Russ Roberts

Scott Atlas on American Health Care

EconTalk Episode with Scott Atlas
Hosted by Russ Roberts
PRINT
Brady on the 2012 US Election... Ober on the Ancient Greek Econ...

Scott Atlas, Senior Fellow at Stanford University's Hoover Institution and author of In Excellent Health, talks with EconTalk host Russ Roberts about the U.S. health care system. Atlas argues that the U.S. health care system is top-notch relative to other countries and that data that show otherwise rely on including factors unrelated to health care or on spurious definitions. For example, life expectancy in the United States is unexceptional. When you take out suicides and fatal car accidents, factors that Atlas argues are unrelated to the health care system, the United States has the longest life expectancy in the world. A similar change occurs when measuring infant mortality--foreign data do not include as many at-risk births as in the United States and the measure of a birth is not comparable. In a number of other areas including cancer survival rates, access to hip replacement surgery and waiting times to see a physician, Atlas argues that the United States is also at or near the top. The discussion concludes with a discussion of access to health care for the poor and the failure of Medicaid.

Size: 28.4 MB
Right-click or Option-click, and select "Save Link/Target As MP3.

Readings and Links related to this podcast

Podcast Readings
HIDE READINGS
About this week's guest: About ideas and people mentioned in this podcast:

Highlights

Time
Podcast Highlights
HIDE HIGHLIGHTS
0:36Intro. [Recording date: July 24, 2012.] Russ: When you combine the public and private expenditures for health care in the United States, you get a pretty big number. That in itself tells you nothing about the effectiveness of U.S. health care. Might be a good thing, might be a bad thing. But you often hear that the United States spends more and actually gets less for its money than other systems. That U.S. health care outcomes are terrible relative to the rest of the world. So, ignoring the expenditure side, at least for the moment, is it true that U.S. health care outcomes are mediocre relative to the rest of the world? Guest: I think this is really one of the biggest problems about misinformation that the public has been hearing. The quality of U.S. health care compared to the rest of the world is actually far superior in almost all chronic diseases, almost all treatments, almost all cancers, screening, access to tests, access to new treatments. All medical outcomes, if you really look at the facts and the scientific or medical peer-reviewed literature, most of the reports don't do that. They take generally very coarse endpoints--life expectancy or infant mortality--things that have contributions from many things outside of health care itself; come up with a ranking, and then publish that as if that is the statement on the quality of care. So, the short answer is: Absolutely not. The quality of U.S. health care is far better than whatever everyone has been talking about; and the facts show it. Russ: Well, let's start with life expectancy. That would, on the surface, seem to be what you want to look at. What would you care about more than how long you live? And of course one answer would be the quality of life. But certainly life expectancy is an important measure of health care outcomes. When you take the United States and put it in world perspective, where does the United States rank as a measure of outcomes? Are we near the top? Middle? Guest: Well, if you go by the rankings that most people cite, which is the Year 2000 World Health Organization Report, I think the United States was something like 37th. Something like that. Russ: Which is disappointing. Guest: Which is disappointing and actually bad, even if you don't consider the expenditures. Because we are a very modern, advanced nation. There is no question about that. Russ: Correct. Guest: But then when you look at the statistic, this is the problem. Life expectancy is not really solely a measure of health care. Russ: No doubt. Guest: There are many, many factors that go into it. And the United States has a lot of--there are many significant differences between the U.S. population and the way we measure things than other countries. Even countries in Western Europe who are just as sophisticated and just as modern. So, the statistic itself is very misleading. Russ: What would be an example? You wouldn't think so. You are either alive or not. It depends how it's gathered--that would seem to be relatively straightforward in a developed country, at least. I understand that in an undeveloped country maybe they are not as careful with the records. In the United States, I think we are pretty good with birth dates and deaths. So, we have a pretty good idea of how long people actually live, who at least were born here in the United States. What else would matter besides that, if we are, say, comparing U.S. life expectancy to Western Europe? Guest: Well, one example that has been published by people at the U. of Iowa is the cause of death. Because, if you are talking about health care, you would think you'd want to talk about the things that are related to health care. And one ranking in this was done to the rankings for all the Organization for Economic Cooperation and Development (OECD) countries, all the most developed countries in the world, ranked the United States toward the bottom for life expectancy. Russ: Consistent with the numbers you said. Guest: Consistent with the numbers I said. Yet when they compensated for differences in suicide and immediate death from high speed motor accidents-- Russ: Things where the health care system is probably not going to able to save you. Guest: Yeah. The health care system has nothing to do with your surviving a gunshot wound to the head, basically. Russ: Or an accident at 60 miles per hour without a seat belt. Guest: Exactly. High speed accidents. And when they compensated for that by giving everyone the same number of suicides and instant deaths like that and then redoing the ranking, the United States elevated to Number One in life expectancy. Russ: Sounds like a cheap statistical trick. But it could be true, the right thing to do. Because, and correctly, suicides and high speed deaths should not indict your health care system. But of course there's some vagueness about how we measure high speed automobile death, I suppose. Some uncertainty about that. Guest: But when you really look at the numbers on suicides, there's an extraordinarily high number of difference in the United States versus these other countries where they are doing the rankings. Russ: Much higher. Guest: Much higher in the United States. Russ: As is automobile accidents per capita. You are saying, I assume. Guest: Yeah. I haven't looked at the exact number of automobile accidents. But another example of that would be just looking at what is the cause of death. And when you look at adults, children, and young adults, through I think the age of 40, something like 25%, 30% of deaths only, were due to an illness. Russ: Right. Guest: Okay, so health care quality is talking really about, generally speaking, about illnesses and diagnosing and treating. So, it's a very, very coarse statistic. Russ: Of course, the other factor, which you mention and talk about extensively in the book is lifestyle differences. So, even with health care outcomes that are related to the health care system, such as diagnosis and treatment of illness, lifestyles, genetics are the two biggest things that are going to affect your health care outcomes, your life expectancy, outside of the health care system itself. So, those are going to matter. Guest: Absolutely. Lifestyle is very different. As the rest of the world is learning with globalization. When you look at the one that probably has the most impact on illness and survival is obesity. And we've heard a lot about this. And when you look at the data, the United States has a far higher frequency of obesity, defined scientifically, as a proportion of body mass, has a far higher proportion of the population that is obese compared to other countries. And with that goes a significant decrease in life expectancy. For instance, it is estimated that from people who survive to age 40--just take people who are 40 and over--obesity decreases the life expectancy beyond age 40 of about 6 years. Compared to non-obese. So, when there is this striking difference, two times or triple, up to triple sometimes depending on the country, the frequency of obesity in the population of the United States compared to others, you are going to have a significant impact on the overall life expectancy.
8:18Russ: Now, of course, you mentioned in the raw data the United States is 37th in the world in life expectancy. The difference between 1st and 37th was 81.2 to 79.6. You might say it's very important. You are suggesting that these obesity numbers actually have a significant impact on both the ranking and that you'd care about it. Guest: Absolutely. You bring up another point here, which is the differences. In the ranking that people use generally as the kind of fundamental indictment of U.S. quality of care was the World Health Report from 2000. And when you really look at their data, you see that the differences between number 3 and number 30 were not even what we call statistically significant. When you look at their own data. And yet they still went on to rank it. Now, what do I mean by statistically significant? Statistically significant, for those who don't understand that, is a difference that cannot be accounted for by simply random variations in the data, that have nothing to do with the question being asked. So, when you say, you are ranked number 30 but your answer is essentially your data point is the same as the ranking of number 3, then why do you report it as ranked number 30? Three equals thirty. Russ: Because it sells the reports, obviously; it is exciting. But what you have in the chart in your book is, basically you have confidence intervals around these overall, weighted measures of health care; and they are quite large. The top of the confidence interval and the bottom encompass large ranges of the entire data set. Which is not encouraging. One thing you don't talk about, I don't think, in the book, which I've always been fascinated by is the trend in life expectancy over time. Which, to some extent, obviously imperfectly, within a country controls for genetics--outside of immigration; at least over short periods of time, gene pools in the United States are roughly constant. The differences are small and can be ignored. To the extent that lifestyle is similar at the beginning and end of a time period; obviously it can go through a time of exercise intensity or obsession with obesity or not worrying about obesity--you can get trends like that. And of course obesity has been increasing in the United States, as officially measured. So that works to shorten life expectancy. The obesity changes over time. The genes, probably the same. And yet life expectancy in the United States climbs steadily every single year. I haven't looked in the last few years, but every year, there is an annual report, news item: U.S. life expectancy increased again. Which, given the challenge of obesity which does suggest that it's getting worse, does suggest that the health care system is getting better. Correct? Guest: Absolutely. And to be fair, life expectancy has been increasing in many of these countries, if not all. Russ: For reasons other than health care. Guest: For reasons other than health care. But also benefiting from the medical advances that often originate in the United States. Which is a separate discussion.
11:42Russ: We'll talk about that in a little bit. But one of the problems with life expectancy data is that--I want to turn to our next topic, which you devote a lot of time to in the book, which is infant mortality. Infant mortality has a huge impact on life expectancy, because you could die--an infant that dies at six months enters into the average as a .5--half a year life. Pulling down the average. And what we often care about--it's really, life expectancy reflects a bunch of things. There's social phenomena, like suicide; but a high infant mortality rate leads to a low life expectancy. But once you survive infancy, it's been a very poor measure of your true expectation. So, we might often care more about your life expectancy conditional on reaching, like, say, age 40, as you mentioned before. But infant mortality is important in and of itself; you've got a statistical impact, and that's another measure where the United States often does very poorly in these international rankings, which is surprising. And you suggest that's very misleading. Why? Guest: Absolutely. And here you get into--there are many reasons, but one of the big ones that we kind of dispensed with regard to life expectancy, is the actual measurement. Because the measurements here are very critical. And this is very different from the United States compared to other countries, including those in Western Europe. Where, in the United States there are many differences. One important one is that we count every birth. And birth is defined as any sign of life. Actually, ironically, using the very strict criteria defined by the World Health Organization (WHO)--we count any heart rate, any respiration, no matter how premature the infant, no matter how small the infant, no matter how immediate the death is. Other countries, including many civilized countries, countries that we think of as similar to the United States, don't count births as live births if the baby does not survive for, say, 24 hours. If the baby does not measure a certain size or weight. If the baby does not survive for 48 hours, or even a week. And so when you throw away the deaths from the most fragile infants in the denominator of a fraction, your calculation is grossly distorted, compared to those who counted even the weakest babies who had on a small chance of survival like with do in the United States. Russ: Having been blessed with four children and seeing the unbelievable medical technology that's available to a newborn baby and a mother--dads don't get, we get a comfortable chair, that's the limit of technology if we're lucky; we don't deserve anything more than that. But, given that, it's hard to believe you'd rather have your baby somewhere that has a lower mortality rate, supposedly, in the data. But there are many countries that do, and you are suggesting that's misleading. Guest: And many other reasons as well, including things like: the United States has a medical culture of doing what I would call the full court press. Every baby. So, we go to great lengths to prolong and hopefully end up having a success of survival--like I said, the weakest and most vulnerable infants. We have a lot of other confounding variables. Like, we are very aggressive about in vitro fertilization, which has a secondary effect of multiple gestation pregnancy--twins, triplets. And these are higher risk--higher risk for premature birth as well as more infant death. There are a lot of reasons where we go to very different extent of trying to get survivals and trying to get births that actually in the end harm our overall statistic. Russ: Yeah. Which is not what human beings should care about. Policy wonks can use those, obviously. The number on in vitro fertilization (IVF) is rather remarkable, actually. I don't remember it off the top of my head; I don't know if you do; but the proportion of U.S. births that are, the frequency of triplets or more, I think is-- Guest: Yeah, it's been expanding dramatically. I don't remember the number either off the top of my head from the graph, but I think the point is, as has recently been in the paper, there have been millions and millions of babies now born by IVF in the United States, and we have been rather aggressive about implanting embryos with the hope of getting a baby. And with that increased risk of multiple gestations. Russ: I've found one stat[?] here that's of interest from the book. This is infant mortality rate within the United States by pregnancy; this is per thousand live births, which as you point out, very premature babies and high-risk pregnancies, but if it's a single embryo, a single baby being born, the number is 5.87 out of a thousand. Very unlikely that the infant dies. If you have quads, it's 146.48. So, it's almost thirty times; and it rises steadily with twins, triplets, quadruplets within the United States because they are more at risk. Here's the triplet birth rate--again, this doesn't compare it to other countries, but in 1980 about 40 out of 100,000 live births were triplets; by 1998, that number had reached almost 200 per 100,000. So a five-fold increase. But that would affect the numbers over time. If you are measuring at a point in time and comparing to other nations with lower multiple gestations, you are going to get a very different number.
17:54Russ: So, the first point, I think the first part of your book, is that a lot of these health outcomes that are used to indict the U.S. health care systems are misleading for statistical reasons. But then you go on to talk about quite a number of interesting other types of outcomes that we would just use in common sense ways of judging efficacy of health care per se. For example, cancer survival rates. So, if you have cancer, if you have lung cancer, God forbid, or some other type of cancer, there's some genetic interactions I assume, potentially with the native population; but a lot of them is going to depend on early detection and treatment. So, how does the United States do relative to the rest of the world on something as crucial as that? Guest: Yeah. Cancer is a very glaring example of the quality of the U.S. health care is superior. In all common cancers, the ones we talk about--lung, breast, prostate, etc.--the United States does better, significantly better, than all the countries of Western Europe and the countries that are held up as models for health care reform. We do better in survival in the rare cancers, as well. And there's a lot of factors that go into it. It's not just early detection--it's treatment, it's availability of cancer drugs; most of the pharmaceutical agents that have become important over the last decade are those in cancer. Interestingly these articles in the peer-reviewed literature are coming out of European authors, who say, right out, forthright, the United States has better cancer care than "we"--the authors say--in Western Europe. So, the facts in the medical literature I think are rather obvious that the United States, if you are going to be sick, and in this case cancer, you'd rather be in the United States than somewhere else. This doesn't just include people, by the way, with super quality health insurance or rich people. This includes everyone. This is overall data, everyone in the system. Russ: So, let's take a few different--I don't know if you know about this, so feel free to duck this question. But if we talked about, say, three different types of patients in America, under our current system. Under our current system, one type of patient might be someone who has access to world class health care. They have a first-rate health insurance program; they don't have to spend any of their own money. They get cancer, they live near a world class hospital and world class doctors; and they clearly get some of the best treatment if not the best treatment in the world, in the United States. Then we might think about someone who doesn't have access to that, maybe geographically, for family reasons, but still has access to what you would call mainstream care. And the third would be a person who doesn't have health care-- Guest: Who doesn't have health insurance. Russ: Right, doesn't have health insurance, would be this third kind of person. Guest: I want to correct--this is another kind of big misconception. Russ: I'm glad you caught me there. Guest: Making it synonymous--no health insurance in the United States means that there's no health care. And that's simply wrong. Russ: If I were a dishonest person I would edit this part out of the podcast, because that drives me nuts, too. So, I meant health care insurance; so this person does not have employer-provided insurance as you and I have; this person doesn't have very much financial wherewithal, period, and has to either presumably get Medicaid and get government aid. So, what's the difference in treatment--and outcomes--but also just technology. Rich people get--and I would add, the other issue I always think about is a heart attack; to take an extreme, a homeless person today gets better care than President Eisenhower got in the 1950s because of technological advances. A homeless person who walks into--can't walk, but gets taken into an emergency room, gets some remarkable stuff. So, am I right there? Guest: I think this opens up a very important topic that has a lot of offshoot discussions. And this is really the difference in care that someone gets in the United States between private insurance and Medicaid, and as a third comparison, people with no insurance whatsoever. No Medicaid, even. And when you look, there are studies in the literature--in fact numerous studies, study after study--that show in a variety of settings, whether it's heart disease, cancer, transplants, all kinds of things: people have better outcomes, better medical care, if you just take the people who are just as sick from each of those populations--private insurance, Medicaid, and no insurance whatsoever. Private insurance has far better outcomes than Medicaid. Even the same sickness of person. It has nothing to do with how sick somebody was when they started. And then, the even more alarming thing is it is very common in these studies--these are peer reviewed studies in the medical literature--that the people with Medicaid do worse than the patient with no insurance whatsoever. Russ: Now, I've got to stop you there, because here at EconTalk we often talk about how unreliable peer reviewed studies are. So, that's a plus, but there's always a lot of choices to be made in how you evaluate, etc.; and people have biases and axes to grind. I want to hear the underlying intuition of why somebody who is on Medicaid is going to get--first, I don't know why, I'd like to know why they'd get worse treatment than somebody who has got a luxurious, as it's often called, private insurance program. And than the person who has no insurance. What's going on in the patient level that would make that happen? Guest: What's going on at the patient level is that when the government--let's talk about Medicaid specifically. Medicaid is a type of insurance policy, just like any other insurance. And there are significant limits to what Medicaid will pay for in terms of diagnostics and treatments, therapies. And so when you restrict the options that doctors have, and in this case we assume that this is the cause because the patients are the same--we've already controlled, the studies have controlled for that. Russ: That's challenging, of course. That's hard to do. We have observable differences between them. Guest: No, not in these studies. Yeah, I mean in these studies, they've controlled for all the confounding medical risk factors. Russ: That they can measure. Guest: Yeah, all the things that can be measured. Russ: They don't have genetic code. There could be genetic differences between the two populations. Guest: You can hypothesize. But what is known in the standard way these things are done, these things are pretty rigorous. Russ: Well, the part I'm willing to concede is--that's why I asked before: x-stage lung cancer, at least a medical diagnosis that's constant across the groups. Guest: Yeah, they took medical diagnoses constant, but also other medical, what's called co-morbid conditions. Russ: Obesity, smoking. Guest: Diabetes. All these things they controlled for. Because it would be unfair to say: This person did worse after surgery, well it turned out they were diabetic. So, you have to control for all--medical, co-morbid conditions are very relevant here. Russ: No doubt. Guest: Once you've controlled for those things, you ask: Why would somebody with Medicaid do worse? It's because the treatment options are worse for the doctors and hospitals. They are restricted in what they are going to do. This is what's available. Now, why would it be better for no insurance? Because with no insurance at all, there are basically two or three things that happen for people with no insurance. They get medical care--they either get it by paying out of pocket or they get it by some other part of the system pays; or the third is it's done for free, literally for free, so-called charitable care. And in the instance of other forms of payment--actually there's something called disproportionate share payments that the Federal government gives to hospitals who treat patients who are indigent and have no insurance. There are ways that monies are shifted around to compensate. But in the end, though, the patients with no insurance have no restriction on what they are going to get once they are being taken care of. Then it simply--this is what we are going to do for this patient, whether we are going to get paid for it or not is a decision of how the person giving the care is going to have to make.
27:33Russ: So, we're sitting on the campus here of Stanford University. Are you saying that a Stanford professor who strolls in--not going to stroll, but who is being treated for lung cancer at Stanford hospital is going to get no better treatment than a homeless person who walks in and is diagnosed with the same condition? Guest: No. What I'm saying is that the group studies that have been published show that it's common, although not universally true, that people with Medicaid insurance do worse than people with no insurance. On any given individual, I'm not so sure that applies. And does everyone get the same care? I don't know how you would--I'm not sure that that would be a realistic statement. Russ: I'm just pressing you because I would love to know what Medicaid rules out that a private insurer supports. Guest: Well, I'll give you an example in general. Medicaid pays about 60% of things that private insurance pays for--60% of the rate. And so the fact of the matter is, money is important, as you as an economist know, a key incentive for what things get done. Health care costs money. Despite what people want to believe, health care is not free. Someone must pay. And things just don't happen for nothing. And so if Medicaid--and actually when you look at the data, which I happened to do yesterday, on the surveys on why doctors don't accept Medicaid patients, a huge percentage of doctors, more than half, do not even accept new Medicaid patients any more. When you ask of these 5 or 6 factors, why? Seventy percent of them say the number one factor is the reimbursement. They're not even going to take the patient, because they are losing money on every patient. Russ: And that's why we need to force them to take them and make the payment more generous, and that way we could all have the same great health care. Or better--or worse, depending on how ironic you want to be--we need to get rid of that private insurance, so we can all have the same quality. Which is often what you hear in defense of Canada. Guest: And this brings up something that I want to point out about the WHO ranking, to circle back, that first ranking, that big study, was that it turned out that equality was a better outcome for quality of care. Equal outcome in that country was higher as a ranking than disparate outcomes even if, even though all the outcomes were better. What I mean by that--it was poorly articulated--even if you had everybody gets A, or B, or C level care in country X, versus another country where everybody gets C level care, country X with As, Bs, and Cs was worse in the ranking than the country that got everybody a C. Russ: Because they decided to rank equality in and of itself. Guest: Equality was more important than quality.
