In addition to being the Medical Director of CareGroup, the corporate parent of six Boston area hospitals, including the Beth Israel Deaconess, Dr. Kim Saal is a practicing cardiologist. Recently named to the post of Medical Director, Dr. Saal before that was President of the Provider Service Network, a CareGroup-affiliated group that represents more than 3,000 Boston doctors in contract negotiations with insurers. His success in managing numbers as well as patients was firmly established when he led the physician medical group representing doctors at Mount Auburn and Cambridge hospitals to several years of profits.

As a physician, my role is two-fold. One is to care for patients, and two is to educate future doctors. I saw that there was a need to educate doctors who are finished with their training. My role really became one of educating practicing physicians, and bringing quality care to physicians.

One part of my job is directly working with the payors, with the health plans, to maximize the revenue that comes into the system for physicians and for hospitals to care for patients. The second and far more difficult part is to work with physicians and hospitals to get them to work in the most efficient way to provide the best quality care for patients.

What is your territory? How big is it, and how many people?

The Provider Service Network represents 3,000 physicians, and seven hospitals. We care for over 400,000 patients.

How do you go about imposing standards and financial discipline on a domain that large?

The way we impose standards is multifactorial. If you cannot measure it, you cannot manage it. So the first thing we do is to develop standards by which we can measure care--how frequently a doctor sees a patient in the office, how commonly they provide mammograms, how often they do tests for diabetes. Given that we know the population of patients that we're caring for, we can then look to see if Dr. X is giving care at the same standard as Dr. Y. Then we give reports back to the doctors that really show their performance against the performance of other physicians across the entire system.

I want to understand how you got into this. Kim, speaking for lots of folks out there, they don't have much idea of what's going on here. What we're trying to do is to open this world up to them. . . . Most of what we've heard in this health care drama is that the health care plans are taking over everything. They're managing it, basically telling doctors what to do, and they were selling some of the patients.

What did you all do? . . . Two or three years ago, why did you all want to take the management back and take the risk along with it?

Years back, health care was pretty simple: the more you did, the more you got paid. It was a cottage industry, with each physician and each hospital on their own bottom line. As health care has evolved more and more, it's become much more of a systems approach, where one looks at the health care of a population. So the health plans then stepped in and developed guidelines--rules by which doctors had to practice medicine.

These guidelines were very often arbitrary. They were set up somewhere outside the local community, with no idea of the patients who were actually being cared for. Physicians, patients, and hospitals were all frustrated. So we developed a model here by which the physicians and the hospitals have taken back control.

We know our patients, and we want to work with them in their best interest. . . . We've shifted that oversight, that management of care, from the payor side--the health plan side--to the physician and hospital side.

Is it working?

It's a long, slow process. It is working. I'm not going to paint this as very, very rosy, and say you can do it right away. It is changing behavior of patients and of physicians, and that's a tough road to go down.

We can see some areas where it is changing. If you look at certain quality measures, such as how often patients are getting mammograms today versus how often did they get mammograms five years ago, it is improving. If we're looking at the care for the diabetic patient, it is improving.

Our goal is to try to reduce the variance in care. Variance in care leads to poor quality and excessive costs. Physicians need information that they can then use to manage their patients. I believe it will not only improve quality, but will also reduce cost.

You mentioned the upside. The upside is a better record on mammograms, and better treatment of diabetics. But you were hesitant when I asked if it's working. What causes you to be hesitant? What isn't working the way you'd like to see it? Where are the holes, and where is the downside?

The downside in managed care is the fact that there are always incentives in medicine. In the 1970's, the more you did, the more you got paid. The incentives today are that the less you do, the more you get paid. What we're trying to do is find that balance between being incentivized and do more, versus incentivize and do less. We really want incentives to do the right thing.

Until one has good measures of quality and incentivized physicians for those measures, then you're not going to have the right solution. So what we're working on now is to develop that balance and experiment. Each year, we're changing it a little bit more to make it better and better.

In terms of financial incentives, what's the difference between the old way medicine was practiced and the new way?

In the old way, when a patient was admitted to the hospital, the health plans paid physicians and hospitals for services that they did. For example, if a patient stayed in the hospital for a week, the physician was paid for seven days of hospital care, and the hospital was paid for seven days of hospital care.

