How Should Washington Control Medicare Spending?

Report Health Care Reform

How Should Washington Control Medicare Spending?

August 30, 2011 40 min read Download Report

Authors: Robert Moffit, Gail Wilensky, Ph.D. and James Capretta

Abstract: With the Patient Protection and Affordable Care Act of 2010 (PPACA), Congress enacted record-breaking provider payment cuts as well as hard caps on the growth of Medicare spending for the first time in Medicare’s history. (Medicare spending has historically outpaced the measure of general inflation, medical inflation, and economic growth.) These new policies, if sustained over the next 10 years, can have far-reaching consequences for doctors, patients, and American taxpayers. Will these measures be effective? How will they impact physician participation and patient access to care? Is there any way that Medicare patients can avoid higher premiums, higher costs, or access problems? Will alternative Medicare payment options, such as premium support models, have a better chance of controlling costs and improving patient access to high-quality care? These and other pressing questions were addressed by three prominent health policy analysts at a panel discussion at The Heritage Foundation on May 19, 2011.

ROBERT E. MOFFIT: Medicare will challenge us for the next several years. The reason: The first wave of baby boomers will start to retire, Medicare enrollment will swell, and costs will soar dramatically.

But something really remarkable has already happened. A lot of us are still struggling to absorb the fact. But Medicare, as we know, has already dramatically changed. And so have the contours of the great Medicare debate.

Let me explain. People talk about whether we can keep Medicare as it is—or was. With the enactment of the Patient Protection and Affordable Care Act, Medicare is already changing pretty rapidly. Since the inception of Medicare back in 1965, the program has been an open-ended entitlement. Whatever the demand for medical services, the supply was provided; and the taxpayer simply picked up the tab, regardless of cost. And the bill got progressively bigger and bigger every year.

A New Consensus. Over the past 20 years, the average annual growth rate of the Medicare program has been about 8.5 percent—very robust. The new consensus in Washington—among Republicans and Democrats alike—is that this rate of Medicare spending is not only undesirable, but it is also unsustainable. So, as we gather today, we are the recipients of a new political dispensation. As incredible as it sounds, especially to those of us who have been through the big Medicare wars on Capitol Hill in the 1980s and ’90s, there is, for the very first time, a powerful consensus on one really crucial aspect of Medicare policy—and that is the need to impose a hard, external cap on the growth of Medicare spending. No more open-ended entitlement.

The central point of disagreement now revolves around not whether we ought or ought not to have a cap on Medicare spending. No, it revolves around how exactly we ought to establish the cap and index it to control future Medicare spending growth. We are, as they say, haggling over the details of agreed upon policy.

A Variety of Options. A traditional point of departure on health care policy, cited by liberals and conservatives alike, has been that, since health care has been growing at twice the rate of inflation, we should somehow tie the growth of Medicare spending to inflation. That begs the next question: What measure of inflation do you want to use? Do you want to use the consumer price index? Do we want to use medical inflation, or a combination of medical and general inflation?

Others say the best way to control Medicare spending is to tie it to the growth of the national economy and index it to the growth of the gross domestic product (GDP). Still, others propose external spending caps with administrative payment changes. You’ve seen a lot of this in the press: bundling the payments for specific episodes of care, paying on the basis of compliance with standards of quality set by the Department of Health and Human Services, or new forms of competitive bidding. In any event, there are many potential indices for growth. The issues are primarily technical, but those details are, as is inevitably the case with health policy, seriously consequential.

There is a related policy question. Besides controlling spending, is there any other purpose behind the adoption of spending caps? Spending caps obviously help policymakers to meet established budget targets. But spending caps can also play a back-up role for an alternative approach to cost control: changing economic incentives to secure value and savings. In other words, the primary source of spending control in Medicare would ultimately depend on our ability to harness market incentives to drive the system in the direction of efficiency in the delivery of medical services.

Market Incentives. The primary reliance on the power of market incentives is the key flashpoint in the current Medicare debate. The premium support option, a defined contribution for Medicare financing, that is being promoted by Representative Paul Ryan (R–WI), chairman of the House Budget Committee, and adopted as a central component of The Heritage Foundation’s Saving the American Dream, has a rich and varied history. While there are and have been real differences, the general approach is embodied in the work of Henry Aaron and Robert Reischauer in the 1990s, both prominent liberal policy analysts; the majority proposal of the National Bipartisan Commission on the Future of Medicare in 1999, chaired by Representatives John Breaux (D–LA) and Bill Thomas (D–CA). Indeed, it goes all the way back to 1980 when Representative Richard Gephardt of Missouri—who later became the House Democratic Majority Leader—introduced the National Health Reform Act of 1980, which called for a premium support system of financing for Medicare. Indeed, though one wouldn’t know it from recent media reports, it has strong bipartisan roots.