30:31Russ: Let's talk about another example that I've been fascinated by because it's an example of where technology in recent years has been I think an incredible improvement in quality of life, and that's hip replacement, which you talk about in the book; and another example, not sure whether you talk about it: knee replacement, another example where older people lose, tremendous loss of mobility and an opportunity to play tennis, just walk or get around because their knees and hips give out. And now we have an explosion in the availability in the United States of those procedures. How does the United States compare to other countries in that sort of thing? Guest: In comparing the countries where more government control is exercised on access to that kind of treatment, national health service in the United Kingdom or Canada, the United States does better. The United States has far more not only access to these life-changing treatments like hip replacement, knee replacement, cataract surgery; but better outcomes. So, what happens is, it's not something that is measured in life expectancy; but it is certainly measurable in quality of life. Now, quality of life is difficult to measure. But when you look at the idea of being able to take care of day to day activities by being able to walk, or being able to see, in terms of cataract replacement--I think it's kind of obvious that you don't want to wait a year to get a cataract operation, even though you are going to live the same, it's better to get it sooner. Russ: Every day is good. Guest: There's no question. Russ: And how big are those differences? Guest: Those differences are huge. And what interesting is they are also economically related. And this is something else that people don't understand. Yes, it costs more money to get the hip replacement surgery, but on the other hand, the indirect costs of not getting it. Russ: It costs more in the United States than elsewhere, is what you are saying. Guest: No. What I'm saying is that getting the surgery versus not getting the surgery. Okay, you are adding costs by getting the surgery. Russ: Correct; it adds to our expenditure on health care. Good to save; you could have avoided it. Guest: But you are also saving money by not losing--let's just say for example that that person can't go to work for three months and needs assisted care. Or can't be productive. There are all kinds of indirect and direct economic losses and expenditures that relate to things. So, actually in many instances the use of so-called expensive technology actually saves money down the road. Indirectly. This has been neglected by a lot of the kind of low-hanging fruit economic studies of the costs of medical care. Russ: Arguing that some of these are wasteful expenditures that our system needs to get rid of, purge, and all that. One other example you give--you know, I've heard this before. You quantify it very dramatically: the time you have to wait to see a specialist for some kind of treatment. You hear horror stories sometimes about other systems; you don't know whether they are representative or not. You find dramatic--you cite dramatic differences in how long till procedures are implemented for people who need these treatments or want these treatments. And how long it takes to see a specialist. Guest: Right. Absolutely. And you look at places like--the best studied ones are Canada and the United Kingdom. But the countries all over Western Europe have these sorts of studies. Where, in the United Kingdom, for instance, they've put this law in place: they do not want anyone to wait more than 18 weeks for referral to a specialist. Okay, in the United States, that doesn't even exist. I mention something about getting a cardiovascular procedure, a minimally invasive treatment, say a balloon angioplasty, even in a non-emergency setting. It's considered unacceptable if it's more than a couple of days in the United States. We don't even consider these things in terms of weeks or months, like other countries do. And in fact these other countries have not only instituted rules to try to deal with these, what you might say are immoral waits for medical care, because of their public, so-called public systems; not only do they try to institute rules, but they also have had multiple lawsuits going to Supreme Court, and now have instituted privatization, all over the world. Countries that started out as more and more government control, what they call national single care systems, are using privatization to help compensate for their severe waiting lists. Russ: Yeah, Canada I know recently there was a court case in the last 5 years, right, that allowed people to purchase services from a doctor, which was illegal before. Correct? Guest: That's right. Russ: Hard to believe. Guest: And this is all over the world. You look at studies from Denmark, all over Western Europe, places that are really the models for our health care reform or from the people who want a single payer system here. But the reality is that the United States not only has the best outcomes, but we have by far the best access. And this has been a gross distortion not only in the press or in the media, but in movies and things like this that rely on anecdotes: the access to care is so much better in the United States. Given that it isn't perfect, given that we need to do everything we can to improve access for everyone, it's by far better in the United States than in these other countries. Russ: So, you are suggesting it's not just the rich people who don't have to wait for a specialist. Guest: Absolutely not. Russ: Do we have any data on waiting times in America? You said the average is lower than elsewhere? But what about different classes of people? We've been talking about that already--Medicaid, people without insurance? Low quality insurance? Guest: The big problem that is kind of analogous to waiting times in the United States is this idea that Medicaid patients now even have problems finding doctors who will take them. So, anywhere from 45-60% of doctors in the United States broadly will not accept new Medicaid patients. And lets just say only 5% of doctors or less than doctors do not accept new private insurance patience. So, there's a difference there. And this is a huge problem for Medicaid patients again stemming from what I said, that patients--doctors and hospitals--lose money on every patient. You can't make up for that in volume. As the old joke goes. But the other issue there is it points out the severe flaw in the logic of the increase in health care reform, which is to put 15-20 million more people into Medicaid. Not only is that a financially unsustainable system, but you are really not offering access to medical care if half of them or more are not even going to be able to get a doctor. Russ: Well, as Hayek said: The curious task of economics is to demonstrate to man how little they understand about what they imagine they can design. So, they will have to fix that; they'll figure that out, too. But maybe not.
38:14Russ: So let me challenge you. The book--you have other treatments of these data elsewhere that I hope we can get some of these charts up on the website. They are really quite striking in terms of painting a picture where the United States looks awfully good relative to the rest of the world. In cases where it doesn't look good, you argue, perhaps persuasively, that there are reasons, statistics or definition, that are misleading. When that isn't an issue, the United States dominates in terms of quality and often dramatically so. What did you ignore? What did you leave out? Are there other measures that critics of you paint a less cheerful picture? Guest: Well, you know, that's making me think quite a bit as I'm sitting here. I think the measures that people cite that paint the United States in a substandard or inferior quality, are really the measures that we've talked about. It's really just about life expectancy and infant mortality. And then there is a third measure, which has been cited, which I go into in the book, which we didn't talk about, is the number of uninsured. Russ: Right. Guest: Because this is really--I do a lot of international traveling, I read the newspapers and speak to people outside the United States, and it is portrayed as scandalous that we have 50 million or so Americans with no health insurance. Which I mentioned is equated with no health care. As if they're synonymous. Russ: As if they are out on the streets if something happens. Guest: But I think here, this is a measure that really has to be scrutinized. And I did in my book. This so-called 50 million uninsured--because when you look who is this population, the raw data, the documents, the U.S. Census Bureau documents, and others, you find out that it's not really 50 million people. After you say: Well, okay, about 10-15 million people are not U.S. citizens in that group--and I'm not saying they shouldn't get health care, but I'm not sure you are going to reform the U.S. health system to get non-citizens insured. Russ: Correct. That's going to be a challenge. Although not that group, but there are others who are illegal who do get health insurance. Guest: Who do have health insurance. Right. And then you take a look at who answered the U.S. Census Bureau survey and said they didn't have health insurance, and it turns out--let's just say, I don't remember the exact number, but about another 10 million or so that said they didn't have insurance that actually were using insurance. And we know that because the Census Bureau people went and looked, looked up and found medical records; these people had insurance that they were using, and mainly Medicaid. Russ: They didn't consider that insurance. Guest: They probably when they answered the question, they thought the question meant: Do you have private insurance? But, be that as it may, this is in one of the Appendices of the U.S. Census Bureau documents, Appendix C-- Russ: Good to know-- Guest: Is that they actually were aware--the U.S. Census Bureau were aware. But they didn't change the response to the question. And then there's another 13 million adults and children--of these 50 million people, 13 million who actually are already eligible for public insurance--Medicaid, a tiny bit Medicare, and the Children's Health Insurance Program (SCHIP or CHIP)--that simply did not sign the paperwork because they haven't accessed the system. So, they haven't used it. Common sense says you wouldn't want to redesign another system to make them eligible for that when they are already eligible for the current public health insurance system. So, you are left with a population of less than 5% of people in the United States who don't have insurance or who are not already eligible for current government insurance programs. I would not call that a crisis in the uninsured.
42:37Russ: So, we're going to come back to this at the end, because the fact that a big chunk of them are covered by the covered system or the current system serves them well--I'm sympathetic to the idea and I think you may be as well--we may not be happy with so-called Obamacare, but the alternative that it replaced, "the current system right now" if Obamacare is fully implemented, is not a great system either, on the expenditure side. So, we'll come back to that. One thing I want to mention, though. One of the criticisms you do hear of the U.S. health care system is that there are very disparate treatment outcomes depending on where you happen to be, which kind of doctor you happen to get. And there's a big push to use data and expertise to standardize health care treatment. Do you think there are lost opportunities within our decentralized, somewhat decentralized system--I love when people say: Well, we know health care markets don't work, look at the United States. Well, we don't have very free market health care right now; it's distorted in all kinds of ways. But some people criticize the result because of the somewhat disparate choices that doctors make and patients end up with. What do you think of that argument? So, we need a more standardized, top-down system? Guest: I think that there's--it makes me a little bit nervous, although intuitively I think it is true that we would like to make decisions based on data, and a lot of medicine is art rather than science, to a great extent. However, a). there are problems with the studies like the Dartmouth Atlas and different expenditures in different regions, particularly because many of these studies are just talking about the people who die. That's a distorting statistic, and you go back and look at their health care, that's a little bit different than talking about the way health care is given in general. But separating that side of things, when you start talking about standardizing care, I'm very nervous about that concept, because when you look at the countries who are standardizing, their outcomes are worse. And so, who is going to standardize this care? I personally don't think that health economists and government appointees are the people who should be standardizing my care. Guest: I don't want them involved at all. Russ: Don't you trust them? Come you. You said health economists. I know some people. I'm sure people are listening now and getting mad for a whole bunch of reasons, not only just this one. Guest: Tongue in cheek, of course. But I think everybody from politicians who push single payer to all the health economists I know, when they want medical care for themselves and their families, that they go and they seek out on their own the best specialist care, the best doctors, the best doctors that they can find. They want the freedom to choose. When you start having a top-down system, with so-called standardizing care, you have to realize that this is all in the context of medical care in the United States, particularly which is evolving very rapidly. Even today. It's not like it's mature now and so all the advances are done. No. And so when you look at what's happening with minimally invasive care, safer treatments, safer diagnoses--now we are in this era where there is a confluence of molecular biology and genetics, entering in medical care-- Russ: Potential customization of care is tremendous. Guest: Absolutely. Everything is changing very rapidly. By the time the government or any other body starts to standardize things, it's already old. If you think computers move quickly, you ought to look at the scientific literature on how medical care works. And so it's an intuitively attractive argument to say that things should be standardized. On the other hand, I haven't found a system with better outcomes that has standardized care. They are all worse. So, I'm not sure it's a direction to go. I would prefer to have a direction where the government gets out of the business of directing medical care. Where government isn't even the insurer at all. Where government helps people who can't afford insurance but doesn't dictate the care by virtue of being the insurer. And I think this is a huge problem with the way Medicaid and other outcomes are, the way things are right now. Facilitate competition, facilitate innovation, but get out of the way.
47:24Russ: We'll talk some more about that in a minute. But you mentioned the Dartmouth Study. Probably the most famous study that I know of on health care is the Rand Study, which purported to find that expenditure is not terribly important in outcomes. It just led to more spending. That there's enormous savings in the current system. Which I think--both left and right appear to justify their own views--but what do you think of the science of that, given that you are suggesting that the expenditure levels of the United States, which lead to earlier seeing of specialists, more access to technology for treatment of cancer, strokes, etc., knee replacements, MRIs. All these things, which are expensive--the Rand Study suggests that it is just feel-good, it doesn't matter for how your outcomes are. It just makes you feel better that you've done something. What do you think of that? Guest: Well, yeah. The Rand data to me shows something different. The Rand data to me shows that when you increase out of pocket expense, and people then make more of a value-based decision, because they are actually spending money-- Russ: They are on money. Guest: They are on money. Or certainly are aware of what things cost as opposed to the current system, where someone else is "paying for everything" and therefore it's free, in people's heads, even though it isn't. Russ: It's nice. It's very appealing. Guest: So, when people pay out of pocket, they make a decision to say: Okay, I'm not going to do it; I'm going to this, I'm going to make decisions; I'm going to save money. And their outcomes didn't get affected. Okay, so, you could say it indirectly shows that some of these expenses are not worth anything. It's probably true overall. There's certainly waste in the system, there's no question. But I think my view is that I would prefer to have people decide, with their doctor--of course, because they are going to need help with this--but in the end it's an individual decision for yourself and your family what you think is worth spending money on. I do not feel comfortable having the government say: Okay, this is not available because it costs too much money. That's a very different scenario, with two different ways of saying: We're not going to spend this much money. Russ: Of course, you know, the standard criticism of that view is that--let me say it a different way. I think one of the great benefits of our system, even though I don't like our current system, the current system which allows people to spend other people's money--is that most people don't have a lot of anxiety about how much money they spend. Turns out, it's an unsustainable system, I suspect. And you have a different kind of anxiety, instead of anxiety whether you can get the best treatment, you have unknown anxiety that you didn't get the best treatment because it didn't evolve and develop. You have to wait, which is an incredibly horrible thing when you are ill and your life is at risk. And of course people die. It's a terrible tragedy. So, what you are suggesting is you want people with their doctors and families made these decisions instead of the government; but not everybody is going to have that freedom. There are going to be poor people who can't afford the treatments. They choose not to, but it's a terrible choice to have to make, and they'll often make a choice that's the best for them, but they don't have good choices. Guest: Well, I think this is where, again, this is a problem with misinformation where people think that there is no alternative to reform other than the so-called Obamacare type of centralization. And in fact there is an alternative to reform. The first statement I want to make is that the real crisis in U.S. health care is the cost. Not the quality. And not the number of uninsured. And so Obamacare, the Affordability Care Act, does not address the cost. That's a fact by all estimates. Now, let's talk about how to address the cost to give people the choices. Because I agree with you: No one wants a system where people who are low income or poor people have really severe problems with access, do not have adequate choices for the quality of care that is available. Given that there is no guarantee that anything is ever equal in the real world, but the reforms to bring the costs down and to increase the choices for people--including poor people--are what we should be talking about. Russ: What would some of those be? Guest: Some of those are the following. I mean, I think there's reforms for private health care as well as government programs, as well as the tax treatment. There's three basic categories of reform. The private reforms are to get rid of the artificial barriers to competition that exist right now in the system. People say: Oh, yes, we know, like you mentioned, the free market doesn't work. Well, the free market doesn't really exist. It's never really been tried. And so these bizarre barriers that are archaic including being unable to buy health insurance outside of your own state--this is a huge and really absurd setup to protect the status quo and really special interest groups in this kind of paternalistic view that people have to be protected. People in New Jersey can't buy insurance from people in Pennsylvania even though it's right next door. And there might be a significant difference in the price. Another problem with private insurance and the cost of insurance I'm talking about-- Russ: Which ends up affecting health care prices-- Guest: Absolutely. Is that the states have put in these mandated coverages, to the tune of 2000 of them in the United States. And it's certainly true that many people are forced to buy insurance that covers things they do not have any possibility of using. For instance, in vitro fertilization. Acupuncture. Massage therapy. Wigs. There are all kinds of things that are mandated that everyone in that state must buy insurance that covers things that they don't want. And what is the effect of that? The effect of these mandates is that by some estimates at least 50% or more of the cost of insurance is due to these mandated coverages that have been built up over the years and are similar to the principle that is in the Affordability Care Act, which is defining minimal essential benefits packages. Instead, people should be able to buy insurance, like catastrophic insurance, that is much cheaper and have these Health Savings Accounts, for instance, to use for small expenses. And then some of the other reforms include reforming the public system. Let people have choices, including poor people. Let people take money that was going to go for Medicaid insurance and use that money to shop around for private insurance, including deductible claims that are cheaper, with Health Savings Accounts. Let's see when insurers, as is true for every other good and service in the United States, historically--when people have a market to compete for, when companies have markets to compete for, there's competition on price and value. And you end up creating a market that is very competitive. And eventually prices come down. If people want to buy something that's cheaper, it will eventually be there. And the same thing has happened, for instance, with computers. People say: Well, I couldn't possibly shop around for medical care or insurance; I don't understand it. Russ: It's too complicated. Guest: Not many people understand literally how a computer works. I think that's a very difficult concept to explain to somebody. But we're doing a pretty good job of shopping for computers. And so, the public systems need some significant reforms: More flexibility, more choice, and more competition--prices come down.
55:47Guest: And then there's other things in that kind of realm, and that is to increase transparency of information. Medical care must be one of the only, if not the only, good or service we buy and use without knowing what it costs until later. Russ: Or ever. Guest: Or ever. We can never decipher what it costs. And so these are things that are kind of this veil of secrecy, I call it, around the price of medical care which allows doctors and hospitals this kind of blanket of protection of what the price is. Let's see when the University of California, Stanford, (UCSF) and neighboring medical centers must put out their prices, let's see what happens to those prices. I guarantee you they come down, just on the basis of transparency of information. And then the third category of reform, really the big category, is tax reform. Because the way the current system is, you are encouraged to spend more money on health care. And by that I mean health insurance and health care. It's endless how much you can deduct from your income on the basis of how much you, what your employer provides. Russ: It's treated as a tax-free benefit. Guest: It's treated as a tax-free benefit. Which is kind of a historical accident, for lack of a better word. And there's no reason that the value of a dollar for health care should be higher than the value of a dollar spent for something else. The system right now encourages more money, encourages health care prices, and is a massive financial burden to the system. Russ: So, it encourages compensation to be given in the tax-free reform rather than the taxable form, and so you would rather have tax-free health care than pay for it out of pocket. But once it's tax free, we want a good one, because it's relatively cheap. Guest: Right. Russ: Do we have estimates on how much that contributes to the cost of insurance? Guest: The cost of insurance per person, I don't know. But we lose hundreds of billions of dollars because of this tax exclusion. That's known. And when you decrease that--there are various ways to deal with that. But basically what I think the important point is: This idea tries to inject some kind of cost consciousness into people. Kind of a value-based decision should be made with everything that money is spent for. Because like we know, but fail to kind of admit as a public, is that health care is not free. It is not going to become free. And I'm not sure that there is any right to health care that is more important, say, than having the right to own a home or clothes or food. I'm not 100% sure why health care is so special, given all these other essential things. Russ: Well, it is a primal part of us. Robin Hanson argued on this program a long time ago that a lot of what we do in health care, a very creative argument, is window dressing. That we get benefit from because we convince each other that we care about from. I'm a skeptic about the theory, but it's a provocative theory, an interesting idea. And there's no doubt that you earlier criticized the World Health Organization rankings--I think correctly criticized them--that equality is a value in and of itself. But there is something emotionally troubling about somebody getting access to a cure, to dialysis, or to a transplant on the basis of income. I think it bothers a lot of people. Whether it should or it shouldn't, what are the long-run benefits of that because of innovation or profits would make up for it--I think psychologically and emotionally there is this tremendous bias in our political system. Guest: And I'm not opposed to that. And I think the answer is to allow people that have less money more choice, more flexibility; and increase, facilitate the competitive forces that will bring the costs down so people can get better access to care. There is no reason why people that are poor have to suffer through the outcomes that their Medicaid insurance gives them. There is no government official that would opt for Medicare. I can guarantee you that. It's inadequate. And it's really kind of an immoral thing to push people into that system when we know the system is a failure. The system of Medicaid insurance, I'm talking about. So, yes, it's true. The question is how to go about improving the system. And I don't think the answer is in the Affordability Care Act, which essentially dumbs down the specialty care and the excellence and the outcomes that we have in the United States in this so-called guise of "fairness." I think that it's a flawed model that will turn back the clock on the medical innovation that has led to the United States having such superior health care.

