Prelude: In his long and distinguished career, Richard “Buz” Cooper has practiced as a hematologist-oncologist, cancer center director, dean of a medical school and founder of a health policy institute, and he is now a Senior Fellow at the Leonard Davis Institute of Health Economics and Professor of Medicine at the University of Pennsylvania. He also is co-chair of the Council on Physician and Nurse Supply at Penn. Above all, he is an independent thinker who looks at health care from the vantage point of a practicing physician and as a student of economic and demographic trends in the U.S. and the world and how these trends impact physicians. In 2002, he was among the first to predict a growing physician shortage, which may be as great of 200,000 across all specialties by 2020 to 2025.Setting the StageQ: Let me set the stage by quoting a paragraph from your 2004 article in the Archives of Internal Medicine (“Weighing the Evidence for Expanding Physician Supply, 2004, 141: 705-714)
“Taken together, the data, forecasts, and signals indicate that physician shortages are upon us and are likely to worsen over time. The picture that emerges is uncomplicated and unambiguous. In simple numeric terms, the number of physicians is no longer keeping up with population growth. The ability to fully service the population is further compromised by the increasing complexity of the care that physicians provide and the decreasing time commitment that physicians are willing to make. These limitations collide with economic trends that predict a growing demand for physician services.”
Does that reflect your current view?
A: Yes.
Cooper BackgroundQ: What are your background and your current position?
A: I started out as a hematologist. I trained at Boston City Hospital on the Harvard Medical Service. I went on to Penn more than 35 years ago to develop the Hematology-Oncology Section and later the Cancer Center. Fifteen years later I ended up as dean of the Medical College of Wisconsin in Milwaukee and did that for almost ten years.
Q. When did you become interested in health policy?
A: During the Clinton health reform effort, there was a lot of talk that half the physicians ought to be primary care doctors, and there should be fewer specialists. That didn’t make much sense to me. One thing led to another, and I wound up getting interested in what kind of physicians were needed, and how many should there be. While I was Dean, I began a Health Policy Institute and led that for ten years after leaving my dean’s position. About five years ago, I moved back to Penn to be a Senior Fellow in the Leonard Davis Institute of Health Care Economics and a Professor in the Department of Medicine.
One Foot on the Ground, the Other on the DataQ: So you’ve had one foot on the ground as a practicing physician and the other foot in the policy arena. Do physicians and policy wonks have different mindsets?
A: For most of my life I was in an academic practice of medicine, and I was head of the Division of Hematology-Oncology, and later a dean, who basically serves as head of a multispecialty group dealing with doctors and hospitals. In the process, I learned how things work and how people think. I came to the health policy world fairly late in life.
What I discovered was a lot of smart people trying to figure out how the health system works from data alone. But you have to live it. You can’t figure out why a baseball team wins or loses from statistics. You have to understand the mind of a baseball player and the dynamics of the game. The same is true a hundred-fold over in health care.
Figuring Out What Works in Health Care On the policy side, the game doesn’t work the way people think it does. Policy folks get attached to their ideas, and they try to fit everything into those concepts. The whole culture builds up around the defense of the indefensible. You can see that in the health plans as they evolve and in health reform. You can’t figure out how to save money without really knowing how the thing works.
Take prevention. It saves lives, and it adds quality to life. But it doesn’t save money.
Or take the notion that physicians cause too much health care spending. There are a thousand anecdotes where physicians churn the system, but in the main, that’s not how the system works. It’s disease that causes health care spending and it is technology applied to disease that increases much of that spending but, mostly, it’s the state of the economy that allows the spending to occur. Economic growth is the fundamental basis for health care spending.
You can look at health care from either end of the telescope, and it will look different. I see it from the clinical practice end, not from the green-shaded end.
Health Affairs Article Predicting Physician ShortageQ: I first came across your work in a Health Affairs article in 2002 in which you said economic and demographic trends were signaling an impended physician shortage. It was prophetic, but it ran against the tide of the health policy community’s opinion, which had long forecast a physician surplus, particularly of specialists. Explain how that article came to be and what the reaction was to it.
