P R O C E E D I N G S
PAUL GINSBURG: I’d like to begin this next session by introducing our first speaker, Dr. Lawrence Casalino, who recently became an assistant professor at the Department of Health Studies of the University of Chicago, after 20 years as a family physician in Half Moon Bay, California, and some significant research accomplished in his spare time, while he was a practicing physician.
Casalino is also co-investigator in the first national survey of physician organizations and the management of chronic diseases, which is linked to the Community Tracking Study.
Larry received his medical degree at the University of California, San Francisco, and earned his doctorate in health services research and organizational sociology at the University of California, Berkeley.
LAWRENCE CASALINO: Thanks, Paul.
I’m going to speak about "Risk Contracting Reconsidered." You can see my collaborators on this paper. Bob Hurley was actually the lead author. I’ll speak a little bit about why risk contracting was thought, by some people at least, to be a good thing, and then what we have found happening with risk contracting, and the problems with both theoretical and as they’ve actually worked out with risk contracting, and then very briefly about policy implications.
To summarize what we found, you’ve already received this impression from what the speakers said earlier this morning, risk contracting is declining in most, I would say by now even, in all markets that we visited. There are fewer plans using risk contracting, with fewer enrollees in risk contracts, and risk contracting is also being modified in ways I’ll speak about briefly.
I just want to make sure that we’re all on the same page with risk contracting. I think this is a fairly sophisticated audience. How many people here do not need me to explain what shared risk contracting is? How many people here understand what shared risk contracting is? Let’s have a show of hands.
And global risk contracting, how many people understand what that is?
Okay. Well, extremely briefly, there’s really been two models of HMO contracting in the states; contracting with individual physicians or contracting with medical groups or IPAs or physician hospital organizations.
In the individual contracting model, which is quite common still, it basically involves capitating primary care physicians only for their services and not really that much else, in terms of risk.
In the group contracting model, which is really what I’ll be talking about, and it’s really what all of the sessions talk about today when people are talking about risk, this will be done either shared or global/full-risk contracting. The health plan gives money to the physician group, always to cover professional services or physician services. And then for hospital services, pharmaceutical services, X-ray, lab, there may be a pool from which those services are paid for. If there’s money left over in the pool at the end of the year, the provider organization and the health plan will share in the savings. If there’s losses, they may share in the losses.
Full-risk contracting, global capitation, all of the money intended to take care of patients, more or less, goes to the provider organization, and the provider organization just spends it to take care of those patients.
So, again, three types of risk contracting—professional risk only, shared risk, and the full risk. I’ll be referring from really to the latter two.
Now the people who thought risk contracting was or is a good thing had several reasons for it. The most commonly cited and probably the most commonly understood is that it involves providers in trying to control costs. So the phrase was always this aligns incentives between physicians, hospitals, health plans and employers, all with an interest in containing cost. It doesn’t necessarily, unfortunately, align interests of providers with patients who, by and large, haven’t had a great deal of interest in containing costs, and that’s been one of the problems.
Risk contracting can also encourage the creation of organized processes to managed care for populations of patients prospectively. I really want to emphasize this. I think that the early savings in managed care, and I think there were early savings, were really the easy pickings, keeping patients out of hospitals who really didn’t need to be there, not hospitalizing patients with low-back pain for traction, for example, as we used to do when I was first in practice. I mean, that was just routine. Oh, your back hurts, we’ll put you in the hospital for traction.
So those kinds of things, eliminating those did generate one-time savings. But to create processes that can really improve the care of patients with congestive heart failure or diabetes or asthma or other chronic diseases, that’s hard, and that’s just barely begun to be done.
Now a capitated medical group, in theory, should have incentives to develop such processes, and in fact they have, and did, start to try to develop some processes. One regret of some of the groups is that, as risk contracting goes away, they wonder if they still will be able to do that.
Finally, and this is a related point, risk contracting can encourage the creation of organization among providers to do these organized processes. So encourage physicians to come together in medical groups or IPAs, encourage the formation of physician hospital organizations, organizations that have the capability both to take risk contracts on because a small group or a solo physician, obviously, can’t do full-risk contracting, and the organizational capability to create these processes, call them disease management processes or whatever you want, to both contain costs and improve quality.
Again, a solo practice, a small group practice, or a hospital alone, for that matter, can’t go that far that way. I would say neither can a health plan go that far in creating these processes by itself, although there’s a role for health plans, I think, because health plans don’t actually care for patients.
Now, potentially, there’s some risks to full-risk contracting, and one is, and this is the best known publicly, the incentives that it gives the provider organizations to not provide appropriate care and also to avoid sick patients. Because payments are not very well risk adjusted to provider organizations, if you draw a sicker pool of patients than average, it doesn’t matter how well you manage care, you’re going to lose a lot of money.
This is a more subtle point, but a very important one, I think. It was thought that risk contracting pretty much by itself gave provider organizations an incentive to improve quality because quality is cheaper. But it turns out that that really isn’t true. Some quality really probably is never cheaper. Other things that improve quality may lower costs in the long run, but the long run doesn’t do any good to a medical group that is investing in improving quality this year.
So, for instance, if a group invests a lot of money in lowering patient’s cholesterol or in getting screening exams for diabetics for retinopathy, that costs the group money this year, but it isn’t going to save any money this year. It’s going to save money years down the line. So this is an example, and there are hundreds, of how global capitation or shared risk does not provide incentives necessarily, in itself, to increase quality for provider organizations.
A very important point, there actually was, and is, a real lack of provider organizations that actually can manage care under risk contracting or any other way. There aren’t very many capable medical groups in the country. There aren’t very many, there may be even fewer competent physician hospital organizations that really have the ability to manage care.
And, finally, I think that full-risk contracting, of which I was, years ago, a very large proponent, it really does go contrary to very fundamental insurance principles. I mean, what is insurance? It’s a big organization, pools risks from lots of individuals or smaller organizations. So that we all are responsible for all of the costs of our own auto accidents, say. We pay premiums to an auto insurance company, and they take care of it, more or less, if something happens.
But in risk contracting, we have the larger, better-capitalized organization, the health plan, passing financial risk, lots of financial risk if it’s global risk contracting, to the smaller, less-well organized, less-well capitalized provider organization.