COMMENTS (140 to date)
Greg G writes:

There were two main themes to this podcast. The first was that that U.S. healthcare outcomes compare much more favorably than we have been led to believe with other modern countries that have more socialized healthcare. I found this claim to be very convincing and well supported.

The second theme was that these good healthcare outcomes are achieved because we have a less socialized system of healthcare. This was simply assumed and not shown at all.

Yes, American cancer care is very good. Cancer tends to be a disease of old people. Most cancer care is paid for by Medicare. Do we really want to simply assume that most Medicare cancer patients would receive the same high level of care they are getting now if there was no Medicare?

Yes, Americans get a lot of hip replacements. Hip replacements are done mostly on old people and mostly paid for by Medicare. Do we really want to simply assume that they would be more available to these people if there was no Medicare?

Many of the people enjoying the quick access to specialists celebrated in this podcast are.....You guessed it: Medicare patients.

Maybe the fact that American healthcare is so good simply reflects the fact that we devote more resources to healthcare than any other society ever has in human history.

Maybe conclusions about what this says about government involvement simply reflect the biases that researchers start with.

Floccina writes:

I think that this study shows that obesity is probably not the biggest factor pushing down USA life expectancy. It could be a big factor in cost.

Jaime writes:

Great example of "Let's just use the information we like to arrive at a conclusion we like." Bunk.

Having lived in a "socialized medicine" country and the USA, my anecdotal experience is that the American healthcare "system" is to be avoided whenever possible. It is riddled with conflicts of interest that result in compromised care.

At least I acknowledge that this is anecdotal and opinion.

Dan writes:

1) A little gripe:

Interesting discussion, as always. However, one itsy-bitsy concern regarding the discussion. What is the "Affordability Care Act?" I know opponents often prefer to label the statute with the derogatory "ObamaCare," but in a more rigorous academic setting, (which, by the way, I consider EconTalk to be) if the speaker is to attack a statute, he should get the name right. I know my point may be a pedantic one, but getting the name wrong hurts the intellectual credibility of the speaker as well as the show; certainly in the eyes of educated outside observers who are not familiar with Dr. Atlas's substantive body of work.

2) Substantive Question:

A question I've always had about health care "markets" relates to information. Is the vast information asymmetry fatal to any free market based solution to the provision of health care? (When I say "free market" here, I mean one which relies less on third party payers and one in which consumers have more exposure to price signals on a per/procedure basis)

Example: Am really I going to second guess the first opinion of my physician when he suggests my young child needs, "such and such" expensive/low benefit test, that may appear to me to be overly costly (what do I know?). Will I then go get a second, third, fourth, etc. opinion, in an attempt to reduce the cost, when all the while my child is suffering? Do I even have the knowledge, training, and or time to make such judgments absent the advice of my physician? In many ways it seems like the market for car repair services, except with many more emotional variables...or, perhaps, this is a more general problem in markets for "complex" services (i.e. those that require specialized training, e.g., medical, legal, IT services, automobile repair, construction, etc.) (maybe a future podcast?)

My intuition tells me that the insurance companies (or governments) can (and do) help alleviate this problem by refusing to cover certain procedures (e.g., based upon cost-to-benefit) or by only reimbursing providers "market" rate for certain things. What isn't clear to me is how making these costs more transparent to the individual patient is going to really help drive costs down all that much.

David Robson writes:

"When you take out suicides and fatal car accidents, factors that Atlas argues are unrelated to the health care system, the United States has the longest life expectancy in the world."

I can accept fatal car accidents as mostly not healthcare related, or at least assume that healthcare issues are not direct causes on most vehicular deaths, but suicides? How is suicide not related to mental health? I thought Mr. Atlas should have been made to justify that statistical reduction more strongly.

drobviousso writes:

Dan -> Why do you assume, if the market where changed in such a way that people would want more information, the market would not respond by finding new innovations in providing information to people? I has happened in the auto industry. I has happened in the auto repair industry. I has happened in the legal services industry (possibly the only industry as heavily regulated as the medical services industry). I has happened in the landscaping industry. It has happened in the child care industry. It has happened in the charitable donations industry. And those are just the one's I know about from dealing with them in the last few weeks.

David -> Perhaps Dr. Atlas believes in a modicum of personal autonomy? Perhaps because the "but for" analysis still puts the cause of suicide on the shoulders of the person committing suicide. Perhaps his reasoning is in the book.

This is the kind of podcast that makes me feel like I didn't learn anything. I learned a lot of interesting assertions, but I'll have to read the book to find out if I can learn anything from them. Hopefully, the kindle version is well laid out.

David writes:

Two rich guys talking about how poor people in the usa get good care. Really depressing podcast.

DJan writes:

Horrible episode.

From the point of an independent observer a health care system that provides good care for everyone (European style) is vastly superior to a system that provides excellent health care for the rich and virtually no health care for those who can't afford it.

I don't really care what economists think about it because economics as a science has a very limited use and this is one of many examples where it leads to absurd conclusions (because it doesn't know human suffering).

Harland Brown writes:

While I do not agree with all of the guests that you have on the podcast, they have all been thoughtful and devoid of flagrant partisan politics up until this point. This so-called spokesperson has an agenda of defending the Republican platform and then discrediting any data that does not feed into that particular view.
I echo the sentiments of an earlier comment, having lived for many years in France and being married to a French MD. The US system is sub-standard and is driven by economic rather than care related considerations. Sure, you can get great care in the U.S. if you have unlimited resources and an acute healthcare issue. The greater problem is that we do a very poor job of trying to prevent people from getting there in the first place. This is evidenced by the very low percentage of MDs now choosing family or internal medicine. It is like driving your car until you run out of oil and the engine seizes up. Then there are a bunch of specialists waiting to take the lucrative work of rebuilding the engine.

Trevor C writes:

In Canada healthcare is handled by the Provinces and not the Federal government, so not as bad as it could be. The good thing about the system (I live in Ontario) is that it is an economy of scale for purchases but the poor thing is that it's not very efficient when it comes to general access to procedures.

There are also issues across the country with provinces not being able to afford the increasing costs of healthcare due to the aging population. It's been a slow motion trainwreck for years and the agreement is that it's not a matter of spending more, it's about making the system more efficient.

Getting something like an MRI or an Athsma test (as I recently had) can take months, getting a non-critical procedure even longer. It is often worth the money to go to a private institution to get these types of procedures.

The areas that are mostly private, like Laser Eye, Dental, Chiropractic, Physio etc (all of which I have used recently ~3 years) are top notch and very reasonable when it comes to price. Based on past experiences I think we would benefit greatly if the reigns were loosened to allow for more private competition.

Floccina writes:

1. @Dan to combat the asymmetry problem, one could select a primary care physician who does very little himself and ask him questions. Most primary care physician do not do all that much high cost stuff as it is, and I find that you tell them you will be paying out of pocket they order less expensive treatments.


2. On comment in that interview that USA has higher cancer survival rates, the superior cancer survival rate in the USA could be due to the fact that we find and treat more cancer that would not be lethal without treatment.

Stevie writes:

Trevor, I'd like to ask you to clarify what you mean by "efficiency." It seems like what you mean is that you have to wait for certain procedures. But is the delay caused by actual inefficiency, or is it simply a matter of basic supply and demand, combined with prioritization? For example, if there are a certain number of MRI machines in your province, and the demand for them is such that many people have to wait several months, the process itself may be extremely efficient but demand is high, making it seem inefficient (if inefficiency is defined as waiting several months). I assume that the more critical cases get higher priority than the more routine cases for the limited number of MRI machines.

And then, I'm sure an argument could be made that having more MRI machines would make the whole process more "efficient." There are different ways of looking at and measuring efficiency.

This may seem like an odd question, but since we're on the subject of efficiency, would you happen to know if the MRI machines in Canada are used around the clock, which of course would maximize the efficiency of the machines? Of course, such a schedule might not maximize convenience to patients. :-) If there are a limited number of MRI machines, I'll make another assumption that evening appointments, at least, are not uncommon.

Stevie writes:

Interesting points, Greg. I haven't listened to the podcast yet. I like reading some of the comments first to "prepare" myself for things to watch for. :-)

There are many thoughtful people in this forum. I have a general question that may or may not be directly related to this topic. Please, anyone feel free to respond.

In the US we seem to have no problem (i.e. no constitutional angst) requiring people who choose to drive cars to pay for car insurance. That's because of the possible externalities. And then there is the entire infrastructure for cars and driving that everyone is helping to pay for, too, drivers and non-drivers alike, and again there isn't any problem with that. Now, assuming that most people choose to live, why is requiring people to pay for a health care system so different, constitutionally speaking? In both cases there are externalities. And I don't know how anyone could argue that cars and roads are not "socialism" writ large. :-)

I have not formed any opinion about this yet. I'm still in the process of trying to wrap my brain around it. The fundamentals are still fuzzy to me. So I would appreciate your thoughts.

drobviousso writes:

Harland Brown, you make two points I've heard in the past but have never had the opportunity to question. Would you mind?

The US system is sub-standard and is driven by economic rather than care related considerations. Sure, you can get great care in the U.S. if you have unlimited resources and an acute healthcare issue.

What does it mean to be driven by "care related considerations" but not "economic" considerations? In what other section of human endeavor do we consider ignoring the cost a good thing?

For example, I like a very, very good cup of coffee every morning. My wife an I have worked out a routine that involves chop sticks, thermometer, specific cups, and multiple timers. But I still don't go buy Jamaican Blue Mountain Special Reserve every month. I try to find an elbow in the price:quality ratio and shoot for that. I try to shoot for the same elbow in all my purchases, including my spending on preventative maintenance on my home and car, and on non-consumption goods.

Why is that a bad idea in the world of medical services?

The greater problem is that we do a very poor job of trying to prevent people from getting there in the first place. This is evidenced by the very low percentage of MDs now choosing family or internal medicine.

You'll have to break down the relationship between successful prevention and the rate of change in the # of doctors in family and internal medicine, because it is not obvious to me. I would think that the evidence would be a little more direct. Also, I have to wonder if internal and family are really the best doctors at prevention. Are they better than nutritionist, physical and occupational therapists, and mental health doctors?

I look forward to your thoughts. Thanks

drobviousso writes:

Stevie - If you are really curious about the constitutionality of Obamacare (and the current administration is officially ok with the name...) I would suggest going to the SCOTUS website and downloading the audio of the oral arguments. They are no more complex than a good episode of econtalk. Then you can find the breakdown of the ruling at any number of outlets.

The basics, though, is that there is no constitutional concept of the externality, per se. Congress may pass a law attempting to minimize or reduce an externality, but it must do so under one of its existing powers (often the taxing power or commerce clause powers are used). Coercing people to get health insurance has been ruled out of the scope of any such power. Encouraging, through non-coercive taxes under the taxing power, has been ruled to be constitutional. I am not a lawyer.

Stevie writes:

Drob, thanks for the suggestion and I'll give it a listen. But it may not be exactly what I'm looking for. There is "Obamacare," and then there is my more fundamental question. The Supreme Court tends to rule as narrowly as it can, but there might be some good tidbits I can use.

I understand that there isn't any constitutional concept of externalities per se; I was trying to keep my comment brief. But I can't help but notice that Congress tends to have more success at exercising its "existing powers" when there are compelling externalities involved.

Also, I can't help but notice that, on the one hand, you say that coercing people to pay for health insurance has been "ruled out," while on the other hand you say that "non-coercive taxes" are okay. Why do you see taxes as "non-coercive"?

You didn't say anything on why people are okay with being coerced to pay for car insurance, but not health insurance. Any thoughts on what the compelling difference is between the two things?

Greg G writes:

David Robson makes a great point. There can hardly be a greater failure in treating mental illness than suicide.

And yet, this week's guest manages to construe our high rate of suicide as evidence in favor of the relative superiority of our health care system. This represents a truly epic level of confirmation bias.

Gunshot deaths are a little more promising way to rationalize our poor life expectancy figures without blaming the healthcare system....except that a high percentage of gunshot deaths are also suicides.

What about smoking? Isn't smoking more common in western Europe than here?

Greg G writes:

Stevie,
Thanks for the compliment. You ask a good question. I don't see the problem with the government requiring people to pay for healthcare but many, if not most, on this forum will disagree. They can, and will, speak for themselves.

One of several very good things about EconTalk is that the worst podcasts still often result in the best discussions.

Justin P writes:

The comments here are as enjoyable as the actual podcast.

If suicides are a good proxy for the mental health outcomes for a health care system, why does Scotland have a significantly higher suicide rate than England?
http://jech.bmj.com/content/early/2012/06/08/jech-2011-200855.abstract
Please correct me if I'm wrong but don't both England and Scotland have very similar health systems? Or are Scotish psychiatrists that bad?

I personally have always found the suicide rate as a proxy for mental health, spurious at best. Mental health is highly subjective as it is. (Brought up numerous times here on Econtalk.) I think most people in the comments, already have their a priori and will cherry pick what they don't like to fit their view, confirmation bias.

Stevie writes:

Greg, I guess that marks you as a socialist. :-)

I don't have a problem with paying for healthcare, either. I'm insured through my company, but of course I'm the one paying for it (with help from tax subsidies). Call me a socialist, but I believe in universal healthcare. Healthcare is so basic and fundamental to the "general welfare" of an advanced society. Economically, it's more expensive not to have universal healthcare, though I acknowledge it's hard to see that with our current healthcare system.

Greg, I know you didn't mean anything by it, but I'm going to expropriate your phraseology to make a point. When you (or anyone, including myself) speak of "government requiring people to pay for healthcare," it's a loaded expression, in the sense that it implies that the government - this external thing apart from the people - is making us do something, sometimes against our will. I try to remind myself that "the" government really is "our" government, and it can serve a genuinely useful purpose. We decide what we want our government to do. In those areas in which there isn't overwhelming consensus we look at things like externalities, sometimes leading to forcing people to do things that they may not want to do. Things like not poisoning the environment or the requirement to have car insurance. Or, chipping in for the healthcare system that anyone would access if the need arose. We all pay for fire protection even if the chances of any individual house burning down is extremely low.

Greg, that's funny what you said about bad podcasts leading to good discussions.

Stevie writes:

Justin, could you say more? If someone commits suicide, I'm inclined to think it is a pretty good indication (proxy) of a mental health problem. In my case, I mean "mental health" in a very generic sense.

Having said that, I'm not so inclined to use the rate of suicides in a given nation as a proxy for the quality of either that nation's healthcare system or its mental healthcare system. I'm inclined to think of it as primarily a cultural issue. Moreover, I'm inclined to assume that a nation with high suicide rates would also be a nation with a high rate of people who feel alienated. Such a nation might have many first-rate mental health counselors and services, but still, if there is a lot of alienation they'd be swimming against the cultural tide.

I don't have any background in Sociology or Psychology, so I may well be full of bunk.

drobviousso writes:

The ruling is narrow, but the oral arguments are not.

Also, I can't help but notice that, on the one hand, you say that coercing people to pay for health insurance has been "ruled out," while on the other hand you say that "non-coercive taxes" are okay. Why do you see taxes as "non-coercive"?

That's the way the courts have ruled. They've gone so far as to say that a tax high enough to coerce, as opposed to encourage, you to get health insurance would be impermissible.

No, no one has any idea what this means. I doubt Roberts, who made the rule, knows. I get a severe case of palm-in-face syndrome when I think about it too hard.

The difference between car insurance and health insurance (legally speaking) is two-fold. One - states mandate car insurance. The states have much broader powers than the federal government. Two - Auto insurance is not required directly. Instead, there is a prohibition in most or all states against driving or owning a car without auto insurance. Obamacare was not formulated in such a way. It was: you must have health insurance. Not: you must only pay for health care via a health insurance intermediary. Not: you must not receive payments directly from patients. Obamacare probably would have been much easier to find constitutional if it was formulated in one of those ways, but it was not. Speculation I've heard is that it would be harder to get all the votes needed if it was, since it only passed by the slimmest of margins as it was.

Chambana writes:

Wow!

If you need a good example of a straw man argument for educational purposes, go no further than this podcast.

Atlas uses allegedly pervasive criticism (???) of the US heath care system to set up a sham argument in which he ‘demonstrates’ inferiority of the universal care in other developed countries. Needless to say, no serious health care economist a or public health official criticized the technical/medical/procedural aspect of the US health care industry as inferior. The problem is – of course – the restricted access, excessive costs, and misaligned incentives within the health care industrial complex.

You would expect such cheap shots from politicians like Mitt Romney, who purposely take out of the context comments of their political opponents to devise false arguments, but not from serious academics whose nonpartisanship we take for granted.

I urge Russ to bring in a serious industry expert who can deliver a rounded (if not completely unbiased) picture of the industry. I propose Jonathan Gruber of MIT, who designed a universal plan at the state level for a Republican governor, but may have progressive leaning.

Chambana writes:

Russ,

I sincerely believe that your comments are a product of your libertarian bias, and not of your views on inter-race relations.

“And that's why we need to force them to take them and make the payment more generous, and that way we could all have the same great health care. Or better--or worse, depending on how ironic you want to be--we need to get rid of that private insurance, so we can all have the same quality.”

However, it is a fact that most white Americans, particularly in Southern states do not like to pay for public goods, including health care, which they consider to be transfers to Americans of darker complexion.

For example, homogeneous groups (i.e., Scandinavians) are more likely to provide public goods. It boils down to how one defines a person. A Swede favors universal care because he does not believe that his white neighbor (who is just like him) will cheat and abuse the system.

http://www.economist.com/node/21525851

[Minor edit, with commenter's permission--Econlib Ed.]

David Robson writes:

As a layperson, I am assuming there is a stronger correlation between health issues and suicide than health issues and car accidents.

It's an obvious question to ask, and the question wasn't asked. In fact, Russ Roberts says, "Because, and correctly, suicides and high speed deaths should not indict your health care system."

It's not clearly correct. If untreated or poorly treated mental health issues are significant factors in suicides, then suicides should continue to be counted against the lifetime expectancy statistics for this argument.

I don't have any evidence either way, it was just something that jumped out at me, especially because removing the data very conveniently reinforces Scott Atlas' point.

Wes writes:

@drobviousso
I wanted to ask about reasoning behind your distinction between car insurance and health insurance.

Leaving aside the issue of federal versus state power, it seems that the difference is that is car insurance is not required "directly," but only for those who own or drive cars. I think your point is that there's a difference between prohibiting people from driving/owning a car and directing them to purchase a product.

This makes sense to me, but it seems that it doesn't touch on the deeper rationale for requiring people to buy insurance, one which is basically the same in both cases. You always have the option to not drive or own a car, but almost everyone will require medical care at some point in their lives. In cases where there is the need for catastrophic care, many states require hospitals to treat you regardless of whether you can pay. And so then there is the problem of cost shifting, where the cost of caring for the uninsured is placed on owners of insurance via hospitals.

So in essence, given the current framework, there's a sense that we nearly all of us are participants in the health care market, even if we are not currently receiving care or planning to receive care. And therefore we need a system to prevent my (potential) costs from being placed on others

Stevie writes:

Drob,

"They've gone so far as to say that a tax high enough to coerce, as opposed to encourage, you to get health insurance would be impermissible."

Well isn't that interesting? By any chance, did SCOTUS set a limit on how much the government could "encourage" people?

Regarding auto insurance, is it the case, then, that a state is free to eliminate car insurance mandates? Federal law doesn't touch this area? Are there any states that currently don't require auto insurance? If you don't know the answers off the top of your head (and I don't expect you to), consider these to be rhetorical questions. I can do some internet research to get the answers.

"Auto insurance is not required directly. Instead, there is a prohibition in most or all states against driving or owning a car without auto insurance."

I know what you're getting at, legally speaking, but to me that's still a "direct" requirement. Sounds like a Hobson's choice.

"it would be harder to get all the votes needed if it was, since it only passed by the slimmest of margins as it was."

I know you know that is an understatement. :-)

keatssycamore writes:

Justin asks:

Please correct me if I'm wrong but don't both England and Scotland have very similar health systems? Or are Scotish psychiatrists that bad?

No need to go in on the Scottish psychiatrists mate as it seems there's a relatively simple answer you've not considered:

The pre-existing level of mental illness is higher in Scotland than other parts of the UK. I'd imagine that mostly explains it. I'd have to guess this is also the reason why Alaska/Montana/Wyoming have significantly higher suicide rates than other parts of the US.

