Oncologist, author, and podcaster Vinay Prasad talks about his book Malignant with EconTalk host Russ Roberts. Prasad lays out the conflicts of interest and scientific challenges that make drugs that fight cancer so disappointing at times. The conversation looks at how policy changes might improve the incentives facing doctors, pharmaceutical companies, and patients.
EconTalk Extra, conversation starters for this podcast episode:
This week's guest:
This week's focus:
Additional ideas and people mentioned in this podcast episode:
Adam Cifu on Ending Medical Reversal. EconTalk. Vincent Rajkumar on the High Price of Cancer Drugs. EconTalk. Robin Feldman on Drug Patents, Generics, and Drug Wars. EconTalk. Robin Feldman on Drugs, Money, and Secret Handshakes. EconTalk. Jacob Stegenga on Medical Nihilism. EconTalk. Azra Raza on The First Cell. EconTalk. Adam Cifu on the Case for Being a Medical Conservative. EconTalk. John Ioannidis on Statistical Significance, Economics, and Replication. EconTalk. Discussion of p-values and other matters. Yuval Levin on A Time to Build. EconTalk.A few more readings and background resources:
Sam Peltzman on Regulation. EconTalk.A few more EconTalk podcast episodes:
Eric Topol on Deep Medicine. EconTalk. Luigi Zingales on Incentives and the Potential Capture of Economists by Special Interests. EconTalk.* As an Amazon Associate, Econlib earns from qualifying purchases.
| Time | Podcast Episode Highlights |
|---|---|
| 0:33 | Intro. [Recording date: February 6, 2020.] Russ Roberts: Today is February 6, 2020, and my guest is Hematologist, Oncologist, and Professor, Vinay Prasad of Oregon Health and Science University. He studies cancer drugs, health policy, and evidence-based medicine, and hosts the oncology podcast Plenary Session. The author of over 200 academic articles, co-author of the book Ending Medical Reversal, which was the subject of an EconTalk episode with co-author Adam Cifu, his latest book and the subject of today's conversation is Malignant: How Bad Policy and Bad Evidence Harm People with Cancer. Vinay, welcome to EconTalk. Vinay Prasad: Thanks for having me so much, Russ. |
| 1:10 | Russ Roberts: We're going to be talking about cancer today--the system that tries to fight it, the cost of that system, its effectiveness. Listeners who want background on the issues we're going to be discussing and listen to past episodes with Vincent Rajkumar, Robin Feldman, Jacob Stegenga, and Azra Raza--we'll link to those episodes and I encourage you to check them out. Now, Vinay, in your book, Malignant, you write the following: For the most part, cancer drugs cost too much and deliver too little. Because of this, their value is poor. Why do you say they cost too much and deliver too little? Vinay Prasad: Well, thanks for having me. It's great to be on this podcast that I've loved listening to over the years. I guess I'd say the answer to that question is, although it's easy to fixate on the exceptional stories in cancer medicine, a drug called Imatinib, which has really revolutionized the disease, chronic myeloid leukemia. Some other really successful drugs are, like, Rituximab and Trastuzumab. And, I think it's easy and natural to gravitate to the success stories. But if you look soberly--sorry--if you look empirically and broadly at many, many cancer drugs, you will find that they offer very marginal or modest benefits. A paper that comes to mind is a paper by Fojo and colleagues in JAMA Otolaryngology. They looked at 71 consecutively-approved drugs for solid tumors. Whey found that the median improvement in overall survival is 2.1 months. So, I think many of us will recognize that 2.1 months is a marginal or modest improvement in survival. And that's the median. That's the typical drug. And you pair that with drug prices that are often $12-, $14-, $16,000 per month of treatment for drugs taken indefinitely that are not curative, and you get value propositions that are quite poor. Listeners of your podcast will know 'dollars per quality-adjusted life year.' I won't need to explain that, but by the-- Russ Roberts: Well, you should remind them. Yeah, remind them, go ahead. Vinay Prasad: All right, I will explain it, then. So, I mean, you know, one metric that health economists and policy analysts use to evaluate value is: How much does it cost to add one good year of life to society? One healthy year of life? And, we discount life if it's in a state of disease or disability, and that's the quality-adjusted part of a quality adjusted life year. And, typically, wealthy industrialized nations have different cut offs on what they think a reasonable dollar per quality figure is. Maybe, the United Kingdom, they'd say something like $60-, $70,000 per quality is the ceiling of what we think is cost effective. In our country, I think there's some persuasive economists who argue it should be as high as maybe $200,000 or $250,000. But that's kind of the upper limit as what we see people endorse. But in cancer medicine, because our drugs offer such marginal benefits in ideal settings and they cost so much, we see dollars per quality that approach a million dollars. Regraph[inaudible 00:04:03] says it is[?] $900,000 per quality-adjusted life year for colon cancer. We see some drugs that may even be infinite, because they cost more than alternatives and they deliver nothing more, such as Aflibercept in colon cancer. So, you know, there's some examples where we are reaching a stratospheric amount of cost per benefit in cancer medicine and the value is, you know, clearly poor. |
| 4:27 | Russ Roberts: One of the many things I liked about your book is that you actually are aware of trade-offs. That's usually the economists' theme. Some might respond to this idea of quality, this quality-adjusted life year, as saying, 'Well, life is infinitely valuable. So what if it costs $250,000 to add a year of life? Or $500,000? Every day of life is--you shouldn't put a price tag on it.' How do you respond to that? Vinay Prasad: And, I guess, you know, you and I will think similarly on that. That sounds lovely. I wish we lived in a world where we did not have to make those kinds of choices where we could simply have any treatment that offered any incremental benefit, no matter how small and at no matter what price. But, you know, as you and I know, the reality is that there is only so much money that we can spend on human endeavors. There's only so much we can spend on health. And what we're doing in medicine is not paying for these low value practices: We are actively not paying for higher value practices. So, there are things we could be doing that would deliver far more per dollar spent, whether it's blood pressure reduction, taking care of people who have maybe cardiovascular disease, things in primary prevention--a number of different strategies that would be better use per dollar spent. And we don't do that. So, the net effect is, in the United States, if you have certain cancer diagnoses, you may be able to get all of the new drugs, but our life expectancy is falling and far less than other industrialized nations. So, I think, it is a trade off, and the choices we make, inevitably take money away from other sectors of healthcare. And we see that. We don't do a good job in many fronts. |
| 6:12 | Russ Roberts: So, we're going to talk about this for a minute, but before we do, I want to clarify one technical point, because as a person outside of medicine and until this last year or so when I became increasingly interested in these questions--and listeners are probably aware that, know that we're going to do even probably a few more. When I hear a number like 2.1 or 2.5 months months of extra life life, it's such a shockingly small number. And then, combined with the fact that, as our episode with Azra Raza pointed out, there's a bunch of people who don't get any benefit. At least half of them, for example, are getting it less than 2.5 months. And they're going to endure some of the most horrific side effects that these drugs sometimes produce. Is that actual number--and of course, it's sometimes overstated because it's, as you point out and others have pointed out, it's often tested on, that number comes from a population that isn't really similar to the actual population that's going to take the drug. So, does that literally mean that half the people live an extra time that is less than 2.5, say, and that half live more? And of the half that live more, how many of them live 3.5 or 4.5? Do any of them live 10 years longer? I mean, what does the right-hand tail look like? What's the 80th percentile? A question you raise toward the end your book for cancer patients when they're talking about treatment: We often focus on that median number. You might be in the right hand tail if you're lucky. You'd want to know how big that is, too. Vinay Prasad: Yeah, that's a great point. So, I guess I would say that--a couple of things. I guess I'd say one is, strictly speaking, it is impossible to know the counterfactual for any individual. So, if you know somebody lived six months while taking the drug, and you want to say that, let's assume the hazard ratio is constant in this study--it's like 0.8 or something. And so-- Russ Roberts: What does that mean? Explain that? Vinay Prasad: Yeah, so hazard ratio is like the instantaneous risk of an adverse event at any point in time. It's sort of a metric of the relative benefit a drug provides. And, you know, some statisticians believe that, you know, 2.1 months, that's the absolute benefit. That's because the drug is providing an 80% benefit and the average survival is, you know, something like six, seven, eight months and that's how you get to that 2.1 month figure. But if you were going to live only two, three months, that 80%, or that 80% survival value, that 20% benefit, that might be even smaller--on the order of a couple of weeks. Alternatively, if you were going to be living a year, because you had a natural version of sort of slow growing biology, that 20% is even a little bit bigger, a few more months. So, that's one way to view it. But, I think that's a naive way. Nobody knows the true counterfactual. And by that, I mean that there may be some people who are destined to live two months without treatment, who might live 10 months; but if you were to assume that, and you know the median is about two months, you have to assume there are people who are going to live 15 months who end up living three months, and it shortens life in some people. So, I think you can take all these kind of different ways in which to think about it. One way to think about it is your idea is sort of an idea that I endorse, which is: What is the 20th and 80th percentile without treatment and with treatment? So, that's the kind of way I often present it to patients. I say something like, 'We have this new drug. We know that people on average experience something between the 20th and 80th percentile. So, they live between perhaps two months and 11 months without treatment, but with treatment that might be four months and 17 months.' Something like that. I find that that can kind of be a helpful way for individuals to make these choices. But, coming back to your question, your question is really about tails. Is there a subset of people who demonstrate really profound benefit from these drugs? And I think that might be true for certain classes of medication, like immunotherapy, which is maybe a hot topic that you'll probably talk to some of your other guests about. That might be true, that there's a subset of people who demonstrate a longer-than-average benefit. But for many of these drugs, it really is kind of an incremental benefit, no matter where you are on that continuum of what your cancer experience might be like. That there doesn't appear to be long lasting, durable remissions. And so, there might not be that sort of tail. And then the last point I'll make is that, again, we're talking really about idealized clinical trials, and everything we know about idealized clinical trials is that the people on these clinical studies, they live longer than people in the real world. You know, I give some examples where people in the real world taking the active drug live half as long as people in the control arm, taking the sugar pill. The people in the real world who are older, have co-morbidities, who are like you and I, people who are not Olympians who are sort of participating in these clinical trials, we, they tend to have shorter survival. And what that means is, that almost surely the benefit of these agents is even smaller, sort of an absolute sense. Russ Roberts: And by comorbidity, you mean other things wrong with them besides the thing the drug's supposed to treat. We talked with Jacob Stegenga about that, this problem: that, it's in the interest of the drug pharmaceutical companies, obviously, to find patients that are healthy and look good; and they tend to be younger, less chronically troubled by other things than the actual people who are going to end up taking the drug. |