In the new way, hospitals and physicians are pre-paid for all the care that the patient receives. Therefore, if the patient spends no days in the hospital or seven days in the hospital, the physicians and the hospital will see the same income under either circumstance. The reality is that we need to develop a methodology.

The incentive, then, is for the hospitals and the doctors to get the patient out of the hospital, so there's a lesser charge against the income they're getting.

Correct. The reality is, how do we develop a system which incentivizes both the doctors and the hospitals to have that patient only be in the hospital for three days, if three days is the right number of days to be in the hospital? How do we get to understand what is the right number of days, and develop a payment scheme that incentivizes physicians to achieve that goal versus more days or less days?

Where do you come into this equation?

Where I come in is that I'm not sure any of us know what is in the best interest of our patients. I'll just give you a personal example. I was a resident in training, and the person who was my teacher told me that this was the best way to do a certain procedure, such as drawing blood.

This is our hospital's way of doing it, which is the best way. Why did that doctor say that? Because that's what they were taught when they did it the year before. You then travel to another hospital, and that person training would be told, "This is the way to do that blood drawing procedure."

Both doctors believed that they were doing what was in the best interest of their patients. Nowhere did one try to compare how it was being done in hospital A to hospital B. The role that I'm playing is trying to collect that data, so we can really compare how hospital A is doing it versus hospital B is doing it, and come up with really what is the best way to draw blood.

What is the role of cost cutting in your job? What part of your job is devoted . . . to making this place more efficient?

Clearly, one of our goals will be to cut costs. The goal we set up for this year was to maintain our expenses at the same level as our expenses were last year. That's in face of pharmaceutical technology, which is eating up a bigger and bigger percentage of the health care dollar.

Other cost savings are ensuring that laboratory tests are not duplicated, and that data is transferred. When a patient registers, they don't need to register 17 times and give that same information, which means you need 17 entry people to collect that data. And then, when people try to do the billing, they get 17 different ways of spelling my last name and then make errors.

So that's waste.

Pure waste.

Why do people say that your job is a killer job?

There are at least two things I can think of. One is that we're not talking about widgets. We're talking about patients, and patients are very, very complex organisms. You cannot say that a patient should be in a hospital for three days or two days or four days.

Let's not put the round plug in the square hole. The second thing we're talking about is physicians, and really what has promoted physicians to live with the kind of care we're giving now in the US is that innovation. If you take away that innovation, you've lost what medicine has really brought to the US.

So there's a complexity of the patients on one hand, and the complexity of physicians on the other hand. Therefore, there isn't a simple plug that hooks up those two pieces. Although we want to reduce variation, we don't want to eliminate variation. If you totally eliminate variation, it's like building a computer. You will lose the innovative forces that are necessary to improve care in the future.

How did doctors respond to your "helpful guidance?"

Doctors' first response was to really appreciate the shift from health plans giving guidance to doctors giving guidance. Over the two years that I've had this job, now I'm looked at as the health plan was. They are no longer very appreciative.

They don't want to lose their autonomy, their ability to make decisions their way. They always know what's in the best interest of their patients. And very often they're quite right about that, so we need to ensure that they have that.

I've always described physicians as similar to trying to herd cats--to get people who are really going in all different directions to read off of the same sheet of music. It's an impossible job.

Is there a way of saying who the problem is?

I wish it was that simple. I wish there was an easy answer to say, "Of our 3,000 physicians, if we just got rid of these three, our problems would all be solved." I think there are two problem areas. One is that in any system, there are some outliers. Those are actually fairly easy to identify, and to develop tools to educate them, or really exclude them from part of the system.

Outliers are abusers of the system, people who really don't provide quality proficient care.

The tougher problem is that, of those 95%-99% of physicians who are really high-quality physicians, there's great variations in how these physicians practice their care. How do you get them to reduce the variability? Say I'm a cardiologist, and I have a patient who had a heart attack three years ago, is fairly perfectly well, and comes to me for a routine office visit.

The question is, what should I do today when that patient is sitting across the desk from me? Is there a standard way? There's a fairly simple problem. It's a person who had an uncomplicated heart attack, is feeling perfectly well. It's a basic example. But you could talk to ten different cardiologists, and they would give ten different ways of caring for that patient.

Why is that bad?

None of those are wrong. But the reality is, if there's a variation, there must be a better way and a worse way. Should we get blood tests on them once a year, once every two years, or never? Should we all give them the same message about quitting smoking? I'm sure we do that. I'm sure we all tell our patients quit smoking.

