remarkable how few get to even phase three. Of all those that get to phase three, a large majority of them get to market. A small percentage of them are stopped, but bunches of things are just stopped right in the get-​go, and so for us to argue what our world would be like, we’d have to say, “Well, what would people demand about using stuff if there were no restrictions?” And we don’t know the answer.

0:18:09.1 Peter Van Doren: I say, the nutraceutical industry strikes me as bad evidence for us because people don’t seem to demand very much at all, and we need to eventually talk about medical devices, which is, even in the regulated world, the pharmaceutical world is very, very regulated and the medical device world is not, so we have… We sort of have a quasi-​Cato world in the medical device world, and that, isn’t that pretty from our perspective either, or from a safety perspective, in that clinical trials are not required, and there are lots of devices on there that have ended up doing lots of harm to people. And again, because they’re not really laissez-​faire, we’re in this mixed world…

0:18:56.0 Peter Van Doren: My wife’s had five hip replacements. If we ever had a conversation with the doctor in which we asked him, “Well, what do you know about this hip and why?” And the sad answer is, even though we’re professionals, and Catherine is a health professional, we’ve never asked those questions. So then you’d say, “Well, why?” The answer is, ’cause in the world we’re in, everyone thinks someone has figured all this stuff out, and no one should be scared of stuff the doctors are practicing widely. Whereas from our opponents’ view, a frightening Cato world actually, it’s certainly possible that consumers would be much more vigilant and much more worried and would require evidence, which in fact, in a new equilibrium, we’d have voluntary provision of knowledge rather than FDA requirement. But there’s no place in the world that actually tries this out, so it’s very hard to answer your question.

0:19:52.4 Trevor Burrus: So what is the process? I know that a big change occurred in the FDA and a lot of the criticisms comes in in 1962, so as you said, phase one, phase two, phase three; at the very beginning, it’s like, Does this poison you? But then you get to this question of, Is it efficacious, and suitably efficacious too, that someone is determining our level of risk and whether or not it is, in terms of some bureaucrat, efficacious enough for us to take it? Which seems a little weird. And that’s where a lot of problems come in, and everything kind of changes then in 1962, right?

0:20:31.5 Peter Van Doren: Agreed. Again, some of our previous conversations about have been science and the limits of science, and I wanna make our listeners very aware that I’m very into science, I love knowledge, I love the acquisition of knowledge, but, I wanna argue that decisions don’t directly follow from science, you need to inject values and costs and benefit analysis, and people differ in that. And yet, instead we have scientists in wrestling matches over trial data, which… I mean, a trial’s a trial, right? It’s just math. And here’s the other thing, which is, what we learn from a trial is relative to what’s called the standard of care, and so ironically it’s not always placebo. If we already have something on the market that deals with that, the control arm of the trial is that standard of care, not just nothing. ‘Cause nothing would be unethical if we already have something that we think works, which may or may not be true. So there is a whole game of, What’s the control arm of the trial and what’s the experimental arm of the trial, and then what kind of people do we let into the trial. Do we exclude some people with pre-​existing conditions? In other words, do we rig the trial so that the likelihood of a good outcome or a bad outcome is in part determined by what lots of people we let into the trial.

0:22:10.3 Peter Van Doren: So then you get a result, and you have this outcome in the experimental arm, and you have the same outcome in the control arm, and then you measure whether the difference in incidence is great enough with some statistical confidence that we then can say scientifically, It looks like this thing solves this medical condition relative to people who were in the control arm. And then full stop. Whether or not it should be approved, or whether or not people should take it or not, depends on risk aversion and costs and benefits, which vary a lot across people. And yet we have a collective declaration on the part of “a scientific advisory committee” and then the FDA itself as to whether, not really about the science, but whether this drug is worth it or not given everything we know and given how rare or not rare all of those other things. Which technically isn’t science, it’s something else.

0:23:19.0 Trevor Burrus: So I know that… I think that one of the first big papers on this that came out, to discuss whether or not this kind of efficacy testing passes the cost benefit analysis, was from Peltzman I think in the early ‘70s. And so when we’re talking about how an economist would look at these efficacy trials and when drugs are blocked or when they’re not allowed out or where people are not allowed to choose their own risk level, Peltzman and many other economists have argued that you add all that up and ultimately it makes people worse off, not better. Correct?

