 I have a couple of goals for this talk. And I'm going to spend maybe five, seven minutes just talking, but I'm happy to get interruptions along the way even during those five, seven minutes. But I want to set up a couple of, I think, big picture issues around the US health care system and what are the sort of big issues that I think about as a practicing physician who studies health care and tries to figure out how to make the health care system work better. I will be honest with you that the first few minutes of it will hopefully leave you a little bit depressed about the state of the health care system in the US. And it is for anybody who has had, and I think if you've had or had an interaction with the health care system, which I would assume most people have at one point or another, if you're interested in this part of the economy, which is about one sixth of the entire economy, there are some very sobering numbers about where things are today in 2008. And then I'm going to talk a little, very little bit about, and I have lots more data that I can get into if people are interested. I don't want to sort of lead with that. But I can get into some of the big efforts that people are trying to do to try to make the system work better. And with the sort of goal in all of this, that I think a major underlying problem of the health care system is the lack of transparency, the lack of good information management. And my hope is that from the sort of folks in the room, and again, interrupt me along the way, ask me questions, but we can sort of generate ideas. I'll tell you, my honest, my goal for being here is to get ideas and input and insights about how the heck we began to make the system work a little bit better. So a couple of the sort of big picture take-home points. I actually counted all those zeros. They're correct. We spend $2.1 trillion on health care. That's how much we spent in 2006. I think by anybody's account, $2.1 trillion is a lot of money. It's about one out of every $6 of the entire economy goes to health care. It's going to get to about $1 in $5 of the economy in the next decade. And just because $2.1 trillion is not a number anybody can understand, it comes out to about $7,000 per person. And if you think about it, a minimum wage worker makes about $11,000 a year. So if a person who works full-time making minimum wage can spend about 65% of their pre-tax dollars and buy health care for themselves. So it's sort of this idea that we're spending unbelievable amounts of money on health care. Questions or thoughts or disagreements on that? Seems like a lot of money, right? It's $2 million, million dollars. And then you sort of ask the question, and I don't want to spend a lot of time on this, but you say, how do we compare everybody else in the world? And if you look at other high-income countries, no one else is close in terms of GDP, percent of GDP, in terms of the fraction of our economy that goes to health care. You have France and Germany and Switzerland get to about 10% or 11% or 12%. We're at about 16.5% in climate. So we spend much more than anybody else, both in terms of absolute dollar amounts as well as in fraction of our economy. So we're well ahead of the game. And the question is, what do we get for all of this? And here, what I wanted to do, and this is the sort of most, hopefully the most painful part of the talk, I'm gonna spend about two minutes talking about the Rand Health Quality Study, just to give you a feel for why most people when they look at the health care system closely will say, I'm not sure we're getting the value that we want for $2.1 trillion. So let me tell you about the Rand Study. Most of you guys know that the Rand Institute, it's this really terrific health, well it's a policy think tank, they do health, they do defense, they do a lot of other stuff, they're pretty non-partisan. And they did a huge study that came out in 2003 that asked the question, how good is care in America? When you go to a doctor or you go to the hospital, how good is care? And part of it depends on what do you mean? How good is care? How do you define good? And they came up with 439 indicators of health care quality. So let me tell you what these things are. And then as I said, I will soon after this and the next couple of slides, I will stop talking and I'd love to hear ideas from people. So these indicators of health care quality are evidence-based practices. What do I mean by that? I mean examples like, I develop chest pain, I go to the emergency room, I'm diagnosed with a heart attack. We know that in the first 24 hours of your heart attack, if someone gives you a beta blocker, you have about a 25% lower chance of dying in the first 24 hours and you have much greater long-term survival. And beta blockers are basically cheap as water. I mean, they cost about a nickel each. So it's the kind of thing that you would think would get done 100% of the times. We've known for 25 years these things are extremely effective. So they came up with 439 such things across the entire health care system from preventive care to chronic disease, diabetes, heart disease, to inpatient stuff, stuff that happens in the hospital. And then they did a pretty comprehensive look across 12 metropolitan areas across the US and what they found was that we were getting it right about 55% of the times. So these are not controversial things. These are not things where there's a lot of subtlety involved. This is stuff that you get 10 doctors in a room and hopefully all 10 say, yeah, that's a no brainer. You gotta do that. And then you look to see how often does it actually get done and the answer is 55% of the times. This is a paper they published in the New England Journal about 45 years ago now suggesting that the care that we get is pretty inadequate. Is this something that people, you guys, I mean, one of the things I'm sort of always interested in is the question of, is this surprising? Did you know this already? Is your sense that the quality of healthcare is pretty good? I'd love to hear people's reactions to what I've said so far. I mean, I think most people know that healthcare is pretty expensive. Why don't we start? Yeah, go ahead. They, yeah, so those are terrific questions. And we know from lots of other studies that there are huge differences by race and ethnicity, et cetera. I don't have that slide here, but the surprise, the one surprise of this study was that if you looked at white, educated, high-income people, that 54.9% was about 58%. And if you were black and poor and uneducated, it was about 51%. So, yeah, there's a racial difference. But no one looks at 58% and says, boy, that's terrific. So instead of getting it right about five out of 10 times, we're not getting it right about six out of 10 times, makes me feel a little better, but still feels a little inadequate. So, yeah, so there aren't, there were surprisingly small differences by race and ethnicity in this study. There were other studies that showed big differences by education, economics, et cetera. Yeah? Do you know the reasons behind this? I'm wondering if the doctors know the problem, but maybe concerned of official laws or something. Right, that's a really good question. And there have been lots of other studies that have tried to look at this. When you ask doctors on these different types of indicators, should you do this every time? They say yes. Do you know the evidence behind it? They say yes. If you ask them, do you do it regularly? They say absolutely. And then when you actually look to see if they do it, you get numbers like this or lower or slightly higher. So there's a disconnect between what doctors think they do and what they actually do. So it's not being driven by a fear that if I do the right thing, it'll cause malpractice, lawsuits or any of that stuff. Some of that can explain 3, 4, 5% differences. It's not gonna get you to down to 55%. Yeah? Yeah, a lot of attention to this. I think where you present it, the way you do, it leads people to believe that one doctor is doing this knowingly. And the recent paper in the New England Journal two weeks ago and lots of others basically explained this by saying that there's seven people on average in the case of Medicare patients involved in the care. And the reason that it's not getting done is because no two doctors know what the other six are doing. Right. And so are there... Yeah, so let me be very clear when I show this number overall, when I say that it doesn't matter whether you're white or black or rich or poor or educated or uneducated, it's sort of under the argument behind it is, and I'm a practicing physician and if you did this for my data, I mean, I feel like I'm a pretty good doc, I work... I'm sure I'm not... Maybe I'd like to believe in a little better than that but maybe I'm not. So the point is it's not about individual physicians at all. It is completely about the system and we'll talk more about that. Let me just finish up the last slide that should add to a little bit of misery and depression about the healthcare system. So now we're spending $2.1 trillion, we're getting it right about 55% of the times but then there's this other thing which is when we admit patients to the hospital about one out of 10 times, patients suffer a significant injury from medical care. So not only do we not do the right things 40, 50% of the times, but we do something that's harmful to people about 10% of the times in the hospital setting, about half of which are wholly preventable things. What kind of things am I talking about? I'm talking about you come in for a heart attack and you develop an infection due to a catheter that somebody put in or you develop a pneumonia in the hospital that had nothing to do with why you came in or you were given a medication that causes serious injury because of a serious side