An Interview with Alan Hillman, M.D., M.B.A.

You might not expect to hear the term “smoosh” from the lips of one of health care’s most erudite observers. But he would argue that the rest of the industry’s vocabulary should be as clear.


Managed Care didn’t reach far afield for this month’s interview, literally or figuratively. Alan Hillman, M.D., M.B.A., is director of the Center for Health Policy at the Leonard Davis Institute of Health Economics at the University of Pennsylvania, and an associate professor of medicine and health care management in the university’s School of Medicine and the Wharton School of Economics. His office, in the institute’s quarters on picturesque Locust Walk on Penn’s Philadelphia campus, is less than an hour’s drive from the magazine’s offices in Yardley, Pa. More to the point, Hillman is one of three members of Managed Care’s 10-person editorial advisory board whose service dates back to the publication’s founding in 1992.

The editors know Hillman as an adviser whose opinions are frank, articulate and informed. Readers, too, have heard before from Hillman. In a June 1995 article titled “It’s Time To Tell Patients the Facts About Managed Care,” he argued that physicians should be forthright with patients about the way they’re paid, and decried HMO “gag” rules limiting what doctors could tell patients about treatment options — well before such rules became a bête noire for the mass media.

We thought it was time to check in with Alan Hillman once again for his varied observations on the state of the health care world.

MANAGED CARE: Do you think the “gag rule” problem you called attention to in 1995 has been pretty much corrected?

HILLMAN: Definitely. Sometimes ignorance of a problem is the reason a problem exists. Once the ignorance is wiped away by education and by people finally reading their contracts — and that’s not just patients, but large employers and whoever is buying health care — the situation improves. The gag rule issue is widely recognized, and I think has been pretty much rectified. There are exceptions, but I see a big improvement.

MC: In the wake of the resignation of the CEO of the Columbia/HCA hospital chain, which had become controversial because of its aggressive ways, there has been speculation about a possible change in the way for-profit hospital companies are run. Do you see that development affecting the for-profit HMO community in the coming years?

HILLMAN: For-profit enterprises in health care are always going to be a concern to observers, because health care is not a commodity in pure economic terms. People are always going to be wondering about the line between treating health care as a commodity and treating it as something personal. But I think Richard Scott’s termination will be a little blip on the screen. I certainly don’t see that one event undermining for-profit health care. I think for-profit medicine will be looked upon more or less favorably in tandem with the way the political pendulum swings. When the country goes conservative and believes that business cures all, then for-profits will be more “in,” and when the pendulum swings the other way, they’ll be more under the microscope.

MC: Speaking of the profit/not-for-profit distinction, Malik Hasan, M.D., chairman and CEO of Foundation Health Systems, has called for doing away with the not-for-profits. Is there any likelihood of that?

HILLMAN: No, because academic medicine will always tend to side with the not-for-profits. This whole distinction between for-profit and not-for-profit is way overblown, in my opinion. Just as the distinction between the Soviet Union and the United States was really one of an economic mechanism, a distinction between for-profits and not-for-profits is a distinction in a couple of financial and economic rules. Both systems need to make a profit or retained earnings — what you call it depends on where you are. In for-profits that profit goes into the infrastructure and to shareholders. In not-for-profits it all goes into the infrastructure, but the not-for-profits can’t go to Wall Street for capital. So which mechanism you choose to keep your bottom line sound is a matter of geography and history and preference. There are more similarities between the for-profit enterprise and the not-for-profit enterprise, especially nowadays, than there are differences.

MC: Let’s put the issue more concretely. Do you see the big, well-known not-for-profits that we know today — Harvard Pilgrim and Kaiser, for example — still being independent and not-for-profit ten years from now?

HILLMAN: Most of them. I would bet you a cigar on that.

MC: What do you think of the critique of people like Steffie Woolhandler, M.D., who talk of profits “taken away” from health care and given to shareholders? [For a related news item, see p. 9.]

HILLMAN: What she doesn’t understand is that those monies originally came from shareholders. The shareholders gave money to the for-profit institution as a loan in order to get it to grow and to deliver high-quality, efficient health care. If those organizations fail in doing that, then those shareholders don’t get their money back. So you can kind of look at the shareholders as taking a risk in trying to help the health care industry.