| 11:27 | Russ Roberts: But the bottom line is this--and this is the hard part to face--is that and so I want to start with this fundamental question, now that we got this other background information out of the way: Who would be crazy enough to spend, let's say $100,000--and that's at the low end for many of these treatments. Some of them are much more expensive--$100,000 for an extra two months of life, let's forget about the median--two months of life that are not going to be so great, by the way. You're going to have side effects, almost always from these treatments. Why would anyone pay $100,000 for that year of treatment, or $10- or $12,000 a month for that treatment, when it is so relatively ineffective and brings a lot of pain? And of course, the simple answer is: Because you're not paying for it. Vinay Prasad: Exactly right. I would say that if you were actually paying for it, very, very few people would choose to do that with their own money. But, we live in a system that has been constructed in a way that you are not paying for it. That society is paying for it. You are indirectly paying for it through your lost wages for a lifetime of employment where you've been paying into high premiums. So, you are-- Russ Roberts: Or your taxes. Vinay Prasad: Your taxes, yeah. You've been indirectly paying for it for your life. And by choosing not to take it, you know, you're helping a diffuse group of people and their future premiums and their future taxes and it doesn't really affect you in the pocketbook. And so, I think you're absolutely right: that if people had to pay that money, the system would be very different. That companies might not pursue these drugs, certainly if they were going to cost so much to pursue them. Russ Roberts: And you say, "The difficult conclusion is that cancer drugs simply are priced at what the market will bear." And then you go on to say that "the market," you put in quotes, "is a broken, convoluted system that exerts nearly no downward force on cancer drug prices." I think a lot of observers to that system draw the wrong conclusion, which is, 'Well, of course, people pay a lot for drugs, because who wouldn't pay anything to save their own life?' The answer is: Well, people all the time forego money to save their own life because they want to, say, leave money for the children or spend it on other things, or care about something else or knows that it might not help or that the risk of other--isn't perfect--the probabilities are uncertain. So, what's really going on here--I just want to make this clear, because it's really important. It's not a market. And it's not a market, not just because these are often monopolists through the patent system. That would be fine. Vinay Prasad: Yes. Yes. Russ Roberts: It's a great thing. It would be fine to give companies who discover wonderful things, the chance to charge, "what the market will bear." But it's really what the political system will bear in many cases or a convoluted insurance-company, pharmacy-benefit manager, subsidized through the tax system, employee-benefit system--which is also messed up. But the point is--here's the part that I've, I've asked this question before. I just want to make this clear, because it's so shocking. It's really whatever they think they can get away with. And we sometimes think that's what the market will bear. But there's sort of a social contract or norm or weird thing that, 'Well, since we're curing cancer, we can charge a lot and we'll set a price that Medicare will kind of swallow.' And, of course, Medicare is not allowed by law to negotiate that price. So, it just gets paid. Vinay Prasad: Yeah. I think you're hitting the nail on the head. I mean, just to kind of push on the fact that this is not a market. It's not a market that you're familiar with; it's not a market that I'm familiar with. To push on that, there are a couple of perverse policies in cancer medicine. One, Medicare has to pay for any approved cancer drug by the U.S. Food and Drug Administration. They must pay for it for the approved indication. They cannot negotiate the price. So, nothing stops a company from coming tomorrow with an approved drug and saying it's going to be $2 million a pill. I think literally the only thing that stops them from doing that is the fact that they're going to be on "60 Minutes" if they do that. Russ Roberts: Yeah, it's just social pressure, and once that price is established, let's say they start off at $100,000 a year for the drug, raising it 10, 20, 15%, the second year, 4%, leaving it the same--there's no rhyme or reason for that other than what they think they can do without getting put on "60 Minutes." Vinay Prasad: Right. The frog doesn't jump out of the pot when you heat it slow, heat the water slowly; and that's what they're doing. And in addition to this problem that every drug approved has to be covered by Medicare, there's another bit of policy that people may not know, which is that many years ago, we allowed groups of experts, who often are in financially conflicted relationships with drug companies to create registries of off-label drug use. And we have wedded Medicare to pay for off-label drug use as recommended by experts at certain levels of evidence. And Medicare here, too, cannot negotiate price; cannot say no. They must pay for these off-label uses of drugs. And that's yet another way in which the market is not a real market--where the payer cannot say no, even if the evidence is inconclusive, flimsy, uncertain, unsound, broken, flawed, as I detail throughout the book. Russ Roberts: Explain what you mean by the off-label use, why that's a problem. Vinay Prasad: Yeah. So, I guess I want to say, you know, I just want to point out that doctors do a lot of thing off-label. It's not necessarily a problem to prescribe off-label. So, what is on-label, off-label? On label means you're prescribing the drug as the FDA [Food and Drug Administration] said it ought to be used. So, the drug is approved for a certain cancer with certain genetic mutation, maybe in a certain age group: You're prescribing it in that range. But inevitably, drugs are prescribed, perhaps for people who might not fit that label, maybe in a disease that wasn't studied. And this is the so called off-label use. Russ Roberts: There's some good things about that, as you point out, because it means that, something wonderful that maybe we didn't know about works, and let's do it. Vinay Prasad: Yeah. In fact, I would say that this idea that you need experts to recommend off-label use, that's also well-intentioned. Throughout most of cancer history, some of these drugs were older, cheaper, cytotoxic drugs, off patent. There may not be a sponsoring company making a lot of money from these drugs, and there might not be an incentive to go through formal regulatory approval, which is a very costly process. So, this system that was well-intentioned, that works for older, cheaper drugs, kind of has been taken over by newer $100,000-, $200,000-drugs that are extremely lucrative to the manufacturer. And, what it has allowed is companies to get sizable market share from expert panels recommending the drug for off-label use. I guess I want to say one more thing. I like to say there's off-label use, and then there's off-label use. There's off-label use that has a clinical study that kind of supports it, and then there's off-label use that really has nothing--anecdote, maybe a faith that it's going to work and nothing more than that. In medicine, we often drift to that latter, off-label use--this kind of really kind of on a wing and a prayer use. And even some of that is codified in the guidelines that mandates Medicare to pay. And if you put all that together, that's the recipe for a non-functioning market. |
| 18:45 | Russ Roberts: So, I want to make a subtle point about this 'Medicare can't negotiate prices.' I'm really glad on one level that Medicare can't negotiate price. I don't want the U.S. government deciding what the profitability of drug companies should be. At least in the abstract. Unfortunately, we don't live in the abstract. We live in a world where almost every other country--certainly all the rich countries--have negotiated prices with the drug companies. So, what it means is, is that the United States taxpayer is--mostly; there are some other contributors, but it's mostly the U.S. taxpayer--is funding innovation and drugs for people around the world. A lot of people, when I say that if we got rid of the profitability of U.S. drug companies, people say, 'Well, but there's a lot of foreign companies.' Yeah. They're profitable because they sell in our market that is relatively unconstrained, where they can charge our prices. I think in your book you pointed out, I've seen a similar number that about half--half--of the world's pharmaceutical revenue is in the United States. Now, I don't know about profit--profits might be even higher, but I'm not sure. It could be lower, too, I don't know. But, we're an enormous driver of innovation. That's glorious, that there's all this innovation. Some of it isn't worth it, a lot of it. That's the tragedy. And it shouldn't be happening. The only reason it's happening is because of this lack of feedback loops where people with skin in the game would normally say, 'Well, that's not worth it.' Instead, pharmaceutical companies--and God bless them, they're trying hard to solve cancer and there's wonderful people in them, as I like to point out--but the current system allows them to put their hand in my pocket as a taxpayer to pay for the drugs at a price they decide, and only they decide, other than through this social pressure. I know this sounds so bizarrely implausible, that they just get to set the price at whatever they want--again, not what the market will bear: What the taxpayer will bear, unaware that this is happening. It's just--I want to get that out of my system. Vinay Prasad: But, I want to ask you a question of this topic that I wonder how you think about. There's an analysis in the last couple years, I think by Peter Bach, who has a Health Affairs blog; and they looked at the difference the United States pays on drugs than the rest of the world. And there's a difference. We're paying more than, I think, the rest of the world put together. And the amount we pay more than the rest of the world is enough to fit the entire global R&D budget, and then some. So, then, their argument is that the premium we're paying in the United States, it's more than enough to sustain all of the innovation globally. That's the argument. What do you think about that as an economist? Russ Roberts: Well, I think it's an interesting question. There is, always--there are two kinds of errors you can make. You can produce too much of something and too little. We have made the mistake in the United States of, I think, providing