effect of that drug that and half the times that should have never been given or should have been preventable, et cetera. About one in four doctor visits where a prescription is written 30 days later, a patient has some sort of injury from that prescription. I think that study is, I think it's a little bit of a high number I've never really believed that, but it certainly underlies the idea that it's very common and the Institute of Medicine, which is a part of the National Academy of Sciences, it's you guys all know the National Academies. The IOM is sort of the medical arm of that. They did, they put together a report in the late 1990s and the report was called to air as human and they estimated that between 44 and 98,000 Americans die from medical errors every year. And just to put that into context, that's more common than car accidents, which is about 43,000 deaths, breast cancer, which is about 42,000 deaths or HIV AIDS, which causes about 16,000 deaths. So medical errors cause between 44 and 98,000 deaths a year and I will, so I think when you put together 2.1 trillion dollars, 54.9% of the times we're doing it right and we're killing between 50 and 100,000 people a year and some of those numbers are a little bit questionable and there's been disagreements, but there's no disagreement about the fact that we have a problem. The value proposition of healthcare is completely, is completely AY. People aren't getting the right treatments, people are getting injured. The high cost of healthcare is a huge threat to the federal budget. When the Congressional Budget Office thinks about the next 20 years for the federal government, the single scariest item is Medicare. If you ask GM and Ford whether they think about healthcare costs, they're getting killed by healthcare costs and 50 million people are unsure that a large chunk of that is because healthcare is so expensive, if you work full time as a minimum wage worker, you're not gonna spend 70% of your pre-tax income on healthcare, you're just not gonna do that. So anyway, so at this point I'm hoping that everybody is feeling at least somewhat despondent about the quality of care they receive and let me be very, very clear that you can't go to the Brigham or Mass General or some fancy place, the Mayo Clinic and escape all this stuff. You can't. I've looked at the quality data from Mass General, I've looked at it from Mayo. They're a little bit better than average, but they're not all that terrific. So this is not a problem that is only for the other half as it were. Questions or thoughts about this? Yeah. I was wondering, at the beginning you started with the slide showing our spending there's a dollar in all countries. Yeah. What's their data coming in at? Are they having the same kind of error rates that we are, are they forming any better than we are? Right, it begs that question, doesn't it? And you see that I didn't show you that data for a reason. Two things, if I wanted, if you wanted to ask me what does this look like for the rest of the world or the other high income countries, I would say I have no idea because those countries in general are far behind us in terms of collecting and systematically analyzing quality data. They all get it, they all get it's important and there's some efforts in this area but they're not doing it as, and we're at least beginning to measure it. They have some fairly sophisticated IT systems in place to do that. Some countries do, you know like Denmark. Yeah, right, Denmark and UK is just kind of a lot of errors. Exactly. A bunch of countries have done a lot of effort in the Netherlands is another place, Finland, et cetera. But anyway, the point is they are not collecting this data and not because they don't want to but they have not had, in some ways we've had about 20 years of work into defining these quality indicators, how you collect it, and we've been doing a lot more work in this area than most of the other countries. But I think they're gonna catch up much more quickly because they do have the IT infrastructure to do it. And this 10% of hospitalized patients data, it's been repeated in Canada, the UK, Australia and their numbers all come out to between nine and 13%. So that suggests that they're doing exactly the same as we are on that, yeah. The classification of injuries from medical care has been published in France according to the number of injuries for each hospital. So there was a list of like hundreds of hospital and the best pupils and the worst pupils so that people could first avoid going where there were problems and hospital could be aware of it and try to get up to the investigation. I had not known that about France and I will look into it. In the US, and that's actually exactly where I want to go next with this, is that people are trying to do exactly that kind of stuff in terms of making that, pulling together data like that, putting it out there and seeing what effect it has in terms of people making better choices about health care. Is there a government or the press collected and reorganized the data according to the service people? And the question is, is that doing stuff like that actually make a difference? Does it make here better? Which is a big interest of mine, yeah. Can you talk a bit more about the indicators and how those were measured? Because what I understood was that we took what the most common, obvious form of care should be available for a certain kind of problem and that's what those indicators are on. These are all 439 things for which there is very good scientific basis. So meaning there's been good randomized controlled trials, there have been many, many studies that say that this medication is highly effective for this condition and it's been done eight times and nobody wants to do that study anymore because it's been proven that this drug works. It's kind of like aspirin for prevention of heart attacks in high risk men. We got it. It works. So it's stuff like that. It's certain types of asthma therapies that we always should give to kids when they come to the emergency room because we have very good evidence that it ought to be done. And these guys spent a lot of time doing a very meticulous job developing these indicators. So if somebody has a contraindication to that medicine that didn't count, they really wanted to come up with a number that was defensible so that the ideal rate here really, excuse me, it really should be close to 100%. And again, I think these are all very clinically based and things that fit it. And I can give you more examples, but it's stuff that I think not too many people really argue with. We've made a question on it. What percentage of the decisions that doctors make over the course of my new day are decisions that are no-brainer like this? So I'm just curious kind of, if this is a small part of the kind of decision? It's a really good question. Yeah, and doctors are making? Yeah, I would say that these are probably 5% of the things that we do. So there are probably 10,000 things that we do across the spectrum of care. But these are the ones that are for which there's not much judgment involved. Because a lot of what we do, another doctor might have done it differently. You don't know. This is stuff that there's not much judgment involved. Everybody agrees, it's just right, you gotta do it. So on one hand, it's a very tiny fraction of the whole thing. On the other hand, this is the part that we ought to, it should be close to 100%, yeah. Having said that, it's a very small percentage of what doctors average do. When we go to the slide where we say, there's a high error rate or high risk rate, is that measuring just all the 100% of stuff that doctors do? This is purely saying how often do people when they encounter the healthcare system, do they suffer an injury basically from that encounter? Do they get hurt because they went to a doctor, they got hurt because they went to a hospital? Hopefully at the end of the day, they get hurt less often than they get helped. Otherwise, you'd ask why am I in this business at all? But certainly, this is how often people get injured overall, based on everything we do, yeah. You can go back to that last slide. I mean, these are no brainers, but they're no brainers in the abstract. How often do doctors actually feel that the decision they're making is a no brainer, or giving aspirin to someone to prevent a heart attack may be the right choice, but that individual may have gastrointestinal issues or other issues. So you need to, so go ahead. I mean, is that an account, does this account for that? They really tried very hard to account for that. Is it to account for it perfectly? I'm sure it doesn't, but they tried to look through all the potential contraindications and say, okay, so you don't want to give aspirin to somebody who has a history of ulcers, somebody who has an allergy to aspirin as, you know, and then you come up with that list. And they really tried to do it in a way that at the end of the day, it was very clinically defensible as these are sort of the patients that we all agree, yeah, these guys ought to be getting that aspirin, or these guys ought to be getting that beta blocker. So do I think they got it perfect? No, there are people who criticize individual numbers, but no one I think clinically looks at this and says, this is actually terrific here, and we're just not doing a good job of measuring it. I honestly firmly believe this suggests care in America is really unacceptably mediocre. And it's unacceptably mediocre no matter who you are. Yeah. So are these numbers gathered? I mean, I'm wondering whether the breakdown is in, when you're sitting in the examination room and you say, I have chest pain and the doctor doesn't prescribe or doesn't give you aspirin, or is it that interaction, or is it sort of from the minute that you