MC: And yet in practical terms, even some private practice physicians who’ve probably never voted for a Democrat will say, “I don’t want Wall Street medicine running my practice. What can we do about this outrage?”

HILLMAN: Yes, but there are a lot of reasons for that battle cry. The life style they want to protect and defend at all costs is a for-profit life style. They’re each running a small business that makes a profit. What we don’t want is micromanagement by overwrought managers who are far removed from the practice of medicine both philosophically and geographically. We don’t want these people, without any intimate knowledge of a patient’s problem or a doctor’s way of dealing with it, making pronouncements and micromanaging. That can happen in both for-profit and not-for-profit plans.

MC: Do you see any examples of such micromanagement happening today?

HILLMAN: Yes. The 1-800 number is a perfect example. You know, they call the 1-800 number for approval for a simple test, such as a CT scan, and at the end of the line is a poorly prepared LPN who knows nothing of the patient. That LPN is asked to follow a computer printout and make judgments about whether the patient adheres to the protocol, or whether an exception should be made — very unfairly, because the LPN has no knowledge of the doctor, the doctor’s style, or the patient. And each patient is different. I think institutions today are moving away from that 1-800 approach.

MC: Toward trusting the doctor more?

HILLMAN: Yes. Not giving them back the complete autonomy they had a generation ago, but using different techniques, such as peer review and having a medical director on site who’s responsible for quality assurance. Having a quality assurance effort that goes through the entire organization looking for ways to improve the delivery of health care, and implementing them at the level of the doctor-patient relationship rather than miles or thousands of miles away.

MC: You didn’t mention practice guidelines. What role do you see for them in assuring quality or standardization in health care?

HILLMAN: I didn’t mention them because practice guidelines have a number of different definitions. Some people call what the nurse does at the end of the 1-800 number the use of practice guidelines.

MC: Backwards?

HILLMAN: Well, maybe she’s following a critical pathway or a protocol. You know, all these words kind of really mean the same thing at that level. Or practice guidelines, obviously, can mean a protocol or a critical pathway that is applied by the doctor or the peer group of doctors or the group practice at the local level. I think that’s where we should put our money. And our effort is to make sure that the guidelines, protocols, critical pathways, algorithms, whatever you want to call them, are developed locally, applied locally and revised and evaluated locally, and that the people who are primary in doing all this are the doctors themselves. That’s obviously totally opposite to when you have an agent in a corporate culture miles away writing guidelines that have nothing to do with the way the medicine is practiced in Oshkosh or the Aleutian Islands.

MC: You’ve given me an opening to ask a question that has become a standard refrain in these interviews. If medical conditions don’t differ very much from place to place, why is locality is so critically important? Is it a real distinction, or is it partly to get the doctors on board?

HILLMAN: I think your question, with all due respect, reflects the reason why doctors are needed in the guideline process. Because small things like the waiting time for a CT scan vs. an MRI, how fast stat bloods are returned, how reliable a Pap smear read is or whether it is to be repeated, how good a particular surgeon’s technique is — all of these things make a difference in whether you go to A first in your protocol or you go to B first and then A. And that’s why the protocols have to be local. You want to standardize things to the extent that you remove unnecessary steps, but the order of those steps and the number of times things are repeated are dependent on what happens at the local level for some of the reasons I mentioned — and many others I didn’t.

MC: Having to do with the strength and capabilities of the facilities that are available.

HILLMAN: That’s one.

MC: Even though the disease isn’t that different, the resources are very different.

HILLMAN: One, the resources are different. Two, the way the environment — the weather, for example, or the barometric pressure — interacts with those diseases produces different ways a patient is presenting. The ways that doctors were trained and the expectations of the culture in a particular area, and how fast or how slowly things move, are all things that affect the way that a guideline is applied, and the reason why a guideline written for a specific disease in Florida may be totally different than it is in Montana. In Montana there are places where patients have to drive 100 miles for an MRI, while in some parts of Florida there’s one 10 steps away. How could you possibly write the same algorithm for both places?

MC: Our cover story this month asks to what extent managed care plans are rewarding the so-called “best” doctors. Roughly half of all HMOs report using some kind of quality incentive in paying physicians, but when you really look at most of these programs, they boil down to two things: patient satisfaction surveys and fairly easy-to-track preventive measures such as immunizations and mammograms. It’s rare for such incentive programs to reflect actual outcomes. What do you see happening in that area in the coming years?