One doctor will go the next step and say, "Let me set up an appointment for you with somebody to get you to quit smoking." The next doctor will just say, "Quit smoking." The next doctor will give you a pamphlet. I assure you that each of those three different approaches will have a different success rate in getting a patient to quit smoking.

We need to learn from one another, and figure out which is the best way, and then ensure next year that we're all giving it that best way.

Medicine is a science-based field. There's a lot of science that everybody has studied. . . . Why are people just now coming to this discovery that there's a best way, and we ought to be practicing the best way?

Presumably, that's what medical schools have been about for the last 150 years. In the past, medicine has been driven by very structured studies, by taking a very structured environment--given half of the patients one medicine, and half of the patients some other medicine, and seeing the difference--looking at a very small population, analyzing them very, very closely, and making some conclusions.

Because of our access to computers today, practicing medicine is really looking at populations of patients where there are so many variables, because you can't control all of those variables. With computers now, we can have access to a big data set, and then analyze the differences between two populations.

When you've got the CareGroup losing $50 million to $80 million a year, and you've got 3,000 doctors, how do you impose financial discipline to get greater efficiency? What are the steps, and what are the tools?

The first tool we use is education. The second tool is to financially incentivize physicians no differently than hospitals are financially incentivized. . . . For instance, the price for us to hospitalize a patient within our system is less expensive than hospitalizing a patient outside the system for managed care. That directs business to the system.

How is a physician financially incentivized to keep patients inside your system? Are they penalized, or fined, or what?

There is not a one-to-one penalty from a patient's care to a physician, but there are group penalties. The physicians have a pot of money from which they provide care. If they overspend that pot of money, they need to go into their back pockets to fill up that pot. If they don't spend that full pot of money, they get to fill up their back pockets.

How do you impose that kind of mentality that you've got to work together to hold costs down?

In this world of managed care and capitation, we have developed a methodology. You have two physicians who are independent physicians, with their own totally separate practices. But for the managed care business, they become virtual partners. Money is provided to this group of physicians, and they share in the risk of that care.

And so, for example, this group of ten physicians will be given money to provide care for a population of patients, say, for 1,000 patients. They will work together to manage the care of those patients. . . . They have a budget. At the end of the year, they will either end up in surplus or deficit in that budget.

When one tries to set a methodology to incentivize physicians, you have this spectrum of incentivizing the individual physician and individual patient care at one end. At the other end of the spectrum, we incentivize 3,000 physicians for 400,000 patients. First, there are concerns that a physician may make inappropriate decisions when they're really thinking about direct cost accountability for the care of an individual patient.

There's variability in patients. The first patient that comes in to see you is perfectly healthy, and the next patient has some serious illness. You can't ask that physician to bear the risk of that patient population. On the other hand, if you're sharing this between 3,000 physicians and 400,000 lives, there's no local accountability.

So we needed to develop a system which was somewhere in the middle. What we have used here is what we call the pod system, which is getting groups of physicians that work together--typically a size of ten to 20 physicians--who really share in their care of patients. They become a virtual group.

They really work together financially. They are incentivized together. At the end of the year, if they're in surplus, they share in that surplus. If they're in deficit, they share in that deficit.

So the pods are really organized as a system of local financial accountability.

Absolutely correct. That's exactly what it is. There's financial accountability, and there's also quality accountability. We look at the data, and look at it at the pod level.

What does your operation take to the pod that's going to improve their performance?

We take them physicians' report cards, generated semi-annually, which do the following things. We have collected the data from multiple plans and aggregated that. We really have enough patients in the system so that we can really analyze how a physician or a pod of physicians is performing.

The second thing that we have done is to say that we don't expect the same number of tests to be done on a perfectly healthy patient as one would expect for someone who's quite ill. So we've adjusted our expectations, based on the health status of the patients that they are caring for.

If one doctor is a specialist taking care of everybody with heart disease, and another doctor is taking care of young, healthy 20-year-olds, we would not expect their utilization to be the same. We make that adjustment. We then analyze that data on multiple levels--use of hospitals, use of radiology, use of laboratories, use of certain testing, use of primary care physicians and the use of specialists.