walk in the emergency room and maybe never see a doctor, or maybe, you know. Other experience. And so a doctor would have said, yes, that person should have gotten aspirin, but that person never really encountered the... Right. Luckily, I mean the way, so if you think about heart attacks, for instance, luckily most of the times when somebody has a heart attack, they end up seeing a doctor. If they go to the hospital at all. Obviously if you just die from your heart attack at home, that doesn't show up here. These guys were only looking post facto. So they found everybody who went to a hospital who was diagnosed with a heart attack. So if somebody went to the hospital and their heart attack got missed, doesn't show up here. So it's a whole bunch of other stuff that doesn't show up here. But if you got diagnosed with a heart attack and you were in the hospital for five days, they go through the charts and say, did somebody ever give that patient an aspirin? Somebody ever give that patient a beta block? If they didn't, are any of the contraindications to that medicine that we know maybe somebody said, eh, maybe I shouldn't give it for one of these reasons? Is that listed anywhere? And then you say, well, maybe people don't document it. You can always come up with an explanation for why some part of it may not be perfect. And I would say that, A, these guys did about as much as one could ever expect someone to do to get it right. And maybe the truth is not 55%. Maybe the truth is 60%. But the truth ain't 90%. And I would suggest that for $2.1 trillion. So for any amount of money, we should be expecting for these tiny number of things, we should be expecting 95, 98%. And anything less than that is unexciting. Yeah. How some diseases related to the indicators? These 44 to 98,000 deaths. You know, this is just an estimate based on the hundreds of studies that have been done in this area. It's an estimate by the Institute of Medicine to say how many times do we think that people are dying due to medical care as opposed to due to the underlying disease that got them to the hospital? And this was their estimate. And again, I don't think it's perfect, but I think it was done based on all the best evidence by a bunch of reasonably smart people. And I think the numbers are probably close to right. You know what the errors are that don't result in death, but maybe if someone's kidney gets taken out, it's probably kidney. And so if people have tried to look at the overall number of injuries that don't result in death, it's somewhere probably north of a million a year. North of a million. Yeah. So, you know, and that sort of makes sense about five to 10% of injuries lead to death, but about 90 to 95% of injuries people survive. So maybe they end up on dialysis. Maybe they end up on, but they don't die from it. You're probably going here next, but my question in the previous slide is that 40% of the time, this best practice thing is not happening and there's a number of reasons you could hypothesize and which of those bear out. Are they, do they know about it and they're just lazy or sloppy? Do they not know about it? Do they know about it and they forgot or do they disagree with it? Yeah. None of them are particularly reassuring, but I don't know. So, this is my transition slide, but it also has a point underlying it, which is that, you know, about 80 to 90% of healthcare in America is paper-based. Yeah. And that gets at, so this is your question. I do believe, and I'm not just saying this because I'm a physician, but I do believe most doctors and nurses are perfect. They're well-trained, they're well-intentioned. Most doctors and nurses wake up in the morning and don't say I'm gonna go out and give bad care and make sure I hurt people today. I think people work very long hours. If you look at the hours that physicians work, if you look at the amount of years of training that goes into training physicians in America, it's much longer than most other countries and I think doctors are very well-trained and I do think they're well-intentioned. So, I don't think the problem is fundamentally that people don't care or they're lazy. So, then the question is, what is it? And I would say a few things. I mean, one is that our payment scheme is completely flawed. So, our system pays for quantity of healthcare. You do more, you get paid more, okay? So, I'm a physician and I get paid by a number of visits and if I provide lousy care, I get the exact same amount of monies if I provide terrific care. And if terrific care takes longer, you know where the incentives are. And that becomes really problematic, I think in other parts of healthcare. You know, the healthcare organizations are, I think, are flawed and what do I mean by that is this. You know, 30 years ago, if you had a heart attack, you went to the hospital, just gonna stick with this example, we basically didn't have any therapies for heart attacks. You know, aspirin as a treatment for primary heart attack was probably mid-80s. So, let's say in 1975 you had a heart attack, you went to the hospital, what did they do for you? They put you in a hospital bed, probably stayed there for two months. You got, you know, they turned down the lights and sort of everything was very quiet and they try to, you know, and mortality rate was 30% within 30 days. Well, 30 day mortality for heart attacks now is about 10, 12% and how do we get so much better? Well, because we do an unbelievable amount of stuff when they come in to people, when they come in with their heart attack. We give them incredibly toxic drugs. We take them to the catheterization lab and while they're having the heart attack, thread a catheter through their groin into their heart, give them either a blood clot-busting drug or open up that artery. This is all high-risk stuff. But we do it and we do it in people who are 90 years old and have six other conditions and are on 15 medications. And most of the time, majority of the time, so we do it and people do well. But the difference is, you know, one of the sort of international experts in patient safety is a guy named Lucian Leap and Lucian always says he's a pediatric surgeon and he trained in the 60s. And he said when he was a resident, there were about 20 medicines, period. 20, 25 medicines that doctors ever used. They could keep all of it right here. You knew the medicines, you knew the doses, you knew which interacted with which, there were 20. We can all remember 20 things. And now, five, 10,000. And the question is, can you keep 10,000 drugs and all of their interactions and all of their doses and half of them sound like the other half? Can you keep it all straight? Can you all keep it all in your head and do this? And the answer is no, you can't. But the healthcare organizations have not, I think adequately responded to the change that has happened in healthcare in the last 30 years. In some ways, we're a victim of our own success. 439 indicators, 30 years ago, maybe there were six. And we could do those six all the time because there were only six things. And now we're at 430 or probably much higher. So here's the part where, again, and this is very helpful and here's the part that I would like to just make one statement about and then I wanna hear from you guys what you think. So a couple of the big problems in healthcare is there's no transparency. How many of you guys have been to a physician or a hospital in the last year? Okay, how much did it cost? Sydney, how much did it cost? No, how much did it cost? It's like a regular visit? Yeah, but you don't know. Okay, anybody know how much their visit cost? Yeah, $700. $700. Did you get a bill for $700? No, I saw something. Right, but your insurance- I got a bill that showed you how much. And do you know how much your insurance paid? No, I think it was the amount. To be sure that it was $700, it's like- I think your point might be made better. I know how much medical care for my dog costs, but I don't really- Know how much it costs for you. Right. Well, so there is this incredible opaqueness because your bill might say- First of all, most people don't look at it, right? But if you look at the bill, it might say $700, and then that Harvard Pilgrim pays $275. You're like, how do they get away with that? When I have bills in any other sector of the economy, I don't get to say, I'll pay you a 40% is that good. I have to usually pay the whole amount. But there's just all of this cost shifting and craziness. So cost accounting is completely opaque to the entire world, I think, except for some small number of people in insurance companies and hospital and doctor billing offices who understand what all this stuff actually is. And Sidney, how good was the care you got? I think it was very good. Yeah. I have no idea, though. Yeah, how do you know? I mean, I was gonna say, how do you know it was good? Well, I got better for the problem that I went in to the doctor about. And do you know that you wouldn't have gotten better if you had- I don't know. Right. Because 90% of the things that I see patients for, they're gonna get better whether they saw me or not. Just true, right? I mean, it's like, people come in with their colds. I've had a fever, I've had this cough, I've got this runny nose, I feel like hell. And I see them and I examine them and stuff. But if they'd never seen me, they'd be better in two days anyway. But they're convinced it was because of me, because I told them, you're gonna be better in two days. You know, here's some Tylenol. So the point is, we don't know. And in any other part of our lives, when we buy cars, when we go to the restaurant, service industry, non-service industry, you know how much things cost. You have some way to judge how good it is. And my general argument is that in healthcare, we don't have that. And that's very problematic. And, you know, the two main things that people have thought about to kind of help regulate the quality of healthcare is regulation. So when, so I'm a practicing physician in Massachusetts, I have showed them that I went to a real live medical school that I did residency, that I have not committed any criminal activities. And then they give me the little certificate and say, you can go practice medicine in Massachusetts. Well, let me tell you, that sets the bar very, very low. Right, because if they don't ask me, do you actually do the right thing for your patients? No, I just haven't committed any crimes. And I guess I went to a real medical school. So the regulation and certification sets the bar very low. And medical malpractice, now what was a little controversial talking about medical malpractice with lawyers in the room, but I'll just tell you that, you know, it's been a dismal failure in terms of thinking about effects on safety and quality of healthcare. Why do I say that? Because about 95% of the times that there are injuries to patients from medical care, there's no malpractice. 