too much--seems impossible, but I think we provide too much innovation, both in pharmaceuticals and medical devices. Because, as you point out, and as you pointed out in your book, and as you pointed out a minute ago, once something is considered the standard of care and is efficacious, according to the FDA, Medicare is going to pay for it. And, so, that's a mistake. That's just, to my mind, a fundamental policy error. But if you're not careful, you'll go too far in the other direction, also. And so, I'm not saying, I would never say, 'Well, let's leave the current system alone,' because I think it's wicked actually, and not just wasteful. I think there are other parts of it, we may or may not get into, that are really horrible. It'd be one thing if it said, 'Well, we have great medical care in the United States, we spend a little too much.' My bottom line is we have great medical care in the United States for a good chunk of the population. For a small but non-significant part of the population, they have trouble getting access to medical care at a reasonable price because the subsidies and system we've talked about, it pushed up the price for them and made it almost impossible. And then there's a bunch of other people who are getting stuff that really doesn't work very well and actually has negative effects, but since they're not paying for it and don't have incentive always to look for the full effects, and that information is rare, because we don't see ourselves as consumers and we trust our doctors--we're going to get to that later about trust; we make some mistakes in that direction. But I think we have to be careful in saying, 'Oh, well, they'll still produce a lot of great drugs, if they make half as much, or a third as much.' I don't know. I would like to see the profitability of the drug industry emerge once the right incentives are set, rather than trying to steer it from the top down, which is what we do now. And when we do that--of course, the pharmaceutical industry has a lot of say in it through the political process, which is a separate problem. Maybe we'll get to that later. Vinay Prasad: I guess, on this topic, I'll just say one thing, which is that: My criticism here is, and I hope I try to persuade the reader through the book, is that it's my view that some of these drugs, it's not just that they're marginal. I actually think that they're ineffective. That they really don't work. I think that they're marginal only in the contrived way you've studied them against a control arm that's not the standard of care, with a design that's flawed in many respects. They're marginal in that setting, in a population that's not typical. But they're actually probably ineffective in the real setting. And what I would like to see, kind of along the lines of what you'd like to see, is I'd like to see the incentives incentivize people to make drugs that help people with cancer in the United States; and that's what I think the goal is. Russ Roberts: Woohoo! Yeah, that'd be good. It's weird that that's not the goal now, and having said that, I like to always mention I have friends in the pharmaceutical industry. They're desperately eager to cure cancer. Vinay Prasad: Yeah, and they're good people who work there, I think. Absolutely. Russ Roberts: So, there's nothing, I think one of the problems that free marketers like myself sometimes make is that they think that the enemy of their enemy is their friend. So, 'If the interventionists hate the pharmaceutical companies, therefore, as a free marketer, I must support them.' That's a mistake. A second mistake is the other mistake, which is to say, 'Yeah, they're just totally wicked, they're horrible. They're evil.' They're not. And certainly the players in them are not; and they've done glorious things to improve human health. Just, not as many as we might have liked for the amount of money that got spent. Vinay Prasad: Exactly. I couldn't put it better. And I don't blame them at all. I really do blame--it's though it's the structure that has allowed the incentives to incentivize ineffective or marginal drugs at tremendous revenue. That's the failure. It's the structure of the system. |
| 25:19 | Russ Roberts: Yeah. So, let's talk about some of that, and you have some fantastic examples in the book. Unfortunately, I didn't--I learned too much, as I like to say sometimes: There are things in this, even increasingly cynical, me, didn't know. We'll get to some of those. But, what are some of the conflicts of interest that decision makers in this space are confronted with and often ignore, and that you want to bring front and center? Vinay Prasad: You're probably referring to the couple of chapters on financial conflict of interest. And I guess--I discussed that as kind of a broader section of the book where I talk about societal forces that distort the narrative on cancer. I think one of the most problematic parts of cancer and the narrative around cancer drug policy is that it appears to me that many of the players, many of the people who could speak on behalf of their constituents in an honest and independent way, many of these people are very conflicted with the pharmaceutical industry. And so, I go through, kind of, 'Where does the money flow?' And the money flows of course from the pharmaceutical industry to, "patient advocacy groups." I put them in quotes because many of them have policies and recommendations that appear to run counter to what patients would really want. One of which is that they're often quite quiet about the out-of-pocket expenses of these cancer drugs and the actual price of the drugs. And that's kind of the devil's bargain. If you are funded by Pfizer, you can talk a lot about a lot of things that matter to patients but you probably want to be quiet about the price of the new Pfizer drug. And you see that reticence in this group. I also talk about the "thought leaders," and I put them in quotes. And, I think it's kind of cliche to call them 'thought leaders.' But there's some truth to the idea that in cancer medicine, a disease that really is hundreds of diseases, there's no one doctor who can know everything about every condition. And in each condition, there are few people whose voice carries tremendous weight. And, those people are heavily, personally, and professionally conflicted with the pharmaceutical companies. And they often are uncritical: they're not very good at statistical methods and critical appraisal. And they often appear to be cheerleaders for pharmaceutical products with tremendous uncertainty, marginal benefits. And, you put this together--you know, experts who are uncritical, who are also being personally paid by the companies; the patient groups who may be reticent about certain topics, always asking for lower regulatory hurdles to get drugs approved; journals, who are also deriving high revenue from the reprints of some of these studies that may be flawed; professional organizations that have large and undisclosed revenue streams from the industry. And, the worry is that all of the players in this dialogue are really kind of speaking on behalf of the industry, or at least part of them are speaking on behalf of the industry. And there is very little pressure the other way, towards sort of serious reform. And I think that's the kind of argument I put forth in that section. Russ Roberts: I'd like you to, it would be great if you could summarize the nature of that conflict because I think it's subtle. It's not that they work for the industry. It's not that they get our check every month--although sometimes they do. But, it's a whole wide range of stuff. As a patient, I always find it interesting: There are two parts of being a patient; and thank God I haven't had cancer, but my father is on his fourth. So, I know a little bit about it personally. As a patient myself, and you notice that this is just the most trivial examples, you notice that your doctor is writing up your prescription with a pen that has the name of a drug on it, and you think 'Well, that's a little weird.' And then they've got all these free samples sometimes they give you and it's like, 'Oh, well, that's nice. Oh, that's wonderful. I don't have to pay anything. I just get to it try to start out.' So, those are trivial. Talk about the range of stuff that a doctor--and again, this is not your average, say, internist in your HMO [Health Maintenance Organization]: we're talking about the leading researchers in the field at the top research institutions. What are some of the benefits they get from pharmaceutical companies, financially and otherwise? Vinay Prasad: That's a great question. So, I guess just one point about the 'trivial' thing. I think you're right: I actually do think getting pens, the occasional dinner, I do think that those are trivial sums of money that are being passed out. Although, I will acknowledge that there's a large literature that shows that even those trivial sums of money have been linked to prescribing patterns, brand-name prescribing for certain classes of medications and things like that. So, although trivial, the industry that's funding that is smart, that they're getting a little bit of return on investment, I think. But, what concerns me far more is this latter category, which is: How much revenue and professional benefits are going to the key leaders in a field, who happen to also be cheerleading for lower standards of evidence and costly marginal products? Some of the ways in which we see that are: One, consulting payments. It is not unusual, and we document in the book, for people to be making $100-, $200-, $300,000 a year in personal financial payments from the industry. Maybe it's $50-, $60-, $70,000, maybe it's less, but there's some people on the higher end who are making, you know, double the average household income in personal reimbursement from either drug talks or continuing medical education talks or that sort of thing, or consulting for the companies. So, that's the personal pocketbook. There are also huge amounts of money that get passed from companies to universities to conduct research. And a lot of that is necessary. That's the only way you're going to get clinical trials to go. But, we cannot discount the fact that many of this funding has significant slush on the top--that the actual cost of doing the research may be much less than what the institution is collecting from the company. And that leftover money often can get recycled in institutions, so that leaders who run certain groups that are very prominent, they may build up war chests of tens of millions of dollars of funding, which can fuel their laboratories, lead to them hiring staff, have luxurious business class travel when they go places, have a slew of assistants. It makes life more comfortable for the academic. And then the final thing is: The companies have access to incredible statisticians and writers, and by working with the companies and collaborating so closely, researchers have many publications, some of which may be sort of gift-authored--not really authored by those researchers, but yet they're getting the credit. These are the kind of ways in which academics benefit greatly from working with drug companies, I think. Russ Roberts: Just got to have a little footnote, and we'll come back to this, but: this idea that they have people writing their papers for them, I didn't know about this. Explain what that is. That's nuts. Vinay Prasad: Well, people who listen to my podcast, Plenary Session, will know that this is something I love to beat on, which is: medical writers are ubiquitous; And in medicine, an author who is the first author of a New England Journal [JAMA] paper--which is sort of a career achievement, like publishing in one of the best journals in economics--that person may not have written much of the paper, that the paper is drafted by a corporate funded medical writer. Medical writers are people who I think are talented: they're