What we looked for are those physicians who are either in the top five percent of over-utilizers, but just as importantly, the bottom five percent of under-utilizers. There are certain physicians who would say, "We have kept our costs really down with primary care physicians," and we found that, indeed, they were correct.

They were in this bottom five percent. They were under-utilizers. They saw their patients less often every year or charged lower bills. And, they said, "We're saving the system money." We then looked at their use of the emergency room. And there, the use of the emergency room was in the top five percent.

. . .

These report cards are presented as part of a pod meeting?

Yes.

So it's done semi-publicly?

It is done semi-publicly, because really more often than not, there's a group learning experience as part of this.

How do doctors respond?

More often than not, the doctors are not aware of what they are doing. Every doctor, in their heart, believes that they are providing the best quality care. I'm a firm believer in that. I think they're not aware of the variation in how they deliver care versus the person sitting in the next office to them.

So, actually, what I expected when we came out with these media report cards, was a great push back from physicians saying, "These are the worst things in the world." Think back to when you were in grade school. You never wanted to get a report card. But the reality is that the feedback thus far has been remarkably positive.

Doctors are looking for this. They're looking for help in terms of how they can improve the way they deliver care. Yes, there are some doctors who really object and fight us every step of the way. But the majority of doctors actually appreciate these report cards.

We have some doctors saying that, ten years ago . . . they could send a patient anywhere in the city, to the best specialist, no matter what institution they're in. Today these same doctors are saying to us, "I've got to keep them in our group, and that hurts me. I don't feel as though I'm serving my patient." You hear that.

Absolutely. I do hear that.

What do they tell you, and what do you say back?

There is a heated dialogue regarding trying to keep patients within the system, versus sending patients outside of the system. Two things need to occur. One is education of what services can be provided within the system, to make sure that our doctors are most knowledgeable about the quality of the services we have.

The second is, if we have holes in our system, fill those holes. I don't think that, long-term, doctors will be told, "No, you must keep your patients in this system," if there's a better doctor outside the system. That's not going to work.

But what is the short-run answer today?

The short-run answer today is that those discussions do occur. More often than not, the patients continue to receive their care outside the system if the better care is outside the system.

In your old physician's association at Mt. Auburn, there was recently a move to impose financial penalties on doctors--something like $250 a day--for each "extra day" that patient stayed in the hospital. What was that all about?

We tried to figure out the right way to incentivize the physicians to provide quality care. One of our units looked at a methodology, which was to ensure local accountability. A patient in the hospital needs an operation tomorrow. The doctor has an office full of patients tomorrow. That doctor says, "I'm not going to change my office. That patient should sit in the hospital for an extra day and get the operation done the next day." There's a cost incurred to have that patient sit in the hospital. There are quality issues. What's the risk of that patient getting infections, or illness, or something in the hospital?

This group decided that we needed to have the appropriate accountability, to have that doctor accountable for that day. We are not talking about medically necessary days. We're talking about medically unnecessary days. We were not talking about the gray areas of medicine, which is, "Should Mom go home today, or go home tomorrow?"

These were very, very black-and-white issues, where clearly a service should have been provided on a given day, and for some reason, the doctor elected to not provide that service on that day. This represents far less than one percent of hospital days, and far less than one percent of the physicians that we are talking about.

The problems again become, "Do you have that incentive given right back to the individual physician?" This was a lot of what the concern was about.

What is the most important thing that patients don't understand about the new system? If you could say something to patients, what would you tell them you've got to understand?

One: the way the care was given in the past was really not the best way to deliver care. Two: we need to work together to develop a better way to deliver care. We need to make sure that we are delivering to you what you are looking for from a health care system.

My belief is that, quite often, the individual patient feels they are getting the best care from their individual doctor. What they don't feel is that they're getting the best care from this system. They are far more concerned about this system than they are about the individual doctor. Individual patient satisfaction of their doctor is actually quite high.

It's distrust in the system, but the reality is that the system is really a combination of all of these individual physicians. I think there really is a disconnect--a distrust of this system--when in reality, physicians, and hospitals, have a simple mission, which is to provide the best quality care for patients at a reasonable cost.

It's the cost factor that's got patients worried.

The cost factor absolutely does have patients worried, and this concept of rationalizing health care. We are rationalizing health care. I'm not sure patients want to hear that. I'm not sure doctors want to hear that. But we need to make a decision. We don't have an open checkbook. The government does not have an enormous war chest of money to provide care.