With the Patient Protection and Affordable Care Act of 2010 (PPACA), Congress enacted a combination of measures of inflation (a blend of CPI and Medical CPI) and gross domestic product—the GDP-plus-1-percent beginning in 2018. More interesting, as you know, the President has proposed a budget option to reduce Medicare spending of GDP plus one half of 1 percent. It, like the PPACA target, would be enforced by the new Medicare board (the Independent Payment Advisory Board) through specific and detailed provider payment reductions. This, by the way, is proposed on top of breathtaking payment reductions in Medicare already required under current law. So, the President’s budget proposal would impose an even tougher cap on Medicare spending.

Make no mistake, the new Medicare debate is just getting underway. The issue of an external spending cap is not simply a technical matter. Rather, the determination of how we do this, and why we are doing it, is tied to broader questions of what we expect in the future from Medicare. We have two of Washington’s best health policy analysts on this question. Dr. Gail Wilensky is an economist and a senior fellow at Project Hope, an international health care foundation. Gail directed Medicare and Medicaid from 1990 to 1992. She served in the White House as the Senior Health Advisor to President George H. W. Bush. She currently serves as a trustee of the Combined Benefits Fund of the United Mine Workers of America. She is an elected member of the Institute of Medicine and has served for two years on its governing council. With an outstanding reputation in health policy, Gail received her bachelor’s degree in psychology and her doctorate in economics from the University of Michigan.

James Capretta is a fellow at the Ethics and Public Policy Center here in Washington. Jim was an associate director at the White House Office of Management and Budget (OMB) from 2001 to 2004, where he wrestled with these issues on a daily basis. He was the Bush Administration’s top budget official for health care, including Medicare. Earlier in his career, Jim served for a decade in Congress as senior analyst for health care issues and for three years as a budget examiner at OMB. He has his master’s degree in public policy studies from Duke University, and is a graduate of the University of Notre Dame.

—Robert E. Moffit, Ph.D., is Senior Fellow in the Center for Policy Innovation at The Heritage Foundation

GAIL WILENSKY: As Bob Moffit has mentioned, the term “premium support” is widely credited to economists Henry Aaron and Robert Reischauer, based on their November 1995 article “The Medicare Reform Debate: What Is the Next Step? in Health Affairs, although as a concept, it has a much longer history. The basic principles of premium support are also the ones underlying Alain Enthoven’s “consumer choice” proposal from the 1970s, as well as policies promoted by the Heritage Foundation’s own Bob Moffit and Stuart Butler, who had been advocating the FEHBP (Federal Employees Health Benefit Plan) as a model for reforming Medicare with an article (“The FEHBP as a Model for a New Medicare Program”) promoting that theme in the same 1995 issue of Health Affairs as Aaron and Reischauer.

An Established Concept. So while the term “premium support” may be a relatively recent term, the fundamental concept has been around for a while. The notion of premium support, when applied to a Medicare beneficiary, is that each senior would receive a fixed subsidy from the government which could be used to purchase a health plan of the senior’s own choosing. The expectation is that the resulting competitive pressures that would develop among alternative plans should moderate increased spending on Medicare. Whether traditional fee-for-service Medicare would be offered as one of the choices and how the government’s contribution would be set—by competitive bids or according to some preset index—are among the many issues that would need to be determined to make the premium support concept operational.

I also want to reiterate another point that Bob Moffit made because I don’t think it’s been properly appreciated to date. Both Republicans and Democrats are now talking about introducing a “hard cap” on Medicare spending, although Democrats have been less explicit about it. The way each side has proposed to implement the spending cap is very different

This is not the first time Republicans have raised the issue of limiting total spending on Medicare. When they took over the Congress in 1994, Republicans proposed limiting spending for each part of Medicare—similar to the spending limits that exist for Part B spending on physician services. It could be argued that as long as a cap on spending already exists for one part of Medicare, as it does now, it would make sense to extend the concept to the other parts of Medicare, such as inpatient hospital spending or spending for home care; but many analysts have raised concerns about locking in the relative share of spending that happens to exist at a moment in time as though that represents an appropriate relative share of spending. There are also legitimate arguments over whether slowing spending on Medicare can be done in ways that are less destructive or punitive than by limiting spending by type of service, particularly when reimbursements are set by administrative pricing, as is the case for Medicare, but unfortunately, these debates have rarely occurred.