Very briefly on what we found. In ’96, well, based on the ’96 survey by the Community Tracking Study, we classified the 12 markets into three categories based on the amount of risk contracting that was being done, and you can see there was really only one, and that was Orange County, in which there was quite a bit being done and very little actually being done in the others. I mean, in the moderate markets, as we call them, only between 13 and 26 percent of patients were in either shared or full-risk contracts.
Now, in ’98, a couple of years later, again, of 41 health plans in the 12 markets that we looked at, we could see that all of them in the one extensive risk market were using risk contracting, and we can also see that, despite the talk about the move away from risk contracting, at least at the time we did our site visits, 2000-2001, only a few plans had actually completely dropped risk contracting.
However, if you look at the moderate and limited markets, really a lot of plans were cutting the number of the patients that they had in risk contracts. In Orange County, especially, there was a great reduction in the scope of risk that was passed in the contracts.
Now I would say this slide is already a bit dated because even in California now there is some move away from risk contracting. I’ve also said that it is the health plans that are moving patients out of risk contracting, and that may be somewhat misleading. It’s probably, more often than not, it’s the provider organizations that are saying, we won’t do it any more, and that’s the reason that there’s a move away, although the plans, also, do not want risk contracts to blow up. They don’t want provider groups going bankrupt. So, in many cases, they’re quite happy to move away from it as well.
Now why the decline in risk contracting? Well, I’ve mentioned some of the theoretical reasons, all of which I think have proved, in practice, to be true why one could, in retrospect, have predicted problems with risk contracting. There are a few other things as well—one I would call "The Empire Strikes Back."
Most hospitals and specialists never really cared much for risk contracting. They wanted to be able to provide pretty much whatever services they wanted to and charge high prices for them and not have anybody bother them, basically.
So, as hospitals have consolidated, and as John mentioned earlier, as single-specialty physicians have consolidated, they have achieved enough negotiating power in many markets to tell health plans, "Why should we put our money at risk? Just pay us." And that’s a big reason for the decline in risk contracting.
There is, also, as is well known, the consumer backlash against managed care. Consumers don’t necessarily know that they are angry at risk contracting per se, but they are suspicious about anything they hear about incentives to withhold care, which risk contracting gives providers, and they don’t like the tight provider networkers and the tight utilization management that were features of risk contracting and which it’s been at least thought that risk contracting couldn’t really be done without them.
There is, also, as I mentioned, still a lack of competent provider organizations. There has been some well-publicized failures, and perhaps even an exaggerated idea of the number of failures of capitated groups. The California Medical Association was particularly successful in convincing the whole country that half the medical groups in California were going bankrupt. This is, actually, it turns out not to be true, although they are having financial problems.
There has been a decline, as you all know, in the number of patients in Medicare HMOs. Medicare HMOs were a big part of the attraction to risk contracting, for reasons I won’t go into now, but that has been a reason that I think we’ve seen provider groups, especially provider groups, less interested in risk contracting, and, finally, again, the probably flawed model of a larger, better-capitalized organization passing risk to a smaller provider group.
Another way to understand this is to talk about the difference between actuarial and care management risk. In other words, if a physician group loses money on a risk contract because they have managed their congestive heart-failure patients very poorly, say, and they’re being readmitted to the hospital again and again, that would be poor care management, and a poor care management risk, and you could make an argument that there should be some financial penalty for the physician group.
But if the physician group loses money because it happens to get a sicker-than-average pool of congestive heart-failure patients, either by chance or because it’s considered to be a high-quality group, and so a lot of very sick patients enroll with it, that would be actuarial risk. It’s not so clear that it is a good idea for provider organizations to be taking actuarial risk. So good models would probably put some degree of care management risk on physicians and hospitals and as little as possible of actuarial risk.
Now, implicitly, there was a fair amount of talk earlier today about what people are seeking as alternatives to risk contracting. I’m not going to go into any detail on this, but one thing is to replace provider risk with patient risk and to increase patient financial incentives which, in itself, is probably a good idea, although there may be limits. There’s talk about an empowered consumer is one with a $2,000 deductible.
[Laughter.]
LAWRENCE CASALINO: Well, the consumer may not think that that’s empowering. At first they may, but when they see what it actually means in practice, they may not feel as empowered. But these are the kind of things, and they were talked about earlier, I won’t go over them, that are being looked at.
Then what wasn’t mentioned as much, and it’s just beginning to happen in some areas and talked about in others—I think Ellen Zane, from Partners in Boston, may talk a little bit about this—plans and employers are trying to find ways to pay providers for performance, and I call that modified risk contracting because, although the providers may be paid fee-for-service, some money is, in fact, at risk for their performance on meeting certain targets. In other words, they don’t have any downside risk, but they’ll get bonuses if they meet certain cost and quality targets, and this does reduce actuarial risk.
And, also, because you don’t have a provider organization holding a pot of money and paying claims, as you sometimes do, especially in California in full-risk capitation, you don’t have to worry that the capitated organization is going to go out of business, leaving lots of claims unpaid, which is a big problem.
Now a little bit in terms of policy implications. One problem with increased consumer financial incentives, and this was mentioned earlier, is that they do penalize people who are poor and ill—poor and/or ill. If you’re both less wealthy, and have some chronic illnesses, and there are increased consumer "financial incentives," you are very likely, and there are more and more studies coming out to show this, to forego care that you really should have. So that’s a problem.
The other problem is that Medicare and Medicaid are very big programs in this country, and because of the way they’re structured by the legislation that authorizes them, it’s almost impossible, if not impossible, as I understand it, for traditional Medicare, non-HMO Medicare, and Medicaid to use this method for channeling patients.
Now, just to conclude, is risk contracting going away? It does appear, even since we finished our site visits about 6 months ago, there’s been even further move away from risk contracting. We heard from Brian Ancell that in Seattle there is now virtually none. There is a move away, I think we have learned that it’s foolish to make predictions in health care, but I doubt that the kind of full-risk contracting or even major shared risk contracting, will come back in the form it was. We probably will move, I hope, to more pay-for- performance kind of initiatives, with some degree of consumer cost sharing.
Basically, something needs to be found to replace what risk contracting was supposed to do. Contain costs, improve quality, those are fairly obvious, but I would emphasize that the key thing that isn’t talked about much is the need to have organized providers. I don’t mean necessarily huge, consolidated, horizontally integrated hospital systems that are formed primarily for the purposes of negotiating leverage, I mean organizations, medical groups or physician hospital organizations, that actually have the ability to manage care.