0:23:53.3 Peter Van Doren: His initial analysis said that, since then, in the notes I sent you, there’s been subsequent analysis of a particular regime change at the FDA. Which was, as conservatives have argued and everyone… And it’s a true stylized fact that the FDA is… We know it’s risk averse and we know it’s slow. And so there’s… In 1992, the Congress passed a law that said, Oh, it’s too slow. We’re gonna speed this up, but we don’t wanna have taxpayers pay for it, instead we’re gonna charge drug companies for, in effect, hiring extra staff to speed the process of new drug applications through the FDA. Well, it did speed up, it sped up quite dramatically, and then an economist, Tomas Philipson, actually said, Let’s take pre and post-’92, and then take all the drugs that were ever withdrawn after the speed up of the FDA process. So they were literally withdrawn. And then calculate the life years lost, estimated life years lost, from those drugs when they were on the market. And then price those life years at the economists’ standard value of a life year, which is somewhere between $100,000–300,000. And then he calculated, alright, this is the cost of the speed up.

0:25:19.4 Peter Van Doren: And it came up to some… The notes tell me how many thousands of life years were lost, I think it was in the tens of thousands. And then you multiply by 100,000 and you get a number in the billions, of the cost of speed up. Then the benefits. And he said, Well, then we know all the drugs that didn’t cause any harm but you got them faster. And it talked about literally through a discount race, speeding up the consumer surplus that came from the speed up. And he came to a conclusion that they more or less balance out. So there was their cost of being faster and their benefits of being faster post-’92, and they seem to be in order. So the kind of Ralph Nader Consumer Reports claim that the FDA has been corrupted by the user fees and the speed up of testing and things like that, the evidence for that does not seem to be true. On the other hand, from our side, there are costs of a speed up, and some drugs are withdrawn because we were a bit too hasty in checking everything out. And, again, the size of the trial and then how long the trial goes on, right? You get this game of, How long do we wait for side effects to show up? And then you need some medical and scientific understanding to make a guess about, This drug affects this metabolic system in this way, and if we’re gonna see an outcome, it’s likely to show up in such and such a length of time.

0:26:57.1 Aaron Ross Powell: How are we measuring benefits, or even conceptualizing benefits, in this context? Because a drug… We don’t… There aren’t many drugs out there that have nothing but positives, you take them and they make you feel awesome and there’s no downsides? Or you take the drug and it just flat out kills you without doing anything nice for you in the time before? We can exclude those because they don’t really exist, everything else is about trade-​offs. And so if we have a drug that, say, prolongs late-​stage cancer by two months but causes pneumonia during that time, is that drug… Do we put that drug in the positive camp or the negative camp? And then a lot of these things may have more… No one wants pneumonia, but we might want it more than we want dying from cancer two months earlier. But then a lot of this is also just really individualized, that you might say, lots of activities that we take bring costs, but you might think of them… You might think that the cost of playing a particular sport is the wear and tear on your body, but I don’t even conceptualize that as a cost because I get so much out of playing the sport that the wear and tear is just kind of an indicator that I’ve been successful.

0:28:30.9 Aaron Ross Powell: So how do we, when we’re economists putting on our economics hat to measure this stuff, how do we even begin to say this is a cost and this is a benefit, or to start measuring them against each other?

0:28:42.9 Peter Van Doren: So the Philipson study does that by taking consumer surplus, and it’s too hard in this kind of discussion to try to describe, it’s basically people’s willingness to pay for something above the actual costs of doing so, of providing that product, that’s consumer surplus. It exists on a blackboard. Economists have various ways of attempting to measure that. Let’s assume for purposes of discussion that they can. Alright? And that’s where this measurement came from in the Philipson study, and then all he said is, So we know what people’s willingness to pay for stuff is relative to cost, he just studied the fact that that was moved up earlier, that’s the benefit of a speed up. Your larger… Let me switch a bit to, you’re asking also, why do we have to collectively consume this decision? And part of my outline that I sent to you guys was Peter’s… My way of looking at it, the meta question that I don’t think enough people talk about, which I want to, is, we have a regime for 100 years now in which we acquire knowledge to inform our decisions by restricting decisions. In effect, we coerce drug manufacturers into performing scientific investigation of things. And why do they do that? Well, ’cause we don’t let them put a product onto market unless they do.