So need to make some decisions. Is it more important for us to care for a young child that has not been vaccinated, or to care for someone for the last five days for the end of their life? . . . Regardless of age, there are certain patients that we know have terminal diseases, yet decisions are made on an irrational basis. What we're asking for is some rationality for that care.

What are the implications for senior citizens in your health-group system in a severely cost-limited or capitated system?

The greatest concern that physicians have today, as risk has been shifted from the health plan's holding the risk for patients to physicians and hospitals holding the risk for patient care is, are they the ones who provide the care? As a patient, it's shifted from one doctor group, which has held the risk, to a different doctor group.

The concern is, where are those patients going to receive that care? If they continue to receive the care by their prior doctor group, their prior hospital, there's no way that this doctor group can manage their care. Medicare is now adjusting the way they reimburse physicians for managed care to a health status methodology.

So, for the sicker patients, the doctors are receiving more money up front than for the healthy patients. Medicare is going through a transition now where, at this point, all patients are allocated the same funds. Therefore, you're far better off on a financial basis caring for healthy patients than sick patients.

You're far better off taking care of healthy patients rather than sick patients.

On a financial basis.

Right. On a financial basis. But under the old system, it was the health plans who had to worry about it. In the new system, it is the doctors who have to worry about it. Is that a good situation to be in? Doctors then want healthy patients, not sick patients.

There has to be that balance. Doctors control 80 percent of health care costs with their pens. Therefore, it is critical that the doctor has some responsibility for that cost incurred. What we've tried to do in our system is to ensure that that individual patient's care is not held at the level of the individual physician.

By really being 400,000 patients, we, in many ways, are like a health plan, where we can share that risk across the whole system. Some of that risk is held at that pod level. Some of that risk is held at a larger group level, which might be associated with only one hospital. Another level of risk is held at the entire system, so that there are levels of accountability in the system.

We talked to a pod last night. A pod's got 500 Secured Horizons patients. Five years ago, ten years ago, patients were younger. They were healthier. Health bills were lower. As you get older, health bills go up. Is there a point at which you get a bad balance, where, financially, you get too many sick patients?

Yes. There is that. What we have done, because the health plans and the government have not yet adjusted reimbursement for the health of the patients, is that we internally take the health care dollars and redistribute them out. Part of the methodology by which we redistribute the money back out to physician groups is based on the health of their patients.

One group would have a sicker patient population, and would receive more funds than a group that has less sick patients. The problem has been that the tools we would use to determine what will be the cost of care for population A versus population B are in their infancy. They are developing.

These doctors in this pod last night were clearly concerned. The phrase wasn't used, but the concern was clearly there, that we may already have too many sick patients in this program and could wind up with even more sick patients. Is that a worry when you're looking at it globally? Do you worry about this?

No. I worry that we will receive patients that we cannot manage ourselves. I still believe that we can go back to health plans, and if we can convince them that we are managing our patients well, and that the dollars we receive are inadequate, they will adjust their premiums accordingly.

What do you say to doctors who say, "This was a bad bargain. We shouldn't have taken on this financial risk. We have too many sick patients."

When doctors make that statement, half the time they are correct--their patients are sicker. Half the time, they actually are not managing their patients as well as some other doctor does. So, one needs to recognize that there are only certain physicians who are best suited to care for patients in the managed care environment that we now live in.

Up to this point, in Boston, every physician has had access to every health plan. . . . What's happening now in Boston is that there is a consolidation, where every health plan will not be offered to every physician. So some physicians are not best suited, and there are some physicians who will, indeed, drop out of a certain health plan.

In today's environment, there are physicians who have not been able to manage their patient population. Given the financial constraints, they have elected to no longer offer that health plan. A physician previously would have Secure Horizons patients that they were caring for, for example, and they would elect at this point to no longer offer that the following year.

Then the patients would have to either stay with their physician and leave Secure Horizons, for example, or shift to some other health plan.

Why are doctors doing that?

At this point, doctors are concerned about taking the risk and financial concerns.

It's costing them too much. They can't manage it?

Yes. If you look in general, and not in specifics for any one doctor, the reality is that today the reimbursement that the physicians are receiving from managed care plans in eastern Massachusetts has been more than adequate to cover their costs for the delivery of care.

Maybe they're right. You really do have the killer job.