Confirmation bias meet kettle. Cherry meet picker. Scottish psychiatrists? As you were.

drobviousso writes:

Wes - There is no deeper rationale. The law is not rational. It just is. There are different school of legal thought that attempt to rationalize it, but none are winning as far as I can tell.

Please note that, for the most part, I'm only making positive statements (as far as my limited understanding goes), not normative ones.

The argument that you'll eventually need medical care was considered by SCOTUS and found lacking. This was due to a bit of reverse induction and the governments failure to provide the court with a line between this concept and the broccoli and GM hypotheticals.

As to emergency and catastrophic cost shifting, I've yet to see a reasonable study that puts the this cost shifting within a few orders of magnitude of a reasonable guess of the cost of Obamacare. It would be much easier, and distort the market much less, to solve the cost shifting problem with a simple usage tax on all medical services. Instead, we have what is literally one of the most complex laws ever passed. The state of PA already has something like this, where they add the tax to moving violations. I have no idea how comprehensive it is or how well it works. I just know I paid it once or twice.

If anyone knows of any analysis counter to this claim, I'd love to read it.

Stevie writes:

Wes,

Your point is well taken. It's true enough that at some time or another, nearly everyone will need (not just want, but need) healthcare services. Another way to put it is to say that healthcare is for the 99% of us, while there might be a very lucky 1% who either never need healthcare or die before they do. :-)

Libertarian-minded individuals like to make individual-centric arguments, which they see as "freedom." But when a libertarian gets hit by a bus and needs medical care, he is like a Wall Street firm that wants to privatize the gains and socialize the losses. Heads I win, tails I win and the taxpayer loses. Just like some banks were too big to fail and they knew they'd get bailed out, our libertarian friend knows that we, the taxpayers, won't let him die.

Libertarians miss the forest for the trees. Some might argue that they don't want healthcare because of the hypothetical possibility that they as an individual won't need it. But in a nation of millions of people it is a certainty that "individuals" will need healthcare services. And just in case our libertarian friend has a bad day, the system needs to be there for him and everyone else. In this way it's a bit like fire departments, as I said in a previous comment. The chances that any individual house will burn down is extremely small, and yet we all pay for fire departments. Even if fire departments were all privatized, we'd still have to pay for them.

Stevie writes:

"If untreated or poorly treated mental health issues are significant factors in suicides, then suicides should continue to be counted against the lifetime expectancy statistics for this argument."

David, I'll agree with that as far as it goes, but I'm thinking that the culture, in particular the degree of alienation in a culture, is a far bigger factor than poor professional treatment for mental health issues. I don't doubt that poorly treated mental health issues can lead to suicides, but I'm guessing that accounts for a small percentage of suicides. I feel the same way when people put most of the blame for problems with kids on public school teachers. There are bad teachers, no doubt, and they do cause some problems, but that is not the main factor.

Stephen writes:

You will have to excuse me as I have only listened to half the podcast so far, but there are a couple of points that are bubbling up in my mind that I would be interested to get some feedback on.

Firstly, I was wondering whether we are getting confused by averages here. After all, if you have a nation that has exceptional healthcare for say 40% of the population and poor healthcare for the other 60%, this nation could surely produce averages of life expectancy, cancer survival etc that actually beat those of a nation where the whole population has access to fairly good healthcare.

Secondly, since innovation and creative destruction are held dearly by in the Austrian economic pantheon (as far as I can tell), what is the attraction of the current system that ties potential innovators so tightly to their corporate jobs instead of launching out on their own to create new businesses and new jobs? Surely there would be more entrepreneurship if people were not so afraid of walking away from their "benefits" thus leaving their families with little or no healthcare?

Thirdly, should it not be surprising that a country that now spends 19% of its GDP (double or more that of equivalent developed countries) has better outcomes? Surely the difference should be much bigger with so much more being spent?

Fourthly, whether the system is better now or not fails to address the issue of future sustainability. Even if a system is currently producing better results, if it is heading for an economic cliff, then how can it really be called a better system? Isn't that like maxing out your credit cards and then gloating over how you have more cool stuff than your neighbors? Sure you may have a moment of glee, but that will soon be forgotten when the bailiffs show up.

Finally, how is it that private healthcare can be available in the UK at a much cheaper rate than the equivalent in the US? The procedures are the same. The private hospitals with their gourmet food are, if anything, swankier. Does that mean that the US providers are price gouging?

Russ Roberts writes:

Lots of great comments. Thanks for joining in. On the not so great side, describing someone's income level or profession or alleged partisan identification is not an educational strategy for disagreeing. If you don't like the conclusions reached by a guest (or the host), try to find some counter-evidence or dispute the quality of the evidence provided by the guest, as many of you have tried to do here rather than dismissing arguments on an ad hominem basis. How poor people or middle-income people fare healthcare-wise in the United States relative to other countries that have even more centralized health care systems is ultimately an empirical question.

On the suicide question, it's an excellent point to suggest that a high suicide rate could be a failure of the health care system and that suicides should not be removed from the life expectancy data. I'd be interested in the impact of suicides is in the adjustment of the life expectancy data. Maybe Scott Atlas can provide some information. I'd also be interested in evidence that access to mental health services reduces the suicide rate.

Stevie, on the issue of how libertarians "miss the forest for the trees" and that they may not claim to want health care, I think you are confusing health care with health insurance. No one is suggesting that you won't want health care. The question is how much does it cost and who pays for it. In a world where some people are unable to afford health care or the level of health care they'd prefer, the gap does not have to be made up by taxation. Charity can play a role as it does even now. The heavy subsidies in the system now drive up the before-subsidy price making health care dramatically more expensive than it otherwise would be. The same is true of education. In both cases, the high price is used to justify the need for subsidy but the subsidy is part of the reason the price is high.

Stevie writes:

My goodness, Stephen, you make entirely too much sense. :-)

You're asking all the right sorts of questions, and even then you're only scratching the surface. That's what gets me a little depressed sometimes, when I allow myself to realize just how little we, as an electorate, even manage to scratch the surface in this nation. Even our hardest-hitting beltway journalists don't ask these sorts of questions of the politicians who are on the respective committees. And even if they did, why should they answer them?

In fact, that was another part of the Brady interview that caught my attention. Brady told Russ that many times a candidate, such as Romney, is better off not addressing certain important issues, such as social issues, because it will only make voters mad and they don't really care anyway. So the strategy is to make up phony stuff, such as "Obama doesn't think entrepreneurs build their own businesses." Or we hear more from the Birther camp. It's depressing!

Greg G writes:

This must be the most thoroughly rejected EconTalk podcast ever. I just read through it again and was struck by a couple more lowlights.

Particularly unimpressive was the suggestion that shopping for healthcare is like shopping for a computer since most people don't really understand how computers work.

And the part where the the uninsured are informed that their lack of coverage is not a crisis.

Also missing was much discussion of how much more we spend to get the good results that we do get. If countries with socialized medicine wanted to ramp up their healthcare spending to the percentage of GDP it takes up here they could fix a whole lot of the complaints about their systems.

Ken writes:

My son lives in British Columbia and he has told me some of the issues he sees. People have to wait for weeks or months for things like heart surgery. Also, when he moved, he had to get on a waiting list for a new doctor.

The MRI example that Trevor brings up is good. Right off people were saying 'just buy more MRI machines', as if the government would not have thought of that idea. Apart from the possibility of other missing ingredients (doctors, facilities,...) there is also the incentive. For a private institution, more capacity increases profit potential. For a government institution, more capacity, increases costs.

The media and politicians have been throwing out the 'US has inferior health care' meme for some time now. It's good to see someone refute that.

Every country is different. Which country best suits your health needs largely depends on what ailment you have.

Justin P writes:

keatssycamore: You are actually making my point for me. Using suicides as a proxy for mental health doesn't work. Like you said if you look at ND vs Georgia, the difference is cultural not health system, its spurious to assume suicides somehow indict the health system. A lot of the negative comments about Atlas' assumption are nit picking that, then saying that it somehow makes everything. He says wrong.
I'm assuming they went into this podcast wanting a more EU style system, didn't like the idea that if you take out suicides and infant deaths, the US system looks great.
Thanks for reinforcing my point. =)

Jim Feehely writes:

Sorry Russ,

This conversation loudly demonstrates the barren narrowness of free market economic dogma - 'a market will fix anything'.

As some of the other comments on this episode say, medicine will never be anything like an economist's mythical 'free market'. The asymmetry of information is only one problem. Are you really claiming that health care decisions of individuals are 'rational' as that term is used by pedlars of the efficient market hypothesis. A person faced with a serious health issue will almost by definition not be in a position to make a rational buying decision.

Public and private health are excellent examples of where a market will not work with any social equity. Look, for example, at one of the catastrophic consequences of Margaret Thatcher's ideological policy of making public hospitals in the British NHS compete. That lead directly to the export of untreatable infection to the rest of the world because competing hospitals suppressed information about the detection of new and scary infectious bacteria so that the carriers were distributed all over the world before the problem became public.

Medicine and public health are perfect examples of where cooperation and collaboration, not competition, will always produce better outcomes.

Public health systems must strive to be effective (not simply efficient) and equitable. No market of which I am aware is the slightest bit concerned with equity. In fact, markets are structurally inequitable because they necessarily assume winners and losers.

You can have your US health system. I will take the Australian system any day because it is demonstrably more equitable.

Economics is not life. And markets are not a solution. Markets are mechanisms that can assist society. But society is not a market. I have not, in 2 years heard such utter nonsense on your otherwise excellent program. This whole discussion demonstrates why economics ought to be downgraded as a guide to social policy.

Regards,
Jim Feehely.

Curt Shrum writes:

I'm not at all sure why suicide and obesity related deaths should not be considered medical outcomes. Suicide is often a mental health issue, and often precipitated by health problems in general, such as fatal and extremely painful diseases. Suicide is often due to financial problems as well, and one of the most prevalent financial disasters is huge medical bills which can bankrupt people. The obesity problem in the US could well be worse than in other countries because countries with government health care have an incentive to be proactive with this problem, whereas in the US people seem to be much more concerned with their 'right to eat themselves to death'.

That Medicaid has worse outcomes than no health insurance at all has a glaringly obvious possible explanation. Poor people with health problems tend to be on Medicaid, whereas many people with means, who can pay for all or part of their medical care, do not have health insurance. This could explain, at least in part, why doctors and hospitals will supposedly provide any and all needed care to those with no insurance, but won't take Medicaid patients at all. Those with means can be made to pay as much as they can, even if they have to liquidate their assets, which may be considerable. Also, poverty itself could contribute to bad health, as well as bad health contribute to poverty. It doesn't make any sense that a doctor would treat a patient who may not pay him/her at all but would turn down a patient who is guaranteed to generate at least some revenue, even if it's not enough to turn a profit.

There is a recent study from Harvard (http://news.harvard.edu/gazette/story/2012/07/expanding-medicaid-to-low-income-adults/) which seems to have been large and well designed which comes to the conclusion that where Medicaid has been extended, in several states, the new groups covered had significantly better outcomes in terms of medical related deaths (6% fewer deaths per year) than the same groups of those in other states who were not extended this new coverage.

It seems to me that you and your guest seem to have seriously prejudiced opinions on this subject, although there is no shortage of prejudice on either side of this issue, and you did point out some possible flaws in the statistics usually cited by the advocates of socialized medicine. Still, it's hard not to come to the conclusion that there was an awful lot of cherry picking of the data in order to reach some of the conclusions proposed on this podcast.

Sebastian writes:

I was excited by a healthcare podcast and somewhat disappointed by the content, in part because unlike many ET episodes it didn't really add much to the discussion.

Things that I would love to know more about(but probably can't since the data isn't out there) presented in bias-revealing form:

a) I think the US is subsidizing other nations by paying extra for drugs and medical equipment. True/False? Magnitude?

b) Medical regulation and the cartel-like nature of the medical profession is a huge part of the spike in medical costs. I don't think I've heard anyone even talk about the supply-side of the problem in any serious manner.

c) How rational is the market? I think we are pumping money into treating the last 2-3 years of peoples lives and comparatively ignoring the first 40-50. More importantly I think we place far too much emphasis on making new things rather than making old things cheaply. At this point US corporations seem so obsessed with creating the next $50,000 a pop drug that we're probably waiting for random university students and/or the Chinese to be the Fords of the medical industry(see cheap blood work kits).

d) Regulation again: the US exports a lot of medical "waste" to the 3rd world where it's put to perfectly good use(various supplies mostly).


I think to a great degree this comes back to patents. Companies are being given weird incentives by their IP monopolies. Why focus on making things cheaply when no one can compete with you on price?


Finally this podcast treated Healthcare systems in a horribly generic manner. There are systems with private healthcare providers(hospitals etc) and national insurance(Canada I think); systems with fully nationalized insurance and providers(NHS, France, etc); and systems with some mixture and variety of national and private providers, national and private insurance schemes(Netherlands comes to mind albeit with universal coverage, US with Medicare but also VA etc). But government, socialism, innovation bleh(these things matter but you hear them on tv every day). An indepth look at comparative healthcare political economy would be fascinating in itself. Heck just compare the Dutch and US hc systems, which are extremely similar post-Obamacare.

Mike Hodder writes:

As a British citizen who worked for many years in the NHS I find this interesting and useful. I am familiar with the views of Wendell Potter and would like to hear a discussion between him and Scott Atlas. That could be quite interesting? The two sets of evidence presented by these gentlemen just do not compare in any way at all. One thing I do feel, and I am not trying to offer any insult, is that listening to Mr Atlas is like listening to an investment banker proposing a complex CDS. That bothers me a bit.

Greg G writes:

Russ

Whether or not access to mental health services reduces the suicide rate is an interesting question but the objections here to the way Scott Atlas uses suicide statistics do not depend on the answer to that question.

There are quite a number of fatal diseases that medical treatment does not significantly slow or prevent death from. We don't remove those from the death statistics. Suicide got special treatment from Scott Atlas precisely because that special treatment produced the result he was looking for.

The same goes for people questioning whether or not suicide rate is a good proxy for overall healthcare. It probably is not that good a proxy for overall healthcare. That still doesn't mean that you get to use a high suicide rate as evidence for the superiority of your healthcare system.

Arne writes:

There is one huge factor of health care costs omitted from the podcast (and from most discussions) - the american legal system.

If I want to insure myself against liability in Europe, the costs are a few thousand US$ (often around 1000). As a breast care specialist, the costs in the US can be higher than 100.000 US$ per year for insurance premiums only. This translates in a salary that has to be at least this amount higher, if I move to the US, and probably some additional compensation for the risk of being sued. So the price of medical treatments could be lowered a lot, by simply installing some additional hurdles for lawyers.

Sri Hari writes:

Russ Roberts,
Another good podcast.
But I have some reservation on how the facts were interpreted by your guest Scott Atlas.

As often described in your podcast, it is foolish to look at the results of a economic activity under different situations and giving empirical value to it.

The outcome of healthcare system of US or Scandinavian countries or UK or Canada - will be different since the variables in each country are not the same. So the very discussion trying to demonstrate US health care system is good or efficient -sounded very lame and defensive.

The measurement should have been- is there any rent seeking activity occurring by the doctors or other service providers. Is the system free enough to have high rate of productivity. So my question to Scott Atlas would be -how does the US system rate on these measurements??

David writes:

Russ,

I think you need to devote a whole episode to the difference between health insurance and health care. Reading all the comments I haven't seen anyone talk about cost, which is the main issue. Does any health care system in the world socialized or not operate without a deficit? Does anyone know how Greece's health system is fairing during austerity?

DevEconHealth writes:

Russ,

While I find the discussion very interesting with well articulated points, working in the Health Economics field and having a bit of training in epidemiology, I do know there are a lot of equally good technical and specific responses and counterpoints to those discussed.
Instead of trying to go into detail here, I was wondering if as a counterpoint in the near future you could get Arron Carroll of The Incidental Economist blog to go on the show and speak on the same issues. Like Scott Atlas he is an MD in the Health Policy field with great expertise and fluency on these matters, and addresses them on his blog though I think he sits on the other side of the aisle.

Would be a great conversation and I would love to see it added to the very informative repertoire of EconTalk episodes.

Thanks.

Pete writes:

I am a bit late to this discussion and admit I haven't read through ever comment yet, but I wanted to make a few points.

1. Echo the comment that suicides should not be removed from the equation. Your brain is an organ just like your heart. We don't remove heart disease, so why remove brain disease, which is ultimately what mental health is? Furthermore, the suicide rate was "far higher" in the US when compared to the other ranked countries. Does this not have direct implications on quality of life? What difference does it make if my hip feels great if life is hopeless? (The suicide data itself brings up another question of material wealth being tied to happiness, which I believe not to be the case. I believe happiness is relative to suffering, but that's another discussion.)

2. Regarding medicare recipients receiving worse healthcare than uninsured... I was excited to hear your skepticism on this Russ. One thing I didn't hear mentioned is that medicare recipients are low income, which is correlated with lower levels of education and free time. Recuperating from surgery or treatment is affected by both. Diet is also absent from the discussion which makes a huge impact as well. While you may control for being a diabetic, you may not control for a high sugar diet.

3. Anecdotaly speaking... my parents lost both their jobs in the 87 crash. We went from top notch health insurance to medicaid (seemingly overnight for a 10 year old...me). We were able to see the same doctors, who spent MORE TIME with us than they would normally because they were required to. When it came to dental care we were more surprised. My brother's braces were covered by medicaid, but not by our insurance. In fact, my parents delayed getting work "on the books" until his treatment was complete because they could not afford to pay out of pocket.

4. Should the government tell us what is healthy? I will refer to last week's episode with Mr. Taubes (one of my personal heroes who tackled some of my favorite topics: corruption, nutrition and cold fusion.)

5. Lastly, I heard no mention of preventive care and access to information.

H Friedman, M.D. writes:

Russ,
Thanks for your continuing series of interesting and thought provoking podcasts.

For the purposes of full disclosure (something rare on the internet) Im a Neuroradiologist, and I know Dr. Scott Atlas personally (we were in the same radiology training program at Northwestern University). He is an intelligent and thoughtful physician, highly regarded in our field.

I think for the purposes of a more though provoking podcast, you should have more strongly challenged him to defend some of his assertions. In particular the issue of Medicaid being worse than no insurance sticks out. While I would never defend the completely inadequate Medicaid reimbursement rates or its bureaucratic deficiencies, there is literature on both sides of this debate, including a recent study covered extensively (NYT http://www.nytimes.com/2011/07/07/health/policy/07medicaid.html, and one of my favorite podcasts Planet Money http://www.npr.org/blogs/money/2012/06/15/155135781/episode-379-does-medicaid-actually-help-people) You may want to consider interviewing Katherine Baicker, a health economist at Harvard, to get another nuanced view of this debate.

I also agree with the previous poster that suicide is associated with a mental health componenet, so completely excluding this from mortality stats is suspect.

Thanks, and the hard work you put in to create a great weekly podcast is appreciated.
HF

Jimmytidey writes:

A little below par?

Normally Econtalk investigates an interesting coutnerargument, and I enjoy it whether I agree or not. This episode is just two guys who agree with each other confirming their beliefs with dubious stats.

Russ's pretend probing amounts to nothing - Surely he'd normally ask more about why no insurance is better than medicade - highly suspicious? Perhaps because most uninsured people are not measured at all? They only appear in the stats if they have the cash to pay for treatment.

No mention of what the US spends on health care ~ 50% more than other developed countries. So even if US health care is as good as the rest, it's still horribly inefficient.

+1 for @DevEconHealth - how about the other side of this argument?

James writes:

I feel bad for Russ. Some commenters seem to be completely misunderstanding this podcast.

David Robson
"How is suicide not related to mental health?"

What a red herring. It is not my job as a doctor to keep Kurt Cobain from killing himself. This discussion was about the curing of clear observable health problems like cancer or a broken hip. Mental health is a whole other ball of wax. Trying to prevent people from making bad decisions (suicide, obesity) runs counter to American culture and most Americans would not regard prevention of those problems as "healthcare". Yes America has a high suicide rate, a high obesity rate, a high imprisonment rate. These are problems, but they are not the topic of the podcast. Whether suicide should be considered a "health care outcome" is a philosophical argument.