The March 1995 Republican congressional budget proposal (a Republican-proposed balanced budget amendment passed the House and failed in the Senate by one vote) received mixed congressional support, even from some Republicans, and produced a strong backlash from Democrats as well as from the public. Not surprisingly, then as now, there were many speeches from Members of Congress and the media about how limiting spending to the degree proposed would “destroy Medicare as we know it.”

The answer to whether the country will now be able to have a rational discussion about the need for a limit on Medicare spending and if so, how to set and enforce the spending limit, is not yet clear. Early experience is not encouraging—as evidenced by the reaction to Representative Paul Ryan’s (R–WI) “Path to Prosperity,” which included a proposal to reform Medicare using a premium support model, and the difficulty or unwillingness of the media or the public to recognize the implications of the Administration’s proposal to enforce limits on Medicare spending through the Independent Payment Advisory Board.

The Entitlement Crisis. Let me also comment briefly on what seems to be a new (or perhaps renewed) recognition that reducing deficit spending and reducing the debt will be difficult if not impossible without tackling entitlement spending. For some of us, this is a longstanding position; for others, it is quite new. What is not helpful is the notion that the need to focus on entitlement spending is a result of the severe economic downturn in 2008 and 2009. It is true that tackling the current deficit will be very difficult if we focus only on discretionary spending, but the “dirty secret” is that the explosive problems produced by current and projected levels of spending on entitlements were waiting in the wings even if the 2008-2009 recession had never occurred.

There are two major reasons for the projections of unsustainable spending on entitlements—particularly on Medicare. The most important reason is that Medicare has been growing substantially faster than the economy for the last several decades—like the rest of health care spending. Ultimately, a spending rate that exceeds the growth rate of the economy as well as the funding sources of the spending will be unsustainable. It will also stress the federal budget since it implies the need for an increasing share of the federal budget to fund Medicare or else increasing resources to the federal government—neither of which are likely as long-term strategies.

The second reason is the changing demographic landscape. As we heard repeatedly this year, the first of the baby boomers started turning 65 in January and they will continue reaching the current retirement age of 65 for the next 20 years. By 2030, we will roughly double the population on Medicare—going from currently 44 million to about 78 million people. The exact number will depend on how long people live during the intervening 20 years. However, the increasing population on Medicare is not the primary problem. If we could get spending to a more sustainable rate, we could deal with the increasing population. The primary problem is what us economists call the “excess spend” on Medicare and health care—that is, the spending rate that exceeds the growth rate in the economy by several percentage points.

If we accept the position that the country has to slow the rate of growth of spending in Medicare, the question is how best to go about slowing it? My observation—based on my experience as administrator of the Health Care Financing Administration (now CMS) and as chair of the Physicians Payment Review Commission and the Medicare Payment Advisory Commission—is that over the last several decades, all pressure to slow spending in Medicare has been placed on the provider community. In contrast, private-sector programs typically try to influence the behavior of both providers and the users or consumers of health care.

The basic strategy used by Medicare to slow spending has been to reduce reimbursements paid to individual and institutional providers. However, reducing reimbursement does not necessarily result in comparable reductions in spending since spending reflects both reimbursement per unit and the volume and mix of services provided. Nowhere has this been clearer than with physician reimbursement under Medicare where physicians bill on the basis of some 8,000 different CPT (current procedure terminology) codes. Over the last decade, fee increases per unit of service have varied between zero percent and 1.7 percent per year at the same time that spending for physician services per beneficiary has increased by as much as 10 percent per year. The impact of volume and mix changes are not quite as serious for areas of Medicare where reimbursement occurs for a larger bundle of services, such as for inpatient hospital services where the reimbursement depends on the diagnosis at discharge, or for home care where the reimbursement is based on a 60-day episode of care.