able to really concisely and clearly explain what happened in trials. Some studies even show that they're more effective at communicating sort of core elements. But, I like to point out on my podcast, sentences that are written that are often ludicrous, that are spun, that are hype. And to me, those sentences can only be written by medical writers. So, I have sort of like a running list on my show. But, it's so different than--I did an undergraduate and one of the things I studied a lot was philosophy--it would be ludicrous for a philosophy professor to have a philosophy writer, or an economist to have somebody draft your articles for you. Yet in medicine, that's considered to be okay, which is something that troubles me just from an ethical standpoint. Russ Roberts: As you point out, and I think--economists point it out all the time, and I think it's important--that many of these people are not nefarious. They don't see themselves in a sinister light the way you and I see this behavior. They consider it that they're doing God's work. They're trying to help people--and of course, along the way, they take some money, but 'What's the big deal? This is what I believe in anyway?' And I think it's a huge problem in general in life, that you fool yourself into thinking you're an objective observer when you may not be. But as you argue, and as economists argue all the time, these have subtle influences on your behavior, whether you're aware of them or not. Vinay Prasad: I wholeheartedly agree. I believe that the academics who participate in these relationships, I believe that they don't see them as problematic, and I actually believe that they're good people. I know these people. They're good people. And throughout the whole cancer system, there are good people. The beauty of, I think, policy, incentives, and economics is that people follow incentives, even good people. We all are good to rationalize why we ought to be pursuing our own interests. We all believe we're ethical. And yet the net result may be something that is not in society's best interest and that really does require reform. And, I do think you get no traction by blaming individuals and asking them to resist incentives. That's not the way to do it. You need to change the incentive structure so that you're incentivized to do the right thing. Russ Roberts: Well said. Wery reminiscent of Milton Friedman's point that the goal of good politics, if we want good policies, we don't want to have a system that requires good people to do the right thing. We want to have a system that encourages bad people to do the right thing. Vinay Prasad: Well put. Russ Roberts: Because we're not going to reliably find good people all the time, and those incentives will work on them, even if they are good people. |
| 35:17 | Russ Roberts: There's a--in Jewish law, it says a judge doesn't take a bribe. And I think the idea of a judge taking a bribe, which I think you reference something like that your book, it's despicable. We wouldn't say, 'Oh, well, it's only a small amount. He probably won't use that in determining his verdict,' if one of the parties to the case is bribing the judge. No. It's unacceptable. It's immoral. It is unethical. So, I think there's an incredible problem that's happened in all kinds of economic life, and I don't want to pretend that economists don't have a similar problem. I work for a think tank, which has an agenda. I have lots of liberty in that think tank. But I am, I assume, under some influence from that. I'm not a freelancer; It makes some kind of difference and you should take everything I say, of course, with a grain of salt. And yet doctors, we'd like to think, 'Well, they're pure humanitarians.' I think it reminds me--I recently interviewed, I think it's probably going to air before this, our conversation, Vinay, so, I think it's probably already out there. I interviewed Yuval Levin about his book, A Time to Build, and he talks about the loss of trust in institutions. And one way to think about that loss of trust is that people in positions of what used to be called authority, in the professions and the media and universities, politics--they had a role to play. And that role was affected by what the norms of the institutions they were embedded in suggested. And in the case of medicine, the idea that we trust doctors, because we assume they're only looking out for our best interest; and when they're not, it's so destructive. And at one point in your book, you talk about how important it is to find a doctor you can trust. That's a tough question to figure out a doctor you can trust. I've used this example before. My mom had a heart procedure done and in advance, I think it before she was having a done, she's in her cardiologist's office and she's, 'Oh, I can't talk for a little bit. I've got to get my EKG done.' I said, 'Mom, didn't you have that done two days ago?' 'Yeah, yeah; but they always do it when I come in.' It's like, 'Yeah, well, they would.' Vinay Prasad: Yeah; they do it before anything-- Russ Roberts: You give examples in the book--which shocked me--of the ability of doctors to get reimbursed for procedures that now are in their office rather than outside their office, under Medicare. I googled around and found frightening, depressing stories. This is not fraud. These are people who are doing the things they claim. But since, 'No, the patient--it's probably better to have the procedure than not; better do the diagnostic tool than not.' People are doing way too many of these; and they have a financial stake in it, and the patient never thinks about that. So, trust is constantly being eroded. What are your thoughts on that in terms of this conflict of interest in the role of eroding trust? Vinay Prasad: Yeah. So I guess I would say, I mean, I share your feelings. I view it as somewhat of a tragedy because I am a physician, and I believe it is a noble profession; and it's meant something for thousands of years. And survey after survey consistently shows physicians are still, although slightly diminished than prior surveys, we're still on the top of trusted professions. And I want to keep us there. But, I think these kinds of activities, incentives, relationships, behaviors that are well documented, they show us: One, that physicians are still human beings like everyone else, which I think comes as a surprise to some people but, you know, physicians are human beings. They follow incentives and they rationalize why they're doing what they're doing. Two, I think it really is a call to arms for us to consider reforming some of these things. People sometimes ask me, I do a lot of talk on conflict of interest, and it's two chapters of this book. People ask me, like, 'Would you ban it? Would you come up with these harsh punishments for conflict of interest?' I actually wouldn't ban it. I would incentivize people not to have it. I think, in 2020, the reality is, you will have a more successful career if you engage in these financial ties to pharmaceutical industry. I think we need to create a system where you can have a more successful career if you don't. And that might mean that there's some opportunities in academic medicine that exist only for non-conflicted people. Such as one would be to serve on one of these guidelines that decides recommended off-label use. Perhaps that should be restricted to non-conflicted physicians. Perhaps the writing of editorials and review articles should be restricted in that way. And these would both be sort of badges or coins that are only given to people who avoid conflict; and I think we have to incentivize people to avoid it as well. That's the real solution. Russ Roberts: And of course, the problem with that is that, as I talked with Yuval Levin about, the institutions themselves are made up of human beings. The people who lead those institutions tend to suffer from groupthink. They benefit from the system themselves. The idea of imposing those kind of costs to incentivize better behavior is really hard for those people because they're kind of part of the system too. Vinay Prasad: Yeah. And in many cases, they are the ones who are benefiting the most from the current lax, conflict-of-interest, policies. One thing you said stuck in my mind, which was the courtroom analogy. And, you know, in the book, in a couple of the chapters, I talk about something that is a lot like a courtroom, which is the Drug Advisory Committee meeting of the U.S. Food and Drug Administration. They don't have it all the time. But, when a cancer drug is coming to market where the risk-benefit profile is truly uncertain--yeah, it has some potential upside, but yeah, there's a lot of concerning downside--we have a Drug Advisory Committee meeting. We have expert academic oncologists who serve as jurors or panelists. We have the U.S. Food and Drug Administration who gives an opening statement of why the drug should be considered but also have these serious caveats. We have the industry who comes and gives, you know, the defense: they say why the drug should be approved. The industry often brings a witness, which is an expert who's typically heavily conflicted as we show in our Mayo Clinic proceedings paper. And then there's a public audience that gets to ask questions. And it's a lot like a courtroom. Both sides have their say. There are people who vote. But the difference between a courtroom is the money. And in the book I trace that every single person, every single entity that is speaking, has heavy financial ties to the pharmaceutical industry. Even the reviewers at the FDA--which is the one thing I'll mention now--although by law they're prohibited from current financial ties to the industry, which is reasonable and good, 60% of them go on to work for the industry when they leave the industry. And that's the so called revolving door politics that has been criticized in many domains. But, I think that that is concerning because if I knew there's a 60% chance that I'm going to work at the University of Pittsburgh, I'm going to take it easy on the University of Pittsburgh when I see their actions. And I think you might see that a little bit in drug regulation. |