DJan
"From the point of an independent observer a health care system that provides good care for everyone (European style) is vastly superior to a system that provides excellent health care for the rich and virtually no health care for those who can't afford it."

Listen to the podcast again. Illegal immigrants with nothing but the shirt on their back can walk into any emergency room in America and get treated, often much more quickly than they would in a socialized system. Read this: http://www.nytimes.com/2012/07/27/nyregion/affordable-care-act-reduces-a-fund-for-the-uninsured.html

Certainly if they are in need of non-critical care they are better off in the U.S. In fact in the American system you are often better off being penniless. It is the people with lots of money but no health insurance who are really screwed.


Chambana
"If you need a good example of a straw man argument for educational purposes, go no further than this podcast.

Atlas uses allegedly pervasive criticism (???) of the US heath care system to set up a sham argument in which he ‘demonstrates’ inferiority of the universal care in other developed countries. Needless to say, no serious health care economist a or public health official criticized the technical/medical/procedural aspect of the US health care industry as inferior. The problem is – of course – the restricted access, excessive costs, and misaligned incentives within the health care industrial complex."

You seem to be unaware that "the U.S. pays twice as much for inferior health care" has become a common headline.

Links for the Google-impaired:

http://articles.latimes.com/2011/oct/18/news/la-heb-us-healthcare-20101018
http://www.reuters.com/article/2010/06/23/us-usa-healthcare-last-idUSTRE65M0SU20100623
http://thinkprogress.org/health/2012/05/03/475941/report-us-spends-more-gets-less-on-health-care-than-other-industrialized-nations/?mobile=nc

That is what Atlas is responding to.

Russ Roberts writes:

I'm enjoying many of the comments but I want to make one thing very clear.

I don't like the current way that health care is structured in America. I think we spend way too much money, much of it unproductive. In a comment to come, I'll sketch out why I think reducing the government's role would be an improvement. But my point is that I have no stake in the status quo. I don't like the status quo. So I have no intellectual stake in the proving how well the current system works. I don't like the current system nor do I think it is sustainable for very long as it is currently structured.

James writes:

Actually, according to Wikipedia the US does not have a particularly high suicide rate:

http://en.wikipedia.org/wiki/List_of_countries_by_suicide_rate

Stephen writes:

I agree with you Russ that the way health care is structured in the US is poor, particularly the government-provided parts. But does that mean that any government-provided solution is therefore going to be poor and suspect?

I have lived in the US, Japan and am from the UK. I have dealt with government bureaucracies in each, and the US bureaucracy has been by far the most obstructive and inefficient. In fact, shockingly so (which gives me some insight into why big government is so detested by free marketers in the US -- what they really dislike is big US government).

I wonder whether a health bureaucracy that ran at the more efficient levels that seem to be possible in similar institutions in other countries could produce more pleasing outcomes?

Stephen writes:

Russ

I would be very interested to hear what you would say to these two comments I made earlier:

Secondly, since innovation and creative destruction are held dearly by in the Austrian economic pantheon (as far as I can tell), what is the attraction of the current system that ties potential innovators so tightly to their corporate jobs instead of launching out on their own to create new businesses and new jobs? Surely there would be more entrepreneurship if people were not so afraid of walking away from their "benefits" thus leaving their families with little or no healthcare?

Thirdly, should it not be surprising that a country that now spends 19% of its GDP (double or more that of equivalent developed countries) has better outcomes? Surely the difference should be much bigger with so much more being spent?

Arnold writes:

Stephen,
what makes you think that "potential innovators" are tied to their corporate jobs? Most of them can probably walk away from their jobs today with a golden parachute that includes healthcare for the whole family. That innovator then walks in to another corporate position where they may have used their previous corporate position to create something innovative to present to a new corporation and thus provide more jobs and business's up and down the supply chain.

Rufus writes:

Sri Hari - - "The measurement should have been- is there any rent seeking activity occurring by the doctors or other service providers."

What we heard and know is happening for sure is rent non-seeking, that is doctors refusing to take new patients for Medicare/Medicaid because the reimbursement rates are below their costs. That seems a pretty significant measurement of the lack of freedom in the system. Either run your business into the ground or stop providing the service.

Floccina writes:

Steve here is some data on life time healthcare spending.

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2645209/

Floccina writes:

@James,
Yes I kept thinking that he meant homicide rate rather that suicide rate.

Kyle writes:

THe longevity study Dr Atlas refers to is in a book "The Business of Health". One of the Amazon reviews critiques the study cited:

"Be aware that one of the most surprising conclusions in the book - that the U.S. healthcare system, when you take away injuries and accidents, actually has the best life expectancy in the world - is based on a very clever twist of the underlying data.

Instead of looking at real-world life expectancy and then taking away deaths due to injury and accident (a method that puts the U.S. at 17th) Mr. Ohsfeldt uses an equation that begins with an ESTIMATE of life expectancy BASED ON GDP PER CAPITA (a measure of how rich the country is), not on actual real-world data. Then he adjusts for injuries and accidents. Of course the U.S. remains at the top of the list, as we are one of the richest countries in the world.

Here's his equation, from the book:
LifeExpit = 50.78 + 3.020 * log(GDPPCit) - 0.077 * [mean(Trans)]
- 0.137 * [mean(Falls)] - 0.133 * [mean(Homicide)]
- 0.0326 * [mean(Suicide)] + year-effectsit

The equation includes "log(GDPPCit)", which is GDP per capita in country i, year t, and it includes factors for transportation accidents, falls, homicide, and suicide. Mr. Ohsfeldt suggests in his writing that the equation and graph are based on real world life expectancy data, which figure into the equation nowhere at all.

I find it deceptive."

Kyle writes:

The longevity study Dr Atlas refers to is in a book "The Business of Health". One of the Amazon reviews critiques the study cited:


“The equation includes "log(GDPPCit)", which is GDP per capita in country i, year t, and it includes factors for transportation accidents, falls, homicide, and suicide. Mr. Ohsfeldt suggests in his writing that the equation and graph are based on real world life expectancy data, which figure into the equation nowhere at all.”

[From: http://www.amazon.com/review/RHM9Z3IN71J21/ref=cm_cr_pr_viewpnt#RHM9Z3IN71J21 --Econlib Ed.]

David Robson writes:

@james -
"What a red herring. [...] Whether suicide should be considered a "health care outcome" is a philosophical argument."

I worded myself more carefully in a later post. Mental health care is part of health care, and if suicide rate is highly correlated with availability of health care, then it was inappropriate for Scott Atlas to remove it.

My critique was that this manipulation of statistics needed to be justified and defended, as it didn't pass the smell test to me.


On an aside, did he remove car accidents and suicides for all countries and reorder them, or just manipulated the number for the US?

Narasimhan Srinivasan writes:

Scott Atlas was referring to patients using Medicaid not being accepted by Doctors. Before 1965 how did people, who are in the same status as those now using Medicaid, accessed health care, when there was not even Medicaid to pay for (however little the payment is)? How the system worked then?

Are there any studies available from prior period before Govt came into the health care system?

Corey writes:

>Thirdly, should it not be surprising that a country that now spends 19% of its GDP (double or more that of equivalent developed countries) has better outcomes? Surely the difference should be much bigger with so much more being spent?

Stephen,

Listeners to Econtalk might be expected to appreciate the increasing cost of marginal improvements in quality. Let me give you a baseball analogy:

Utility infielder A bats .273 while Utility infielder B bats .301 (a 10% superiority). All other statistics i.e. , stolen bases, slugging percentage, errors are equal. I think A may be paid much more than B, maybe twice as much, even though he has only a 10% higher chance of bringing the winning run in from second base with 2 outs in the bottom of the ninth. I guess it is because the owners think that winning a game is so important it is worth the extra money. How much more is a 10% higher 5 year survival in breast cancer worth? I think it is likely worth a lot to those who have just been diagnosed with the disease.


>I appreciate the subject and information. As a physician interested in economics I am heartened to hear these criticisms of National Health Care discussed. Atlas is correct in asserting that Obamacare will do little to control the cost of care. It will control the amount of care delivered.

[Commenter's formatting symbol (a less-than symbol rather than a greater-than symbol) that cut off the display of part of this comment has been fixed--Econlib Ed.]

Ken writes:

@ David Robson -

Do you honestly think that suicide rate is 'highly correlated with the availability of health care'? The causal density for suicide is extremely high.

Life expectancy in general has too many contributing factors to use as a yardstick for the quality of a health care system.

Patient outcome of patients with similar predispositions is the only criteria that could give meaningful results. In such a broken down comparison, establishing a 'winner' would be difficult, but that too is more realistic. Which country would be best for you depends on your own situation.

David Robson writes:

@Ken -
"Do you honestly think that suicide rate is 'highly correlated with the availability of health care'? The causal density for suicide is extremely high."

I don't know. I strongly suspect that suicide is highly correlated to mental illness. I'm not at all sure if access to mental health care lowers the suicide rate. Based on my first assumption, I had to wonder how Scott Atlas decided it was OK to remove. I didn't think it was at all an obvious choice, like murders or car accidents.

I found this on the WHO "Suicide prevention" page:
http://www.who.int/mental_health/prevention/suicide/suicideprevent/en/index.html

There is compelling evidence indicating that adequate prevention and treatment of depression and alcohol and substance abuse can reduce suicide rates, as well as follow-up contact with those who have attempted suicide.

That's enough to make me think the decision to remove suicide from the life expectancy statistics needs to be justified well.

"Life expectancy in general has too many contributing factors to use as a yardstick for the quality of a health care system."

Agreed. At best it's a pointer or a barometer of overall health care. It's certainly not specific enough to make meaningful conclusions, especially considering the relatively small spread between the top countries.

Sebastian writes:

In fairness US suicide rates are actually below a lot of western Europe, I would think it boosts the gap between Europe and the US to remove suicides from the equation(what helps the US is getting rid of car accidents and homicides).

As for the Medicaid vs uninsured debate, was that not a case of adverse selection? I would have liked a lot more detail on how the surveys made sure the groups were the same. Were the uninsured college students with well off parents to support them(which is what would make sense to me)?


Oh and the claim by the guest that 50% of Medicaid patients don't have (or wouldn't have) a doctor is just disingenuous. 50% of doctors don't accept Medicaid patients does not mean 50% of Medicaid patients don't have a doctor. That was just silly, unless he had actual statistics on Medicaid patient registration.

Greg G writes:

@James

Whether suicide should be considered a health care outcome is NOT a philosophical issue. It is a medical and empirical issue. People used to think infectious diseases were caused by evil spirits. That did not make infectious diseases a "philosophical issue."

Mental illness been shown to have very real physiological signatures in the brain and extraordinarily high correlation with suicides.

There are many diseases we don't have effective treatments for. Should we remove all of them from this analysis or only the ones that produce the result we are looking for?

DevEconHealth writes:

Russ, et al.,

I think that Ken's earlier point should not be lost in the shuffle of topics. I think this goes at the heart of the veracity of Dr. Atlas' claims, if true which it seems to be.

In reading his articles rather quickly and going over the interview, it does not seem like he doctor really took an honest approach to this scientific endeavor. Instead he seems to have chosen his conclusion and sought to justify them with whatever evidence he could find or create without trying to take the evidence in the field as a whole into account.

I re-qoute Ken:
"Instead of looking at real-world life expectancy and then taking away deaths due to injury and accident (a method that puts the U.S. at 17th) Mr. Ohsfeldt uses an equation that begins with an ESTIMATE of life expectancy BASED ON GDP PER CAPITA (a measure of how rich the country is), not on actual real-world data. Then he adjusts for injuries and accidents. Of course the U.S. remains at the top of the list, as we are one of the richest countries in the world.

Here's his equation, from the book:
LifeExpit = 50.78 + 3.020 * log(GDPPCit) - 0.077 * [mean(Trans)]
- 0.137 * [mean(Falls)] - 0.133 * [mean(Homicide)]
- 0.0326 * [mean(Suicide)] + year-effectsit

The equation includes "log(GDPPCit)", which is GDP per capita in country i, year t, and it includes factors for transportation accidents, falls, homicide, and suicide. Mr. Ohsfeldt suggests in his writing that the equation and graph are based on real world life expectancy data, which figure into the equation nowhere at all."

DevEconHealth writes:

One last point, which seems to be getting a lot of discussion. While suicides and traffic accidents seem strange for inclussion in healthcare measures from a clinical perspective, this is not true from a Public Health perspective.

A well organized health system with health workers, community health programs, and public health measures can reduce diabetes rates and related diabetes complications in a population through community education, community level health monitoring, good practices and follow-up; all of which can occur outside of the clinical context and outside of the world of the typical MD's daily patient interaction (though they do operate at times to refer patients into such programs/resources). I am pretty sure everyone would agree that this is an aspect of the health system.

In the same way, mental health aspects of a health system can in turn reduce suicide rates through education, health worker support, and basically all the same type of mechanisms for the disease of diabetes (even leaving out the discussion on the role of medication) in the field of Public Health as well as allow for the diagnosing and treatment of underlying mental health issues in most cases.

Vehicle safety as well is dealt with in Public Health as a public health education and intervention issue that can be greatly improved and reduced by the health system. Not to mention the fact that vehicle accident survival rates are greatly correlated to the health system organization (interconnection between emergency call system- first responder system- hospital ED), the training and expertise of first responders (who are part of the health system and represent part of its quality), as well as the skills of doctors and/or surgeons. I can say that survival rates from traffic accident fatalities here in Cambodia are far higher on a per accident basis than those in the States, and that is entirely a result of the health system in this country.

So, in summary, I think its a bit premature to just throw off these indicators as "obviously" not dealing with health because in actual health practice and system planning they form an important aspect of the health system, particularly in the Public Health segment of that system. The fact that Dr. Atlas wants to throw them out of the equation so matter-of-factly without any detailed justification makes me wonder if he has really considered these indicators in detail or perhaps is operating with a "clinical medicine" tunnel vision that does happen to the best of us.

(One final note. Suicide is listed as a disease complication and indicator in relation to mood and depressive disorders the DSM IV-TR and treated similarly in medical field in general. It has its own classifications, risk factors, epidemiology, and treatment. Whether it should or should not be part of the health system may be to some a point of philosophy, but in modern medicine NOW it is a major aspect of mental disease, a fact which any doctor should know and if they fail to follow treatment and referral guidelines for such with their patients they should rightfully have disciplinary action taken against them by their Board. In the same way that high blood pressure is indicative of underlying pathologies and thus requires treatment, so also does suicidal tendencies and the act of suicide, and both are thus used as indicators in the health field of disease in the population.)

Sorry for the lengthy dialogue. I know I said I wouldn't, but I did.
Still loving the show and discussion
DevEconHealth

Mark Egge writes:

Professor Roberts--

I appreciate the careful distinction in this show between having health insurance versus having access to healthcare.

That said, I was disappointed that you let slide the statement by Professor Atlas that "you are really not offering access to medical care if half of [those insured by Medicaid] or more are not even going to be able to get a doctor" (based on his assertion that over half of physicians are not currently accepting new Medicaid patients).

If a Medicaid patient only has access to half as many doctors as a patient with private insurance, this is not at all the same as not having access to any physicians at all. (It may be the case that there are differences in quality, wait time, etc. between physicians who accept Medicaid versus physicians who do not, but this is another matter entirely.)

I second Dr. H Friedman's suggestion above that you consider the findings in the recent study published by the National Bureau of Economic Research on the effect of Medicaid enrollment (http://www.nber.org/papers/w17190). The study finds that, compared to the uninsured, Medicaid participants had higher healthcare utilization, lower debt related to medical bills, and better self-reported physical and mental health. (A future expansion of the study will report on actual outcomes.)

I also second the suggestion that Professor Katherine Baicker would be a great future guest of the program. A show on Medicaid would be thoroughly interesting!

(As an aside: I really enjoy the podcast. That's so much for all the effort you put into getting great guests and conducting insightful and balanced interviews!)

DevEconHealth writes:

On the issue of whether medicaid or no insurance is better, here is a concise summary of the data and what it actually shows with reference to the studies and experts.

http://whatifpost.com/surprise-medicaid-is-better-for-your-health-than-having-no-insurance.htm

HT to Prof. Frakt of The Incidental Economist for the timely link.

Jakob Engblom writes:

Very interesting podcast. Being from the outside, I have always believed that the US health system was both:

* The greatest in the world if you could pay for it

* Very unequal in access

And that statistics like infant mortality and average life span reflected the second point. This podcast, while a bit suspect in terms of motivation, does seem to put the second point in question.

The life expectancy statistics tricks do not sound genuine to me. Suicides are a part of overall health, as is obesity. They reflect on how society as a whole takes care of all its people, and I do not see the US as particularly good in that respect. Also, the idea that the US counts live births differently than places like Canada or western Europe smacks of some kind of anti-abortionist agenda. I simply find it easier to believe that poor people in the US have worse access to prenatal care. And that an overuse of IVF leads to too many triplets and quads that do impact the overall survival of children.

Still, very interesting to hear that even people without medicare or insurance could get access to health care. Basically, that means that the insured sponsor the uninsured... so why not just make it all a tax on everyone and be done with it? It should in the end get cheaper for the current group of insured who seem to pay for most of the system anyway.

Clearly, the US system is great on access to care (if you can afford it). Here in Sweden, I sometimes realize that we would have gotten to specialists much faster in the US. On the other hand, we do not need to worry about whether we are insured or not, and how any existing conditions would impact the cost of insurance. Difficult tradeoff.

I also see how private doctors tend to cater to the wishes of their patients, to the detriment of the greater good. One particular point is the overuse of antibiotics in China, India, France, the US, and other places where private doctors feel a need to keep their customers. This leads to resistant bacteria and viruses, in the end risking the life of many more people. A more rigid system can simply tell people that "no, you won't get pencillin to recover faster from a simple cold, as that would hurt other people in the long run." Sometimes you DO need a pretty dictatorial approach to handle the big issues. Not all things are best handled in little local transcations - externalities are hard to price in, even in healthcare.

Kyle writes:

Actually, here is the better critique from Aaron Carroll at The Incidental Economist.

http://theincidentaleconomist.com/wordpress/how-flawed-is-life-expectancy/

Stephen writes:

Podcast was horrible.

Discussion/comments were a far more enlightening and engaging dialogue.

Russ asked a few good skeptical questions early on, but turned to confirmation bias way too soon.

At least Russ did timidly argue when the guest said,
"I'm not 100% sure why health care is so special, given all these other essential things."
Russ: Well, it is a primal part of us.

James Strong writes:

These healthcare debates usually pit the U.S. as "free market" healthcare vs a subset of european countries that implement government-run UHC programs.


1. There is nothing "free market" about the U.S. healthcare system.

2. I really wish people would go beyond comparing U.S. to Canada and include Singapore and Switzerland in the picture, particularly Singapore, which implements a voucher system.

swernga writes:

I listened to the entire podcast, started reading the book and read all the comments. I have also spent 25 years in the healthcare industry on the supply chain side, trying to help hospitals address the rising costs and utilization of medical devices, supplies and pharmaceuticals.

Just as Dr. Atlas argues in the beginning of his book, a lot of really smart people have definite opinions about the US healthcare system, but very few of them are based on facts or statistics. The comments here support the point Dr. Atlas is making. With the exception of a few posted links, I didn’t see a lot of empirical disagreement. I did see a lot of people attacking the method of data analysis used by Dr. Atlas, and I assume very few people here have spent any time at all reviewing the World Health data and the analytical methods they used to publish their findings. Why would anyone blindly trust their methods?

You need to have car insurance to protect other people from your actions. You want other people to have car insurance so you are protected from their actions. The analogies between health insurance and car insurance don’t work.

I would agree with the inclusion of suicide with respect to life expectancy but not as part of the healthcare system. If someone recognizes they have suicidal tendencies AND they seek the counsel and guidance of a healthcare professional, but they still kill themselves, then you could argue the healthcare system failed them. People who kill themselves in isolation without seeking support have not accessed the healthcare system and cannot be included in the evaluation of the effectiveness of the system.

Most people in the US agree with Russ that the system is broken and needs to be changed, but the quality of care and outcomes is the real focus of this podcast. If the quality is so poor compared with the rest of the world, why do so many people, including Canadians and dignitaries from around the world, come to the US when they need serious healthcare? Why do so many medical students flock to the US to receive their medical training and accreditations?