98% of the times, some very, very high number. And a good chunk of the times that there is malpractice, there's a malpractice lawsuit. A bunch of the times there wasn't even an injury from medical care. A lot of the times it wasn't preventable, certainly wasn't negligent. And so it's the classic example of when the AMA says the trial lawyers are doing blah, blah, blah, they're right. And when the trial lawyers say the doctors are just being defensive, but there's a lot more out there than they're willing to admit, they're right. It's just that there's very little overlap between bad care and malpractice. And so for me, as a physician, I think of malpractice as a relatively random event that has almost nothing to do with the quality of care I provide. And there are some things I can do to help control it, like be nice to my patients. So when I do injure them. You can use the term malpractice, you mean actual litigation. I mean like the lawsuit and then the, and then the, and I'm speaking as a non-lawyer, so I don't know, but that's what I think about. I've gotten sued. Right. I just meant, yeah, as opposed to a doctor doing something wrong. Right. Which is, right. That's a separate category. That's a separate category. That's the category you're just thinking about. But for it to work, you have to change your behavior somehow, you need to be more careful because you're worried about being sued for doing something wrong. And you're saying it doesn't factor in. Doesn't factor in. Doesn't factor in. Is there any knowledge about the fear of malpractice leads to a lot of extra tests or extra defensive care or being defensive? Yeah, there's a lot, as you might imagine. This is an area of great political debate, right? It depends on which side of the political spectrum you come down to on is where you end up usually pulling data. The truth in my mind is, does malpractice lead to defensive medicine? Absolutely. Is it a huge factor in the 2.1 trillion dollars? No. You know. Injuries because a lot of the tests and all could be actually invasive and could actually lead to some of it. Is there any study? There's very little data on this. I think there's very little doubt in my mind that some of that defensive medicine leads to bad care because you end up doing tests you really never should have done. And all these tests will have false positives that then you have to go chase down. And in general, it pushes people to do too much and that actually leads to bad care. Other questions or thoughts about this? So what I'm gonna do is tell you about one specific part of the response and that's the New York State Cardiac Surgery Program. How many of you guys have ever heard about the New York State Cardiac Surgery Reporting Program? One person. Can you tell me anything about it? It's fine, if not. I don't know the details about it, but there's on a public record so people can evaluate how different doctors have done based on some baseline of different procedures or successes and failure rate. Terrific. So, exactly. What New York State did, and they did this in the early 90s, but it's really becoming a model today for healthcare, I think. This is the part that sort of is the response that I'm somewhat hopeful is gonna make a difference. What they did in the early 90s was they looked at the data for cardiac surgeons doing heart bypass surgery in New York and they found huge variations. Let me see if I can just show you. Here it is. 1991, cardiac surgery in New York. 31 hospitals. This is risk-adjusted mortality rate. So this is doing really good, and they got the national experts on risk adjustment. They put them in a room, said, how would you do this? And they did it in a way so that it sort of leveled the playing field. So hospitals that took care of sicker patients should have had the same risk-adjusted mortality rate as hospitals that took care of relatively healthy people. There were 31 hospitals in the state of New York doing heart bypass surgery, and what they found was that the risk-adjusted mortality rate ranged from 0.5% to 7.3%, about a 15-fold difference between the best and worst hospitals. And that means about one in 200 patients here died of heart bypass surgery and about one in 14 died here of heart bypass surgery. They looked at this data and they said, this is unbelievable. These are all surgeons in the state of New York. Only 31 hospitals. This is unacceptable. And what they did was they started publishing data about cardiac surgery mortality for every hospital and every surgeon, making it publicly available. And you can imagine that in the early 90s, the way you got this information, if you wanted it was you wrote to the Department of Health and they would send you the latest report. And as of about 2000, all the stuff is on the internet. So you just go, you can download every file from 91 through 2005 or 2006. So that's what the data looked like in 91. And then here's what it looks like in 2003. Still a bunch of variations, but let me remind you one thing. Here you have a bunch of hospitals that are at, so majority of the hospitals are between three and four. Bunch of hospitals, four, five, six, seven percent mortality for heart bypass surgery. In 2003, on the same scale, it looks a little different. So there are three hospitals, sorry, this is getting cut off, three hospitals that are over four percent, but the statewide average is now a little over two and almost everybody's under three. So clearly something happened over these 12 years to do this. I would say this is not terrific. There's still a lot of variation, but it's a lot better. Yes? Thoughts about this? How do you think this will happen? Let me see if I can just show you. Here's what happened to overall mortality rates from 89 through 2005. The statewide average dropped from about three and a half percent to a little under two percent, which is nice. It's a significant drop. Any thoughts about this? Why would just creating these reports, putting them out there, do this? Yeah. So what would hope maybe the one thing it could do is to put the bad surgeons out of business, right? And how would that happen? A voice in? Yeah, so you can imagine that the worst hospitals and the worst surgeons would see big drops in market share, right? People would stop going there and then all of a sudden they'd say eek and they'd either go out of business or they'd get better. So that would be one mechanism. Any other thoughts? How might it work? Well, let me just, since we're talking, how does this work? So here's some data. We took about 15 years, 10, 12 years of data. We did this for hospitals and for surgeons and I'm just showing you the hospital data. So what I did was we said, okay, report card comes out if you're in the top 10 percent, if you're one of the really good hospitals. This was your market share, all the hospitals here. This was your market share before the report card came out. Here's what happened to your market share after the report card came out. And what you see, and this is not statistically significant, this is just random noise. But at best it looks like the best guys are losing market share and the worst guys are gaining market share. I don't actually think that's true and the statistical test suggests that there's no thing, no effect. But it certainly doesn't look like the people who are the bad performers are losing market share. Everybody's going, oh my God, Columbia Presquitarian, terrible mortality rates, I'm not going there. So unfortunately, that's what you would have hoped for and it didn't actually happen. Other thoughts, other theories? Yep. If the price of care are changed. Price of care. Right, for employees. Right, so you can imagine that these guys maybe started getting, it became more expensive. That sure would be nice. But there are people who've done the work of talking to other health plans, use this data to maybe pay more to the good guys or remember, most people don't pay for cardiac surgery themselves. So if you're a surgeon or a hospital, you can't just raise the price, we cost 15,000, now we're the best in the state, we're going at 20,000. People aren't gonna pay that, the health plan has to pay that and that didn't happen. Yeah. Technology would have changed in 10, 12 years. Right, so tell me more. Well, that might drop everything down and flatten it out. Fair enough. And I did skip over this, but cardiac surgery mortality, as the whole country, the mortality rates got better, there's pretty convincing data that it fell much faster in New York than it did in New Jersey, Pennsylvania, Connecticut, et cetera. So there is a sense that this fell faster and more so than other places. Well, presumably the technology changes were the same. Yeah. And as you were saying in terms of