Payment Cuts. Once again, under the Patient Protection and Affordable Care Act, all of the pressure to reduce spending in Medicare is on the providers. In addition to physicians, who are already scheduled for substantially reduced reimbursements as a result of the sustainable growth rate provisions of the 1997 Balanced Budget Act, all major providers of services in Medicare will see reductions in reimbursement. Unfortunately, the incentives to provide more and more complex services that are inherent in the fee-for-service oriented traditional Medicare system remain in place—just reimbursed at a lower rate.

There are limited provisions in the PPACA legislation that also provide for some changed incentives for providers (rather than only reducing reimbursements), such as the value-based purchasing reimbursement provisions for hospitals and nursing homes, or the accountable care organizations (ACOs) that would allow hospitals and physicians that are not part of formally integrated plans to work together to coordinate care and share in the resulting savings, provided they meet a variety of requirements. However, as the preliminary rule for ACOs has made clear, even when the organizational change is strongly supported by the Administration, the implementing regulations may be written in a provider-unfriendly manner. Basically, Medicare remains a fragmented, fee-for-service system with separate silos of payment for hospital inpatient services, outpatient services, physician services, home care and nursing home services, outpatient drugs, etc.

Given Medicare’s history that reimbursement reductions frequently do not result in spending reductions, especially a few years after the reductions are in place, it is not surprising that the Administration has backed up the reductions with authority to enforce spending limits by granting the power to impose further limits on reimbursement through actions by the Independent Payment Advisory Board (IPAB). Again, all pressure is limited to providers, with the IPAB having no authority to change the age of eligibility, limit or change benefits, impose new or additional cost-sharing measures on seniors, etc. It can only change reimbursements to providers.

It is within this context that I find Representative Ryan’s proposal to be so interesting. The proposal, as it was originally outlined, becomes much more interesting with a few key parameter changes—at least to me as a policy person and former Medicare director. Under Ryan’s initial proposal, everyone who turns 65 in 2022 or later would receive a fixed subsidy from the government. Initially, the amount is set at $8,000—the amount that the Congressional Budget Office (CBO) estimates that Medicare will spend on a 65-year-old in 2022. The subsidy would grow at the rate of inflation—the specific measure Ryan used is the consumer price index for all urban consumers (CPI-U)—and would be greater for people who are sicker or older. Low-income beneficiaries would receive additional support through a medical savings account. A wide variety of private plans would be offered and the government would monitor the plans to make sure they provide a minimum set of benefits and don’t discriminate against individuals who are older or sicker. All people currently 55 or older would be able to continue enrolling in the Medicare program that currently exists, or they could choose to enroll in the new program.

The changes that I have proposed (see my article “Reforming Medicare—Toward a Modified Ryan Plan,” in the May 19, 2011, New England Journal of Medicine) would increase the subsidy to a rate of about GDP plus 1 percent, include traditional Medicare on a defined contribution basis as one of the options, and ensure that at least one plan would be available in every area at whatever percentage the original subsidy had been intended to subsidize when the program started. Traditional Medicare assumes premium payment, deductibles and co-insurance, and whatever subsidy is provided to new 65-year-olds.

The point of this type of strategy is that plans would have strong incentives to come up with delivery reforms that would keep premium costs low, or risk losing seniors when they have an opportunity to change their enrollments—just as Part D plans have strong incentives to behave similarly under the current Medicare program. Those who prefer to remain in the traditional Medicare program should have an opportunity to do so—but with the same subsidy that would be available to them to choose any plan. If Rick Foster, the CMS actuary, is correct and the reimbursement reductions in current law are likely to cause access problems, perhaps being able to change incentives and the way health care is organized without passing new legislation, like private plans would be able to do but Congress would not, would produce better outcomes. But if no plans are available at an acceptable cost, the government will need to reconsider the subsidy provided to seniors.

If we can get people to focus on these kinds of policy choices, perhaps we can have a serious dialogue about the future of Medicare. This is clearly not happening yet.

JAMES CAPRETTA: One way to enter into this conversation about which cap would be more effective is to ask, Would it be better to have a cap that was enforced by, as Gail was discussing, the IPAB, the Independent Payment Advisory Board, or some administrative structure? Or would it be better to have something along the lines of what Paul Ryan has proposed—premium support—which has a long, long history? Perhaps the way to start thinking about this is to ask, What are we trying to achieve here? I would say that the President and his people, when they pushed the health care law through last Congress, did use this phrase that I kind of agree with, which is that the key to everything is delivery system reform. I put it slightly differently, which is, What process will bring about continual and rapid productivity and quality improvement in the delivery of health care services?