| 42:21 | Russ Roberts: Let's talk about the FDA [Food and Drug Administration]. It's interesting for me, as a older economist, because I've seen a lot of ups and downs in the FDA's reputation and how economists look at it. In the early days of this conversation, Sam Peltzman wrote a provocative piece that suggested the FDA was killing a lot of people, because they took a long time to approve drugs that could save lives. It's a very tedious bureaucratic process. It's very time consuming. And in response to that, the FDA and other political pressures, the FDA has things like accelerated approval, and there are people allowed to take drugs as part of trials if they're near death. So, there's been some improvement in that. The irony is, is I think we've, and it's hard for me to say this as a former TA [Teaching Assistant] for Sam Peltzman, but I feel we've gone in the other direction as well. The FDA approves almost everything, these things that have very marginal benefits. And I think their response would be, 'Well, it's not our job. Our job isn't to decide whether things are worthwhile. Our job is just to figure out whether it's works at all. Whether it's harmful. How harmful. It just has to be safe and efficacious.' The problem with that, of course, is that they're embedded in a system where their stamp of approval means that a pharmaceutical company gets to put their hand into my pocket as a taxpayer. Vinay Prasad: Exactly. Right. Russ Roberts: So, now, so one of the challenge is then again--it's an interlocking system. You don't really want the FDA to go rogue and decide whatever they want. But, what might the FDA do differently? How might it be restructured? You have a number of suggestions in the book that would at least, given the current system, make them a better participant. Vinay Prasad: That's a great question. I guess I'd say a few things. One, just give your listeners some of the landscape. If you look at 100 consecutive drugs that are approved by the Oncology Drug Products Division, you would see this kind of breakdown: One third of them, we know they improve overall survival, or health-related quality of life, or a patient-reported outcome--something that's intrinsically meaningful to people, for about a third. One third of them, we know they delay something called progression or disease recurrence, which sounds as if it's clinically meaningful, but for a lot of reasons I get into in the book is probably just the illusion of being clinically meaningful. It has to do with sort of arbitrary radiographic cut-offs that don't make a whole lot of sense, that have a convoluted history, and that don't really tell us how well people feel or function. So, that's a third: they delay progression-free survival [inaudible 00:44:59], something like that. And a third of them are approved on the basis of studies that have no control arm. It's just, 'We gave the drug to 40 people, and we found that of the 40 people we gave it to, 20% of them had their tumors shrink 30% or more.' And that's the so called response rate or measures of tumor shrinkage. Response rate and progression-free survival, they have some correlation with overall survival, which I detail in the book, but it isn't a perfect correlation. And so we know that by approving drugs based on these endpoints, it is likely that many of them simply don't improve survival and don't improve quality of life. So, that's one thing. That might even-- Russ Roberts: So, you suggest in the book--correct me if I'm wrong--that the FDA is free to not approve those because that trial is a lousy trial. Vinay Prasad: Yes, they're free. Yeah. Russ Roberts: That's a trial where the horse race was between a drug that a pharmaceutical company wants to sell and a control group that is not relevant because it's not best practice, say: It's not what would be done as the alternative. And that to me seems like the FDA should just not approve those. Why are they approving them? Vinay Prasad: Yeah. Well put. I guess I'd say that in the trials that where they do improve survival--and progression-free survival--you have the problems of: Is it tested against the right control arm? Did the trial use appropriate standard-of-care after the trial ended as you would get in the United States--and we see many of these trials are run globally in places where standard of care is not reflective of the U.S. standard of care. And such trials often informe neither nation. They don't answer whether or not the drug works in the United States, and they don't help the other nation because the drug isn't priced to be purchased in the other nation once it comes to market. But anyway, that's true for the two thirds of drugs that have a control arm. But one third has no control arm. And sometimes you don't know whether or not it's better or worse than what we're already doing. The FDA is free to say no; but, in fact, what we have seen over the last 20, 30 years is a consistent drift towards saying yes--and we can talk about why that might be. The next thing I want to talk about-- Russ Roberts: It's because 'they care about people.' Vinay Prasad: That's the stated line. I certainly think that--well, we can come back to that. The next thing I want to say is the speed. I guess I agree with Sam Peltzman that you do not want a bureaucracy or regulator delaying needlessly drug approval. So, I am a proponent for speed. But, there are two ways you can have speed. One is you can streamline how trials are run, you can accrue more patients per unit time. Simply by including more people on clinical trials, you'll get faster results because it's proportionate to event rate over time. The other thing you can do for speed is to lower the bar the evidence hurdle that has to be cleared for drug approval. What we have seen is: Speed, people use that primarily to mean lowering the evidence hurdle for drug approval. That isn't just speed, that's also uncertainty. It's not just speeding a drug to market, it's approving a drug that may or may not work. And that's been historically how people have interpreted this speed argument. And as I talk about in the book, we have actually done some empirical work, where we argue that lowering the evidence hurdle doesn't really improve speed as much as you might think. There are a number of reasons why that is that have to do with the technical aspect of the surrogate endpoint used. So, I guess I want to make that point. But your real question is: What can the FDA do to reform us? I guess, we have a number of-- Russ Roberts: [?] Vinay Prasad: Yeah, reform itself. Before I come to that, let me just say one more thing. I would say, it would even be okay if the FDA had uncertainty at market entry for many, many drugs. That would be okay if we resolve the uncertainty 10, 5, you know, 15 years on the market. At some period of time, if we resolve the uncertainty, you know, it was okay to have uncertainty upfront, if that meant a little bit extra speed. Russ Roberts: Meaning improve it now, and if we learn later, it doesn't work, take it off the market. Vinay Prasad: Take it off the market. Russ Roberts: They never, almost never do that. Vinay Prasad: Exactly. That's the point. Not only do they never, almost never take it off the market, they have a poor track record of making sure those studies get done--the post-market, confirmatory studies. Sometimes when they get done, they give the full approval when the drug shows it improves the same exact surrogate endpoint. So, 20% of the time, they convert and accelerate it to a full approval based on the same exact surrogate endpoint, which really doesn't eliminate uncertainty. It just doubles down on the uncertainty-- Russ Roberts: And, a surrogate endpoint, meaning some measure that we like to think might be correlated with what we really care about, like the size of the tumor, but it could be that it's not going to actually extend life. Vinay Prasad: Yeah; I guess I'd like to quote my co-author of the other book I wrote, Adam Cifu, who has been a guest of your podcast. And he likes to tell me that, 'A surrogate endpoint is an endpoint a patient didn't know was important until a doctor told him it was.' So, the classic is: You didn't really care about your LDL [low-density lipoprotein] level until the doctor told you it mattered a lot. You don't-- Russ Roberts: Cholesterol. Vinay Prasad: Yeah, your cholesterol level. You don't care about your blood sugar until the doctor said that that matters. And similarly, you don't care necessarily if your tumors have a uni-dimensional--sorry: You don't care necessarily if the sum of the product of the diameter of your tumor shrinks 30% or more, but that is in fact, the definition of response. So, that's why you're a responder or not a responder--a very arbitrary definition. Russ, let me take a minute to say this definition. In the book, I tell a story about where this definition of tumor response comes from. And, this is a true thing. There was a study done in the 1970s by Charles M oertel, where he invited 16 oncologists to a table. On the table, he put a bunch of marbles and he rolled out a sheet of foam rubber, and he asked them, 'Using your calipers, tell me which marbles are bigger and which marbles are smaller.' Measure all the marbles. And the cut-offs that we use to this date derive from the amount of change in a marble that can be reliably discriminated by two oncologists measuring through foam rubber with a caliper. That's where we get the cut offs. And they've just been codified and passed forward through historical accident. They are not cut-offs that have been selected because they are telling us people live longer or live better when you have this amount of tumor shrinkage or deferred tumor growth. Russ Roberts: Another example of this point, which is one you've written about eloquently on Twitter--and I recommend for those listeners who are on Twitter to follow Vinay. He's a great, great player on Twitter. He's funny, and insightful and wise. And, one of your themes on Twitter is the failure of screening--something we've talked about a lot on the program as well. But, what could be wrong with figuring out whether you've got a tumor or not? And the answer is, of course, is that 'Well, if it doesn't kill you, all it does is invite tests and interventions that are actually bad for you.' And so, sometimes, it's better not to know. It's better to make a different kind of mistake. And I think this point about surrogates is really important because often, tragically, it's all we have--to look at progress. But we really do have to care about what the ultimate effect is of that surrogate endpoint to see if it's a good surrogate or not. |
| 52:03 | Russ Roberts: Now, one of the suggestions you make in the book is that the FDA--excuse me, that the clinical trial system is not very healthy. It's not very good. And in particular, one of the problems is that it's done by the companies with an incentive to have certain outcomes, the pharmaceutical companies themselves. It's a very common example of how the political process does something that's really not a very good idea, but it has political incentives. So, we probably--we could--as taxpayers pay for those trials, but instead, we've "imposed" the costs--and I say "imposed" in quotes--we've pretended to impose the cost on the pharmaceutical companies. Of course, we may end up paying for that indirectly in the price of the drug. It's not free. But as a result of them doing the trials, they have a lot of control over things. And we're not talking about fraud here, although that does happen, but rarely. We're really talking about some of the shenanigans that you can play with, some of the factors we already talked about: the population to do the testing on; you can do multiple trials, which is basically a form of p-hacking--that is a way of hoping that a result comes out that might end up being random, but it doesn't look random because you've done a bunch of them. And, all that stuff is--just seems to me to be very poorly monitored by the current FDA system. And you suggest that we should get someone else to pay for it and be in charge of it. Vinay Prasad: Yeah, I guess I would say that one of the final arguments in the book is that we take for granted that the company that develops the product is the company that designs, conducts, analyzes, reports the registration trial--the trial that test fundamental efficacy. We take that for granted, because that's the way it's always been. But, it doesn't necessarily need to be that way; and I hope in the course of the book I give you enough reason to fear that anything short of that won't really reform the problem. And I guess what I'd say is that, if the difference between a -pvalue of 0.49 and 0.51 is $10 billion in expected earned revenue--that's the incentive difference for the company--that, even within all of the rules of clinical trials, you will do everything you can to win. Whether that's: pick something to go up against that is the absolute weakest drug; whether that's: run your trial in a setting where, when people progress, their disease gets worse on therapy. There may be in a setting where there is no further therapy, so, that might exaggerate any differences observed within the trial, but also make it not really reflective of what would have happened in United States; whether it's run, trial in populations that are younger and healthier than average patients; whether it's gaming drug dosing. So, one of the things I talk about in the Epilogue is trials where the starting dose was equivalent but dose reductions were skewed, so that the competitor drug was penalized at every dose reduction. I mean, there's so many ways in which you can kind of put a thumb on the scale of a clinical