your own practice, I mean, people, doctors aren't necessarily aware of what they're doing. They just would have hard data to present themselves and say, this is where we're really at as opposed to your self-perception of where you're at. Right. And it would give them a partner engagement before. Right, so that I think is a major factor. It's hard to quantify that and study that, but I think that's a big factor, yep. Well, two probably related things. One is you then put in self-monitoring systems to actually gauge where you are and what you're doing well and what you're doing not. Or perhaps more importantly, you look at the people who aren't losing so many people when you go over and talk to them, buy them a coffee and say, help them. Why is your mortality at half of mine? Right. Again, that's what you hope happens. And there's good anecdotal evidence that some of that actually started happening. That surgeons started getting, did they go to conferences? Yep. And guess what? The surgeons were paying close attention to this information. Patients are not paying any attention whatsoever, but the surgeons are paying very close attention and they know, why is Frank, like why is his mortality so low? And they go talk to him at the conference and say, what do you do? You know, and all that stuff begins to happen. And then here's the one other slide that's I think kind of dramatic, which is what we did was we had said, what are the chances that you're gonna stop practicing in the state of New York after a report card comes out? And if your performance is in the top second or third quartile, basically about three to 4% of surgeons leave. And they leave because they retire, they're moving sort of a natural rate of people stopping practice. And then we looked at the fourth quartile and we said, well, these are the bad performers. The ones who got rated badly, about 20% of them quit over the next two years. Did they quit or did they move to Mississippi? All right, that's the truth, right? Jersey, okay. I went to high school in Jersey, so it was personal. And so what we actually did was we actually tracked all these guys down. And the answer is some of them just quit. Some of them moved. And you can imagine they're potential confounders. Like maybe the surgeons who were older are likely to have high mortality rates and maybe they're more likely, they're clearly more likely to quit because they're coming up to retirement. So we did a bunch of adjustments with just for surgeon age and surgeon practice and stuff. And what you find is that the guys in the bottom are about three times more likely to quit. And what it says is people really do pay attention to this information. You know, a lot of times what it was was the chief of surgery came to the person and said, listen, your mortality rate sucks. You're making our hospital look bad. Can you do something about this? They either could fix it or they just decided I wanna practice in this environment and left. So my point on this is people clearly do pay attention. All right, so in the last sort of 10, 15 minutes that I have, I'm gonna not, what I wanna just make the last sort of transition point is what has happened. Is the New York state, yeah. I'm sorry. No, please. Just a quick question. You said that the survey didn't really have much of an impact on the patients. Yeah. I would imagine that if I knew something like this, I would feel very aggressive in a situation where I felt like the person where I was seeking advice from was a lower rated professional than his coworkers. And I'm just surprised that this didn't, I completely agree. And there was a survey that somebody did of patients undergoing cardiac surgery in New York and asked the question, did you know this report card existed? Majority of them didn't. Then they said if you knew that this report card existed, at first of all, it's free, would you be willing to pay for it? How much would you be willing to pay for this information that tells you whether your surgeon is a 7% mortality or 0.5% mortality? And most people said about 20 bucks or less. Which sort of boggles my mind to suggest that you're not willing to pay 50 bucks to find out that the surgeon who's gonna perform open heart surgery on you might have a mortality rate that's 7% as opposed to 0.5%. What is going on? However, isn't a lot of people, especially with HMOs, they have no say in who they get referred to? There is some of that. That's the insurance angle of that. There is an insurance angle. And, but people usually have more than one choice. Usually, not everyone, but most people, there's a network of maybe six cardiac surgeons in your network. But if those six, at least you can look at their mortality rates and have some sense of, is this guy at least average? Right, but you need a referral for a lot of people with HMOs. And what if your doctor's just saying, no, this is my friend and here's where you're going? Right. And they surveyed actually cardiologists who do the referrals to the cardiac surgeons. And cardiologists, they were asked, do you believe in the validity of these report cards? And the cardiologist generally said, yep, they're pretty good. We believe in them. And then they said, do you use them to refer? No. Exactly. And so there's this huge disconnect between like everybody going, yeah, these are pretty good. Do you use them? No. Because I've always referred them to Dr. You know, Dr. Jai. And Dr. Jai is a great guy. Forget the fact that his mortality rate is often. But you're still holding on to the well-intentioned doctor model despite this? I mean, I think they're well-intentioned on some level. There's clearly a problem. Yeah. But the drop in mortality in New York has made policy makers sort of around the country say, we've got to have much more transparency in the health care system. It clearly has an effect. We don't know why it works the way it does. We can't figure out why it is that patients don't even use this information. So part of it is if you go to the New York State Report Card website, and that's one of the links that I gave to Alma, and you download one of these report cards, they're not all that easy to understand. Well, that's what I was going to say. I mean, I've been to a lot of these websites. Massachusetts has some under- Yeah, Massachusetts is now up and running. And you don't look at throwing them. It's awful. You have to go digging through spreadsheets. And just trying to find some simple metrics. It's horrible. It's horrible. It is. And so this is a place where I realize with seven or whatever, 10 is I'm getting to what I'm hoping is something you had. I mean, this is a place where the internet has been transformative in lots of other places, in lots of other industries. And in health care, it's so far has been doing nothing. Yeah. Well, I'm specifically transparency. The legal question that comes up is why hospitals are not required to give patients the records in the electronic format where they could be shared or aggregated on the internet in order to derive these things. Specifically, the Harvard hospitals, both of them, have these records available electronically because they send them to the Social Security Administration. Yet these same things are not available to any individual patient. Right. Couple of thoughts. I've spent a lot of time thinking about exactly this topic. Some facts and data for the country. About 90% of hospitals don't have an electronic hospital. There's one second. OK, so hold on. There's one second. First of all, 90% of hospitals are paper-based. Fundamentally, that's a problem unto itself. What are we doing in 2008 with paper-based records? But fine, let's talk about the Harvard hospitals that are electronic. Let's talk about Brigham and the BI. 50 yards from each other. I was attending at the Brigham about a year ago, saw a patient who came to the Brigham one night because the ER at the BI was full. And she's gotten all over care at the BI. And I saw her the next morning. And I wanted to get her records. The BI has an electronic record. The Brigham has an electronic record. So how did we get the BI records? Got a piece of paper. She signed it, faxed it. They printed out the electronic record. And they faxed it. And then we had to create a paper binder and put it all in there. It worked. I got the information I needed, but we have a problem. But that's not the problem I'm asking for. No, but let me just finish the thoughts. So there are two issues around sharing of information. One is you can ask, well, why doesn't the patient control this information? Why doesn't the patient get it? There are some laws. And the laws say patients always have a right to get their record. And if you want to pay some nominal fee, 10, 15 bucks, I'll print the whole darn thing out for you. The second is that it's very clear. And again, I'm getting a little beyond my own expertise. So push back if I'm getting this wrong. But it strikes me it's very clear that the hospital owns the record. The doctor owns the record. When I see a patient and I'm writing a note, it's my note. Maybe about you, but it's my note. I own it. You know, Amazon owns information about my shopping patterns. Maybe about my shopping patterns. But Amazon owns that information. And I can't say to Amazon my shopping information, give it to me, at least with records you can do that. I'm asking a legal question of people here about this. Regardless of ownership, or at least my understanding of the legal issues of ownership, if there's a law that says, MGH has to give me my records, which there is. And they choose to give them to me on paper in order to make it hard for me to correlate these things myself over the internet with other people who have had heart attacks or been treated, what part of the law applies to that particular straightforward thing? It's not ownership