0:30:18.9 Peter Van Doren: Well, is there some other way, do we have to do it that way? And the answer is no, we don’t. Knowledge has public good characteristics, once it exists it’s difficult or impossible to restrict consumption to those who pay, so knowledge has the classic public good definitional aspects to it that one learns in microeconomics. Alright, so we could have NIH run lots and lots and lots and lots of trials, we could just raise taxes, we could spend money, and we could collectively run science, but we just do it separately from decisions, right? We don’t have to answer whether something should or shouldn’t be on the market, it’s just smart people do trials, they then write labels or they write circulars or they write articles in the New York Times, and it says, Well, as best we can tell, you’ll get this benefit and you’ll get this cost, and it costs this much, and if we take a normal value of a human life it may or may not… You know, blah, blah, blah, blah, kind of all science nerd-​dom forever, and then Consumer Reports and Channel 7 and Channel 4 try to distill it down to help people make decisions, right?

0:31:38.2 Peter Van Doren: The second way of doing it is we don’t have publicly funded trial knowledge at all, we just have laissez-​faire. And then through the kinds of arguments we’ve been talking about today, we think through backwards induction that, Wow, lots of people who worry about things would not consume products if no knowledge were provided, so then we’re back to companies would compete on the basis of the investigations they had done, and they would develop brand names over the acquisition and provision of information, and blah, blah, blah, blah, blah.

0:32:15.8 Peter Van Doren: And then, in Cato equilibrium, we’d say, Well, how different would the world look like in terms of knowledge generation than it would now? If you’re left of center, you’d look at the nutraceutical industry and you’d say it’s hopeless. There’s far little, too little demand from an upper middle class point of view, on the part of most people, for knowledge. And there’d be lots and lots and lots of stupidity and harm out there in Cato world. And then if you’re us, we point to the examples we think contradict that and we would say the opposite. And probably both are right, right? And then we get back to some people are risk-​averse and risk very hesitant to do stuff, and other people are not. And we would then have to hope our institutions would live with the consequences, and in effect would we let disasters… In other words, there’d be incidents, and just like in 1937, the so-​called Elixir Sulfanilamide scandal that led to the 38 amendments, basically a solvent was put into the first sulfur drugs to make it easy for kids to ingest when they had colds, and hundreds of them died ’cause it was a poison. And some people knew that at the time, the chief chemist who made it for this firm didn’t.

0:33:42.6 Peter Van Doren: And, oh my god, the you know what hit the fan. This thing killed kids. And, in fact, the chemist involved committed suicide, he was so distraught over what he had done. So can we live with these kinds of things and not have the political system intervene in our world? The answer is, historically the answer has been no. We just… There’s… Anyway, you keep asking me questions about this…

0:34:15.2 Trevor Burrus: But on the other side of this, a few years ago there was a case at the DC Circuit called Abigail Alliance, which came out of the death of a young woman named Abigail Burroughs who had a rare form of spine and neck cancer, wherein she was told she could not use a drug to treat that cancer that had promising indications, but had not yet been approved for efficacy for that type of cancer. And she eventually died and a lawsuit in her name asked the question of whether or not there is essentially a right in the Constitution, or protected by the Constitution, to try something to save your own life. And it seems weird that you would tell someone like that, This is too risky for you, when they’re looking at two months to live. And that’s a big controversy, that’s Right to Try, and I know you and others at Cato, like Michael Cannon have disputed this, but to me it seems pretty obvious that, in those situations, taking someone’s risk and telling them how risky it is to you, if it’s worth it to you, is really crazy.

0:35:27.9 Peter Van Doren: Yeah, I’m not disputing that. What the argument for… Again, people who like trials, and I’m one of them, and Mike Cannon and I debate this, which is… The people who defend trials and the knowledge that’s acquired through trials do not believe we could run trials unless we can restrict access, because if we don’t restrict access, we never would get anyone to be in the control arm of an experiment. In other words, if there’s a promising treatment, and it’s basically everyone involved is dying, how do you randomize people into the control arm so that you figure out whether or not this thing really does anything or not? If nobody’s willing to be in a control arm, we’ll never learn. Now, Mike Cannon says, What we need to do is pay people enough. In other words, we need to have control arms, not as mandatory guinea pigs because we don’t have market access because we don’t have trials, but knowledge is useful, let’s pay people to acquire knowledge. And the so-​called challenge trials in Britain, other countries have… We don’t, but in this vaccine, the development of the vaccine we have in Europe, they had some challenge trials.