trial. And, companies will do so as long as it's not explicitly prohibited. At the same time, when companies seek drug approval, they go to the FDA with a user fee, which has been codified into law, of usually a couple million dollars, and the results of the clinical trial. And they say, ask the FDA to go through the results of this clinical trial--'Take our money'--and to decide whether or not the drug warrants approval. The solution that I proposed is that you go to the FDA with your drug, and you say, 'I'll give you access to X amount of units in my drug; I'll manufacture it for you. Here's all my preclinical data. I know my drug works well in triple-negative breast cancer, for instance. I know it has a lot of activity there. But, and, instead of $2 million, I'm going to give you $32 million dollars, or $52 million dollars. And, what I want you to do--or the FDA or a third party, impartial agency, is you design the control arm, the clinical trial, the population, the post-protocol therapy.' You design all that and you run the trial through the same perhaps contract research organization. And then what happens is the moment that DSMB [Data and Safety Monitoring Board] calls you and says this trial is successful, you approve the drug. You don't even need a review on the back end, because you've already signed off on the protocol. Russ Roberts: It's a form of pre-registration, the way we've talk about sometimes in academic work, that, rather than letting somebody run the thousand regressions, statistical analyses till they find one that happens to by chance be correlated with what they want to prove, instead, you have to say in advance what you're going to run, and if it doesn't come out, you're done. Vinay Prasad: Exactly right. And you get the FDA to pre-register the statistical plan that they will agree is a favorable risk-benefit profile. And then the moment that trial is halted, that's it: It's approved. It's around the market the next day. That to me, I think, would actually even save time. It has the potential to save a year of time that's typically lost in regulatory consideration. It also has the advantage of removing the conflicted party from the design and conduct of the clinical trial. A couple years ago John Ioannidis, who has been on your podcast, Adam Cifu, John, and I wrote a paper in JAMA where we said that, 'Asking the sponsor to design and conduct their own study is like asking the painter to judge the painting contest for which they've submitted a piece of artwork.' The temptation to game the system, put a thumb on the scale, is too hard to resist. And, I think the only definitive solution to, you know, what I discussed in the third part of the book, which is all the problems in cancer clinical trials, is to remove the conflicted players from the design and conduct. |
| 57:29 | Russ Roberts: So, there are obviously issues there. I like that idea. One of the issues there is trying to decide which drug should get to that trial stage. What's the mechanism by which that third party is going to do the game? Of course, now all of a sudden they might be careless because they don't have skin in the----it's not going to be a perfect system, but it might be an enormous improvement over the current one. I want to step back, and now look at a more set of philosophical questions. I want to start with an observation: Which is that I have interviewed a number of doctors on this program. You, your co-author, Adam. I'm not going to list them all. There are many. They're wonderful people. I have immense respect for them. I've mentioned on the program, I have a lot of friends, personal friends who are doctors. They're extraordinary people. They're unbelievably giving with their time for their friends and family. They don't say, 'Well, if you want that advice, go see me in the office. I'll bill you for that.' These are gloriously kind, compassionate people; and overwhelmingly, they're aware of these issues. I talk to them about them explicitly about these conflicts of interest, incentives to over-test. They are aware of them. I'm sure they don't deal with them perfectly, but they try. Obviously, they are not--the whole profession isn't like that. Without naming names, I'll link to this Wall Street Journal story that looked at, I think the test was for an ophthalmologist could check for the saltiness of tears of their patients. One doctor had managed to do either close to a million or over a million dollars' worth of those tests. Basically--I think that person just gave it every single person. I'm sure there was a story to tell--another doctor--to that patient--and other doctor--to that patient, 'This is better than--yeah, it's good to know'--was billing Medicare for those salt tests, probably weren't well correlated with anything that was of interest of the patient. Not fraudulent, actually did the tests. But managed to convince the patient and themselves that this was good medical practice when I'm pretty sure it wasn't. So, that's going on in the right-hand tail. This handful, and it's probably, let's say it's 10% of doctors who are accounting for an enormous portion of over testing and abuse of the fact that Medicare doesn't check every patient to see if it was really advisable. So, one way to improve this system--obviously, you and I agree that there's incentives that are at the heart of the problem--one way to improve the system is to try to change the cultural norms that doctors face, sort of the badge of approval for the doctor who doesn't take the trip to the nice place for the conference that's paid for by the pharmaceutical company. Very hard to brag about that. So, it would also require a lot of internal, philosophical, conscience-oriented, Adam Smith-impartial spectator stuff that would give people pride in that, and more than they have now. It's a pretty weak way to fix it, but it might be the best way and it might be the only way. The alternative, which always appeals to me--and I'm alone, there's about seven of us--the alternative is to say, any system that doesn't have skin in the game, which is the fundamental problem, this system, that the people who are paying the money aren't the people who are accepting the cost of benefits--the people who are making the decisions aren't the people paying the price, the people who, except for the fact that of course, yeah, sometimes you do, you pay a health price. But at least the financial price is distorted here, then until we get to a system with more skin in the game, all of this, it's just lofty talk. We're not going to make any progress. Do you have any thoughts on that? Can you come with me a little bit in that direction? Vinay Prasad: I guess, I mean, I might be a little bit more sympathetic to you, although I don't share your view. But I am sympathetic in the sense that what we have now pleases neither of us. Those of us who think that the right regulations could make things better, we're not happy because we see the regulations are distorted and encourage chasing marginal drugs or even ineffective drugs at lofty revenue, where there's nobody able to exert downward pressure. We're not happy. We think maybe a little few more regulations or different regulations might change that. And then I think people on the other side who think fewer regulations would help things, they also have a point. Because it's only through the arbitrary construction of the rules, as they are now constructed, that we are able to have this. And if we really did have a true, you know, free market where people had a lot more skin in the game, I just don't believe people would be willing to pay so much money for such marginal drugs, and drugs with so much uncertainty. I mean, as a doctor, I know, because I talk to patients. I--you know, we didn't talk about this, but how do I reconcile sort of a health policy book, which is about what we as society should do, with actually going in on Monday and actually doing this on an individual level? And the way to reconcile it is that, you know, people need to be empowered to help make decisions that fit their views and their lifestyle. And my goal is not that everyone makes the same decision about every drug that I have about that drug, but that they give all the information so I feel like I'm as honest as I can be. And if they're willing to accept uncertainty, so be it; if I'm not willing to accept uncertainty, so be it. And if someone else isn't, so that's what I try to do in the office. But, I guess back to your--I forgot how I read your philosophic question. |
| 1:03:15 | Russ Roberts: Well, I was asking you about whether--actually, I'm going to complicate the question. So, let's think about three different ways to make the world a better place. One way is, we take a lot of the implicit subsidies out of this market that gives people the opportunity to distort the feedback loops in their own favor. Vinay Prasad: Sure. Sure. That's what you'd like, right. Russ Roberts: That's what I want. I want to get skin in the game. The second is to say, 'Well, we're stuck with this complicated system. Let's reform the FDA. Let's try to encourage'--actually, my second thing is let's try to encourage norms or behavior that are more attractive despite the incentives. Right? Despite the incentives. Let's encourage people to fight against those norms--I mean, excuse me, to fight against those incentives. That's a little bit of a quixotic fantasy. But, the third way, I guess, the third way is change the policy and hope that we could lobby and all that. And, I'm in favor some of that, although, part of the problem is if you fix one piece of it and you don't fix the whole thing, it's not clear you can really make it better: you get unintended consequences, and so on. I'm going to add a fourth one, which I know everybody's lost track because they're too subtle, but I apologize. The fourth one is: As patients, you and I, when we are confronted with these choices, can ask our doctors things like, 'Is there a generic?' Too many times I think, as patients, we just do whatever the doctor tells us. And that's changing somewhat through the internet. It's changing through an awareness that doctors have this conflict of interest. When doctors tell me I need a test, I always ask, as you suggested we should do in the book, 'How is that going to change what we're going to do?' That's true for when I'm helping my parents who are in their late 80s navigate through the system. I've talked about my mom before: She's very uncomfortable asking a doctor for a second opinion. That's a problem in a world of conflict of interest. In a world of conflict of interest, get that second opinion. Ask what the drugs could actually achieve. Ask what the probability is if you do nothing. Don't always have an answer, of course; but, ask. Vinay Prasad: I agree with the way you've outlined things. I mean, I think that the only thing that I would strongly vote against is the status quo. I hope that in the course of the book, I hope I've made the case for the policies that I have tried to put forth as, like, why this would pull things in the right direction. But I think one could consider fleshing out a big theory of how skin in the game might improve things. Of course, the concerns there are: How do you actually do it in the setting of a really vulnerable population? So, that's the challenge for the other view. Russ Roberts: Yeah; no: obviously, and for people who can't afford it, it's not a trivial problem. Ultimately, you have to subsidize something, almost certainly because of the fact that there will be people who can't afford the out-of-pocket costs, either through philanthropy or the clinics that provide things at discounted prices. Even if the prices were much lower than they were currently, there's so many people who couldn't afford every form of medicine and intervention that we might think is good for their health. So, that is the challenge to my viewpoint. I just think it would be a much better system than we have now. We'd retain the incentives for innovation that's effective, get rid of the incentives for the innovations that's much less effective. |