Risk did generate some movement in that direction. Pay-for-performance could, but that’s a big if, and only strong pay-for-performance initiatives would encourage that. Otherwise we’re just going to go back to the kind of fragmented system that we still have and had the in the past.
I think that policies that strengthen Medicare+Choice will tend to make it more possible to have modified risk contracting—it’s easier to say. I think that things that weaken medical group infrastructure, a collective bargaining bill for physicians, for example, would tend to have physicians not organized into medical groups. I think that that would not be desirable.
And, finally, really, the probably most fundamental lack of physician groups, and physician hospital organizations, and hospitals, for that matter, is the kind of information that would be necessary to really manage care, to improve quality and contain costs. I think that, at current rates of change, it will be a long time before hospitals have adequate systems and a very long time before medical groups have adequate systems.
Crazy as the thought may seem in the current political environment, perhaps there does need to be some thought to the federal government becoming involved. I called it a Hill-Burton bill here. It may not be the best thing to call it, but some push that would give some financial incentives and maybe assistance to create information systems.
I would just conclude by saying that there’s a lot of reference to managed care and managed care not working. I think it’s a little bit of a mistake to say that. Gandhi was once asked by a Western reporter, "Mr. Gandhi, what do you think about Western civilization?"
And Gandhi said, "Well, I think it would be a very good idea."
[Laughter.]
LAWRENCE CASALINO: I really think that the same thing would be true about managed care. It would be a good idea, and we haven’t seen it yet really. What we’ve seen—well, we know what we’ve seen.
Thanks.
[Applause.]
PAUL GINSBURG: Next, I would like to introduce Kelly Devers, who is a health researcher at HSC, and she helps oversee the design and implementation of the Community Tracking Study site visits and is the provider team leader.
Kelly received her doctorate in sociology from Northwestern University and was a Robert Wood Johnson scholar in health policy research at the University of California, Berkeley and San Francisco.
KELLY DEVERS: Good morning. It’s really a pleasure to be here, and I’d like to acknowledge my co-authors, Linda Brewster, who, unfortunately, was not able to be here, and Larry Casalino.
Hospitals’ competitive strategies have a significant impact on the health care delivery system and on key outcomes of policy interest, such as cost and quality. What are the primary strategies that hospitals have been using to compete over the past 5 years and how have they changed, if at all?
Today, I briefly describe findings from the CTS site visits about how hospitals’ competitive strategies have been changing and the policy implications of these strategic shifts.
What we found is that hospitals are reviving and reemphasizing retail strategies. By that we mean strategies designed to effectively compete for individual physicians and their patients, the ultimate consumers of health care. While hospitals must still compete for managed care contracts or wholesale business, strategies for success in this arena are currently be deemphasized.
In order to attract and retain physicians and patients, hospitals are investing in a wide array of services and their associated facilities and amenities. Oftentimes, these services are added or expanded to match or improve upon those offered by competitors. They also are designed to make it more convenient for patients to access the services.
Finally, hospitals are aggressively marketing to consumers. Hospitals have always marketed to consumers to create brand-name image and build patient loyalty. However, hospitals are increasingly marketing specific service lines to the most desirable segments of the patient population. This current strategic emphasis is somewhat surprising for two reasons:
First, hospital consolidation was supposed to minimize the need for retail strategies, particularly service duplication. If there were fewer hospital competitors, hospitals would not have to invest as heavily in similar services and advertising. In essence, there would be fewer Joneses to keep up with.
In addition, consolidation provided hospitals with an opportunity to capitalize on economies of scale, while still keeping patients in their system. Service duplication among system-affiliated hospitals could be minimized, if not eliminated.
Second, managed care theoretically made hospitals more sensitive to the costs and the impact of new services on clinical quality of care. Through selective contracting, managed care plans would essentially purchase hospital services wholesale for their enrolled population, giving them both greater incentive and leverage to negotiate better prices and to ensure that services had a measurable impact on the clinical quality of care in addition to patient satisfaction.
In 1996 and ’97, there was evidence that managed care plans, as wholesale purchasers of hospital services, were reshaping hospital strategy. In particular, our respondents reported that hospitals were focusing on strategies for success in a selective contracting, full-risk environment. While this type of managed care was not prevalent in all of our markets at the time, it was expected to dominate the landscape in the near future.
For most hospitals, the key strategic response was to develop an integrated delivery system attractive to managed care plans and capable of managing care and risk for a portion of the plan’s enrollees.
Given this overarching strategic vision, hospitals continued to assemble the pieces of integrated delivery systems. Specifically, hospitals devoted significant resources to mergers and acquisitions of hospitals, other health care organizations and physician practices. They also invested resources to develop a risk-contracting infrastructure.
In this environment, hospitals selectively consolidated and added services. Consolidation helped reduce cost. However, services were selectively added in an effort to attain must-have status in health plan networks. For example, hospitals added obstetric and neonatal intensive care units attractive to young managed care members or other services that enabled them to provide one-stop shopping to managed care plans.
By 2000-2001, the emphasis on strategies for success in a wholesale managed care environment had waned, and hospitals were back to traditional retail strategies. Our respondents reported that hospitals were focusing on strategies for success in broad provider network moderate-risk environments. In many markets, selective contracting and full risk never developed to the extent anticipated or, as we have heard, was undergoing significant change due to the managed care backlash, provider failures, and pushback.
Although remnants of the integrated delivery system model remain, hospitals were reemphasizing traditional retail strategies. Resources that previously may have been directed to system development are now being freed up. Hospitals have deemphasized horizontal and vertical integration, particularly mergers and physician practice acquisition, and are shedding unprofitable lines of business.
Hospitals are investing much of these resources in a wide variety of inpatient and outpatient services attractive to physicians, in particular specialists and their patients. Broad provider networks increase patient and provider choice and more moderate financial risk decreases sensitivity to the associated cost increases. While hospitals remain under financial pressure, they view investment in these services as the key to growth, market share and profitability.
As this table indicates, hospitals are adding or expanding a variety of highly specialized services. Inpatient and outpatient specialty care centers were most frequently noted by our respondents serving as a mechanism to increase specialty service volume.
Niche specialty services were the secondly most frequently noted expansion. Often these services were useful for attracting star physicians and could be directly marketed to consumers. This strategy was utilized by academic medical centers, as well as community hospitals.