A: The article grew out of interests I had developed during the Clinton health plan. It started in an innocent way. I was looking at in data from COGME (Council of Graduate Medical Education), which had been constructed by the Bureau of Health Professions. It indicated how many physicians there would be in the future and how many physicians per capita there would be. And it predicted a surplus of about 150,000 physicians.
Well, I am a data wonk because I had done a lot of research in hematology. I have a quantitative mindset. So I did a simple calculation to determine what the population was likely to be in 15 or 20 years. For the COGME model to work, the population had to stay constant. That didn’t make sense, so I called the Census Bureau. They said, oh no, we thought that the before the 1990 Census but we don’t think that way anymore. So they sent me their projections in an envelope – not an email – email hadn’t been invented. I took their numbers and plugged them COGME’s physician supply projections, and sure enough, there was not going to be a surplus of about, but a shortage.
Presentation at AMA Meeting I presented those data in an AMA meeting, and Phil Lee was on the podium. He was the Undersecretary for Health at the time. That was his first knowledge that the COGME data was fallacious. He called the powers that be and told them they had gotten it wrong.
So they developed new model based on demand. Lo and behold, their new model again showed a surplus of 150,000. They came up with the figure of 150,000 too many physicians no matter how they did it. Like Carnac the Magnificent on the Johnny Carson show – they had the answer – all they needed was the question.
Policymakers have been producing phony numbers all the way back to the GMENAC (Graduate Education National Advisory Committee) in 1981. Now the Dartmouth Group is producing the phony numbers. The root idea is to prove there are too many specialists and to do it in as creative and elusive a way as possible. It was hard to figure out what GMENAC and COGME had done wrong. Dartmouth was an even bigger challenge, but I’ll get to that later.
Coming Up with the Right NumbersQ: How did you come up with the methodology for projecting the demand for physicians?
A; Well, I was given a contract by COGME to figure out how many specialists would be needed. I thought that there would be something to draw on to build a model from previous workforce studies. But there was nothing. COGME’s way was no good. The Lewin Group’s way was no good. Other consulting groups used variations on those methods and arrive at the same fallacious conclusions. The group that I had assembled was left with a contract, a date to report the findings and no way to proceed.
The First Principle – The Money Available Q: So how did you figure it out?
A: I went back to first principles. If you look back to chemistry, you ask, what is the limiting factor in a chemical reaction? So, what is the limiting factor in health care? There are infinite ways to take care of patients and more ways invented all the time. And patients have vast needs and even greater desires for health care. So, neither of those is limiting. But money is limiting. Health care is determined by how much money is available. It’s a hard concept for physicians to accept – it was for me. But it’s true.
If you look back over the entire period beginning in 1930s, you find that the growth of health care spending tracked the growth of the economy, but with a lag of about four years. Tom Getzen, an economist who has taught me a lot, showed that relationship. And if the economy slows, health care spending slows – we’re seeing that now.
The number of patients hasn’t changed. The amount of disease hasn’t changed. The number of doctors hasn’t changed. But there’s been a decline in hospital utilization of about 8%. The effect of economic growth on health care spending is slower. It plays out over 3-4 years, as benefits plans chance, hospital staffing changes and patients feel more comfortable committing more to health care. More is available to treat disorders that were previously untreatable, or if treatable, unattended. And innovation builds in the wake of economic growth.
As a rich nation, we tend to think we can have as much health care as we want. But we can’t, and we’re seeing that right now. We’re no different than Sub-Saharan Africa. They have an AIDS epidemic but they don’t have sufficient funds to do all that is needed. We have more money, so we do more. But we still can’t do all that is needed.
So, ultimately, it isn’t how much patients need, or how much they want, or how much technology is available to care for them, or what doctors might want to do for them. It’s what society is able to purchase. That’s the discussion that is going on now.