0:36:53.8 Peter Van Doren: And basically you pay mostly young people a lot to be guinea pigs, and they’re in the control arm or etcetera, or in the experimental arm in the case of a vaccine where we don’t know what’s going on. And so I’m not against access, per se, but I worry… Or people who defend trials, it’s extremely difficult, and in fact impossible, I’ve been been in debates with people a lot over this, so the… Ezekiel Emanuel came to Cato, and it’s a famous Cato discussion, his and most other people’s position is that you cannot induce people once the horse is out of the barn. Once something’s available, you then actually cannot conduct a trial. Because once word is out, there’s tremendous attrition in the arm that’s not receiving treatment and the trial collapses.

0:37:56.9 Peter Van Doren: And anecdotally that’s true in the current world. My wife has run a number of trials and once the word is out, tentatively that, something that the committee has revealed some results and there’s good news, it looks like, but we have to keep going to make sure everyone flees the trial. You just can’t random… You have tremendous attrition and you don’t learn. And so then the question is, if we offered people enough money, would they stay and continue with the trial? And then we’re back to, who would come up with the money? Taxpayers, or would companies do this, even in a world of free access? And the honest answer is, we don’t know ’cause we’ve never tried this kind of world.

0:38:42.8 Aaron Ross Powell: What’s the cost of that? If we live in a world where a little bit of evidence comes out that this thing might work, but we don’t know for sure, and then everyone flocks over to take it. If it turns out that it’s then really dangerous, let’s talk in the case of, that Trevor mentioned of the woman with the terminal cancer. And it turns out it’s really dangerous and it kills people who take it in that situation faster than they would have died, it kills them in a month instead of two months or whatever. As that starts to come out, we’re gonna get an equilibrium back in the other direction because people are gonna say, “Oh, it turns out I’m not going to take… I’m not gonna jump into this thing that… ” So you get more people who are willing to not take it, to test against. But if everyone rushes in and then it turns out, the worst that it seems is that a bunch of people are then taking a drug that isn’t doing anything, or doesn’t have as much of a positive effect as we had originally thought, which is potentially a cost, but there’s also that we have now in the interim, potentially allowed people to be taking drugs that are helping them when before we would have been excluding them because we needed this control group.

0:39:57.8 Aaron Ross Powell: And so if we just opened it up to everything, it might be that there are more drugs out there that aren’t as effective as we thought, but there would also be more drugs out there accessible to more people that are effective. Is that a problem? Or… Because it seems odd to just necessarily say, “We’re willing to potentially sacrifice people’s lives in order to meet some knowledge threshold that we think is important, because we’re talking about people’s lives here.”

0:40:25.6 Peter Van Doren: Oh, I would have to unpack all this… Let’s see, how can I… Remember, I said, what the current system does is prevent lots of stuff from being out there. Okay? So, of the trials that are done on procedures and practices and all of that that are already out there, we almost exclusively find, not that they cause harm, but that the positive effects are very low. So this is the Robin Hanson… We’re wasting somewhere between a third and a half of medical expenditures, even in the system we have now. So then we have to talk about, alright, in Cato land, what would insurers… No one actually pays their own money for expensive stuff, so this is… So then in a laissez-​faire world, what would insurers do? Well, insurers, basically now, there’s a world where the FDA approved stuff. Alright, so then, everything that’s FDA approved still isn’t covered by insurance. Then Medicare goes through a process to try to say, “Well, we cover it,” well, then guess what? All the privates follow what Medicare does. So we’d have to undo that entire world we have now, in which Medicare approval of expenditures for something is the gold seal for everyone else to follow.