as far as I can tell. Well, you're getting beyond my area. No. Ownership is almost irrelevant, because even though he owns it, you have a right to have it. You want it for free in an electronic form. I didn't say free. Or you want it in an electronic form. I said kind of $15. I'm not sure. Why don't you already have access to it in an electronic form? Because they don't want to give it to it. They have it for the Social Security. If I file a disability claim with the Social Security Administration, they will take that PDF file. They will send it to the SSA. But they will not give me, from either the BI or MGH, that same exact file. I have no electric file. BI has an electronic system that patients have access to. It leaves something to be desired. And so does the Brigham. Patient gateway. It is not in a form to be used on the internet. It's not easily exportable, changing. It's a portal. Yeah. It's a consumer review. Your question is an interesting one. In the old days, when there was only paper on Freedom of Information Acts, and you knew there was some paper in Washington that you wanted, they would make you go there and pay for a Xerox in costs. So I think you're asking a question that hasn't really been answered. But I've looked at access to electronic medical records, because I'm interested in electronic medical records. I can't understand why. The only one I'm familiar with is the BI, which has all kinds of problems, but at least it provides you with free electronic access. It's sad to say it's one of the best on the country. Let me just say this about that. If you take the fuel research data, the largest reason why people want to see their records is to look for medical errors. The way to look for medical errors, the top-ranking reason, the way to look for medical errors to have the complete record, the one you would get under the hyperprivacy laws as a PDF file or whatever else they're willing to give it to you. So having a portal, if that's what you meant by having your records, or a thing, does not serve the purpose of looking for medical errors. Let me just cover it by the hyperprivacy law. There's some circumstances it doesn't, other circumstances it doesn't. Let me take a step back and give a broader picture. I think there is little doubt that we all agree. I certainly agree, and I think everybody agrees, that the ideal is that there is an electronic record. I, as a patient, have control over it that has information about every doctor visit and every hospital visit I've had all going in one way or another, feeding in, so it's comprehensive, it's complete, and it will make care better. It will get rid of duplication. It'll get rid of all the bad stuff. I think it'll make care cheaper, hopefully, if we get rid of the huge amount of duplication that gets done in the health care system. The technical, cultural challenges to achieving that vision are many. And when you ask the question, why on earth, when I was seeing that patient at the Brigham, why did I have to go through this crazy process, for instance, is because at the end of the day, both the Brigham and the BI are convinced that it's to their advantage not to share this stuff electronically with each other. Well, it gives mobility to patients. I agree with the second part, but the first part is completely, the first part of what you said, the technical and cultural challenge, is not at all true. In other words, the patient has a right to take their information from the Brigham and give it to you at the BI. And that's not a cultural or a technical issue. It's only the second thing, which is the strategic. But there are cultural and technical issues, and it sounds like there are people who know far more about electronic medical records than I do, but I'm gonna wade in anyway. Just correct me as I get it wrong. There are technical issues around metadata, how to represent, what metadata counts and how to represent it. There are cultural issues around how much of the information, does doctors notes, is that part of the record? There are cultural questions around privacy in two regards. One is, does a patient whom you say has complete control over, well, okay, is the patient entitled to not pass along certain data that she may find embarrassing or harmful or not want the doctor to know that she tried for other doctors and how much control do we want to let patients have over it? And finally, there's this gigantic privacy issue of coming up with a way of identifying patients that can let two records be associated without having a unique identifier that can be tied to the real world person, which raises all sorts of issues about, for example, your right to protect your privacy about AIDS or STDs or whatever it is you don't want people to know. So I disagree. I think there are gigantic technical and cultural issues. Let me just respond to that. Yeah. Those issues apply when you have provided to provider communications. All of the issues you've said are huge technical and cultural problem. None of the issues that you brought up apply to an individual, as an individual, seeking healthcare services from a doctor because there's nothing networked about that. When I go to the BI, they can identify me biometrically. So the next time I come back, they know I'm the same person. I can't screw them up. They have their own copy of the part that they own, so I can't besmirch them or they can digitally sign stuff if they wanted to. So all of the issues having to do with patient privacy, identity management and all of these things do not apply relative to the way patients are getting together on the internet. They only apply with a provider-to-provider interaction. So I fundamentally agree with, and you're getting into the issue of personal health records that people can control. Why don't we go ahead and just, yeah. So I wanted to get to, I'm confused now between where you see the mechanism for this transparency having an effect on the system. Myself, the individual patient having access to my records versus a collection of data that exists about that you were presenting before. And so there's a, one is, there seems to be this mechanism about that the patient makes smarter choices versus the hospital's name smarter than that name. Right. In the smart system. Yeah. So I'm proposing a specific mechanism whereby this transparency actually causes change. Great question. And you're right, that those are two wholly different things related but pretty different. Let me give you a couple of thoughts about them. One is that I have some evidence on the hospital side, the data that I showed you. There's some other data that suggests that when you start making this stuff publicly available, hospital surgeons, doctors start paying attention and start doing things with this. It is slow. The response by these healthcare providers is hardly the kind of response. I mean, cardiac surgery 12 years later, yeah, it's a little bit better. It's hardly terrific. There's still a fourth eightfold variation. So my point is that that's a pretty slow but important mechanism for improvement. The other part that you bring up, the issue of consumer engagement, consumer access to information, this stuff is wholly untested until very recently. And I'd say even as of today, most people have no idea. Most people don't know what's in their electronic record. They don't know what medicines they ought to be getting. Now I'm not proposing that every patient should become a doctor and know exactly what all the right therapies are. But I am suggesting that the one way in which the world has changed very dramatically between when I was a resident in 1998, 1999, and now that I'm practicing is that the information with which patients come to see me is dramatically different. In 97, 98, if they had a disease, they maybe knew a couple of things about it. Now people come in with reams of information, tons of data that are out there. And right now it's mostly, most majority of doctors just find it annoying. Be perfectly honest, because people come in and say I read that I have blah, blah, blah, blah, blah. And then you have to just take a step back and how to work through it, et cetera. But to me that's just because that's where we are with information on the internet today is that for a lot of healthcare information, it's out there, but the sort of categorization, waiting, kind of figuring out what's good information, what's not so good information, that stuff hasn't matured as much as it could. And certainly it has not had the effect that I can figure out in terms of patients engaging with the healthcare system in a positive light. But I think that's, we're on the cusp of that really changing, but I don't know. I mean, this is the part I really want to get feedback on. I mean, what do you guys think? Is it one of the fundamental problems here that, and I think the problem with the economic choice, the consumer-based choice model is that, even the cutting-edge economists are turning away from the rational choice model as being flawed, that these are not rational choices that we're making, that decisions about our health implicate all sorts of ridiculous fears, not ridiculous, actually quite natural fears and concerns that we have, that our decision-making isn't grounded on number crunching our rational chances of survival, but whom we trust and people we know, and in some cases, people go into the worst case scenario and they automatically assume that they have every conceivable form of cancer, that people assume that they're indestructible and refuse to see the doctor. But either way, I mean, these are not rational choices. So what I'm asking about in terms of how the transparency of information will help us is, will it? Because it's not like my discovering that this TV is cheaper at this online retailer than this other online retailer. That