HILLMAN: We’ll move more toward the use of outcomes to reward quality. But the reason it is so hard to do is that statistically what one doctor does can’t really be connected to the outcomes of care. For example, a surgeon can do all the proper things in a coronary artery bypass — a technically perfect surgery — but the patient dies on the table. Or you can have a horrible doctor who does terrible surgery and some of his outcomes are going to be good. So statistically you need to have doctors at the extremes in order to say they are “bad” or “good.” But the clearly bad ones, one hopes, have been kicked out by their peers. Most doctors are in the middle; they have some good outcomes and some bad outcomes. Should one doctor be rewarded for having three more good outcomes than another, when in fact it may just be the result of statistical scatter?

MC: They say they try to adjust for severity.

HILLMAN: Well, severity adjusters will account for only 30 percent at best of the variance in outcomes. If we had a perfect adjuster, then we could more easily relate outcomes to process. But we don’t have a very good method of doing that. If an outcome is good, can you say reliably that that is due to the intervention of the doctor? You can’t. But what you can do is see if the doctor’s processes — the actual techniques that a surgeon uses or the cognitive intervention that a non-surgeon uses — are technically correct in terms of an algorithm or protocol or guideline. It’s very easy to throw around the term “outcomes” and say you’re going to relate outcomes to payment. But that can only be done with validity on large group practices, not on single doctors. And even then you’re going to be rewarding some doctors who are not good because their group on the whole is better than another group.

MC: Do you think that is an affliction that the vocabulary of health care is suffering from? Terms that are easy to bandy about, but that don’t necessarily reflect how much is actually there?

HILLMAN: Absolutely.

MC: Is disease management one of those terms?

HILLMAN: Absolutely. As you were saying that, it was crossing my mind. Disease management, case management, protocols, algorithms, practice guidelines, critical pathways. We give names very easily to things, but we don’t really define them. I challenge you to go out to ten different “experts” and ask them for a definition of managed care, or of quality, or of disease management. You’ll get something totally different from each person.

MC: How can we rectify that problem? Standardizing the vocabulary while we’re standardizing health care?

HILLMAN: Yes. To the extent that it’s a major problem, that should be done along with the refinement of health care. In other words, as we find out what’s right for certain kinds of patients, we should apply the vocabulary and the appellations that are used in medicine. On the other hand, I’m not sure how much real impact on the quality of health care this naming problem has.

MC: Let’s focus on disease management. There is an awful lot of talk about it, and in some cases it’s terminology used in marketing initiatives by pharmaceutical companies.

HILLMAN: Absolutely.

MC: In some cases it’s being embraced by managed care organizations.


MC: How do you define disease management?

HILLMAN: My definition is when a team approach is used to take care of a patient with a disease so that we don’t just look at the patient-doctor interaction when the patient visits the doctor, but we follow the patient outside the doctor’s office. We know how he lives, and we have home health care when he’s at home and follow him through the course of his disease. And we institute preventive steps. For example, in asthma, steps A, B and C are preventive, because you don’t want to wind up with an asthmatic kid in an ER. Disease management would assign a case manager to make sure the patient is taking the appropriate medicines and is avoiding things that set off the asthma. So that may mean visits with the family once a month, or once a week, or whatever. Then, when the patient visits the doctor, there is communication between the nurse and the doctor, so it’s a team approach. And you could easily have called disease management “team management” or “team medicine.” Other people will tell you disease management is the application of protocols to diseases in hospitals, and it’s the protocol that’s the central theme of disease management. Still others will say it is the value-added services given to managed care companies by pharmaceutical companies for the use of their drugs — not necessarily involving a comprehensive look at the patient.

MC: Is there any danger from this diffusion of different interpretations of disease management?

HILLMAN: The danger is if you are a buyer of health care, whether an employer or a patient, and you’re told in your booklet that a certain organization has a focus on disease management, and you have one idea and they have another idea about what disease management is, then you don’t know what you’re buying. You can be duped or unintentionally misled.

MC: In the initial visions of managed competition, it was thought that there would develop exclusive teams of HMOs and their dedicated hospitals, primary care doctors and affiliated specialists. Instead, the prevailing reality now is all these physicians out there contracting with all these HMOs at the same time. That raises questions about how to standardize health care to improve quality. How does an HMO influence the behavior of physicians who are also affiliated with all of its competitors?