Health reform is about assuring that everyone is covered by some health plan, but after that, it is about how to rein in spending to what the nation can afford. Not what is needed clinically or desirable personally – it is what is affordable collectively. The struggle in the political arena is whether more will go to those with lower income or not – how much will we as a society share? Just today, the Wall Street Journal had an editorial on SCHIP, the children’s coverage bill, and said it would lead to single payer system, the ultimate sharing, and they opposed it. Our country is divided over how much to spend but even more over how much to share.
Over long periods of time, the total amount spent on health care has increased about 1.5 times as fast as the growth of the economy overall, as measured by GDP on a per capita basis. That makes sense. Health care is a growing part of the economy, and many other things are not growing. Food, clothing, transportation, household goods are not growing parts of the economy, so how does a nation grow its economy? Growth is in electronics and in leisure and travel. It is in new inventions. And health care. Not more of yesterday’s health care. The growth is in new health care – stents, MRIs, and other things that were not in our vocabulary 20 years ago.
Another important thing is that the US is not a homogeneous country. It is large and economically diverse. Health care spending is distributed in odd but predictable ways. Since growth is driven by economic growth. It should not be surprising that growth in health care spending is greatest in areas of the country with greater wealth. But the oddity is that those same areas tend to have a lot of poverty – think of dense urban centers – affluence and poverty side-by-side. Wealth creates the capacity for health care. But it is low-income individuals who use the most health care resources. Wealth is a source of health care creation; poverty is a source of health care consumption.
Two Fixed BeliefsQ: Two of the fixed beliefs of the policy community are supply-induced demand, meaning doctors drive demand and costs, and another is that there ought to be equal distribution of services without regional differences. Comment please.
A: There’s no question, those are the dominant views. I tremble every morning when I open the newspaper, because some reporter or editorial writer will voice and reinforce those views. The New York Times buys in 100 percent – doctors are at fault, and we ought to make health care uniform everywhere for everyone, and if we do we will save 30% -- the 30% solution. Easy money -- sounds like Madoff. Well, it is not much different. A lot of people have bought it. But it’s not there.
The supplier-induced demand theory was spawned in the late 1950s with the notion that the number of hospital beds relates to the amount of utilization and the number of surgeons relates to the number of surgeries. So the notion was the surgeons cause the surgery and beds cause the utilization. Not long after, David Dranove, an economist at then at Northwestern, published a paper showing the number of births was directly related to the number of obstetricians. If you looked only at the data, you might conclude obstetricians cause pregnancies. The cart was before the horse.
Where Doctors Go to PracticeDoctors tend to practice where there are resources to support medical care. Physicians go where there is demand. That’s where they are recruited to. Few just hang up a shingle. Do they induce demand? Undoubtedly examples exist of physicians who churn the system. But in the main, they go to where there are resources to pay for care – that is, where there is demand for care.
By the late 1990s, most people gave up on the idea of supplier-induced demand. When people looked at Medicare data, they found that there was little increase in service with decreases in fees. It surprised the Medicare people that the volume adjustments they built into their models did not materialize as they expected.
Target IncomeThe notion of supplier-induced demand is associated with another notion called “target income,” the income that physicians expect to achieve. The thought was that if physicians were not earning enough, they would have patients come back more often or do unnecessary procedures so that they could reach their target. That undoubtedly happens to some extent.
Somewhere, sometime, some place, some doctor is doing something to make an extra buck. But it tends to be evanescent and it is not pervasive. There is a moral imperative. And there is peer pressure. And there are watch-dog activities through insurance claims. Or the opportunity to do so disappears. More often than not, it isn’t the target but the unconscious enthusiasm that physicians have for what they believe is good for their patients.
Medical effectiveness studies sometimes prove them right, and sometimes wrong, and the system changes. There is a lot of uncertainty and a lot of pressure for more uniformity. It is a dynamic process. The system is imperfect, but it strives to be more perfect. Information technology will certainly help to smooth the unevenness.
In an interesting study of target income, folks at Thomas Jefferson University Medical School did a follow-up study of all of their graduates, and what they found was that most doctors achieved their target income – some a high target and some a lower target. Among those who did not, three things changed: 1) they provided less charity care; 2) they did less teaching; 3) and they changed specialties. That’s what doctors do. Churning the system isn’t a very popular avenue.