Finally, hospitals reported adding cardiac surgery programs. These programs were added by hospitals that previously never offered such a service, as well as systems that already offered it at a system-affiliated hospital.
Hospitals also reported a variety of facility expansions. Most prevalent were outpatient facilities that were primarily joint ventures with specialists. Seventeen new hospital physician joint ventures were reported in our communities in the recent round of site visits. Emergency and operating room capacity, including intensive care units and other inpatient capacity expansion, was also noted. In some cases, hospitals reported previously misestimating need in these areas, but in other cases, these additions were primarily designed to match and improve upon what the competitor was offering.
Finally, some new hospitals were being built. This included general acute care and specialty facilities, such as heart hospitals. Again, in some cases, these new hospitals were designed to meet demand in rapidly growing communities, but there were also instances of new hospitals being built to please specialists who had offers from for-profit specialty firms.
Does this hospital competitive behavior signal a new arms race, similar to what was observed during the fee-for-service era? The term "medical arms race" refers to service mimicking and one-upmanship. Hospitals added a service because a competitor already offered the service or is expected to do so in the near future.
The medical arms race appears to have been rekindled. Mimicking and one-upmanship is returning as hospitals compete more intensely for physicians and their patients, and rapid technological change continues to present opportunities for such competition as in the past.
However, the medical arms race of today is new in the sense that the players and dynamics are different from the fee-for-service era. First, there are fewer larger hospital competitors because of consolidation. Only 2 of the 12 CTS markets remain unconcentrated using the U.S. Department of Justice and FTC guidelines. As noted, consolidation was supposed to slow service duplication, but it appears that fewer larger systems may intensely compete rather than complement one another.
Systems have also had difficulty consolidating services within their own system. There are also new and more viable nonhospital competitors. More services can be delivered in new settings by highly specialized organizations, such as for-profit ambulatory surgery, diagnostic and treatment centers.
Finally, there is a greater cost pressure than in the fee-for-service era, clearly. However, because of the evolution of planned products and payment, managed care plans may have fewer mechanisms for controlling costs than was once anticipated.
In addition, capacity constraints were reported, however, in some of our markets. While marketwide occupancy rates do not appear to support this assessment, respondents noted a variety of recent developments that may account for this view. On the demand side, population growth in key geographic submarkets and rising utilization due to changing demographics and looser managed care products were noted.
On the supply side, prior misestimation of capacity needs, the need to replace aging facilities, the continued desire for short travel distances and convenience, and the nursing shortages were all cited.
The new medical arms race or even a modest uptick in service additions have several important policy implications.
First, retail competitive strategies have the potential to drive up costs. The duplication of services that may result increases the overall cost of health care through excess capacity, and services may result in supply-induced demand.
Second, these competitive strategies may actually threaten the clinical quality of care. Low volume results in poor outcomes. Even if there are marginal improvements in outcomes, including patient satisfaction, who is going to pay for the service?
Finally, the competitive dynamics may result in a misallocation of capacity. By that we mean an oversupply of some services and the undersupply of others of value, such as prevention and disease management.
There are a number of potential market and policy responses to these developments. Already, a variety of efforts are underway to make consumers better purchasers. As we have heard discussed, tiered approaches to cost sharing are designed to make consumers more sensitive to the cost of additional or enhanced hospital services, hopefully, without negatively impacting access to needed care. Specifically, consumers would pay more for utilizing hospitals or systems that are more expensive.
Another possible activity is to continue working to provide credible, comparative, and useful information—by that I mean understandable—about clinical quality to patients and consumers.
A second major approach would be to reconsider state and federal policy. Many certificate of need policies were relaxed in the hope that competitive market forces and managed care would lead to a more rational allocation of services. The ability to curb the expansion of services attractive to physicians and their patients, however, may be more than we should have hoped for for markets in managed care.
Public and private purchasers or other policy makers may also want to try to facilitate greater technology assessment to understand the cost effectiveness of many of these new and expanded services.
Finally, another potential response is to revisit and reconsider antitrust policy. Antitrust cases have been less aggressively pursued in the health care sector because consolidation was expected to have positive effects, unlike many other industries. This was particularly true, given the nonprofit status of the majority of community hospitals which have some obligations to act in the community’s interests.
However, consolidation appears not to prevent or slow service duplication to the extent anticipated by itself. In addition, the behavior of not-for-profits and for-profit hospitals may be blurring in the current competitive market environment. Each of these potential market and policy options have strengths and weaknesses which key market actors and policy makers will need to assess.
[Applause.]
PAUL GINSBURG: Now we’ll switch to panel discussion. On my left is Ellen Zane, who is network president of Partners HealthCare Systems, Incorporated. She is responsible for the developments of a provider network that features Massachusetts General Hospital, Brigham & Women’s Hospital, community-based physician groups and community hospitals throughout eastern Massachusetts. She previously served as the CEO of Quincy Hospital, a 290-bed community hospital in the Boston area.
J.B. Silvers is the Treuhaft Professor of Management at the Weatherhead School of Management at Case Western University in Cleveland. Until l997, he directed the Health Systems Management Center there. He also served as president and chief executive officer of Qual Choice, which is a Northeast Ohio HMO. In addition, he was a commissioner of the Prospective Payment Assessment Commission. He earned his doctorate in finance from Stanford University.
My first question is to Ellen Zane. In noting that Boston is a market that at least was heavily involved in risk contracting, I was wondering if you could give your perspective from at least Partners or from Boston as to how far it got and what’s been happening lately.
ELLEN ZANE: We had quite a few full-risk, global-risk contracted patients. In the Boston market in the mid to late ’90s, there were relatively flat health care premiums, where the HMOs were largely in a price war looking to grab market share from each other.
While that was going on, however, there was true underlying medical inflation, and ultimately, around the year 2000, it hit the wall where providers, like Partners, essentially said, "We will no longer take capitation budgets that don’t reflect the real cost of care."
So we drew a line in the sand and refused to take some capitation. There does remain capitation in the market. There are several hundred thousand patients or more just in the Partners system and probably close to a million in Boston that are still capitated. But we are seeing a large movement toward fee-for-service medicine, where the health plans simply don’t want to restrict choice and access for consumers.
Most recently, on the Partners front, we negotiated a pay-for-performance contract with a very large HMO, where we do not have capitation risk, but we are essentially bonused, as Larry said, for particular parameters of care that are likely to improve health care in general. So it runs from everything to increasing percentage of generics prescribed, to making sure that patients get eye exams that are diabetic, to CHF programs and so on.