Vijay writes:

Russ,

Patients on Medicaid do worse than others simply because they belong to a lower socioeconomic stratum. In my 10 years of practice I have never observed any test, procedure or treatment denied to patients on Medicaid. The observational studies your guest mentioned were obviously not controlled as patients with different levels of social support and educational levels were being compared. The real comparison would be to take the same group of uninsured people and put some on Medicaid and then compare their outcomes to the control group. That study has been done and was recently published in the NEJM http://www.nejm.org/doi/full/10.1056/NEJMsa1202099#t=articleTop.

Your guest is quite fond of reminding the listeners that lack of insurance does not mean lack of health care. He implies that they still somehow get reasonable health care (magically?). What a lack of insurance really means to an individual patient is that you will forgo all standard preventive care. If you do get really sick, you will get world class healthcare and then be made to pay twice the rate that insured patients pay. If that bankrupts you, which it very often does, you may now be eligible for Medicaid! Health insurance as it currently exists is not just insurance. It is a means by which patients can assure that they (through their insurance) are liable for the actual cost of the care received, and not the inflated rates that hospitals bill them for. By expanding insurance coverage, the government ensures that a serious illness does not automatically lead to financial ruin.

Jakob Engblom writes:

One more comment on the good outcomes in the US and access to things like hip replacement. It is really what you would expect...

* The US has a higher GDP per capita than all large countries in Western Europe.

* The US income is a bit more unequal

I.e., the US has a large population with comparatively high incomes and purchasing power compared to Europe

* In other fields such as automobiles and electronics, this income advantage manifests itself as a higher consumption of higher-end goods.

* In healthcare, you would expect to see the same as there is some market aspect to it. Richer people should mean more expensive drugs and treatments are used, simply because on average, more people can afford it.

Thus, Western Europe do fewer expensive things simply because there just isn't as much money thrown at healthcare. Just like in other areas, the difference in income makes a difference in consumption. There is nothing market vs socialist vs voucher at play, it is mostly a matter of how much money is thrown at the problem.

The other question when comparing countries is how efficiently the money is used, and that is a much harder question to answer in an unbiased way.

Sebastian writes:

@swernga: I'm not sure why methodological disagreements are somehow less valid than empirical ones?

Here's my methodology for calculating life expectancy: ignore all deaths that happen on US soil, now calculate life expectancy. Suddenly we're expecting to see Honest Abe walking about DC, since "officially" he didn't die.

Atlas' methodology was suspect. He did things with the data that don't make sense, therefore it is perfectly valid to criticize him for it.

For instance, there are plenty of things that health providers can do to combat suicide: for instance ad campaigns encouraging people to seek help, providing crisis hotlines, educational initiatives in schools to teach people early on what the warning signs are and so on. This is all stuff that can be done well before someone has a gun in their hands.


Finally, note how virtually no one attacked his claims about child mortality, since a)it's harder to do(given the nature of the claims) and b) they actually seemed reasonable.

Greg G writes:

Swernga writes:

"People who kill themselves in isolation without seeking support have not accessed the healthcare system and cannot be included in the evaluation of the effectiveness of the system."

Many, many people die of first heart attacks and strokes without ever having realizing they had a developing disease. They die without ever having "accessed the healthcare system" to treat the disease they did not know was developing.

Does this mean these strokes and heart attacks should be removed from consideration by Dr. Atlas when comparing mortality rates? If not, you are being inconsistent. If so, I'm pretty sure he didn't so that.

MichaelM writes:

People arguing about public awareness campaigns for mental health issues as part of the health care system are conflating this system with the wider social model. It's perfectly possible for non-health care related organizations to run these campaigns.

This is like saying its unjustified to remove car accidents with immediate deaths because a good public transportation system would reduce car accident deaths.

Really I'm coming to absolutely loathe every form of over-aggregated data. Is like to see an international study of incidence rates of various illnesses and injuries, times from diagnosis/injury to treatment, and time from treatment to recovery. Too much research these days is done with over-wrought, extremely aggregated '''empirical''' work, and very little is done with actual empirical study.

That was the nice thing about the Coase podcast. He got this. Essentially no economist goes out and does what Menger did and studies actual economic agents in the real world anymore. I learned more about price rigidity in a few months working retail than a few semesters of macro could have taught me.

DevEconHealth writes:

@MichaelM
Actually, I think you are getting a bit confused on what public health is, which is an aspect of the health system.
As it is commonly defined:
"The science and art of preventing disease, prolonging life and promoting health through the organized efforts and informed choices of society, organizations, public and private, communities and individuals".

Public health promotes the education on health and health related safety and the reduction of avoidable negative health effects. Whether or not you have your public health campaign run by a hospital or an ad agency does change the fact its an organized health campaign. Additionally, if it is brining awareness of signs of suicidal tendencies for people to seek help or proper infant positioning to prevent SIDS, it is still in the health system arena and can be measured by indicators.

With this analogy, I think it is sounding a bit like a straw man though I see the point you are trying to get at. It is a false analogy as public transportation is NOT a health or safety intervention though it would have some of those positive externalities (not to mention possible effects on obesity rates). Seat belt use and education is considered a public health intervention as well as child safety seats and airbags. Their purpose is safety and they were implemented in response to epidemiological studies on the type and incidence of injury rates seen in the population. Public health failing to get people to take on healthy behavior with safety belts are no different than failing to get women have regular breast cancer screenings, people to stop smoking, or individuals to follow preventative measures to avoid complications during diabetes. I know from the outside it may seem there is no obvious line, but on the ground we do see it though of course there are grey areas as for all fields and topics especially in economics.

As for the over-aggregated data point, fair enough, but this really is just a general indicator that is used by those who quickly want to do a comparison without needing to go into the field in detail. For more detailed comparative indicators you can start with the OECD health system indicators (http://www.oecd-ilibrary.org/social-issues-migration-health/health-at-a-glance-2011_health_glance-2011-en), in which the US does ok in a few indicators but for the most part ranks badly in many indicators such as post-operative sepsis. While you can say the US is better than one or another in this or that indicator, as a whole we rank rather badly and we definitely rank nowhere near the top in terms of the best health systems.

Even these OECD indicators are rather basic and for complete detailed analysis I would recommend the detailed discussions and critique of data and research on The Incidental Economist. There is real world analysis in the health sector and a lot of us are doing it. But just because the interviewee chose not to delve into it and stick to the Health Econ 010 statistics should not be held against the field itself. There is a lot more than gross aggregates, but it will take a lot of time and learning to understand and delve into them; something that cannot be accomplished in one hour of discussion.

Ken writes:

@DevEconHealth I just went to the OECD site and it looks like it's a bunch of measures that would have too many non-healthcare related contributing factors to be valuable as a metrics. They look like they could be useful for comparing genetic and cultural aspects of health, but not health care.

Greg G wrote:

Many, many people die of first heart attacks and strokes without ever having realizing they had a developing disease. They die without ever having "accessed the healthcare system" to treat the disease they did not know was developing.

It's not that they did not 'access the healthcare system'. It's that they did not 'present with symptoms'. They could of course be screened, but from what I've read most men above the age of 25 have significant blockages.

Does this mean these strokes and heart attacks should be removed from consideration by Dr. Atlas when comparing mortality rates? If not, you are being inconsistent. If so, I'm pretty sure he didn't so that.
He should not have made any comparison at all, since his position was that mortality rates is not a good metric. I'm surprised that anyone even has to point out that mortality rates is a poor metric for measuring health care. It's like trying to decide when it's time to change the oil in someone else's car based on how dirty their windows are.

We spend a ton of money on healthcare in the US. 50% of it is currently spent by the government. The way we pay (through premiums and with little exposure to price in the interaction) removes any incentive to carefully choose how much to consume. I'm of the opinion that we over-consume healthcare.

What really worries me is that at a time when medical research suggests we should be moving towards personalized health care, we are likely to move instead towards 'standardized health care'. Also, I believe that surgical techniques have progressed more quickly than medical treatments due to less stringent restrictions on the former.

EurMik writes:

Sorry for repeating what others have already discussed/commented (but with some new points) on a podcast that was not really at the usual (high) level of Econtalk-quality, but…

1. I had a very hard time understanding how the adjustment with respect to suicides and high-speed road fatalities could lead to the stated effect of placing US life expectancy at #1. This considering that, according to OECD Health Data, several of the countries with higher life expectancy compared to US also have higher rates of suicide (even though most have lower road fatality rates). But, apparently the adjustment is based on predictions from a model (which seems far from convincing) and not on actual data. This makes the argument rather weak (even though I do not quite see the point of the argument, anyhow).

2. It is difficult to accept the exclusion of both suicides and road-fatalities from the calculations of life-expectancy. Mental illness is the leading burden of illness in the developed world and arguing that (preventive) mental health care should not be seen as part of the health care system is, I argue, an outdated argument. Further, the survival of a high-speed road crash will (I would assume) also be at least somewhat related to the health care quality (e.g. access to quick helicopter/ambulance services, quality of hospital surgeons, other types of health care resources etc.). There are other causes of mortality, if we are to play that game, that seems more relevant to exclude in order to create a life-expectancy that is a better indicator of health care quality.

3. There is a slightly irrelevant, I argue, discussion about testing for statistical significance in life expectancy differences across countries. We could have a philosophical debate about hyper-populations and so on. But, life expectancy data is not (at least in the developed world, as far as I know) based on samples of the population. It is based on actual (register data) of the full population (more or less). So any difference in observed life expectancy between countries is also a statistically significant difference.

4. I think few serious academics or policy “wonks” argue that health care quality in the US (for the average person) is inferior to the quality in other western countries (even though I don't really know what the domestic US debate looks like). I would say that most serious people believe the oppoiste; quality is higher in the US.

I mean, looking at PPP adjusted spending per capita in US dollars, the US spends 8,232/capita, which is about 55% more than a richer country such as Norway and a slightly less rich country such as Switzerland (5387 and 5296) and about 90 to 140% more than less rich countries such as the UK and Germany (3433 and 4338).

So it would be very surprising if health care is not of higher quality in the US. But, the question is of course: Is it really worth it?

Lucy writes:

I usually post as Rufus, but since my other dog Lucy just died yesterday, this post is in memory of her.

It also has an economic story. As we try to compare % of GDP spent in different countries on healthcare, it really isn't comparable. For example, I'm almost certain the US is at the very top, perhaps extremely so, in paying for the healthcare of our dogs and cats. (I think Europeans may spend a lot on their pets, but still, they have nothing like PetSmart or other big box retail stores dedicated to pets). Considering vetrinary services are effectively a free market, what explains our willingness to spend so much money on our pets? If we had a more free market in healthcare, who is to say that we would not end up spending the same % of GDP simply because that is our cultural preference?

Making the choice to spend money in one way simply means it cannot be used for some other purpose. It does not mean we're paying too much, or, more accurately, it would not if people could not vote to appropriate other person's wealth.

Instead, if we are spending at an unstainable rate, then we are simply going to face difficult decisions down the road. At least with my dog, the decision came down to what I felt was the ethical choice and within my own financial means. That's what's missing from the bureacracy of Medicare and other government systems. They set arbitrary standards of treatment and quality that encroach on our liberty in the name of the "public good", when really there is no such thing.

Trevor C writes:

Stevie - I am not specifically talking about how many procedures are done per hour with each machine, I don't know the number. It's more of an issue that it takes a long time to get a reference, the guy that sits beside me waited months also getting a time that works is an issue and even longer to get in. I am fairly certain that they do not run at night since none of the people I know that have had the procedure were given the option (I know a few).
I do know of one person that have had the procedure privately late in the evening so the private MRIs may go all night.

From his first Dr appointment until he had his meniscus repaired my friend waited 18 months.

I guess I should have used the word ineffective not inefficient when describing the system in general.

Pete writes:

Probably the best argument for US healthcare is that America even ranks as high as #17 with our patently detrimental diet and lifestyle.

Also, the very definition of mental illness means that you probably don't know you have it. I'm too crazy to know I'm crazy.

Want to fix health (not healthcare)? Make it more profitable to keep people healthy than sick. If Merck makes $X on every pill it sells, what's the incentive to stop selling pills?

Jason writes:

While I appreciated this podcast and it challenged my underlying assumptions I did get the sense Scott comes at it from a bit of an ideological place. A few issues in here are that:

As mentioned earlier the high rate of suicide is atleast partly connected to the mental health system in this country and that should be more highly considered.

More importantly the type of health care Scott seems to be talking about is treatment when one is sick for cancer, heart disease etc... He completely leaves out the importance of preventative care which MA is moving toward with a recent bill that was passed. To me that is directly connected. Health care is about so much more than how we treat someone AFTER they are sick. More focus needs to go into prevention in the first place.

Another issue is that the interviewee thinks people will make rational choices in the marketplace. He discounts the issue that many people will not consider the possibility of a catastrophic illness and therefore someone who is seemingly healthy won't buy enough insurance. This belief that people act rationally in the marketplace is a farce and unfortunately people do make the wrong choices on these types of things far to often. All you have to look at is how poorly people with 401k's have saved for retirement and you can understand that there are some things where we make poor choices.

One of the biggest issues is that the interviewee was not challenged about his attack on medicaid and forced to back it up. A recent study in Oregon brings into question his assertions that medicaid is a failed program. http://www.nytimes.com/2012/06/23/health/oregon-study-reveals-benefits-and-costs-of-insuring-the-uninsured.html?src=me&ref=general&pagewanted=all

The interviewee also ignores how people without health insurance will be less likely to go for treatment until things are very bad.

Finally, I don't have the numbers but it would be interesting to look at the difference in health care outcomes for those from socioeconomic and racial backgrounds. That felt like it was mostly ignored or assertions in the podcast were not really backed up. As someone else said, this felt like a podcast with two well of white guys who don't have an understanding of what people of color go through in this country.

Eva writes:

The understanding of "health care system" presented here seems to be one that excludes 1. prevention and 2. mental health.

Saying that our healthcare system isn't as bad as it seems, it's just that we have more sick people, seems odd to me.

In response to commenter drobviousso, I would like to point out that the cause of suicide lying with the individual is at least to a degree incorrect insofar as it is most often the result of depression or other mental health issues - which are preventable and treatable. Assigning "responsibility" to the individual rather misses the point.

Some interesting work has been done on the "economics" of suicide and it seems plausible to frame it as a choice in the sense that there are "benefits" and "costs" that can be weighed (although I would argue that the underlying mental health issue often skews valuations of outcomes, leading to a choice that, in a state of mental health, the individual would not choose to make).

But putting the onus on the individual in this way, ironically maybe if referring to drobviousso's argument, actually emphasises the role prevention plays in framing choices and affecting outcomes.

I do not subscribe to the view that "doing nothing" (not working towards better suicide prevention) in a world where there already are existing structures that frame choices is a neutral option - it merely re-inforces the status quo, which is in itself a choice (at the social level).

It seems a convoluted conclusion to say that 1. most people are covered already (with most of the presumed uninsured covered by Medicare), 2. Medicare etc. provide awful coverage, 3. Medicare should not be extended. The more logical argument, if one were aiming at improving outcomes, would be to either reform Medicare etc. to ensure adequate treatment is available to all, or to come up with a new system that does better.


Finally, I would like to invite the guest to reflect on his use of the term "civilized country", which I am spelling with a "z" out of respect for his probable American cultural background.

PBT writes:

Re the lifetime data...I'm sceptical that somehow the USA is a more suicidal place than Western Europe. I'm also highly sceptical that the US driver is more dangerous. I mean a US
55/65 mph speed limit vs. driving on the Autobahn or Autosrade...sorry this just doesn't pass the smell test.

Pete writes:

@PBT

Skeptical based on what? At least sight some evidence. I could easily say Americans drive more because outside of the major metropolitan areas, you HAVE to drive everywhere. Europe is much more geographically centralized. Even lumping all of Western Europe together is suspect as driving in Germany is much different than in France, which is different than Poland. There are just too many factors to be so speculative. You have to rely on what we can actually measure.

Ken writes:

@ PBT

If you believe that car deaths are a good measure of how dangerous the driving is, then why do you think it should be used to measure the quality of health care?

According to actuaries, age is a big factor in dangerous driving and car deaths. Would you use a comparison of car deaths to show that the population in one country is younger than another? That is the mistake people make when they try to use life expectancy as a measure of health care.

John Berg writes:

A stimulating and exciting podcast and set of comments, all done without the final, written form of Obamacare which some estimate to exceed 15,000 pages of regulations.

John Berg

Robert Wiblin writes:

I also came to say that excluding all suicides is an embarrassing howler. Many or most of those suicides would be averted with a better mental health system. What you might reasonably exclude are the extra 'successful' suicide attempts that occur because people have easier access to guns in the USA.

Instantly fatal car accidents could be excluded, but not obviously those where the person makes it to hospital alive.

What I would like to have heard was whether the US is getting good value for money from its extravagant healthcare access. How much more do you have to pay to get an extra year of life through medicine, relative to getting an extra year of life from better gun control, diet, exercise, traffic regulation, etc?

David Blair writes:


Hello Russ:

Even though I can't say I agree with much of what your guest had to say, it was, as usual, an informative and well-structured presentation. I'd love to see a follow-up episode treating possible bottom-up, evidence-based solutions using the models discussed in these links:

http://www.annfammed.org/content/10/3/250.long

http://www.newyorker.com/reporting/2011/01/24/110124fa_fact_gawande?currentPage=2

Again, thanks for your work here. You've made Monday morning one of my favorite parts of the week.

All the best,
David Blair

Samuel Metz writes:

Dr. Atlas’s opinions on American health care certainly qualify as unconventional. They discount, dismiss, or ignore data from institutions (the Organization for Economic Cooperation and Development, the Commonwealth Fund, etc.) and journals (New England Journal of Medicine, Health Affairs, Journal of the American Medical Association, etc.) that other writers consider credible.

His opinion that America’s is the world’s best health care system overlooks statistics demonstrating otherwise. America ranks at or near the bottom in maternal mortality (there 40 other countries where a pregnant woman and her baby have a greater chance of surviving the pregnancy), foot amputations per 100,000 diabetics (we amputate three times as many necrotic feet as the OECD average), lives lost to treatable diseases (these 44,000 Americans would have survived had they lived in another OECD country), and medical bankruptcies (absent in other industrialized countries – a diagnosis of lung cancer in the US carries a 7% chance of bankruptcy within five years).

Dr. Atlas discounts our meager life expectancy by blaming lifestyle considerations. However, torturing data is this fashion always compels the data to confess; usually the confession is exactly what the interrogator wants to hear, not what is true. Dispassionate analysis of our relative ranking in life expectancy after correcting for smoking, obesity, traffic fatalities, race, and homicides makes no difference in our relative ranking.

He is correct the US defines “live birth” differently from many other WHO countries. This makes our world ranking for infant mortality and life expectancy at birth look disgraceful. However, when we compare US infant mortality only to those countries with the same definition of live birth, we still retain the highest infant mortality rate among our peers. It is still disgraceful.

He implies American neonatologists make humanitarian (and usually futile) efforts to save premature or underweight neonates, efforts not made in other countries. If so, we expect a lower American life expectancy at birth than these other countries. This is true. However, after these desperately ill neonates die young, we expect a rise in comparative life expectancy at ages five years and above. This does not happen. The life expectancy ranking of the US compared to other countries does not change at any age from birth to 60 years old. Curiously, at age 65 our rank rises from 30-35th until we are 14th at age 85. This may suggest an effect of Medicare on health care access.

His claim that Americans enjoy universal access to health care lacks substantiation. Most data corroborate that financial barriers account for our last place ranking in the above and other public health endpoints. If the Center for Disease Control and Prevention included “insufficient access to care” as a cause of death, it would rank #10. Unlike US citizens, no one in other industrialized countries is denied care because they lack money.

As an economist, Dr. Atlas is aware no “free service” is ever “free.” “Free” emergency room care is paid by higher insurance premiums, higher cost-sharing, and higher taxes. Physicians like Dr. Atlas and myself are frequently compelled to work without reimbursement. It is no more justifiable to compel physicians to provide free services than to compel insurance companies to sell money-losing insurance policies. All this “free” care makes the US the most expensive nation in the world for health care.

We should note that “free” care in emergency rooms is only for urgent care of sudden illness and acute complications of chronic illness. No one gets primary or preventative care, free or otherwise, in emergency rooms.