Regional VariationsEver since Wennberg and his colleague published their now famous article in Science in 1973 on small area variation in Medicare services, people have asking: Why is the level of care different in different parts of the country?
The singular answer is that it’s because there are different numbers of physicians in different areas and physicians induce the demand for what Wennberg and his minions call “supply-sensitive services.” But after studying this for almost two years, it’s very clear to me that the underlying phenomenon is not caused by physicians – it’s caused by economic dynamics like those that we have already discussed.
Regional variation is a product of regional differences in wealth, overlaid with differences in poverty. It’s not generally appreciated that health care expenditures for people in the lowest 15% of income are 50% to 100% greater than for people of average income. There’s also a difference at the high end. The wealthiest 15% also consume more, but only about 20% more. So there’s greater utilization at both ends of the income spectrum, but for different reasons and with different outcomes.
More spending at the high end improves outcomes, not simply for a specific condition but across the board, because the care consists of a broader spectrum of beneficial services. More yields more. But among the low-income patients, outcomes are poor despite the added spending. In fact, the added spending is because of poor outcomes – more readmissions, more care for disease that’s out of control.
And these differences are exaggerated in dense urban environments, like Detroit, Chicago and Philadelphia. Now, when you blend all of this into “regional” studies, which average rich and poor, urban density and ex-urban comfort, racial and ethnic groups, you get just what you’d expect. High costs with average outcomes in urban areas (the average of excellent and poor outcomes at different ends of the income spectrum).
A good example is the Dartmouth study of academic medical centers. You find that one group of academic hospitals provide more care than another group. The Dartmouth folks say that Mayo is more “efficient” in resources used per patient or in number of doctors devoted per unit of patient care than in LA, Philadelphia, Miami, Chicago, and New York City.
But the so-called “inefficient” hospitals are all in dense urban centers, while “efficient” hospitals are all in smaller cities, often college towns liked Madison, Wisconsin or Columbia, Missouri, or in places like Rochester, Minnesota, where Mayo is located. Rochester is 90% Caucasian with low poverty. But in fact, Mayo is the most resource intensive center in the upper Midwest. Among peer institutions in similar socio-demographic environments, Mayo actually uses more resources. But you can’t compare Mayo to Los Angeles, where only 30% of the population is non-Hispanic white and where you have tremendous pockets of poverty.
The Dartmouth group doesn’t acknowledge the fact that there are enormous social differences between populations served by academic hospitals in various cities and even in the same city, where patients distribute in a non-random way. If you ignore these fundamental considerations, you can make the numbers fit the preconceived notion that there is more spending where there are more doctors and doctors cause the spending. You can “prove” that it’s the fault of specialists. But in the movie of “The King and I,” Yul Brynner, the King of Siam, tells his son to watch out for “people who try to prove that what is not so is so.” And they do. But that’s because they ignore the complexities of social structure and get it backwards. Doctors go to where they are needed, and the needs in urban centers are huge.
Fitting Conclusions to Fit Preconceived TheoryQ; Are you telling me Dartmouth comes to conclusions that fit their theory?
A: In my view, they are so committed to the “30% solution” that they don’t want to know more. But they must know more. It’s too easy to observe. For example, most of their studies depend on Medicare expenditures, which they assume represent health care spending overall. But it doesn’t. There’s no relationship between Medicare and non-Medicare spending in communities. And Medicare spending doesn’t correlate with the volume of care in a community – or even in a hospital. But total spending – spending from all sources -- does, and it’s the only valid measure. So their famous map of Medicare spending is not representative of health care spending overall. A good example is their well known papers in the Annals of Internal Medicine in 2003 – they are the ones that are quoted the most, even by Daschle and Baucus and folks who are part of the Obama team.
Where the Most and Least is SpentBy slicing and dicing and then mixing and matching, Dartmouth collected areas around the country with high Medicare spending, but some also had high spending overall while others had low spending overall. They then took these and constructed a “high-spending quintile.” It was comprised of most of America’s major cities – Chicago, Detroit, Pittsburgh, Philadelphia, New York, Boston, Houston, Dallas, New Orleans and also Los Angeles.