0:41:57.4 Peter Van Doren: So then again in Cato land, would we have people using their own money to do stuff, and the answer is no, not much. What they would say is, they want insurers to pay for everything that’s available, and since everything’s available, we then would need knowledge. And so then, where would the knowledge come from? And we’re back to, well, we either publicly fund it and then insurers base decisions on that, and then they don’t cover and do cover and then insurers would vary, but then, oh my God. We went through this, right, with the patient bill of rights in the early ‘90s. When insurance companies start to crack down and say that something’s not approved and we’re gonna constrain you, all hell breaks loose because Americans like healthcare. Again, the Cato world that you describe would eventually have to run into insurance, and then, describe what would happen there, and I don’t know what would happen there.

0:43:01.4 Trevor Burrus: I wanna go back to… The human challenge trials is interesting because in the pandemic, we learned that we had the vaccine in January of 2020, and then we spent…

0:43:14.9 Peter Van Doren: Well, sort of… Let me put in my editorial comment, which is, do you know what Europe did made sense and what we did made absolutely no sense, in terms of diversified portfolio… If I’m a risk management officer for the United States, would you wanna place operation work speed on a never proven messenger RNA vaccine technology that had never ever worked? Whereas J&J and AstraZeneca were using vaccine technologies that were totally wonderful, and had been shown to be useful in other contexts. So Europe, what that now looks like it’s behind the eight ball, actually was very sensible. The US was an extremely high-​risk, absolutely stupid bet which happened to work out. Alright. So be careful in saying that… I am… If the world had gone the way we had thought it should have gone, given our knowledge a year and a half ago, the New York Times would be saying how stupid the Trump admini… Only the Republicans would make a stupid bet like this, right? Do you notice how no one is saying this? And yet I think, from a financial portfolio way of thinking about things, the claim that we knew we had something that worked in February 2020, I would say, absolutely no way, Jose. Not true, we just lucked out.

0:44:46.7 Trevor Burrus: Okay. Well, even then, let’s say a few months later when things are looking more promising, but this idea of human challenge trials, which at this time, there would be a lot of people probably willing to say, “I’ll try it out.” And it’s interesting because, we pay people to take risks, we pay people to strap them to rockets and blast them into space, and… One of the problems with NASA, one of the reasons NASA is not terribly good at what it does is ’cause it cares too much about astronauts, that’s one take. Right, it’s like you just said, there’s someone out there who’s like, “Man, we’re gonna blast you to Mars, and it’s a 50–50 chance you’re coming home, but you can have a hell of a ride on the way.” We do that, and I think we should do that more with space, but why we should do that in times like this with pandemic we should definitely be paying people, and I don’t think it would be that expensive or difficult to do.

0:45:40.6 Peter Van Doren: Yeah, fair… I mean, I’m not…

[chuckle]

0:45:41.7 Trevor Burrus: No dispute there?

0:45:42.7 Peter Van Doren: I have no views… I’m saying, let’s try… It’s…

0:45:47.3 Aaron Ross Powell: Just to clarify then. When we’re talking about the need to pay people to participate, or in the scenario we’re just talking about that need to pay people not to take this, this medication, how many people are we talking about? If we need to find an effective medication and we need to pay some group not to take it, do we need 50% of the people who might benefit from it to not take it? Do we need 10%?

0:46:15.5 Peter Van Doren: Oh, you need… I thought you were asking a different question, which is, how large should the end of the total trial be, and then the ans… And then you always divide in half, there’s just, half are assigned to some treatment and half are assigned to nothing or standard of care. And then based on some science that we have, we then unblind the results at some time in the future, and we see what the results are, where is the… And the difference in means, sufficiently large enough, so then we have to insert the non-​scientific question is, how confident do you want to be in this result and what level of side effects in terms of percent do you want to worry about? And then, once you tell me that answer, I can tell you how many people we need to pay. And a chart I gave you in the outline is, if you wanna worry about things in the 1% range, you gotta have thousands and thousands of people in the trial, and that’s why… And no one… That’s just a lot, a lot of money. I gave you some data, I think, which is, I’ve always… In these discussions, there’s a lot of papers out there that say how much it cost to bring a new drug to market in the US, under the current regime. And it’s in the billions of dollars.