So the jury is still out. It is largely new in our system, but we’re hopeful that if the capitation budgets cannot be sufficient, that there are other incentives that we can put in place to incent physicians to think about quality first. My concern is that there may not be enough money in the incentives to change behavior. So the jury is still out on that.
PAUL GINSBURG: You mentioned that you began to resist capitation just because the rates were too low, and I guess you felt that you got more adequate rates in fee-for-service than you did in capitation. In a sense, I guess, could capitation have had a shorter life span because it came at the wrong time? It came at a time when plans were really squeezing providers. In a sense, if the rates were higher, could you imagine a different course today?
ELLEN ZANE: I think that’s an important question. Many of the physicians that work with me believe that managed care is good and that capitation provided incentives to do very creative and innovative things for patients. But if the dollars were not sufficient or the patient population is not an actuarial sound large enough population, it is bad. So, as a result, I think if it had come at a different where people were truly looking to manage care properly, it could have had a different outcome.
PAUL GINSBURG: A question for J.B. Silvers. If you could compare Cleveland’s experience in capitation risk contracting to Boston, is it different?
J.B. SILVERS:: Cleveland had, I don’t think it ever had the degree of physician-based risk contracting that we’re talking about in Boston or California that Larry mentioned earlier. Most of the risk had been retained at the system level, the hospital system level, and we’ve taken the all-American approach of consolidation into only two major entities in Cleveland, so it’s a little like "Brave New World," with constant battles going on among relatively equal parties, but those two parties had taken risk during the consolidation period.
Unfortunately, they and other PHOs, hospital-based PHOs, had never figured out or chosen to pass that risk on to the doctors in any meaningful way. So, in one contract that we had with an independent hospital system, the PHO was retaining all of the bonuses that we paid and paying nothing out to docs, so they didn’t see any incentive whatsoever. It sort of messed up the idea. You know, it didn’t really work quite like it did.
What has happened in the last year, basically, maybe 2 years, is that the oligopoly that we have now is beginning to act like an oligopoly and has pushed back and has said not only we want more money, but we don’t want to take risk.
So I believe the Cleveland Clinic is almost completely, if not completely, out of the risk-bearing business. University Hospital still is in it because they own the insurance company that I ran for 2 years. But other than that, they’ve taken themselves out of the risk business almost totally, including the county hospital, by the way, that took Medicaid risk contracts. They’ve canceled those.
PAUL GINSBURG: No talk about the pay-for-performance that Ellen was mentioning?
J.B. SILVERS:: Well, in the plan that I ran when I was on the board before I became CEO, we had had withholds and things of that sort with docs, that kind of sort of backward pay-for-performance. I always thought that was stupid to tell the doctor it’s your money, but we’re going to hang on to it.
We have flipped that around and made it here’s your fee and here’s a bonus for additional performance. And very early on we had split the fee so that half of the bonus was due to cost considerations of the panel and the other half was due to quality considerations of the panel. I thought that made a lot of sense, and we continued with that. In fact, I think there’s some empirical evidence that, in fact, it works pretty well. So we’ve been doing that for quite a number of years.
Other plans, I think, have been much slower in developing that methodology, but I think it’s definitely a promising way for us to go.
PAUL GINSBURG: Ellen, in Boston, is this something that is just you’re pioneering with one plan or are you seeing it elsewhere in the market?
ELLEN ZANE: The pay-for-performance?
PAUL GINSBURG: The pay-for-performance.
ELLEN ZANE: We have pioneered this with one large plan that made a fair amount of press in the market and others are attempting to replicate it because we’re seeing trends spiraling out of control and something needs to be done.
PAUL GINSBURG: I have another question for each of you. How would consumer financial incentives that were discussed in the previous panel, how would they effect risk contracting, both its current or future states.
ELLEN ZANE: It’s very difficult to take risk when there is no, and I tried not to use the term "gatekeeper," but when there is no navigator, when there is nobody really at the point to manage the care, and it’s basically open access. It’s very difficult to reinstate risk in any major way. Even if the consumer has more skin in the game, it’s probably still an open-access-type product. So I think they’re working in different directions.
PAUL GINSBURG: Sure.
J.B. SILVERS: I think you’ve seen some movement toward three-tiered models, particularly around copayments for pharmaceuticals. We hadn’t gotten anywhere near the way Minnesota had moved, which I think is obviously one of the considerations for the next step, to get that in there.
What I think we found is some interest in defined contribution, not necessarily a medical savings account, but an interest in somehow creating a marketplace where individuals could choose among various competing plans. It is being redubbed employee choice, partly for marketing reasons, I think, and that hasn’t happened yet in Cleveland—a lot of interest.
I was on the board of a start-up company trying to get some of that going, and it failed. I think the timing was wrong. With a tight employee labor market, no one was going to mess with it. In this labor market, I think we’re going to be seeing some movement, and you mentioned Textron, I think. I’ve heard that. I think you’ll find some other key examples and then a lot of companies will go to that, which does create some consumer incentive, financial incentive, very strong.
PAUL GINSBURG: What’s your notion, say, looking 2/3 years down the road? Do you see some form of risk contracting being, in a sense, returning?
ELLEN ZANE: The medical directors that I have been working with have been very clear with the physician population that we need to make pay-for-performance work. Because if we don’t, somebody holds the risk, and that risk has to be mitigated and these cost trends have to be controlled. So, if this doesn’t work, I believe that we will be back to some type of heavy risk on providers and maybe also on consumers, but it won’t be very enjoyable. So it’s in our best interest to make this work.
Some of the models we’re seeing coming out in the market for consumers in Boston, at least, are not refined yet and need some tweaking for sure.
PAUL GINSBURG: Sure. Actually, do you have a comment on the announcements of the surcharges for academic medical centers?
ELLEN ZANE: As a matter of fact I do.
[Laughter.]
ELLEN ZANE: Given that the Mass General and the Brigham are both major academic medical centers, we are not against, in any way, the idea of consumers having more responsibility and that the right care should be provided in the right place.
However, in the City of Boston, as many of you may know, there are a large number of academic medical centers. And for 6- to 700,000 people in the Greater Boston suburbs, academic medical centers are their community hospitals, and we don’t believe that people should pay a surcharge for accessing care if it is, indeed, their community hospital.