And then they average everything – north and south, rich and poor, good quality and poor quality, high total spending and low total spending. All these areas had in common was high Medicare spending. So that’s the high-spending quintile. You’ll never guess what it was compared with. The comparison group was the entire area extending from Alaska through Washington and Oregon to Wyoming, Montana, , Kansas, Nebraska, South Dakota, Minnesota exclusive of Minneapolis, Wisconsin exclusive of Milwaukee, and then across to Maine, Vermont, and New Hampshire. The northern tier. Sparsely populated. White. Non-urban. And cheap.
And you would expect that things would be very different in these vastly different “regions.” But everything was the same. Quality, access, satisfaction, even mortality. When things were average in each of these heterogeneous groups, it was all the same. Differences were not discerned because differences were not discernable. But, then, if they had, what could be made of it. Why would anyone want to know how Newark compares with Nebraska?
Most people I went to school with would have said, “Oh gee, there are no differences, I must have done something wrong.” But not the Dartmouth crowd. They said that because differences were not found despite all of that extra spending, the extra money must have been wasted. And if health care could be the same in both regions, the US could spend 30% less. And everyone believed them! Remarkable! They did it! They proved that what is not so is so. And so, as the sun sets on America, we can all sleep comfortably, knowing that if only Manhattan could be like Montana, all would be well for health care – and we’d save 30% in the process, enough to pay for all of the promises of health care reform. Dream on.
Specialists and Quality of CareBut even worse than Dartmouth’s 30% solution are the studies in states that were carried out by some of their associates at Harvard. The famous one liner that came from that is that “states with more spending and more specialists have poorer quality health care.” It’s quoted everywhere – twice in the current issue of Health Affairs. But if you look at their study in Health Affairs a few years ago, you’ll find that the state with the most specialists and the most Medicare spending, and also the poorest quality, is Mississippi.
Q: Mississippi?
A: Yes, it’s Mississippi, the poorest state in the nation. It does, indeed, have poor quality, but how could it have the highest spending and the most specialists? The answer is it doesn’t. Mississippi, as you know, has the fewest specialists, and although it does have high Medicare spending, it has very low health care spending overall. It’s not surprising that low total spending and few specialists are associated with poor quality. In fact, when all of the states are examined, more total spending and more specialists are associated with better quality – just the opposite of the Dartmouth-Harvard message but just what you would expect.
You might wonder how they arrived at the opposite conclusion. Well, they never really measured how many specialists were in Mississippi or anywhere else. They did some statistical maneuver where everything was converted into residuals, and I guess that Mississippi has a lot of residuals. It just doesn’t have a lot of doctors.
I published my observations about these studies in two papers in the December 2008 issue of Health Affairs online. But much to my surprise, they were accompanied by two rebuttals from the Dartmouth crowd, each with summary statements by the editor that said I had simply reconfirmed the Dartmouth work.
But it all made sense when I learned that the new editor of Health Affairs, Susan Dentzer, is a Member of the Board of Overseers of Dartmouth Medical School, the former Chair of the Board of Dartmouth College, a former Trustee of Dartmouth-Hitchcock Medical Center and winner of the alumnus of the year award from Dartmouth. She has a profound conflict of interest which she failed to reveal in her editorial – an egregious ethical breach. So, it all made sense. And it all is rather remarkable. Fortunately, truth has a way of surviving, and the truth is that states with more health care spending and more specialists have better quality health care.
The $640 Billion Dollar Question Q: I’d like to finish up with a few concluding questions. The first is a $640 billion question. How will this deep recession we are in influence the physician shortage?
A: Well, it won’t influence the shortage long term unless the recession continues for years. If it does go on for many years, it will influence everything, and we’ll have a country that we won’t recognize. The assumption is that growth will pick up again in 6 or 12 or 18 months. After that, we’ll be back on track of GDP growth of about two percent a year. Averaged over a decade, growth will be at the historic rates, and it is these broad averages that determine the needed supply of physicians. Not that the short term changes don’t matter. Physicians will be busier if the economy has a spurt, just as some are now becoming a little less busy as the economy sags. But it takes a long time to train physicians, and the training decisions have to be based on long term trends.