Dr. Atlas does not address the question of patient choice. Most Americans obtain private insurance through their employer. Their employer chooses the insurance company, the insurance company chooses the physicians, and the rest of us do what we’re told. Allowing patients to choose insurance companies does not permit them to choose providers, unlike patients in other industrialized countries.

Dr. Atlas expects enhanced competition among insurance companies to reduce cost and improve public health. He may be right. However, this is hypothetical. Other industrialized countries provide better care to more people for less money than we do, and none use free market private insurance companies as Dr. Atlas proposes. If we contemplate radical changes to promote better outcomes at lower prices, we might model ourselves on working examples rather than hypothetical examples generated by unproven hypotheses.

As an aside, European health insurance companies do compete and do provide better care at lower prices than we do. However, they operate under rules American insurance companies find unacceptable. (1) An insurance company can charge whatever it wishes, but it must sell the policy at the same price to every patient regardless of sickness or health. (2) Insurance companies cannot refuse a policy to anyone, sick or healthy. (3) Insurance companies cannot drop a patient from coverage for any reason. (4) Every policy must cover every treatable disease, so no patient can inadvertently pick a policy that risks bankruptcy or death if they acquire a treatable disease. (5) If the government finds an insurance company has still managed to cherry-pick healthier patients, the company pays a premium which subsidizes companies with sicker patients.

As a result of these rules, European insurance companies compete with lower premium prices, added benefits, and improved customer care.

American insurance companies have no such restrictions. Therefore, they compete with higher prices to discourage sick patients from purchasing policies; with restricted benefits to reduce financial liability; with increased patient cost-sharing via higher deductibles and co-pays; and with delayed or denied provider payments. These strategies reflect the simple fact that no business can make money selling adequate insurance policies at affordable prices to people who are sick.

Allowing insurance companies to sell across state lines, to set whatever prices they care to, to limit benefits, and to continue denying 30% of all first claims are hardly likely to provide better health care outcomes at lower overall cost.

The contention that cost-sharing patients make smart consumers of health care has been refuted by almost every study conducted since the original 1984 RAND study. Cost-sharing reduces health and increases costs. Dr. Atlas fears customers will consume unsustainable amounts of health care if not restrained by cost sharing. This is partially true in that citizens of other industrialized countries with little or no impediments to health care see their physicians two to four times as frequently as US patients, and spend more time in the hospital than US patients. Clearly they are indeed consuming more primary care than Americans. But these non-US patients spend half as much as we do and enjoy better outcomes.

One reason why Europeans consume more care but spend less is that patients rarely demand the most expensive care: They do not cry out to be placed on ventilators for a cough, to spend a week in the ICU for a simple infection, or to have their chest opened and their coronaries bypassed in the absence of disease. It would appear that allowing patients to consume as much (inexpensive) primary and preventative care as they want greatly reduces their subsequent need for (very expensive) intensive and emergent care. At least that is suggested by experiences in every other industrialized country.

As can be seen, health care does not follow conventional rules of supply-demand economics. As costs of health care increase, needs for health care do not change. In contrast to flat screen televisions, everyone has a minimal need for health care. Once that need is addressed, there is little incentive to consume more. Few people look forward to painful procedures or hospital food. Whatever Maynard Keynes and Friedrich Hayek said about free markets, little of it applies to someone with a sick child.


Like many of those unfamiliar with the vast variety of non-US health care systems, Dr. Atlas labels all of them as “government-run.” Few of them are. Many use formats that superficially resemble ours, with insurance mandates, two tier systems, fee-for-service, capitated payments, salaried physicians, etc. Government involvement runs from complete (Cuba) to merely enforcing business rules (most of central Europe). However, what all successful health care systems in the world (including a few in the US) have in common are: (1) Everyone is included, with no discrimination against the sick, (2) Healthcare is encouraged with minimal cost-sharing, and (3) Financing is performed by publicly accountable, transparent, not-for-profit agencies. In most countries you can make a profit providing healthcare, but we are the only country that depends upon profit-making to finance healthcare.

Dr. Roberts made the most important comment during the interview. He stated universal, equal access to health care is a worthwhile goal. That is corroborated by the experience of every successful system in the industrialized world.

Dr. Atlas’s thoughts on the American healthcare system are certainly interesting, controversial, and thought-provoking. However, they should not be mistaken as mainstream or, on some points, credible.

Samuel Metz, MD
Portland OR

(please see www.samuelmetz.com/1204253q9a.htm for references)


Phil Langton writes:

My habit is to listen to econtalk podcasts as I cycle to or from work, a journey of about an hour. Generally, I am entertained and learn something. Generally, but not this time.

I urge my students to be sceptical but Professor Atlas is only sceptical, it seems, of data that is inconvenient to his prefered argument. I was astonished that he thought disregarding suicide was acceptable as concerns over money and health are very prominent motives (alongside relationships and mental health) evidenced in suicide notes.

So, sceptical as I was within the first 15 minutes, I was floored by his assertion that one is better off without medical insurance (in the US). I would be very interested to see the data on suicides for individuals who have for years paid for health insurance only to find their claims denied and/or seen their premiums rapidly rise to the point they can no longer afford the premiums.

The last part that I took issue with was the assertion that the patient should be the arbiter of their treatment. There is good evidence that when clincians become patients they prefer NOT to make treatment decisions, especially in areas outside of their own expertise, but wish to defer to 'expert opion'. I've also read with interest that clinicans who have a financial interest in diagnostic equipment or diagnostic labs, order far more 'revenue creating' tests (several fold more) than clinicans who have not such interest (source - Practical Wisdom, Barry Schwartz).

This was one of the intellectually weakest arguments that I've had to suffer through in the two years of avid listening to econtalk. I would love to know if Prof. Atlas enjoys any financial support from companies providing (or lobbying for) private health insurance in the US.

txslr writes:

Did no one notice that:

1. The study that left out suicide was not acutally done by Dr. Atlas. This was not his choice, he is merely citing a study done by someone else.
2. Suicide rates in the U.S. are well BELOW the OECD averaage. While not dispositive, this suggests that leaving out suicide would improve the U.S. position versus its peers on life expectancy.
3. The study that Dr. Atlas quotes excluded suicide, auto fatalities AND HOMOCIDES. Suicide is not likely to skew the data much, certainly not in the direction and to the extent that most commenters seem to think.

Given that U.S. suicide rates are well below OECD averages, would the critics among the posters be willing to concede that U.S. mental health care is better than average?

I didn't think so.

dave writes:

I found the speakers arguments to be weak, and as others have commented, it sounded as if he had an agenda to prove, rather then evidence to discuss.

Many of the comments above touched on thoughts I had, but there is more to get off my chest.

Firstly, I was astounded that the high numbers of obesity in the US were used as an excuse for the poor life expectancy numbers, rather then an indictment of it. In countries with more socialized health care systems there are incentives for public outreach and policy to encourage money for sport and other physical fitness activities. This is PART of our health care system, and why we are less obese and live longer.

On infant mortality your speaker's data may have been out of date. At least in regards to Canada. In Canada the way we measure live births was brought closer to the American style way back in 1993 and our infant mortality rates, while higher then before 1993, are still lower then the US. This is most often cited as being the result of prenatal classes and more doctors visits during the pregnancy of Canadian women vs US.

I also heartily agree with one of the early comments that part of the high cancer survival rates might be due to over treatment in profit motivated clinics. Consider the fact that routine PSA screenings are no longer recommended because they do more harm then good.

If 100 people were incorrectly "diagnosed" from a bad PSA test, and 20 of them died (from the cancer or the treatment), it would be reported as a 80% survival rate, even if had they been left alone they all would have lived!

I am a Canadian and yes the vast majority of us love our health care system. And no we are not heading off a financial cliff. Far from it in fact. No only did we weather the recent financial crisis pretty much better then every other nation, it has also just been reported that on average Canadians are richer then Americans! (net worth per capita).

So to all those that say socialized health care will drive a country into financial ruin, consider that as strong evidence to the contrary.

The stories of our waiting lists make me laugh the most. I once hit my head in a floor hockey game (yes hockey). A couple days later I was feeling funny so after work I walked into the nearest medical center (not an emergency room).

I saw a doctor after waiting about 30 minutes. She conducted a brief neurological exam (which I failed). She booked me for an MRI the next morning. Immediately after my MRI I was instructed to go to an office across the street where I meet with another doctor who discussed the results and I learned all about what a concussion is.

The whole episode was less then 24 hours! And this is not unusual. The care that me and my family receive has always been prompted and excellent.

Yes there are some waiting lists, often localized due to temporary staffing issues, or because the resources are being used for life-or-death cases first. Private health care interests in Canada tend to make a big noise about it.

For me it come down to a basic human ethic. We in the west are more then rich enough to care for our sick and our elderly. Health care should be provisioned in the same manner as fire safety and civil policing. Without care for your race, religion, or financial status. Because we can, and because its right.

BTW: One subject I would love to learn more about is how much US medical technology comes from Veterans Affairs hospitals? Is there a hidden subsidy going on there?

J writes:

For those who support a system of health care that is more free market in its approach, I wonder, are there are currently countries that provide a model that can be examined?

Thanks,

J

Jen writes:

Dr Atlas criticises critics of the US health care system for over-simplifying and sensationalising but he is no less simplistic when discussing other health systems. Universal health care or universal insurance is not the same as 'government-run' and does not mean that 'the government decides what health care I get'. His descriptions of the problems associated with universal health care systems certainly do not reflect the situation in Australia (where I live).

Also, agree with the comments above regarding the decision to exclude suicide as though it had nothing to do with health care. It is similarly wrong to measure the efficacy of a health system the way he does, e.g. by looking at life expectancy for a specific diagnosis. the most important indicator of life expectancy for most cancers is stage at diagnosis so a better indicator of a health system's effectiveness is to assess what proportion of treatable cancers are identified and at what stage. Basically, Dr Atlas makes a case for the US providing good emergency care and cancer treatment - these are definitely important parts of the health system but there is much more to health care which Dr Atlas appears to ignore.

Perhaps Dr Roberts could look for a guest next time who has a better knowledge of European and other health systems and who can offer a more informed and less ideologically-driven comparison.

txslr writes:

Sure mental healthcare has some impact on suicide rates. Would anyone care to tell me how much of an impact it has? Please explain how you came up with your answer.

I’ll wait.

Hint: The suicide rate in South Korea is about 10 times higher than in Greece. How much of this difference is explained by the relative quality of mental healthcare in those two countries? An answer correct to 2 significant digits will suffice.

Greg G writes:

txsir,

No one is arguing that the suicide rate is a good proxy for the quality of the overall healthcare system. Many, if not most, individual causes of death are not a good proxy for the quality of the overall health system. Dr. Atlas chose to single out suicide for special treatment because he was looking for a particular result from the start. That is the point.

Consider heart disease. The U.S. has higher death rates from heart disease than many third world countries. That doesn't mean our medical treatments for heart disease are worse than third world countries. That doesn't mean we can calculate how much of the death rate for heart disease is due to medical care. The causes of any given disease are often extremely complex.

The measures that Dr. Atlas decided to use in his analysis were "his choice."

Interesting that last week's podcast is still attracting more comments than this week's a week later.

Sebastian writes:

At some point patient choice in treatment was compared to shopping for a computer. That bothered me at the time but I couldn't put my finger on the problem. I think it's this:

When shopping for a computer no one gives a hoot about how it works, what we care about are the outcomes of the purchase: will it run X, Y software, how heavy is it, how does it look, what's the user experience. You can literally go from knowing nothing about computers to picking the right one for you in 30 minutes flat in a shop with a decent sales rep(if you had a list of software that you cared about in advance).

Where is the conclusive on-line list of outcomes for taking drug X for condition Y(for some drugs you don't even have access to decent 3rd party studies, let alone easily understandable infographics)?

Cause and effect in medicine is probably the fuzziest relationship to determine outside of the social sciences. Why don't the tools to help patients make hard decisions exist now? Don't insurance companies already have that incentive? Do patients even want this responsibility?

txslr writes:

Greg G.

People here are arguing that suicide should be a piece of any good metric for quality of heath care. Let me recap.

Dr. Atlas says that longevity is a poor metric for the quality of health care. As an illustration of the problem he points to a study done by two other health researchers who estimate that, if you were to leave out deaths by suicide, homicide and automobile accidents, the U.S. would jump to number one in the world in longevity. Since there are legitimate questions regarding the degree of impact of the health care system on deaths by suicide, homicide and automobile accidents; causal density and comparability across nations, including them in a measure of the quality of health care (i.e. longevity) is highly questionable.

Many of the responses here have run to: "How dare Dr. Atlas leave out suicide, which is impacted (at least on the margin) by mental health care? He must be biased!"

Answer: He didn't. The researchers who did the analysis he quoted did. I don't know why. Perhaps because, while there is some health care impact on death by suicide, it is likely swamped by other factors, such as those that result in a suicide rate in South Korea that is 10 times that in Greece.  Perhaps if Dr. Atlas had done the study he would have left it in, but then, since the suicide rate in the U.S. is below average for OECD countries, it is not clear that including it would have undercut his "bias". Perhaps someone would have accused him of bias for taking out homicide and auto fatalities (which removals make the U.S. look better) while leaving in suicide!

Regardless, Dr. Atlas' point stands - if a metric for the quality of a nation's healthcare system jumps from below average to the best in the world by taking out deaths by suicide, homicide and auto accident, one has to question the quality of the metric.

Yossarian writes:

Great podcast, Russ (as evidenced by the flurry of highly emotional responses from those on the political left who now have to wage a much more difficult battle inside their own heads in order to maintain the cognitive dissonance needed to keep parroting this OECD nonsense).

Let me say that I agree with Russ w/respect to the deficiency of the pre-Obamacare healthcare system, although I won’t speak to this as this podcast was more about the quality of healthcare delivery than the efficiency of healthcare finance. I give Obama more credit than his GOP counterparts simply because he brought the problem to the forefront and ignited a nationwide dialogue. That being said, I hope he loses in his quest for more centralized control and financing.

I found myself in complete agreement with Scott Atlas’ critiques of OECD rankings. Why anyone would think that auto accidents, other accidents, and homicides should be included in life expectancy statistics in order to measure the quality of healthcare- or the merits of universal insurance- is beyond me. Also, as far as I could find, they do not control for ethnic groups as certain groups are statistically more at-risk than others.

Nor do I think suicides should be included since, as far as I’ve seen, there is no clear relationship between mental health policy (whatever that means) and suicide. In fact, my conjecture would be that more access to the Psychiatrists chemistry set (anti-depressants) is likely to increase the rate and/or lower the quality of life. I’m skeptical that mental healthcare as delivered via the health insurance system can address the suicide problem.

But even if we include the suicide rate, the US rate doesn’t seem to be that high, relatively. OECD even seems to be calculating the rate for a specific at-risk group and controlling for ethnicity- why doesn’t it also do this for life expectancy, etc.? And I challenge you to look at the suicide rate statistics and draw some sort of conclusion about healthcare systems and mental health quality. Southern Europe has lower suicide rate than Northern, Asia greater than Europe, rural greater than urban, etc. Also, soldiers have been committing a suicide/day, LBGT have higher rates, rural have higher rates- all these facts definitely skew the US suicide rate higher. More tolerance for LBGT and less war are policies I support (perhaps a minority in the GOP but not amongst the more libertarian-minded classical liberals).

This podcast showed the difficulty- perhaps impossibility- in measuring “healthcare.” Open up the OECD statistical spreadsheet with an objective mind and you will be left with more questions than answers.

Seth writes:

DJan wrote: "From the point of an independent observer a health care system that provides good care for everyone (European style) is vastly superior to a system that provides excellent health care for the rich and virtually no health care for those who can't afford it."

Straw man alert.

Who would or does support such a system? Better yet, where does such a system exist today?

Pete writes:

@ Phil

I listen while cycling too!

The comments are very telling. So much confirmation biased on both sides (probably myself included). I certainly see a fundamental disagreement on the relationship between suicides and health. One would think, Brain=Organ=Heart, but I guess if you subscribe to a higher "self" calling the shots, it's not so clear cut.

I also think Health Care is a misleading term the way it is most commonly applied. To care for one's health includes much more than just treating illness and disease. There's prevention (which you think you could measure with instances, but then you get the "but we're living longer argument") and overall quality of life, as well as lifestyles. It makes no logical sense to make our children sick by raising them on highly processed foods and then pat ourselves on the back for figuring out how to treat all the illnesses we brought about (while arguably diminishing quality of life) having spent billions on research all the while touting the efficacy of our "Health Care" system.

And to defend Atlas, who is selling a book, by highlighting the fact that he didn't actually do the research himself is a stretch, no? He chose which data to sight and which to omit, and told (sold) a story.

Greg G writes:

txsir,

I don't know which overall healthcare system is best or which has the best life expectancy properly understood. Therefore I made no claim about that. Dr. Atlas suffers from no such uncertainty.

He endorses an analysis in which "the United States elevated to Number One in life expectancy." That analysis relied on removing suicide deaths. This was one of the very first things Dr. Atlas chose to emphasize. Russ thought it was important enough to highlight in the intro paragraph.

Dr. Atlas justifies this by suggesting that suicide deaths are not, "related to healthcare." Many here disagree. We think that most suicides are caused by mental illness and that mental illness has medical treatments. As with a great many diseases, the treatments available are often ineffective.

No doubt any system at all for calculating life expectancy is problematic. No doubt deciding which system delivers "superior healthcare" overall is even more problematic. It is hard to understand why the second task would be easier than the first.

Yossarian writes:

@Greg G:
Please explain to us the manner in which the Greeks, Turks, and Mexican treatment of mental illness is superior to the Korean, Japanese, and Finnish approach? The former have the lowest suicide rates while the latter are the highest. But I'd much prefer take suicide out and evaluate mental health (if that's even possible) in a separate way...

Greg G writes:

Yossarian

I answered that issue in an earlier comment but I can certainly understand if you didn't read through over 100 comments.

No one at all is claiming that better treatment necessarily results in a lower death rate. Not just many, but most, diseases have other confounding factors that may affect death rates even more than good treatment.

For example, the U.S. has higher death rates than many third world countries from heart disease. We don't use that to conclude that they are better at treating heart disease than we are and we don't use it to take deaths from heart disease out of our mortality rates.

Jimmy Tidey writes:

This is by far one of the most active comment threads on the podcast.

Is there any chance of some comeback? I do think there might be something of an obligation to acknowledge everyone's feedback.

Hearing from someone with a point of view on healthcare that doesn't coincide quite so neatly with the hosts would be very welcome -- or just a brief mention of the issues at the beginning of the show.

Stevie writes:

[Comment removed for rudeness. Edit option declined by commenter.--Econlib Ed.]

I have a few editorial remarks about the now 112+ comments in this thread.

First, we appreciate the interest shown by the many comments and commenters. We read all the comments. Of course, it's infeasible to respond to each comment individually. Plus, for the record, an extended, forward-looking comment Russ composed about halfway through the thread was accidentally lost. He had some tech problems last week, so recreating the lost comment on top of handling other tech matters was not in the game plan. Such are the whims of life. Rest assured, though, that Russ reads all the comments and takes them to heart.

Second, as should be evident, we welcome all points of view. Disagreeing with the points of view expressed by Russ, the invited guests, or other commenters is welcome.

On the other hand, the drum-beating tone of indignation, snideness, and disrespectful denigration of others displayed by many of the comments on all sides in this thread has not been conducive to productive discussion or debate.

We do get it that people are frustrated and angry, holding disparate and possibly irreconcilable views about political and social matters that bleed over and cannot help but color the comments about the topics of our podcasts. However, we remind you that letting anger, snideness, or rudeness get out of hand is against our long-standing civility policies. We expect EconTalk comments to rise above the fray. I've allowed a lot of comments in this thread that are marginal in terms of both tone and content. There is obviously a lot on people's minds. Probably it's better to get it on the table all at once. Civility violations will not be so easily tolerated in other threads.

EconTalk prides itself on providing a space where even the podcast guests sometimes feel comfortable and are excited about contributing to the comment section. Comments that display a tone of indignant self-righteousness--i.e., you obviously can't touch my argument because I'm so obviously right and you are so obviously wrong--do not invite engaged discussion by other commenters, much less by the guest speakers or by Russ. Good writers might write that kind of emotional, self-indulgent material as a first draft, but then self-edit the material before it sees the light of day. We have a Preview window to help with that self-editing moment.