is, that's apples, apples, numbers, and dollars, whereas you're talking about something that's both technically difficult to understand, emotionally difficult to grapple with, and where people aren't gonna be in the position to make these rational choices. Yeah, no, it's a totally fair point. And this is in fact the place where maybe the model of patient or consumer engagement with information as a transformative figure in healthcare won't work out. Again, for me, I'm not convincing will or won't. I'm really trying to find the empirical data. We don't have much yet. But I'm sort of thinking about how do we think that this information will begin to affect care? Well, one of the things that I'm gonna talk about is the whole role of the health plans and payers that are involved in this homemade multi-stakeholder A-tricks when we start looking at quality and transparency. And there's now a big push kind of moving towards what are called consumer directed health plans, where you have a high deductible couple with an HRA or HSA. Have you looked at, or is, I mean, these are so new, is there just not any data yet to really define, okay, now that consumers are taking on great responsibility for a high deductible plan, are they using this information differently? Or are they using this information, or are they starting to go to health grades or any of these other sites and starting to collect this information and look at it and say, okay, I'm gonna make my decisions based on their quality as well as their cost. So, right, great question. And basically, let me just sort of provide a little greater context for those of you who don't know the underlying issues, the underlying issues are one of the, when people on the right of the political spectrum look at this stuff that I just presented, they don't disagree with any of it. They say it's very obvious what the problem is, which is consumers are not behaving like consumers because they don't have any skin in the game. They don't have, it's actually a bad phrase, but that's what they use it, that's what they use. So, the argument goes, you know, you have no idea how much things cost. So if it costs 400 bucks or 200 bucks, why do you care? Your copay is 15 bucks. If any other part of the economy ever worked that way, yeah, costs would go through the roof. People would always buy more expensive stuff or they wouldn't care a certain. And so what we need is transparency. And when you have transparency in terms of quality information linked to having more skin in the game, you actually have to pay for the entire visit out of your own pocket. If Sydney had to pay for her visit, she would know exactly how much it costs. But she has some big notion that it costs about 200 bucks. Don't mean to pick on you, right? She would know exactly how much it costs and if it costs 200 bucks, she might actually say to the doctor, 150, and try to negotiate. And maybe that would affect things. The problem is maybe about two, three percent of Americans have these high deductible health plans and they're an atypical bunch. And we don't know how this is gonna play out. My personal feeling is it's not gonna be big. It's, these people are not in it because they wanna be activist consumers who wanna, who think they can get a better deal by doing, they're in it because that's the only thing their employer offers, because it's cheap. And I'm not sure. Based on really the arguments that you've made, I'm not sure people are gonna act like rational consumers in this context. But it seems to be growing in terms of the number of plans and the number of jobs. It is growing, but I, you know, do I think it's gonna ever hit 20%? I doubt it. I think they're gonna peter out. The plateau pretty early. I'm very optimistic about this, despite Jeans. I'm not. That's what I'm gonna say. So tell me about your optimism. I'm excited about optimism. I think there's, I think there's someone that depressed you. I think there's a huge role for intermediaries in this and that the intermediation is not working well right now and that you go in and you have a certain person who gives you information and you have no way to gauge whether that information is good. And then the secondary referrals that this person is giving is good or not. And then additional layers. It could be better doctor to doctor referrals. It could be better management of hospitals. It could be better weeding out of poor doctors. It could be an entirely independent intermediary that looks at healthcare outcomes and advises people like me that couldn't figure it out whether what works and what doesn't work. And I only see upside in this. I think it's great. The other thing that I would like to see is it might be able to bring the notion of quality into healthcare, which I don't, if it's there I don't know how to get it right now. I mean my healthcare is coverage and procedures which is a really a bad proxy for good healthcare. Correct. And that good doctors is a much better proxy for that than procedures. Yeah, or I would say good systems. Yeah, probably the most important thing. No, I obviously on some level I'm optimistic about all the stuff you said too. It's really my take is it's the beginning of a very interesting journey about how transparency, better information. Because if the bottom line is that someone like you can't navigate the health, I mean, I can't navigate the healthcare system half the times. And I'm like a physician who studies the healthcare system. Like this is terrible, but if smart educated people can't navigate the health, you can imagine for the rest of the country what a disaster it is. But it does provide this opportunity. And I always like to remember that $2.1 trillion leaves plenty of interesting business models and cash for people to come in and act as effective intermediaries. Yeah. I think one of the interesting things, I'm looking at other places where the internet has had sort of one way or another some kind of transformative effect is that there had to at some point be a really kind of disruptive critical mass of sort of freaky data getting out there and then people get comfortable with it or it gets, we sort of set norms about that. And I think, I don't know. I mean, I'm optimistic this is really interesting, but I think people are so concerned about healthcare data being aggregated, being identified with an individual person and affecting every part of your life that I think it's gonna be slower or maybe scarier on any other. Yeah. No, I think that's right. I think there are some special issues around healthcare that have to do with privacy and confidentiality and that are real. I also think that my take has always been, and I was talking a lot about this just before, is I always sort of feel like that stuff is solvable. Well, it certainly points more towards better sort of hospital to hospital data sharing is that I'm more optimistic about that than I am about something sort of disruptive and on a internet sort of grassroots level happening. Yeah. I'm less optimistic about the hospital, the hospital sharing. Maybe. No, for cultural reasons. I mean, again, I always come back to the BI and the Brigham who are two Harvard teaching hospitals 50 yards from each other. You can throw a stone from one and hit the other, but you can't electronically send a file from one to the other. There's a problem. Yeah. Where can you see the possible impact of care worse? Yeah, you know, we don't have a single care, but we have a government paying for 45% of healthcare. So pretty from the government, it's already the big payer. And Medicare has been doing some innovative stuff around some of this, but it's sort of drops in the bucket and the stuff that they're doing. I don't necessarily see. They're doing P for P this year. They're doing P for P? Just introducing it. They're just introducing it and it's tiny amounts of money. And yeah, it's little disolts. I can actually address that the feeling and the people that talk about this that I get is that Medicare is actually one of the most evil things that's going on in this category. Yes, because everything in Medicare is set up as fee for service. And the thing you mentioned before that drives a lot of it's either quantity or quality thing is actually part of the reason this problem is so intractable to the extent. And this is not something I care a lot about certain things I would do for privacy. This isn't something I care about. I think I'm a pretty objective reporter of this. What people who study this basically say is that it's never gonna be fixed until Medicare starts paying for preventive services, for having a medical home, for having things on a subscription basis that take care of your health as opposed to those things which are fee for service. And increasingly the private insurers are going in the direction of doing what Medicare does. Because Medicare is so dominant, they're actually less likely to innovate in their own right. Let me offer some, I don't actually disagree with too much of that but I do disagree with some of it. HMOs were an attempt to do exactly that stuff, offer more preventive services, do all the stuff that fee for service doesn't because they had a motivation which is keep patients healthy, manage costs, et cetera. I'm not sure, I mean they certainly did a pretty good job on quality actually. The quality scores for Kaiser are much, much better than they are national. Be very clear about that. But it hasn't worked out so well with the American public. And part of it is because right now we have a couple of models. We have fee for service or capitation. Capitation is I as a doctor get 50 bucks a month per patient and I take care of everything. And then the incentives to get set up, well maybe Sydney doesn't need that CT scheme because it's expensive and it comes out of my pocket. And that creates its own dynamic in terms of the relationship and a lot of people, doctors and patients really struggle with that. But there's a third way which I think the gentleman mentioned which is that you have, you don't have to have either of those two. You can have an independent third party that aggregate patient interest or that serve that are not providing services that play that intermediary role. And that hasn't really been seen outside of their medical tourism and buying drugs from Canada space. Oh I'd love to see it happen. And concierge medicine. Right, I'd love to see it happen. Okay, is it time we're doing okay? Any other questions or thoughts? Yeah. What do you think about the presidential candidates' proposals and prospects in the next four years? Yeah, one of my favorite things about all this of course is that everybody talks about one issue, the uninsured. Now don't get me wrong, I'm not here suggesting that it's good to have uninsured Americans. It's awful. But it's one part of a complicated set of issues and there's a reason we have 50 million uninsured. 