The Quest for Quality. That’s really what we’re after. You have to get that; you have to have rapid productivity increases in how services are actually delivered to patients for us to be able to slow the pace of rising costs without harming quality. If you try to do something that slows the pace where you don’t get more efficiency and more productivity, the only way to really cut costs at that point is just to erode the quality—make people wait longer, pay people less so perhaps you get less effective interventions, etc. Really, the holy grail of what we’re after here is more bang for the buck, so to speak, from the health sector so that we can slow the pace in a way that still retains value and, in fact, improves value for the patients.

Now, then the question is, What’s the big hindrance to all this? There’s been a surprising amount of agreement that’s led the notion for a cap, but I think there’s been a really surprising amount of agreement around what the main problem is: Why don’t we have a more productive system? It comes back to Medicare—that Medicare fee-for-service is the dominant payer in most markets. (Not the largest single payer all the time in terms of segments—there can be a combination of employers that are sometimes larger—but as a single payer, Medicare is the largest in most places.) And the system can’t really function if those who provide the medical services don’t maximize Medicare revenue. So the whole system kind of gets built up around maximizing what Medicare pays through, mainly, its fee-for-service structure.

Two years ago, the insightful writer Atul Gawande wrote a very influential article for the New Yorker about McAllen, Texas, and how costly medical care is there. President Obama read it and gave it to all his staff. It was a great article with some very interesting insights into what’s going on in American health care. But the one thing the article never really mentioned was, Who paid for all this in McAllen, Texas? Who really facilitated the build up more than anybody else? It was the federal government—through the Medicare fee-for-service program.

So there’s been a surprising amount of agreement, although you don’t really hear it from the proponents of the law that passed, that to get at this productivity issue—how to make health care better and more effective, higher value for the dollars spent—you have to change Medicare. And the way you know that even proponents of the health law agree with that notion, is if you look at how they are going to fix the cost problem in American health care. If you go back to the law, the Patient Protection and Affordable Care Act—where are all the reforms that they tout as the solution to the cost problem? They are in Medicare. There are those things called accountable care organizations, bundled payments for what hospitals and physicians get for taking care of a certain diagnosis. There’s a push to bundle those together so that you get one payment instead of five or six. There’s a new center for Medicare and Medicaid innovation—$10 billion to test new things. And, of course, IPAB, which we’ve already discussed to some extent.

The Goal of Efficiency. The goal of all of this, really, is to try to make all health care in the U.S. more efficient by using the levers of Medicare. The Administration is going to try to engineer more effective health care throughout the U.S. by leveraging Medicare’s power over the marketplace. What’s the big hurdle to this? It’s really a political economy issue, and I call it the Lake Woebegone effect, which is that, to really drive a more efficient and higher productivity system-wide, you have to do some things that are a bit tough. One of the toughest things you’ve got to do is exclude the low-value providers from what might be called the preferred network. You’ve got to tell them, You know what, you cost too much and your quality indicators are not good enough, so you’re not giving us good value, therefore you’re not going to be a preferred part of this system anymore and we’re going to steer our patient population to people who are providing more bang for the buck.

The federal government has a very, very difficult time doing that. In fact, the reason why Medicare fee-for-service looks the way it does is because the political operation that oversees it really doesn’t want that to happen. It’s much easier for a politician to say to all the people that are providing services in a community, Hey, we’re going to treat all of you equally and just pay all of you less, than it is to go to one individual physician group or one individual hospital and say: You’re out. You’re no longer good enough. We’re going to steer all our Medicare people to the higher value hospital down the road.

What’s happened over time is that efforts that have been pushed—many by Gail, while she was overseeing Medicare directly, and then through MedPAC—to try to move Medicare along toward demanding higher value for the payments that are made and trying to measure that and build that into the system really have foundered. They don’t get very far. I use this one as an example that goes back a long time—I think Gail was probably involved with it—it’s called Centers for Excellence. This was an effort to designate certain facilities as essentially the higher end, the people who were actually delivering the best outcomes at the lowest cost. Medicare was going to tiptoe into the world of making distinctions and designating Centers of Excellence, and then, unsaid but thought, to eventually try to steer the patient population their way because they were delivering more for less.