HILLMAN: Fortunately, physicians don’t distinguish between and among patients. What’s been shown in good scientific literature is that physicians will take an amalgam of all their contractual obligations and kind of smoosh them together to develop a method of practicing that reflects a composite of all their contracts. So they deal with each patient the same, and don’t look at the chart and say, “Oh, Blue Cross. Then I’ll get a chest X-ray before I get an MRI instead of getting an MRI before the chest X-ray.” That means the managed care companies in an area really would be better off getting together with the physicians and writing guidelines as a group, rather than each plan’s doing it on its own and thinking that the doctors are following its guidelines. Because they’re not.

MC: From some points of view it would be alarming if they did, anyway.

HILLMAN: Absolutely. So if they got together, it would probably save money and result in better pathways in the long term.

MC: In other words, the imposition of certain quality guidelines should no longer be a competitive feature among HMOs. They should get together and say, “We’re going to change doctors’ behavior, but we’re all going to do it together, because all the doctors are with all of us.” Is that not being done now because HMOs are shortsighted?

HILLMAN: I don’t know if it’s shortsighted, but they’re just competitive. If they spend so much money on the development of protocols with physician experts, they don’t want other managed care organizations to get the benefit of that. Really, there are no examples of what I’m recommending that I know of. Instead, it’s each managed care organization for itself.

MC: Well, here we sit at the University of Pennsylvania, and you’re an authority on this stuff. Let’s tell the nation how we’re going to begin this process of HMOs getting together to work in a collaborative way. Another thing, of course, is turnover: Your patient today may not be your patient 10 years or 20 years from now when he has his heart attack. So that limits what is cost-effective for you to do, and if you do things beyond what is cost-effective, you’re not doing your job well according to the bottom line.


MC: So how do we get around or above that?

HILLMAN: Well, I think some entity needs to rear its ugly head and take leadership of this effort. And if it’s not the National Committee for Quality Assurance or the Joint Commission on Accreditation of Healthcare Organizations or Managed Care magazine, it will will have to be the government. First, we have to deal with the antitrust implications. Second, we have to get managed care to agree to abide by the pathways that are put together by managed care groups working together, and not to keep proprietary secrets about these things. I hope the leadership will come from the private sector, but if that fails, then it will have to be the government. Or we’ll just continue to have 10 different guidelines sent to doctors that are — well, not totally ignored, but smooshed together to form a method of practice.

MC: I think that concept of “smooshing” is a story for us. So do we call on someone here today to start the process of HMOs getting together?

HILLMAN: Yes, let’s call on the NCQA to do that. But they don’t have an unlimited budget. And it’s easy for me to call on them to do it, but that’s not why they were put together or who they are. Also, some people will say, “Don’t we have more pressing problems?” I would say everything we’ve talked about is very important, but we’ve still got 40 million Americans with inadequate health care. That will hopefully be reduced with the new budget deal, but I would venture to say that the kind of health care that is going to be provided for these kids is not the kind of health care you would want for your own kids. If I were the health care czar in this country and was told to prioritize the problems and put some people on each one, I would say the primary problem is access. Costs have more or less come under control — although I think they’re about to explode again with the new genetic techniques and the new technology that’s on the horizon. But you can put a band-aid or a short-term solution on costs. Quality, for people who can access the system, is pretty damn good. It is in this country that I would want to have my leg set, or my appendix taken out, or my brain operated on. Problem number one is access.

MC: When we interviewed Michael E. Herbert, vice chairman and co-CEO of Physicians Health Services Inc. in Connecticut recently, he said some health plan executives are beginning to think in terms of turning the accreditation function over to a government agency like the Securities and Exchange Commission. Might that be the entity to standardize guidelines?

HILLMAN: I am a proponent of that. That’s a whole other discussion, and I haven’t brought it up, because although the Securities and Exchange Commission is extremely successful today, the way it got successful is a very bawdy tale. If we just turn authority for health care over to an SEC-like commission, will it have to go through the same process the SEC went through 50 years ago? But I do believe that’s the correct way to go — a public-private consortium of scholars, providers and patients that is initiated by the government and given some regulatory authority but not complete regulatory authority. We want it to be able to regulate some things, but not to be able to tell the doctor in Tallahassee exactly how to practice.

MC: Thank you, Dr. Hillman.