| 1:06:36 | Russ Roberts: Let's talk a little bit--there's a fascinating example in the book, and it's a common conversation I have related to this topic of policy change: the United Kingdom versus the United States in their cancer drug policy. A lot of people tell me, 'Well, the United Kingdom and Europe and elsewhere, they live just as long as we do. In fact, they live longer and they spend less.' Well, that's true. They're also not as fat. They don't drive as many cars; they don't have as many guns. When you correct for that, they kind of live the same length of time. Maybe we even live a little longer. But, we still spend way too much. So, but a lot of people will say, 'Well, look at the United Kingdom. They're doing fine.' They have some issues, though. So, talk about the difference in how they treat cancer drugs. It's fascinating. Vinay Prasad: Yeah. I mean, I guess I'd say the predominant way they do things is that they prioritize things based on cost effectiveness. But as we talked about, based on the cost of these drugs and the benefits provided, and the uncertainty around those cost effectiveness estimates, there are a lot of things that just don't make the cut. One of the things I talked about in the book was how, about a decade ago, they created something called the Cancer Drug Fund, which was a fund that was meant to bypass the traditional cost effectiveness, which was a fund of money-- Russ Roberts: Because of political pressure. Which is fascinating. Because, despite the fact that some of these drugs probably are a bad idea, they gave in and said, 'Yeah, okay.' Vinay Prasad: Right. They said that we're going to keep a special pot of money on the side; and we're going to pay for drugs that don't meet traditional cost effectiveness. To which the Lancet, I think, appropriately said, the moment they created such a fund, it's already intellectually bankrupt. Which is what it is, because it flies in the face of the entire idea that with what money you have, you ought to maximize health outcomes, with what money you have. And it ran counter to that from the outset. So, I think the challenge they faced was, of course, they ran that fund. It ballooned out of control; they ended up dissolving it. And so, that was a challenge that was not sustainable: that you couldn't pay sort of an infinite amount of price in the U.K. system. Of course, they spend far less on health care per capita than we do in the United States. Now, the challenge I think they face is that there are many drugs that are new, sexy, in vogue, that don't get covered. And that creates, I think, personal stories that are quite compelling. A story of a patient who didn't get a drug which is quite compelling in the media. What you don't hear so often is maybe about all the people who are getting access to, sort of a more basic service that provides a better value. That's not as compelling a narrative. And so, I think that's something that the United Kingdom struggles with. But one thing I want to say is that it's so difficult to talk about, you know, 'How do other countries do it? Can we learn from other countries?' For the reason you alluded to in the beginning of the podcast. Which is, we're talking about half or more than half of cancer drug spending is our nation. What the United States does around cancer drugs, how we decide to approve drugs, how we pay for drugs--this prevents any other nation from straying too far from what we do. The EMA [European Medicines Agency] across the pond, they pretty much approve every drug we approve-- Russ Roberts: What is that? Vinay Prasad: The European Medicines Agency, the FDA equivalent in Europe. They pretty much follow lockstep with the FDA's philosophy--on everything. From surrogate endpoints, to which drugs, to drug approval, to clinical trials. The same trials , often state[?] drug both[?] requirements. They are looking to us for guidance. Japan. Canada. They all look to studies that are primarily done for the goal of getting market share in the United States. Similarly, because we have allowed the cost of cancer drugs to reach sort of astronomical heights, they are limited in terms of the starting point of negotiation: how much the drug price could go down. They're fundamentally limited. So, I think it's very difficult to look to other countries for lessons when the United States is such a big player in the game. |
| 1:10:13 | Russ Roberts: There's another philosophical issue that I think has to be confronted, and you talk about it very beautifully in the book, which is: Life is precious. And when I was talking Azra Raza about her book, The First Cell, one of the things in that book that just is heartbreaking and powerful, is the willingness of people with devastating diseases to endure terrible treatment, because it might work even when it's a long shot. And that's a human urge. We can't ignore that human side of this. And you're telling, talking about the United Kingdom, that's really what was driving this, and you give an example of the book of, 'But what about when it's me or my parent, who's got this terrible disease? There is this drug that's not on our approved list,' if you're in the United Kingdom, or in Canada or wherever. It's not on the formulary. And you're told: 'You can't have it. Because we don't think it's worth it.' That's really hard to do for a parent. It's hard to do for a political system. It's kind of a fundamental aspect of this, that, as we've gotten really good at this extraordinary, difficult task of understanding the human body, which still horribly far away of understanding it fully, but as we made progress, the idea that we wouldn't use an opportunity to try to live a little longer is really hard for people. Vinay Prasad: I think you're right. I think it's really hard for people. I also think that, I mean, I want to take a step back and just say that as an oncologist and having participated in and seen so many of these kinds of questions, and also having a little bit knowledge of the United Kingdom, I think when the United Kingdom is saying you can't have access to a drug, one scenario is that drug offers a very, very modest benefit at tremendous price and that's why you can have access to it. I think that's the one that's the hardest scenario. It hurts in the way you've described. I think what it under-discusses is the fact that, far more often, the scenario is: There's this new drug, and it's not that it has a marred modest benefit. Instead, it actually might have no benefit at all. Or the benefit is just so uncertain. And the toxicity is certain, and it's not pleasant. And it's horrible. And, they're denying you access to that drug. And in those cases, I think if people actually had perhaps a clearer understanding of what is known and not known about the drug, many people themselves might not want the drug. And I think that's in part what the book is trying to educate, which is like what do we know about this drug and what don't we know about this drug? So, I think one of the concerns is that, that the drug may not even work. It may not even be a marginal drug. It may be this sort of totally inert and useless drug. And then the other thing is, that's hard to describe to people, is that, there are worse things than death. And I think it's painful to have to say, but there are people who die from taking experimental drugs or drugs with uncertainty, that die worse than they would die if they didn't take the drug. And as doctors, we see that from time to time. And we know that they are worse ways to die than just dying of cancer. And, part of our policies is to protect people who are sick and vulnerable, from having to make those desperate choices. You know, we once lived in an era not that long ago in this country, where people would participate in barbaric things that offered no possibility of benefit. In fact, maybe the history of medicine is a history of such a thing. The grand 1,000 year arc of medicine is a history of things people tried, often barbaric, that had no value. Russ Roberts: Or harmful. Vinay Prasad: Or harmful. Or and worse than death. So, I think policy has to reconcile with that aspect of it: that human beings are, as Alexander Pope says, 'Hope springs eternal in the human breast. Man always is, but to be blessed.' You know, we're always optimistic, but at the same time, we know that biology is difficult and that you can tinker with biology and make things a lot worse. Russ Roberts: One of my hopes is that by talking about these issues, and I think sensitizing people who listen to the role of probability in these types of situations is helpful in changing the culture of medicine, as I think Eric Topol has talked about--people taking charge of their own medical information in this digital era. And I think that's really important. That we as consumers be aware that just because something's FDA approved, doesn't mean it's a good idea to put it in your body. And I think that's really hard for people. It requires a change in how we look at the medical profession and the medical system. But I think it's the right way to think about it. There's uncertainty. Some things are going to be harmful. Some things have costs that are not worth their benefits, even if they're "approved," even if they were recommended by my doctor--and especially when my doctor has a conflict of interest. And I think that part of it, unfortunately, being sensitive to that, is important. Vinay Prasad: I absolutely agree. You know, sometimes I think about--what I hope for this book to achieve. And I guess I hope that what I've done in the book is, I hope that when you walk away from this, that I've educated you about cancer: that you understand cancer and how we do drug development in a way you didn't understand it; that at a minimum, the book is educational, even if you don't agree with my policy prescription, and even if you have an alternative policy prescription--as Russ you might. But I hope that we agree on the problem, and we agree that that's a problem. The next thing I hope, more than anything, is that when you finish this book, you cannot tell me that this could have been an essay in The Atlantic. That, this is in fact a book that's book length that needs to be book length. And I think that's one of a key goal of the book, is that there are enough themes and lessons here that this could not have been condensed. And if anyone tells me after that this could have been a blog post, I will be very unhappy. So, those are my two goals with this book. Russ Roberts: I think it's pretty successful. I want to say--I like to let my guests have the last word. I'm going to read a quote from the book as the last word. You're free to add to it, but it's a beautiful summary of, I think, what a reader will learn from this book. And it's probably, it might be short enough to be in a tweet actually. So, you might want to use this a few times. Here's the quote: The magnetic pull of profit, regulatory capture and hype have resulted in policies that lead us astray from the best interests of people with cancer. Much of that pull is subtle, hard to see and easy to miss or to ignore.... End of quote. That's a beautiful summary to me of what you tried to do in the book and I think achieved and what I hope we achieved in this conversation. Vinay Prasad: Thanks for having me, Russ. I'm glad that you took the time to look at this and I'm glad we finally got the time to connect after all these missed opportunities over the years. Russ Roberts: My guest today has been Vinay Prasad. His book is Malignant. Vinay, thanks for being part of EconTalk. |
READER COMMENTS
Dr Golabki
Apr 20 2020 at 11:23am
First let me say, I haven’t finished episode, but early on there was a discussion that I think is a bit of a mistake that can be dangerous.