But if you pick one, if you say, Hey, these guys are a Center of Excellence, suddenly the people that are not a Center of Excellence want to know, Hey, wait a second, why am I not on the list? Then they go to Congress and through the political process and claim the data’s bad—“you didn’t properly risk-adjust it, we have sicker people than the guy down the street”—and the whole thing becomes a quagmire of politicians competing, the regulators are a little bit uncertain, and it usually ends up foundering.

Hard Choices. When push comes to shove and you’ve really got to make some hard choices, the political process recoils from it. It is very difficult for politicians and regulators to make distinctions and what they end up doing is falling back on the tried and true mechanism, which is, instead of making distinctions on value, they just pay everybody less. That’s really the pattern. You see that quite a bit already in Medicaid, as Gail mentioned, and you see it more and more in Medicare, as Medicare’s chief actuary, Rick Foster, has very capably pointed out on numerous occasions over the last two years.

Now, the alternative vision is, then, back to enforcing a cap where we’re not relying on the political process to try to drive higher productivity, but relying on a consumer-driven process to do it. In most parts of the American economy we don’t need the government to drive productivity. Productivity happens because people have scarce resources and they shop and find the best value for the limited amount of money they have, and so that drives people supplying the services to do more for less. That’s the basics of getting more out of our economy every year.

Can we replicate that in American health care? That’s the key question. I’m not a person who thinks that the government doesn’t have a role in this. The government does have an important oversight role, as any of these plans would require—including Paul Ryan’s plan. The government would approve the plans that are offered, making sure they meet minimum standards. Secretary of Health and Human Services Kathleen Sebelius said the other day that if your voucher runs out, you’re going to die sooner—really inflammatory kind of language. That is ridiculous. The point of a voucher is that you’re going to be buying into plans that guarantee you out-of-pocket protection on your financial exposure. So every single person would have a real insurance product. The insurance plans would protect the beneficiaries’ out-of-pocket expenses. That would be the whole point of the premium support credits. So that kind of language from the Administration is really out of line.

My point is that the government would have an oversight role to make sure this is an operating and functioning market. But the crucial decision about which plans get business from the beneficiaries will be driven by consumer preference, and so automatically, the way to get more market share is by pleasing the customer. That, I think, is something that is not as prevalent in American health care as it really needs to be.

Part D Experience. We have a partial prototype for this that is working very well. It’s maligned for other reasons, but the basic issue in Medicare Part D, the new drug benefit that started in 2006, is that it’s structured this way. It has a defined-contribution system, basically, where the beneficiaries are entitled to a fixed amount and they can use it to apply to a number of competing plans. When Medicare Part D was enacted, there were predictions that it wouldn’t work, insurers wouldn’t participate, the beneficiary population wouldn’t like it because it was too confusing and there were too many choices, and that costs would soar because the government wasn’t holding prices down—all dead wrong.

Every one of those assumptions was false. Massive participation by competing insurers; with the beneficiary population, like with any other market, you have 10 percent of the people who are market leaders and who figure out where the real value is, and the other 90 percent of the people follow. That drives value. You have tons of consumer surveys showing they like the program; costs have come in 40 percent below projections; and what’s really going on here is that these plans are driving seniors to take generic substitutions in a way that people never could have done if they tried to do it through a law. People are signing up for plans that offer generic drugs that are identical to branded drugs. If you take a generic, it will cost you almost nothing. If you pay for the branded drug, it will cost you a lot more. Generic substitution in this population has gone up astronomically in a five-year period, so people are saving tons and tons of money as a consequence.

So, markets do work; they can work in health care, and the alternative is not more efficiency through a cap that’s enforced by the IPAB, but price controls that will have a deteriorating effect on quality.

Questions & Answers

Question: You talked about the problem of making choices between physician groups or provider groups. The insurance industry has been doing that for a long time. They let people be preferred providers or they don’t. Sometimes they do it on the basis of individual services, but sometimes they do it on the basis of just yearly performance. Now, if they can do it, isn’t that a model that the government can use, or could the government get the insurance industry to perform that function?