The goal of EconTalk is not: Seek and obliterate the opposition. It is: Seek and find an engaged path for mutual learning and interaction, even if it's just one small step at a time. If you write a comment of an indignant nature that to your mind smashes your opponent into oblivion by your brilliantly killer argument, I suggest that before you post it, you read it in the Preview window and edit it for tone. Ask yourself: Would I open myself up and respond to a stranger who addressed me like that? Remember that a great and gracious argument does not require a snide or indignant tone. An argument that sounds like a complete rout to you may have a weak underbelly. If you have to resort to haughtiness to intimidate or humiliate your opposition, your argument is probably too weak to stand on its own. Generosity is the foundation of strength.

If you are just scraping by the norms of civility, you will only attract respondents who also only want to scrape by. If you reach for the stars, the stars will pay attention.

Pete writes:

"Good writers might write that kind of emotional, self-indulgent material as a first draft, but then self-edit the material before it sees the light of day."

So what did the first draft of this post look like? :)

Excellent post Lauren. Thanks.

Clever question, Pete.

My first draft was indeed way more indignant. Yep, I edited it. A whole lot. :) Mostly, though, you can never edit enough.

Rheal writes:

I enjoyed this podcast because it brought me back to watching Sicko in high school. At the time, I had a feeling that something seemed a little off in the statistics. I was willing to believe that the United States didn't rate first in the world with regard to healthcare, but I also wasn't buying our ranking as 37th. While I didn't agree with everything that Atlas had to say, I did appreaciate some of the clarifications with regards to the WHO statistics. As usual, I feel a little more enlightened.

Jimmy Tidey writes:

[Comment removed. Revision to come.--Econlib Ed.]

loveactuary writes:

Russ I felt that this podcast was weaker than almost all previous ones. There were many assertions from Dr Atlas that deserved challenging, which you were capable of:

a) (as noted by many others above) Suicide may not properly be excludable altogether when calculating life expectancy as health care $$ spent towards mental health may go a ways to reduce these deaths. Do any countries spend more per capita on mental health care (all else equal :) and incur fewer suicides?

b) stating that the Medicaid population receives worse care than the uninsured. While Dr Atlas I believe mentioned that other variables were controlled for (age, gender, risk factors, etc I certainly hope) it's clear that for those uninsureds making the choice to be uninsured, they clearly have some conception that their morbidity risks are lower (what us actuaries refer to as selection), and thus the quantity and quality of care needed for them is less than otherwise. Not to mention the moral hazard effect of those who have Medicaid coverage and their propensity to use it.

I am a free-marketer and consistent listener of this podcast, and probably align more closely with Dr Atlas on many of these issues. But by keeping up with The Incidental Economist I have seen that many of these assertions that Dr Atlas makes are likely more nuanced.

I'm with @DevEconHealth: I would like to see a counterpoint podcast with Aaron Carroll or Austin Frakt (both of whom I struggle to agree with 4 times out of 5) just to rinse the taste of unchallenged assertions out of my mouth.

loveactuary writes:

ah right - I left out from my post above, which has also been discussed earlier:

c) The laptop analogy from Mr Atlas does not pass my sniff test. "We don't know how a laptop is made and yet we feel comfortable purchasing it on the free market" implies that: we should feel comfortable purchasing complicated healthcare on the free market (I do, in fact, think that we should have a free-er market for healthcare, but I also think better analogies are called for here).

In fact we do have rational expectations about the outcome of our purchasing this laptop, and our range of confidence about the purchasing decision is quite narrow (will the computer function? how fast? can I write emails and use a spreadsheet software?) With healthcare, our range of confidence about purchasing, say, brain surgery is enormous. For example (and I make up results for illustrative purposes): there's a 5% chance that we could lose our life, a 10% chance of permanent brain damage, and an 85% chance that we'll be just fine. Not quite on par with the decision I'm making when I go to BestBuy or click 'purchase' on Amazon. As you put it in the podcast (and I feel you should have continued to emphasize) there is something primal about health care and more psychologically-riveting than other products which we purchase.

Russ Roberts writes:

I have so many reactions to these comments that instead of getting them all down here, I will try to devote a part of a future podcast to discussing some of them.

I do want to make a couple of quick points.

Life expectancy is a lousy way to measure the quality of a health care system. Too many things going on--genetics, life-style, urban density, etc.

As some have noted, the life expectancy numbers that Atlas discusses were not generated by him. He was drawing on work of others. They tried to control for a smattering of factors and showed that when you control for those factors, the US ranks first. This does not prove that the US has the best health care system or a system that doesn't need improving. It proves that there are lots of factors affecting life expectancy. Which factors you should control for is another question. As some have pointed out, the US suicide does not have a particularly high suicide rate so that is not the explanation for why the US moves up in the rankings. As txslr points out in his comment, the fact that the US has a lower suicide rate than France or Ireland says nothing about the quality of mental health care in those countries relative to the US.

There is an excellent discussion of the life expectancy numbers here, by Carl Bialik in the WSJ.

Pete writes:

The laptop analogy is flawed simply because consumers are TERRIBLE at selecting the best computer. They are constantly mislead by marketing. The proof: Apple. :) (Actually consumers make terrible choices on most everything they buy. What neuroscience is just discovering about human persuasion marketers have known for decades.)

Yossarian writes:

In a world where you don't see your bill because it's paid from your taxes and bundled with defense spending, social security, infrastructure, etc. and where every procedure (from chiropractor to rehab to antidepressants to acupuncture) is covered and every doctor available, isn't a consumer equally free to be misled into poor treatment choices with the only difference being that someone else- often the consumers making more sensible choices- is picking up the tab?

Greg G writes:

Russ

The fact that U.S. suicide rates are not that high is very relevant to this discussion but not something anyone would conclude from listening to this podcast.

To quote your intro paragraph:
"When you take out suicides and fatal car accidents, factors that Atlas argues are unrelated to the health care system, the United States has the longest life expectancy in the world."

So why then does Dr. Atlas conclude that taking out suicides improves our relative position?

Derek writes:

A couple of claims I wondered about

1) Car crashes are unrelated to health care? Huh? When you're in serious a car crash, guess what service best improved your chances of survival? If somebody gets to the ER and then dies, that would be a death due to the crash. Obviously a better system would change that rate. I was also confused as to whether Atlas was also taking out the same variable from the other systems he was comparing to.

2) The US does count every birth with any signs of life as a birth, but we also do this in Canada. The WHO definitions which are used to compare countries are followed in Canada, and are exactly the same as those Atlas identifies for the US. Yet, we have a lower infant death rate, and only a slightly higher stillbirth rate (3.3 vs 3.0).

3) I disagree with those talking about suicide as significantly impacted by health care. The suicide rate is higher in socialized countries, first. And secondly its not clear health care really changes this. It is VERY difficult to examine the impact of care on suicide rates, since depression is a diagnosis based on subjective responses by the patient, not objective testing. Diagnoses that rely on no hard evidence (such as 95% of psychiatry) should be excluded from any analysis.

4) The comparison to buying computers is nonsense. You have lots of time and rational-minded opportunity to make the best choice and do lots of research. You can't wait with many health care choices, and you often make them in a compromised, pain or anxiety filled state. In addition, people are actually quite bad at picking the best computer. However, computers are cheap because they can be made cheaply. Not so with health care. Large amounts of money are wasted in marketing confusion in health care, just like in any other market. Hospitals and doctors are incentivized to provide the most profitable intervention for themselves, not the best value for the customer. This often means getting a $10 000 procedure for a 86% success rates vs a $1000 procedure for a 85% success rate.

5) The inability of insurers to participate across state lines is a serious problem. Absolutely right on that. Furthermore hospitals or insurers should be allowed to bargain for prices as a whole hospital system. This is forbidden by law, for unknown reasons.

6) Lawyers are probably one of the largest inefficiencies in the system. The threat of litigation creates incredible waste both in payouts and defensive medicine. Without significantly cutting lawyers out of the equation (capping payouts at $1 million + medical costs for example), they will remain parasites who suck enormous value from the system for everyone.

Russ Roberts writes:

Greg G,

You have misunderstood Atlas's point or he has communicated it poorly. I too thought that he was taling about "taking out" suicides and fatalities from car accidents. But that isn't what Ohsfeldt and Schneider did (the people who did the research that Atlas is citing.) What they did is try to predict life expectancy based on different observable factors such as suicide rates, homicides, and car accident deaths to hold those constant. As Carl Bialik describes it (read the whole article):

Instead, Dr. Ohsfeldt and Dr. Schneider performed a statistical calculation, called a regression, to estimate how much mortality rates from homicide, suicide and accident influenced mortality, on average, from 1980-1999 in 29 of the 30 developed countries in the Organisation for Economic Co-operation and Development (they skipped tiny Luxembourg). Then they adjusted life-expectancy stats to get a rough handle on what life expectancy would have been like had the rates of these deaths been the same in all 29 countries. Their result: The U.S. would have ranked first, at 76.9 years of life expectancy — an increase of 1.6 years. Meanwhile, Japan fell from 78.7 years to 76 years, indicating it had been benefiting inordinately from low rates of accidental deaths and homicides.

That doesn't prove anything other than there are lots of factors that affect life expectancy other than the health care system and you can debate whether suicide and homicides are auto deaths are caused in part by the quality of the health care system. But it does appear that including suicide makes the US ranking worse not better. So you can't condemn Scott Atlas for that one--it's not his analysis and it works against his conclusion.

Greg G writes:

Guest: Consistent with the numbers I said. Yet when they compensated for differences in suicide and immediate death from high speed motor accidents-- Russ: Things where the health care system is probably not going to able to save you. Guest: Yeah. The health care system has nothing to do with your surviving a gunshot wound to the head, basically. Russ: Or an accident at 60 miles per hour without a seat belt. Guest: Exactly. High speed accidents. And when they compensated for that by giving everyone the same number of suicides and instant deaths like that and then redoing the ranking, the United States elevated to Number One in life expectancy. Russ: Sounds like a cheap statistical trick. But it could be true, the right thing to do. Because, and correctly, suicides and high speed deaths should not indict your health care system. But of course there's some vagueness about how we measure high speed automobile death, I suppose. Some uncertainty about that. Guest: But when you really look at the numbers on suicides, there's an extraordinarily high number of difference in the United States versus these other countries where they are doing the rankings. Russ: Much higher. Guest: Much higher in the United States.

Russ Roberts writes:

Greg G,

Good catch. I don't understand that given the numbers I posted from wikipedia. Maybe they vary over time or maybe Atlas just got it wrong not realizing that the suicides actually work against the United States. My other though is that the US rate is not lower than all other countries.

txslr writes:

When I heard Dr. Atlas' statement regarding the level of suicide in the U.S. I assumed he simply misspoke and that he meant to say homicide.

Greg G writes:

Thanks Russ. It's really David Robson's catch. On my first listen I was more focused on the points about obesity and infant mortality which really did change the way I think about comparing mortality rates. When I read David's comment I was a little disappointed in myself for not picking up on his issue the first time through. Perhaps I overcompensated because I have been banging on the point ever since. I find the discussion fascinating.

Maybe Dr. Atlas simply misspoke. If so, he mentioned suicide three times without ever really framing the issue correctly. Or maybe he was so eager to accept a conclusion that supported his political views that he accepted the conclusion without really understanding how it was arrived at.

I think there are at least two other reasons, beside the fact that healthcare policy is so controversial, that make this a hot button issue.

First, it is simply assumed here that prevention should not count as part of a healthcare system. That is a very dubious idea that, at least should be argued for rather than simply assumed. It is not like our method of mostly ignoring prevention is working so well after all.

Secondly, people who have worked in the mental health field and people who have family members with mental illness are very sensitive about the misconception that mental illness is not a real illness (or, as was said in the podcast, not "related to healthcare").

Due to the fact that this is such an emotional issue I think Lauren was wise to give commenters more leeway than usual. By the standards of the internet this is still a very polite discussion. The respectful tone you always set in the interviews has a great deal to with that, as does Lauren's moderation.

As always, thanks for thanks for providing these podcasts and this forum.

Frank writes:

@Lauren [Econlib Editor] et. al

First-time poster. Very interesting podcast as always and a great discussion -- in particular Dr. Samuel Metz' response.

I concur that there is a level of confirmation bias by the guest and host in this podcast, and that there is degree of confirmation bias in some of the opposition comments. It is clear that even if the discussion in the first portion of the podcast is true, it does not necessarily mean that (at least heavily) government-funded universal healthcare systems are inferior, and the claims to suggest this were indeed suspect. There are just so many variables for all healthcare systems that are not held constant in a simple ours-vs.-theirs comparison, not to mention the fact that ours is part-private, part-public. None of what I'm saying here is new insight compared to the previous comments.

Just an observation: I did, however, want to point out how ironic it is that a libertarian-hosted podcast would ban posts based on uncivility. As a libertarian on social issues (but not economic), I generally dislike censorship, even of the most heinous or ideological or unintelligent comments (which is, of course, most of the comment section on any other website). I just find a "civility policy" to be quite contrary to libertarian thought, and somewhat humorous in this case.

Russ Roberts writes:

Frank,

I have a different definition of censorship. I reserve the phrase for government suppression of information--either spoken, printed, or on the web.

EconTalk is private property. It's not a public square. We have all kinds of restrictions on that property that we think enhances the product. We might be wrong of course. But deciding how private property is used and restricting certain types of access or use of that property is not censorship. You can call it censorship, of course, but if it is, it is not in conflict with liberal principles.

You have no right to enter my house. You enter at my pleasure. it's my decision to allow you to enter or to refuse you entry. If I refuse to let you in, that's not repressive or a suppression of your rights. That's my exercise of my property rights. There's nothing illiberal about my decision to keep you out of my house if I choose to do so.

Brian Gibson writes:

As always, thank you providing one of the great podcasts on the Internet. However I find that this podcast was one of your weaker efforts.

Dr. Atlas clearly has an axe to grind. His entire agenda is to dispel every derogatory claim made against the United States health care system while accepting virtually every derogatory claim made about all other systems.

This is evident in how he cherry picks his comparisons to other health care systems. He spends a great deal of time in challenging commonly accepted statistics when it relates to the United States but accepts those same stats when talking about other nations. Wouldn't the logical assumption be that if the stats are wildly off with regards to the United States they are also wildly off with regards to other nations? Would you ever accept such a clear case of one sided scrubbing in other cases?

Dr. Atlas' claims about health care for the poor are also dubious. While virtually everyone would agree with giving the doctor more freedom to make sound medical choices, there are still costs associated with those choices. Costs that the poor generally cannot afford. How does the free market fix this? He doesn't really say. It seems mostly focused on simply beating up on the conservative straw man of the faceless bureaucrat.

Dr. Atlas also criticizes the fact that state laws often require certain types of insurance. His assertion is that these mandates are not needed by everyone thus they create cost inefficiencies. However his examples didn't further his point. Mandatory coverage for in vitro may not be something a person needs today but they may need it in 5 years. Or they may never need it. That is the ENTIRE POINT of insurance. Dr. Atlas also doesn't explain how excessive coverage would cause significant cost overages. Insurance coverage isn't defined by the number of people paying for the insurance. It is defined by how many are receiving benefits. How does expanding the base of people paying for coverage cause an increase in NET health care spending?

Rufus writes:

@Brian

"How does expanding the base of people paying for coverage cause an increase in NET health care spending?"

Simple: New people who are covered by Medicare/Medicaid are eligible for additional services funded by the taxpayer for services they would otherwise not receive.

Basically, no one is paying their fair share, so adding to the pool increases spending, but at a deficit in proportion to the payments.

Play around with this calculator and see if you can find a single demographic that isn't taking out more than they are putting in.

http://www.usatoday.com/news/graphics/debtcalculator/flash.htm

Brian Gibson writes:

Rufus,

That would make sense if Dr. Atlas was talking about government provided services. But he was not doing that. He was saying that state mandates on private insurance were increasing costs. So perhaps we should talk about that.

Or perhaps you were trying to take my statement and apply it to an entirely different situation, namely government provided health care. If that is so then I will certainly agree that government provided health care services absolutely increase the costs. That is the price we pay to at least pretend that our health care system is not entirely created to provide care for the wealthy.

David Mitchell writes:

I really enjoy Russ's style of discussion on his podcasts, which (as a physician) gets me out of the medical mindset for a while, and is truly enjoyable. He really helps bring complicated topics to an understandable level, and doesn't cloud concepts in academic lingo.

Although I've never taken a course in logic, I must say that the lack of logic in this guest was even baffling to me, and very uncharacteristic of the show. Russ always treats guests with great respect, but I was surprised the guest wasn't called into question on more issues. I don't have time to list them all.

One, in particular, regards the failure of Medicaid. He seems to assume that "no insurance" is better than Medicaid. He states that because 45-60% of doctors don't accept Medicaid, half of patients with Medicaid won't be able to find a doctor. Crazy logic. No, they just go to the doctors who accept Medicaid. That's like saying that because half of car dealerships don't service Chevy's, half of people with Chevy's won't be able to get their car repaired.

He quotes the percentage of doctors that do not accept Medicaid, but does NOT quote the percentage of doctors who do not accept "no insurance", which I can guarantee is much higher. In fact, these patient often have to make an upfront deposit before being seen.

Very unusual guest. But I suspect he is very welcome in certain circles.

Bogwood writes:

Depending on the context,queues can be a logical form of rationing. A medicare patient stopping for gas in Miami may get his cataract done immediately if she is not cautious. A patient in Vermont may wait until the worse eye is 20/50. If enough seniors wait for real disability a certain percentage will never have to be done. With the current aggressive approach to cataracts waiting up to a year is rarely a problem.

SteveS writes:

Dr. Mitchell,

I think that the guests point was that in some states, Medicaid patients can't find a doctor at all. And this problem will become much worse if 12M people are added to the rolls.

Steve

SteveS writes:

Jason,
For another perspective on the Oregon study, read this column by Avik Roy:

http://www.forbes.com/sites/aroy/2012/08/01/economists-claim-medicaids-health-outcomes-are-great-or-do-they/

Steve

Harland Brown writes:

A few short responses:
The distinction between economic-driven vs. care focused health care is certainly not an absolute one. Cost is a factor in any health care system. But if the decisions about choice of specialty, cost of medical education and fee schedules are overwhelmingly driven by a desire to optimize one's economic gain, the result can be very irrational in terms of health care delivery. Care will be focused on those areas that are most likely to provide economic gain, regardless of whether that produces improved outcomes.

When you have several hundred thousand people doing medical coding in this country to comply with the insanely complex pricing model that the private/public system imposes in the U.S. it is nearly impossible for all but the most diligent patient to truly understand the costs involved in their care, much less to negotiate them with their providers. In economics, I believe that this is referred to as an "information asymmetry". It is as if that cup of coffee that you enjoy could cost you $0.30 or $300, but you won't really know until you receive the itemized bill from your insurance, assuming that you can make sense of it. This brings up an additional economic issue that is largely overlooked, namely that the health care is a three-way arrangement (provider, insurer and patient/consumer). I would contend that the conventional mechanics of supply and demand are largely distorted.

My remarks about GPs or internists are not so much focused on what medical specialty is best equipped to do preventative medicine, but rather on the benefits for most patients to see a doctor on a routine basis to catch medical issues before they become acute problems. The huge focus on specialists in the current system is very ill-suited to catching medical conditions early on when they can be treated at lower cost and with better outcomes. The current system in the U.S. seems to be focused on responding to acute or chronic health care issues once they appear and on doing a very poor job of preventing patients from getting to that point in the first place. More and more private practices are actually calling upon administrative and nursing staff to do much of the work that a doctor used to do (while charging the same fee as a visit to an MD).

As an anecdote, I will point out that I have never met a single person in France who looks upon the provision of health care in the U.S. with envy or as a place they would like to go to get care. This includes a great number of medical professionals who would consider themselves on the right in the French political spectrum.
For some historical background on French medicine, I recommend the chapter on American medical student in Paris at the beginning of the 20th century in McCullough’s The Greater Journey.
http://www.amazon.com/The-Greater-Journey-Americans-Paris/dp/1416571760


Comments for this podcast episode have been closed
Return to top