47 million. And it's because healthcare is extremely expensive. There's some chunk of Americans who have made the absolutely rational choice not to buy healthcare. It's too expensive and they're pretty healthy and they don't need to buy it. There is the safety net problem of but if they get into a car accident they know that the government's gonna bail them out because they're gonna pick up the $50,000 tab of the three weeks of hospitalization that might come out of them. So that is sort of problematic. I think the presidential candidates are, I don't know if John McCain even has a health plan. I've looked at both Obama and Clinton. We both personally have one. Yeah, exactly. You should have done that. I don't know if he has one for the country. And I think that at the end of the day most of the health insurance experts I talked to suggest that Clinton's plan is a little bit more realistic in terms of getting us to where we need to go in terms of most number of people insured. But ultimately, just like Massachusetts all of this falls apart if we can't get a grip on healthcare costs. Because as long as healthcare costs continue to be as high as they are and rise twice the rate of inflation it becomes untenable for the government to keep doing bigger and bigger, bigger subsidies. It's just very difficult. So I, and I don't think either of them has, either Obama or Clinton has gotten serious about healthcare costs because they don't have to right now. They're in the campaign phase. Why would they get serious about how to, so they both talk about containing costs. But you know, sort of to me it's like, sure, we all want to contain costs. We all want to improve quality. We all want to improve access. How on earth do you actually do it? And I don't think either of them has talked seriously about that. How do you actually do it? How do you actually? How do you actually contain healthcare costs? Oh man. You have a pen in your mouth. No, I don't. Title of the talk today is not containing healthcare costs. So I do think that in the longer run things like electronic health workers are going to be very, very helpful. There is a ton of waste in the healthcare system. A ton of duplication of stuff that will get better with better information. I think payers have to start paying more for outcomes and less for individual things. So instead of saying you'll get more, if you do four more tests, it's just you get a certain amount of money to manage this patient. But yeah, it creates that tension that I talked about that, you know, and then there's a lot of, there's this whole movement now, somebody, what's called comparative effectiveness, which is, okay, so you got a guy who's had six months of low back pain, okay? Disabling low back pain. The MRI doesn't show any major problems. What do you do? Well, depends on if you live in Boston versus New Haven versus New York City, you're gonna get dramatically different therapies. Because it's whatever the local surgeons think you ought to do. And you see eight 12-fold differences in rates of back surgery in towns right next to each other with identical patient populations, because one town has a back surgeon and the other has two back surgeons and the other town has one back surgeon. And everybody's gotta be fully employed, right? So it's payers starting to take much more of an active role in saying, we're just not gonna pay for the back surgery for which there's good data, this is not gonna help the patient. But payers have not wanted to get into that because you can imagine the politics of it gets very, very complicated. That's the key point that we see being rational about this stuff. Yeah, I've had six months of back pain and my insurance company won't pay for my back surgery. This is an outrage. Even though I can convincingly tell you that back surgery is not gonna help you, doesn't matter. It still feels like an outrage and people are not. And there was that whole thing that's same now with the young woman and the operation that they approved at the last minute but basically the critical evidence was. Yeah, and you know, she was only gonna get a couple more months of life anyway. And you know, it is a complicated set of issues. I mean, there was a whole thing in the 90s around bone marrow transplant for breast cancer. And we were, and you know, the insurance company said this stuff is experimental. And you can imagine the 38 year old woman who's got metastatic breast cancer. She's got three small kids at home. Her only chance of surviving is a bone marrow transplant and the big mean old insurance company says no experimental. Well, you can imagine how well that goes over politically. It's a disaster. And so insurance companies just all just lie down and said, fine, we'll pay for all of this stuff. Part of the, let me just finish, until somebody did a good randomized control trial that showed that it didn't work. Part of the answer to your question is that the amount of money that's spent out of 2.1 trillion in the last six months of life is shockingly high. And if you count the amount of money that's not spent in the first 30 years of life on things that would potentially increase, keep people well and completely out of the insurance and a health care system, it's even scarier. So the amount of distribution issue in terms of the way the incentives are set up, be it single payer or not, insurance are phenomenal. The question is how do you in real terms start fixing this? And there are two thoughts. And one is that a lot of my folks at Dartmouth have done all the really good work on the payments of the last two years of life stuff. And I'm good friends with them, they're terrific people. I always remind them, I as a doctor don't always know what the last two years of life is. And sometimes you do, but you don't always know when this is the terminal event. So it's always post hoc analysis that they do. Someone's died, then they look in the last six months. Okay, if I had that sort of clarity of vision as a clinician, my life would be a whole lot easier. I don't. The second problem is there is a huge issue around expectations and management of people at the end of life. You know, patients come in, they're very sick, families have not grappled with the fact that this patient has a terminal disease and is gonna die. And they want the patient to go to the ICU. And it's very hard for me as a doctor to say, no, I will not, I'm gonna have the patient die right here in the emergency room. I generally can't do that. I'm not laying that out at your seat. No, no, no, I'm not, it's a cultural issue. And so it's very hard to, so from a policy point of view, how do you begin to fix this stuff? Well, you gotta educate the American public. Good luck, you know, change people's expectations about healthcare. Terrific, I mean, these are great platitudes, but how do we actually do it very, very difficult? You know, the more the people on the right would argue, whether it's the first 30 years of life or the last couple of years of your life, at the end of the day, you know, people have to take personal responsibility. I spend a lot of time thinking about what do my kids need and not need in terms of healthcare and vaccinations and all that stuff, personal responsibility. And the second thing is, the reason we spend so much money at the end of life, part of it is because no one, the people who are paying for it is the government's paying for it. And if individuals actually had to pay more of it, they would think about, is this valuable? Do I wanna bankrupt my family for something that's completely useless? But those are very hard issues to talk about. And, you know, and do you want your 85-year-old grandmother making the decision of, no, I think I will die because I don't wanna bankrupt my family? That stuff is not the kind of conversation that we in America have been willing and open to having. And we just say, no, no, no, just pay for it. You're rational. You know, it's sort of like the old thing of like, you know, people talk about why does healthcare reform not ever come to be, even though every five, 10 years we talk about it. And the argument is, because the status quo is everybody's favorite second choice. So everybody has their first choice? Single payer, please. House savings account and consumer directed healthcare, please. Pick them. And then if you say, okay, you can't have single payer, what do you want? I'll live with status quo. And everybody arrives at status quo because it's their second favorite choice. As long as it's their second favorite choice, we don't move. Am I gonna need depressed again? The ending was supposed to be uplifted. Internet will change the healthcare system. Nothing will change. Yeah. Status quo rules. Anyway, all right, that's it. Thank you so much.