0:47:43.8 Peter Van Doren: Alright. So, what percent of those costs are the trial cost, versus, let’s call it the science R&D costs? In other words, if we had no trial requirement and we just had laissez-​faire, then we still had R&D, how much would drug prices go down, or how much of the current drug price is R&D versus trials? And the answer I found was in… Basically in the last 10 to 15 years, 62% of drug costs are trials and only 38% are the scientists doing that, what’s called pre-​clinical R&D. And so the trials are expensive, and the… So how rare an event do you wanna worry about? And in the Vioxx case, we found that 1% of heart attacks weren’t visible in the end, at… Before the FDA approval, and then in the much larger trials that occurred by accident actually, it was… Merck was greedy and wanted to claim that this, that Vioxx was the best thing since sliced bread, so they did a trial versus Naproxen.

0:48:55.7 Peter Van Doren: And then there was some theory, and it was reasonable that it might actually reduce colon cancer, that Vioxx would reduce a recurrence of polyps in people who’d had polyps in a colonoscopy. People hate colonoscopies, and colon cancer is an ugly death. So wouldn’t it be wonderful if Vioxx could stop that? So they did two big trials after it was approved, that Merck paid for, called post-​marketing trials, and they had 8000 people or so, and they found 1% heart attack rate because the end was sufficiently large, whereas the trials before approval were much, much smaller. So, we’re back to how, and is 1% heart attack rate worth it? Well, not if… ‘Cause everyone was taking Vioxx, 300 million people… My wife was taking Vioxx. Vioxx was wonderful, it reduced arthritis pain, beautifully. Well, it causes heart attacks. It had a great… Not horrible, but at a rate greater than standard of care or Naproxen, or… And since a quarter of people that have heart attacks die, that was… That was real.

0:50:10.9 Trevor Burrus: Yeah. So much with our discussions with you Peter, when, they’re always enlightening. And if people are looking for the bottomline, the economist, do you wanna… On the other hand question here, but is there a bottomline in terms of the way we should think about the FDA as libertarians or just sort of… We shouldn’t just say it’s a necessary thing, the government does, everything that it does is correct because as you’ve pointed out, politics of the FDA are pretty risk-​averse, so we should probably be aware of the fact that they will pay costs for those heart attacks, but the unseen are all the people who died because they didn’t get the drug, and that’s a really hard thing to pinpoint. Is that just something we have to live with or what can we… Is the FDA…

0:51:04.2 Peter Van Doren: Well, what I… Again, what I said earlier is, I would like to shift the Cato discussion from good FDA, bad FDA, to… One, we should be for the separation… We should not try to acquire knowledge through banning decisions. In other words, we ban products being on the market, as a back-​doorway to pay for the acquisition of knowledge, because we make companies pay for the acquisition of knowledge, they try to gain the system so that the trials design will help them, not hurt them. So in the Vioxx case, Merck excluded people with heart attack risks from their original trials because there was some metabolic chemistry that their scientists were well-​aware of, that the way this drug worked would in fact might have cardiovascular risks. Alright. So then you want Cato designing the trial or somebody… You don’t want Merck designing the trial. So that the current system gets knowledge, but it doesn’t get totally unbiased wonderful knowledge, it gets knowledge that companies want to help them in their effort to sell stuff, and make stuff approved.

0:52:26.9 Peter Van Doren: So, I think Cato should be for inventive ways of trying to separate the acquisition of knowledge from the decisions to use or not use things, and rather than say the world would be wonderful with the FDA, it’s just that we should be both for knowledge acquisition because it’s useful for consumers to make decisions. And we shouldn’t interfere with decisions, as the first order of course, totally agreed. The puzzle is, because we tend to just be anti-​FDA in my view, we haven’t gotten creative in our ways to get people to realize that we’re not against… We’re not for just everyone taking stuff and dying all the time. No, that’s not what we’re for. We’re for the acquisition of knowledge, but we don’t want to interfere with people’s decisions either, that’s also a big priority for us. And the world hasn’t been very kind to us in allowing us to navigate those two goals at the same time, because most of the world has linked the acquisition of knowledge to constraining decisions, and very few experiments have been conducted other than challenge trials to try to de-​link those too.

[music]

0:53:53.9 Aaron Ross Powell: Thank you for listening. If you enjoy Free Thoughts, make sure to rate and review us in Apple Podcasts or in your favorite podcast app. Free Thoughts is produced by Landry Ayres. If you’d like to learn more about libertarianism, visit us on the web at www​.lib​er​tar​i​an​ism​.org.