Q; You have highlighted other factors that aggravate the physician shortage. For example, you point out that by 2020, 60% of medical students will be women, and women spend 20% to 25% less time in practice than men.
A; Yes, and the figure of 20% to 25% may be overly optimistic. Women physicians these days are increasingly dropping out of practice altogether in their 40s or early 50s. And men are seeking better lifestyle arrangements, too.
Q: You also have pointed out the physician shortage involves all physicians – not just primary care doctors.
Primary Care – Yesterday’s Concept A; Yes, but I think “primary care physician” is yesterday’s concept. Primary care is something that many providers engage in. Generalist physicians, as I prefer to call them, have special roles in primary care, and these roles are still being defined. Some generalists are hospitalists. Some are rural practitioners. Some supervise teams of non-physician clinicians who provide much of the uncomplicated care, like treating upper respiratory infections, following mild hypertensives and stable diabetics, doing well baby exams and dealing with a lot of acute, usually self-limited disease.
A minority of what generalist physicians in urban settings now do is chronic disease management, but I believe that aspect will grow and that generalists of the future will be dealing with panels of patients who are sicker and more demanding. What I’m not sure of is how many will be needed. The problem is that there will be too few of all specialists who take care of chronic illness – oncologists, cardiologists, urologists, and many others. And we’re all interdependent.
So at this point, I think that the emphasis in generalist medicine should be to define roles for the future and construct training programs around those roles. My choice would be a track for rural medicine and a track for urban generalists who focus on chronic disease management and on overseeing teams of non-physician clinicians, but the leaders in the field prefer the model of the “medical home.”
Q: Doctor Paul Grundy, Director of Healthcare Transformation at IBM and champion of the medical home, calls such a doctor a “comprehensivenist,” and specialists sty “partialists.”
A: Whatever you call them, we’re going to need high level generalists – fewer than the number of primary care physicians that we now have but thoroughly trained for challenging practices.
Medical Home Concept
Q: What do you think about the “medical home” concept?
A: I think it’s applicable to children, but probably not to adults. It can probably work for employed groups, but I don’t think it’s applicable to the Medicare population or to low income patients. For some adults with chronic disease, generalists will be home base, but those generalists will be fully occupied with such patients and I don’t think there will be much time for wellness care and other tasks. There just won’t be enough physicians.
For other patients with chronic illness, specialists will be their home base, probably assisted by a high level nurse practitioner who provide much of the general care for the specialty patient. But even if the medical home works, it can’t work for everyone. Physicians are going to be compensated to provide more care per patient and to spend more time with their patients; it’s what I call concierge “lite.”. But those physicians will not be able to care for as many individuals. I can spend more time with you, but then I can’t spend as much time as with your neighbor – or maybe no time at all.
Q: A final question. What direction do you think health reform will take, and do you think government alone can reshape the system?
A: Health care reform will take one of two directions.
One approach, which was the Clinton Health Plan approach, is to do everything possible to restructure everything for everybody and to re-design a theoretical system that won’t work. I hope that doesn’t happen.
Alternatively, government can concentrate on the one thing it can do: create an insurance system that puts everybody under the insurance umbrella and leaves everything else alone. Daschle and Bacchus and others have conceptualized this approach. Each has a similar way to create federal oversight that somehow makes it possible for everybody to be insured. If they can do that, and it will be a Herculean effort, it might work.
But they also have other ideas. They want to impose regulations for medical effectiveness panels, P4P, different kinds of reporting systems and so forth, and that will sink it. Not that there aren’t things that they could do. For example, they could foster electronic medical records and support research in medical effectiveness. But remember, if the new health plan succeeds on the insurance side, it will confront what Massachusetts is confronting – the doctor shortage. Health care reform is absolutely on a collision course with the doctor shortage. Something has to be done about it, and it is spelled GME.