GAIL WILENSKY: Many major insurers have become much more aggressive over the last five or six years when it comes to designating preferred institutions and physicians, whether by individually identifying high-quality, low-cost providers, or by establishing networks of preferred providers. Initially, the designations focused almost exclusively on price, but as measurements have improved and have been subjected to rigorous evaluations, insurers have included both quality and efficiency measures. They have also been doing more to encourage patients to shift to preferred providers by offering lower deductibles or co-payments and occasionally providing lower premiums for healthier lifestyles.

The problem for Medicare is the one that Jim mentioned: Medicare has had enormous difficulty whenever it has tried to differentially reimburse among duly licensed providers. The political process does not easily tolerate any differential that is not objectively defined. Take, for example, DRGs—diagnosis-related groups—that pay for a discharge from a hospital. Once the adjustments are made to the base payment for a particular DRG to include wages paid in the area, whether the institution is an academic medical institution, whether it’s rural (which is sometimes gamed in order to get a higher payment), all hospitals receive the same payment for that DRG.

Why is that? It is because the political process has been unable to sustain a differential payment based on quality or other differences. Competitive bidding was supposed to set the price for durable medical equipment (DME), the most widget-like” part of health care, as of 2009 and some limited competitive bidding had begun to be embraced by Medicare in this area before 2009. But the year it was supposed to be extended to all DME, Congress did exactly what Jim said it does—rather than exclude some suppliers and get a best price by increasing the volumes paid to others, it just reduced the reimbursement amount across the board. I’m a trustee for the United Mineworkers Health and Retirement Fund and it has used competitive bidding for DME for many years as a way of achieving lower prices for its retirees. Other payers do similarly, but not Medicare.

When I was heading Medicare, we ran a demonstration project that mimicked something that the private sector had been doing for the previous 10 years—pay a single price to cover all of the physician and hospital expenses involved with a coronary artery bypass graft (CABG) or a valve replacement, which are categorized in adjacent-numbered DRGs. At the time, these two heart procedures represented the highest-cost, high-volume procedures that Medicare paid for. Although this had never been tried in Medicare, some employers had been using a similar type of payment for their employees throughout the 1980s. This allowed employees to be treated at such nationally renowned places as the Texas Heart Institute, the Mayo Clinic, and the Cleveland Clinic—and at a sufficient discount that the patient could have a family member or other caregiver fly out with him and still save the employers money over having it done locally.

The demonstration was initially aimed at accepting three hospitals with their associated physicians, but so many good proposals came in and so many hospitals wanted to participate, that after the first few years, the number was expanded to five, and then, ultimately, to 10 sites. At the end of the demonstration project, an evaluation was completed that indicated lower spending for these hospitals, higher patient satisfaction, and, as best as could be measured during that period, comparable or higher quality. Unfortunately, what happened after that is what has frequently happened with demos in Medicare—nothing happened. Payments to individual physicians, separate from payments to hospitals, continued as the payment norm.

So the question that Jim has posed is key: How do you best encourage different behavior, assuming that’s your objective? Is it by having the government impose specific changes? Or are there enough (or could there be enough) interesting innovations going on that could make a difference if new fiscal incentives started to drive the adoption of the innovations? The danger of assuming that government-imposed or -encouraged innovations would be sufficient may be more obvious now given the strongly negative reaction that has been produced by the proposed rules for accountable care organizations (ACOs). ACOs allow physicians and hospitals that are not formally integrated to share the (auditable) savings that result from working together. They have been widely promoted as a major strategy to move away from the negative effects associated with current fee-for-service Medicare and many hospitals and physicians have been preparing to form ACOs, in anticipation of the rules that were supposed to be released by CMS last January. The release of the proposed rules in late March has caused many of these same groups to step back and reconsider the desirability of this strategy.

The proposed rules were clearly written to safeguard the government from potentially losing money. Among the most controversial provisions were those that set the minimum savings between 2 percent and 4 percent, despite a large physician practice demo that indicated savings at this level could not be achieved by half the groups that had worked together for many years, and also requiring all groups to go “at risk” for both losses as well as gains by year three of the demonstration, even though the groups would not have received information on the patients assigned to their ACO before the start of year three. The Administration has been taking steps to make the proposed rule more attractive, but, again, it goes back to the fundamental question of the best strategy to get more innovative delivery systems in use in Medicare. To me, the inclusion of a variety of private plans that have government oversight competing with a government option but competing using the same rules, regulations and subsidies provides the most promise.