Secondly, these incentives have provided so that if an individual wants to access an academic hospital tertiary center for certain DRGs, for transplants, for burns, then they won’t be charged a surplus, but when we look at the data, people use academic medical centers for far more than that. So we have some concerns about access for people who live in the city, what is tertiary care and emergency care?
PAUL GINSBURG: Good. Now, in Cleveland’s, this is one of the sites where we saw both intense competition among the two systems, and behavior that kind of resembled the medical arms race. Could you kind of describe what’s going on in Cleveland?
J.B. SILVERS: Well, I mentioned the "Brave New World" earlier. That seems to me like the best, if you remember Huxley’s novel from many years ago where there were three countries that were constantly in battle, never winning, always in some sort of equilibrium as a way to control the population. I don’t think we quite are that far along, but I see this competition taking on some very interesting and maybe bizarre, if you think of it in a broader market context, characteristics.
Bragging rights have to do with things like the number of sites, which I find very interesting. The University Hospital Health System now advertises 150 access points. Fifty of them came because they bought Union Eye Care, which is basically they sell eyeglasses.
[Laughter.]
J.B. SILVERS: How do they suddenly go from 100 to 150? And then it dawned on me, well, we bought a—okay. I don’t quite understand that strategy, but it sure sounds good.
The Cleveland Clinic is in the business of providing commercial locations, so they have very good real estate agents that find things near intersections of the Interstate, so they can put the CCF logo up nice and high. It’s a very pretty building. It’s very nice. It reminds me of a combination of Starbuck’s, Home Depot, and maybe the hub-and-spoke strategy of United Airlines all mixed together in some sort of a strange way.
[Laughter.]
J.B. SILVERS: Not clear how successful it is, but it looks good. And I think in a world where you had enough cash coming in, it could look—
Well, I think what you’ll find now is some real refining of these strategies with lower margins, unless the systems are able to push back well enough and hard enough to get the margins back up from private payers. Medicare is just fine, thank you. They’re doing just fine with that. But if you can get more money from the private pay to subsidize this continual arms race, if people like to use that term, or commercial venture, if you’d like to think of it that way, then I think it can go on for a long time because it looks good in a commercial sense. You’re approaching the market, you’re getting lots of exposure, you feel good about it, it works well.
If, on the other hand, employers start pushing back the other direction, demanding much tighter managed care or doing defined contribution, which then makes the consumers much more price aware, then I think you’ll have to find a refining of this investment strategy, which does have some fairly bizarre characteristics to it.
PAUL GINSBURG: One thing that’s always perplexed me about Cleveland is that I could see some of the vigorous competitive strategies by the two major systems to have been suitable in the managed care environment of 5 years ago, where, in a sense, there was selective contracting. So that if they could serve all areas well enough so that they could be the sole contractor with a health plan, that would be a real advantage, but, of course, that’s not the model today.
To what extent do you think maybe they’re kind of pursuing a model that doesn’t make sense today?
J.B. SILVERS: Now you’re asking me to make a statement that would get me in big trouble if I said it doesn’t make sense.
I think there’s an argument that says that. I wonder, however, I think in this competitive environment there’s almost a natural instinct for competitors to want to have some equilibrium, to have some balance involved.
I, remember, I was just reflecting with somebody earlier, a study I did very early and maybe in the ’70s, where I was looking for two hospital markets. This was during the CON period. I was looking at two hospital markets, where one hospital had a very large high proportion of the volume, and the capacity and the volume, and another one had a much lower one.
I was frustrated because I couldn’t find very many. They almost always, in the Midwest at least, tended to be sort of roughly equal. And then I started looking at some CONs, and I found statements in there about, well, we’ve always had 42 percent of the obstetrics market, and so and so has just added 10 new beds, so we have to add 11 new beds. I mean, I literally found people saying things like that.
[Laughter.]
J.B. SILVERS: I thought, well, that is an interesting form of competition that we might be going back to, sort of competitive equilibrium in a funny kind of sense that’s not really market based, but it’s capacity based, in some sense. That’s part of what’s going on, and it’s the strength of the systems, and the egos of the people involved and things of that sort too.
PAUL GINSBURG: Yes, speaking of egos, could you envision, if there were different people in the lead, whether, in a sense, this could be more of a cooperative thing, where they maintain the status quo and compete less vigorously?
J.B. SILVERS: Now you really want to get me in trouble.
[Laughter.]
PAUL GINSBURG: I’ve got to stop this.
J.B. SILVERS: Oh, sure, of course. But I think they are very strong leaders, and I mentioned the last time I did one of these things with you a few years ago that if you didn’t know the leaders in Cleveland, you really wouldn’t know the game. It’s very much tied to the individuals involved, one of them gone now, two are still around, in terms of the two systems.
But in terms of whether it would be different, I think what you found was a little bit of the "Russians are coming" phenomena. When Columbia HCA came into Cleveland, it looked like this wave was going to take over the independent hospitals, and the two large tertiary centers said, "Oh, my God, this is terrible. We’re going to get eaten a live. We better respond," and they responded so much that even though it was almost like the Russians are coming, Columbia was not that big a deal, in retrospect.
The reaction was just phenomenal, in terms of everybody needing to choose sides. So we had this huge game of tag in the backyard, where everybody chose sides, and then they found out this isn’t so bad. Now I can actually use some market power. And then they started figuring out, well, I know what it’s like to be a monopolist now. I can sort of act like a monopolist, and so we had a strange phenomena.
The question is, is there any other credible threat of entry that would discipline the market, and I think part of what you see here, William Ballmer wrote a book quite a number of years ago now, where he talked about instead of just looking at concentration ratios, you need to look at credible threats of entry.
I think what you find is in specialty niche services, the kind you found, there still is a credible threat of entry from your own docs, as well as maybe from outside groups. Therefore, you find institutions responding to that, a very logical economic process of response to that.
PAUL GINSBURG: Ellen, do you see similar things going on in Boston? Particularly, is there a threat from physicians organizing their own outpatient facilities to compete with your system?
ELLEN ZANE: Interestingly enough, we’ve seen very little of the for-profit entry, particularly organizing physicians to compete. However, the arms race between the large medical centers and the community hospitals is very much alive. I heard a story the other day of about a 150-bed community hospital wanting to acquire a PET scanner. You know, you just have to ask yourself why.