Vinay comments, “what I would like to see, kind of along the lines of what you’d like to see, is I’d like to see the incentives incentivize people to make drugs that help people with cancer in the United States; and that’s what I think the goal is.” And Russ agree’s with a “woohoo”, which has to be an EconTalk first.
So here’s the current incentive for curing (even one small subsegment) of cancer: you make 10s of billions of dollars and you become a famous hero. I think that’s plenty of incentive for any person or company.
It’s true that the incentives for incremental improvements in the current system are too high, but it doesn’t follow that you’d have more breakthroughs if you fixed that problem.
It’s like saying that the reason we didn’t have iPhones in 1900 was that there was so many distorting incentives for Bell Telephone Co. You’d be right that there were big distortions in the phone market in 1900, but that’s not really central to why there weren’t iPhones. There weren’t iPhone because making an iPhone was really really hard and required ~100 years of additional innovation to achieve.
The same, unfortunately, is true of cancer. This matters because Vinay’s framing can lead people to believe there’s a free lunch.
Ajit
Apr 20 2020 at 1:38pm
Terrific Episode. Can’t recommend it enough!
However, at every point in the conversation, I came back to the same depressing thought. The reason this system persists, at least to me, isn’t due to nefarious lobbyists and crony capitalism. It goes back to Bryan Caplan and his explanation for why bad economic policies get passed in the first place – by popular demand!
All of this hand waving and pearl clutching and talks tweaking the incentives will not solve this problem. And I am not naive enough to believe this kind of awful incentive structure is limited to oncology.
In any event, here is the reality – we either go to a full government run system with heavy rationing and much lower innovation, or we go to a mostly private system where there is inexorably some(maybe a lot) of discrimination of care based on price(which is politically toxic).
Right now, we have chosen neither option because both would effectively deny care to a segment of the population. It really is as simple as that. We’d rather pass the burden to the next generation until they end up bankrupt. Much like how moral hazard is the next generation’s burden to bear.
This goes right back to Jim Cogan’s book, “The high costs of good intentions”
[I think you meant John Cogan, not Jim Cogan. You might want to check out our EconTalk episode with him at https://www.econtalk.org/john-cogan-on-entitlements-and-the-high-cost-of-good-intentions/. –Econlib Ed.]
Alan Goldhammer
Apr 21 2020 at 12:57pm
There is a simple pricing solution to all of this. Back when i was still working at PhRMA we had a committee that looked at health outcomes and QALY. We could not talk about pricing as that would violate anti-trust laws. However, it occurred to me that a different pricing approach was needed for oncology drugs. Since patient response is all over the map, and Dr. Prasad has nicely framed this, pricing of this class of drugs should be based on performance.
There should be a relatively low entry level price, perhaps based on simple recovery of R&D costs which would then escalate for every month the patient lived past the median. The better the drug works, the better the profit stream for the pharma company. Insurers, including Medicare (which is made more complicated because oncology drugs may fall under Parts A, B or D depending on where they are administered) would like this as the up front costs would be lower, rising only if there is a successful medical outcome based on absolute patient experience. I’ll leave it to the economists to figure out the right formula.
There was a real life example of this in the UK some years ago. I cannot remember which company it was but their oncology drug received an unfavorable review from the National Institute for Health and Care Excellence meaning they were likely not to get reimbursed by the National Health Service. The company said they would give the drug for free providing reimbursement came when there was sufficient evidence from use. My proposal above is a similar approach.
Jerry
Apr 22 2020 at 10:00pm
Hi Alan,
The drug you are thinking of is bortezomib (Velcade) which is now generic in the US. It ultimately was used in the UK and the manufacturer essentially gave the NHS a money back guarantee. If the patient did not respond the drug was free. Since this was for myeloma, the question that then became heavily debated was of course define “respond.”
Doug Iliff, MD
Apr 22 2020 at 10:34pm
Toward the end of the interview, Russ and Vinay wander off into a general topic of incentives, norms, and skin-in-the-game. There is no more important topic in American healthcare, but in the present convoluted and conflicted non-system, solutions are partial and unsatisfying. I’ve experienced all of them.
Today I did a couple of office treadmills— the old $225 variety, not the $5000 nuclear versions. One was for otherwise unexplained fatigue; the other was for a bigeminal rhythm which resolved in the office yesterday. I was 99 44/100% sure that both would be normal, and they were. So why did I? One reason was to reassure the patients; the other was to keep them out of the emergency department after-hours, where the reassurance I provided might have cost a $10K overnight in the hospital.
In another system— Britain or Canada, maybe—the expensive ED visit would not have been an option, and my simple words of comfort might have sufficed. But that’s not the world of American healthcare. There is always a place offering more expensive comfort, so I do the best I can.
Saving system money in the process of treating patients with no skin in the game has been my passion for 45 years. I always prescribe generic drugs unless there is no option; I educate about the wonders of GoodRx; I sign for expensive samples so I can give them away to the Rescue Mission or patients who can’t afford them. Still, there are so many recommendations I make which do result in my personal financial benefit. “The heart is deceitful above all things, and desperately wicked,” Jeremiah says. “Who can know it?”
Meeting the Friedman test in America is hard. One man’s rationing is another man’s cost-effectiveness. The agent with skin in the game holds the rationing power. What are the alternatives?
For the most part, fee-for-service in all of its mutations makes the insurer the skin-holder. Everyone agrees that this has not worked very well. Insurance companies are too far removed from deliberations in my exam rooms, and susceptible to manipulation by the big players.
During the 1990s we brought health care inflation to a halt by giving the skin to doctors, in the form of full-capitation plans where I had to approve every expenditure. I got a monthly per patient fee, and a cut of the money I saved the system. That worked until malpractice attorneys found it an effective tool to accuse physicians of withholding needed care for personal aggrandizement.
I spent five years in the Army Medical Corps. That was very cost-effective, because the government held the skin. We did a good job with what we had, which was sometimes substandard, but we were free of legal second-guessing and complaints had to travel up the chain of command.
There is only one other option: give the skin to the patients in the form of health savings accounts. This isn’t perfect, because the patient is only holding the skin up to the level of their high deductible; however, especially when they see their HSA accumulate assets, it is the most effective way I have experienced to align incentives with good practice. I still think that if the country had adopted HDHP/HSAs 15 years ago, with refundable contributions for poor people, we wouldn’t be experiencing many of the travesties which now beset us.
Jerry
Apr 22 2020 at 10:40pm
Thanks for this episode that sheds a little light on issues surrounding oncology drugs in the US. I am an oncologist and am acutely aware of many of these issue. I would hope, though, that Russ’ listeners don’t get the impression that there have been limited advances in oncology care in general. I do not think that Dr. Prasad significantly highlighted the significant cures that have been achieved by drugs such as Herceptin, Yervoy, Keytruda, Opdivo, Rituxan, and many more – these are e essentially the drugs that Dr. Prasad classified in the top third.
What we really want as a society is more effective drugs and less marginal drugs. So, how do we keep the top one third and not the marginally effective drugs? The biggest change that I have seen over the last 20 years in this regard (really starting in the late 90’s) is that the FDA begin using less rigorous requirement for the definition of “efficacious.” Instead of needing to statistically prove survival other metric such as “progression free survival” and “response rate” were introduced by the FDA and of course companies looking to get their products approved used the lower hurdles and more drugs were approved. These changes at the FDA were a direct cause of political pressure from patient advocacy groups. This is again another manifestation of the “bootlegger and Baptist” problem. The politicians and pharma industry were completely aligned with the patient advocates to make cancer drugs easier to approve. We need to really think about how much value these end points really represent.
The other problem that has driven the lowering of the efficacy bar is that over the years essentially the “types” of cancer have increased dramatically as we have gained greater knowledge of the molecular basis of disease. Instead of lung cancer being two types (small cell/non-small cell) it is know many types (EGFR+, ALK+,ROS+, PD-L1 greater than 50%, less than 50%, etc..) based on the genetics of the tumor. Because the number of patients diagnosed with more specific diseases is much smaller because of this slicing and dicing it then becomes much harder to recruit enough patients on trials to power the studies effectively. When only 1500 patients a year are diagnosed with a disease then efficiently running a trial with sufficient power to prove overall survival is daunting! This issue of small populations drives a lot of the need for surrogate endpoints such as progression free survival.
The final issue is that value is in the eye of the beholder. Is two months improvement in survival worth it? Well maybe it is if that means you can go to your daughter’s wedding or attend a grandchild’s college graduation. That’s why these questions are so tough!
Comments are closed.