Until very recently, cardiac surgery was largely done at academic medical centers in Boston and recently through the CON program. That will now be outmigrated for six additional sites in Eastern Massachusetts. So there is definitely a thirst for people to get in the game.
PAUL GINSBURG: We were hearing before about the tiered copayments and tiered networks. If this actually developed, to a large extent, would this have an important effect on the competitive dynamic in hospital markets?
J.B. SILVERS: I think it probably would. However, when you get to this level of consolidation, as we have in Cleveland, it’s a little hard to see how—it’s like the fellow from Seattle was speaking, it’s a little hard to see how you’d do much tiering. And particularly since Cleveland started with the best basis for that of all that I know of, at least, in the Cleveland Health Quality Choice data, which John had mentioned earlier, very good data.
It basically was dumped as one of the first acts of an oligopoly by the Cleveland Clinic. It’s no big secret. They pulled out, and they pulled out for technical reasons, but mostly because they didn’t do exceptionally well. They did fine. They were, clearly, a good, quality operation, but the best facility was Mount Sinai Medical Center which, bless its heart, is now closed.
[Laughter.]
J.B. SILVERS: So, you know, it sort of worked out the way the two big systems would like to have it work out, and then we got rid of the quality measures, so now we can just go with reputation.
It’s sad. I understand completely the competitive issues, and that’s probably what I would have done in that situation, too, but the data to do tiering is not really there, I don’t think.
PAUL GINSBURG: Do either of you think that some—is it going to be noticeable the extent to which these medical arms races affect costs or the use of services and could that trigger a policy response?
ELLEN ZANE: It’s already quite noticeable. I was chatting with someone this morning about radiology costs. They are the new pharmacy. It’s a true "build it and they shall come." We’re seeing MRIs and CTs proliferated, and when we look at utilization, it is skyrocketing. There can be no other outcome, except that it tremendously ramps up the cost of care.
So this technology getting better and better and more sophisticated, and often better for patient care, and we have to remember that as we evaluate what costs are end-to-end or by episode, but, nonetheless, technology is a huge driver.
PAUL GINSBURG: Massachusetts tends to be one of the leaders in regulation of health care. What’s the talk among Massachusetts policy makers about a policy response to these developments?
ELLEN ZANE: Massachusetts was always the Peoples’ Republic of Massachusetts, but now we have had, for the last decade or so, Republican governors, and so there is no talk of the—we are deregulated now, and from time to time someone in the legislature will try to place a new requirement into a plan, and they are often successful, and it does increase cost.
Prudent Layperson is a good example. While none of us, it’s motherhood and apple pie to want people to have access to emergency rooms, but through Prudent Layperson, we’ve looked at our emergency costs go way up. So there is a cost to all of this, but there is no talk of bringing it in.
PAUL GINSBURG: Sure. I don’t know if you have anything to say—
J.B. SILVERS: Well, I don’t—
PAUL GINSBURG: Before you do, I’m then going to turn to the audience, so if you could start coming up.
J.B. SILVERS: There’s no question it’s driven cost up. We’re seeing double-digit increases in Cleveland, like everyplace else. The real question for me is, is that really bad? People, employers clearly think that’s a problem because it’s driving their costs up, but they’ve absorbed it so far, and I think they’ve absorbed it partly because of tight labor markets, but also because they feel that the employees value these services so much.
It is not clear to me that in a world where employees are making more choices themselves, that they would necessarily check—not pay more for services. I think people, in fact, there’s some evidence that’s coming out now and some interesting studies that show that people, that, in fact, technology has provided great value. Drugs, for that matter, have provided great value in terms of cost issues, but also in terms of perceived value.
So we may be just providing the market more of what it wants, and I think it’s a little too early to say that we’re going to throw the brakes on this so fast.
PAUL GINSBURG: Anne?
MS. GAUTHIER: Anne Gauthier of the Academy for Health Services Research and Health Policy.
I enjoyed the presentations, all of them. And while one certainly doesn’t want to predict the future in health policy, it’s still fun to speculate a little. I noted, with interest, today that two very old types of regulation, not rate regulation, but two very old things, certificate of need and even Hill-Burton came up today in conversation.
I would be curious in some speculation on how systems might respond under this current medical, "medical arms race," if there were new types of Certificate of Need restrictions or parameters put in place, and I guess you might want to say what those might look like in today’s world even before answering.
PAUL GINSBURG: Who would like to answer that?
ELLEN ZANE: Certificate of Need didn’t work, at least in Massachusetts, they didn’t work. So I don’t see them getting a life all too fast again and wouldn’t want to predict on what it might look like. Because when we did have regulation, I saw lots of lobbyists making lots of money to work with lots of legislators to get around it. Ultimately, at the end of the day, they did.
PAUL GINSBURG: Anyone else?
J.B. SILVERS: The only issue I can think of is CON is not so much for hospitals, but for other facilities. It’s a little, I’ve been toying with the "Harry Potter" thing, you know, the sorcerer’s stone. Harry got it because he didn’t plan to use it for bad. If you haven’t seen the movie, don’t worry about it.
[Laughter.]
J.B. SILVERS: The issue I’ve seen is the number of cases where physicians have been co-opted and—co-opted, where they’ve been bought by cooperative ventures, and it doesn’t matter if it’s for-profit or not-for-profit, they still are getting bought off, frankly—Start, notwithstanding—through joint ventures and other arrangements that effectively let a medical director make a lot of money or whatever.
I think there are some fundamental issues when the person that directs the traffic has that kind of financial interest. I don’t think I want my primary care doctor to be a capitalist, to be blunt, and I teach in a business school. I think we’ve got some fundamental regulatory issues about those sorts of questions, in terms of who makes the capacity, the utilization decision and how we pay them. Now we haven’t wrestled with that approach.
PAUL GINSBURG: Yes?
MS. FRIEDEN: Joyce Frieden from Physician’s Weekly.
I was wondering if Ms. Devers could expand a little bit about how the medical arms race now is different from in the fee-for-service era. I’m sort of, in some ways, it’s surprising to hear that it’s going on at all because everyone is complaining about the cost of health care, but I am wondering if the things that people are getting, the machines or whatever, are different from what the arms race was.
KELLY DEVERS: I think we certainly are seeing more activity in the outpatient areas. As J.B. Silvers mentioned, that there is more threat of entry in that area. I do think that the new twist here is the consolidation on the hospital side, and so we have fewer major hospital competitors that appear to compete just as intensely as may have been anticipated when there were more competitors in the marketplace.
So the consolidation hasn’t necessarily slowed at all the kinds of competition between hospitals and coming back again to the notion that there are more potential competitors on the outpatient side.
PAUL GINSBURG: There’s a question over there.
NORA SUPER: Nora Super with the National Health Policy Forum.
I work closely with folks who work on Capitol Hill. I have to tell you, in the past couple of years, all they hear is the reason for any changes in the hospital industry has been the Balanced Budget Act, which none of you all seem to mention, which I found telling.
I wonder how much you think the new Prospective Payment Systems have or have not affected hospital strategies in dealing with the private sector.
ELLEN ZANE: Let me say very clearly that the issues around the BBA are underlying all of the negotiations that at least we have had and my colleagues within Boston have had over the last 18 months to 2 years. As a matter of fact, we were down here lobbying, and someone very high up at HCFA at the time said to the CEO of the Partner System, "Don’t come crying down here about the BBA, when you’ve got lousy HMO contracts that don’t cover your costs back home, so go home and fix it."
That, basically, was the catalyst that had us go back and very much draw a line in the sand about these capitation contracts. So the BBA is a very significant driver in what we’re seeing, at least in our market today.
J.B. SILVERS: The MEDPAC had an interesting chart in the June report this year, 2001. They showed payment-to-cost ratios, in other words, profit margins, for the last 8/10 years. There was no question, I mean, you just look at the data and it’s obvious what the BBA is—nothing. If you look at going from about 90-percent payment-to-cost ratios for Medicare and about 80-percent for Medicaid at the early part of the decade, they’ve become basically break-even businesses.
Where you look at private pay, it’s gone from about a 30-percent additional margin, 130 percent of costs being paid out, down to perhaps half of that on the average. The problems that hospitals have had have not come from Medicare, they’ve come from private payers, and it’s been this competitive market that we’ve talked about today.
Now the BBA clearly said put up or shut up. It said, "Look, I’m not going to do this any more. I’m going to level that out," and that was a catalyst then to make them go back and talk to private payers, which they’ve done.
PAUL GINSBURG: If I could do a follow-back, that throughout our site visits, we’ve often heard about that hospitals and often physician entrepreneurial outpatient facilities tended to go very much into orthopedic, cardiac, certain areas which always were described as these are the most profitable areas.
Given a system where, obviously, the plans would not want to pay more for some services in relation to costs than other services, why is it that this has persisted so long? I mean, if you feel it’s correct that certain services are more profitable than other services and there’s a strong incentive for hospitals and outpatient facilities to put capital into them.
ELLEN ZANE: If I understand correctly, it largely follows the dollars. It’s about how we get reimbursed for those particular services, and they are fundamentally more profitable.
PAUL GINSBURG: That’s what I meant. Why is it that the reimbursement system has made them more profitable and not changed in reaction to noticing how profitable they are, relative to other medical services?
ELLEN ZANE: Part of it could be an artifact of history, but also because of the requirements to provide the service. Although orthopedics is a good example, it’s become more, and more, and more a secondary level-type service, but remains very profitable. So I would sense, and I may be wrong, but I would sense that a fair amount of it is an artifact of history.
LAWRENCE CASALINO: If I can just add, I think a lot of it is just rates were set at certain levels for certain kinds of services, and there’s been a lot of inertia there. But, also, again, certain specialties have a relatively easy time organizing, either legally or, in very many cases around the country I would say with little doubt we saw illegally, where you only need, in a fairly populace county, 10 or 15 orthopedists really, maybe less, to, in effect, negotiate collectively with plans or, for that matter, negotiate with an IPA and just say, hey, you want orthopedists in your IPA, this is what you’re going to pay us, and they can keep their rates high much more easily than primary care physicians or physicians in specialties where there are more physicians.
If I may, I just want to comment one thing about the BBA. That’s always talked about in terms of hospitals. There is the debate which some people don’t consider a debate, I guess, about whether Medicare+Choice was paying health plans too much for taking care of Medicare patients. If they weren’t paying them too much, it’s a little hard to understand why all of the plans are getting out.
But I want, again, to look at it from the physician group side. The average capitation rate for a Medicare patient is 3 or 4 times that for a commercial rate. Since Medicare patients tend to be sicker than commercial patients, the opportunities to generate both higher quality and more savings for managing care well are much more.
So, insofar as there exists Medicare HMOs, there is potentially, if the rates are reasonable, potentially more incentive for organizations, provider organizations, to engage in risk contracting or some form of pay-for-performance than there is on the commercial side, where rates are much lower capitation rates and where the rewards of managing care are probably not as high because the patients aren’t as sick.
J.B. SILVERS: Just one more thing on the BBA. Picking up on Larry’s comment, if you look at the prospective payment mechanisms for nonhospital services, that’s where there’s a very legitimate concern among the hospitals because they had backward integrated into—forward integrated into everything else, skilled nursing, a lot of outpatient services, obviously, and then on down the line in home health care.
If you look at the margins that they are getting for Medicare on those nonhospital services, they’ve taken a complete bath. They’re just down the tubes. So that hospital systems that had integrated across the board and a continuum of care, all of those nice words, have basically, for financial reasons, I might add, wrapped in, because you get paid well for that, wrapped in the rhetoric of patient care and economy of care, I think quickly are going to disintegrate around those services for exactly the same reasons, because you’re not going to get paid.
So, you’re going to find them spinning off all sorts of things and getting rid of services, and they probably didn’t make sense to begin with, from a patient care point of view. But that’s a serious issue, the BBA. Do you want an integrated health system that tries to cross those boundaries, in which case we were doing it, or would you rather have independent entities out here doing these things, in which case we have coordination issues, which is where we’re going now, for financial reasons through the BBA?
PAUL GINSBURG: Jonathan, I have to beg your pardon that we took so much time, and the people running this conference will shoot me if I don’t let you go to lunch now.
What we’re going to do is there is food right out the doors, it’s a buffet. I would urge you to go there quickly, bring your food back to the table, and when you’re all settled down, then our luncheon speaker, Janet Corrigan, will begin.
I want to thank this panel for a really great job.
[Applause.]
[Whereupon, at 12:02 p.m., the proceedings were adjourned and reconvened at 12:37 p.m.]