Q: Right now, what’s happening to us as seniors, it takes forever to get an appointment to see a specialist that you know is supposed to be the best in the state, for instance. I have lived in Maryland most of my life and had used the Annapolis clinic for surgery, and it’s a hard drive to go back and forth, particularly when you need therapy and all that stuff afterwards for a couple weeks, so I was trying to find something comparable in Delaware that would handle all the needs as the Annapolis Sports Clinic did for me over the years. But they’re cutting that out. They’ll do my knee and do it under Medicare, but they won’t do it if I have a back problem. The compensation is not there. In fact, he said what Medicare would pay would hardly cover the cost of the people that were in the operating room to do the surgery, much less any profit for them whatsoever, and they were not a charitable organization; they had a lot of money invested in this clinic.

So how do you see this; when we are just not going to have these doctors that are going to perform it? Then what is Medicare going to do?

GAIL WILENSKY: What’s happened over the last decade is the worst of all combinations. Medicare has kept the fees per unit of service almost completely flat, but of course, costs have not stayed flat over the same 10-year period. You might think that’s not really fair but at least Medicare must be saving a lot of money—but you would be wrong. Unfortunately, Medicare isn’t saving a lot of money because the volume and mix of services being provided has grown substantially, and grown in such a way that Medicare spending on physician services has grown at rates as high as 10 percent to 12 percent per year during this same period, which is clearly unsustainable.

The problem is that physicians bill Medicare using approximately 8,000 different codes, which makes it effectively impossible to reward those that provide good quality care efficiently. We know from other data that the U.S. makes use of very aggressive treatment compared to other parts of the developed world, which is in part driven by the inclination of the American public to “do something” even when it isn’t always clearly that doing something will provide a better clinical outcome. But the aggressive behavior is also driven by the way physicians are reimbursed as well as concerns about potential liability challenges, since physicians rarely get sued for being too aggressive but may get sued for being too conservative, especially if there is an adverse outcome.

This tension is likely to get worse because of all the pressure to lower reimbursements under Medicare. The hope is that physicians and hospitals change how they organize and deliver services and if they are reimbursed differently, there will be less focus on reducing the reimbursements of individual units of service and more emphasis on rewarding good clinical outcomes that are provided efficiently. But there is no indication that this type of change is happening very quickly. The Center for Medicare and Medicaid Innovation (CMI) is supposed to encourage pilot programs that will foster this type of change, but there is increasingly a sense of dismay that not much seems to be happening with the center.

I think what you are reporting is just the beginning of potential access problems under Medicare and seniors aren’t going to tolerate it.

Q: Is something going to happen that these doctors are going to be forced to do this?

GAIL WILENSKY: Economically, many physicians may be forced to continue accepting Medicare. Many doctors say they won’t accept Medicare if the rates don’t improve, but depending on their specialty and the size of the non-Medicare population and their insurance coverage, physicians may or may not realistically have the option of refusing Medicare. The question is not only whether services will be available, but whether the particular physician you want to see will be accepting Medicare. It is more likely that there will be appropriately licensed providers around to provide services, but not necessarily the physicians you prefer to see. That’s a different kind of access issue.

Q: It seems to me that those who were supporting the Ryan premium-support avenue have a problem of marketing the idea to the public, so that there is a base of public confidence in premium support. As I understand it, the new scheme that government employees have for health care insurance is a premium-support-type of operation. Is it a useful idea to begin to bring out some of those government employees who are using this premium-support-type idea and begin to make their size-up known to the public, to build confidence among the public at large that this is an acceptable and tested approach?

GAIL WILENSKY: That’s how I frequently describe the Ryan plan—it’s like the federal employees model (FEHB). The subsidy would be set differently, but, like the FEHB, it offers a variety of private plans whose benefits have been approved at premium rates that make sense given the benefits offered, a subsidy that doesn’t increase if the person chooses a more expensive plan, etc. During past health care reform debates, people frequently said “I want what they have,” meaning the health plan their Congressmen have which should give people some comfort. FEHB has been around for about 50 years and covers about 9 million people. Again, the analogy isn’t perfect but I think it’s a close enough analogy to be useful.

JIM CAPRETTA: Bob Moffit is a true expert on FEHB and has made this point many, many times over the last 20 years that this should be the model and the way to describe how we go about thinking about this. So I think you’re exactly right. There are some small differences between how the Ryan plan and how FEHB works, but those could be explained and maybe modified or thought through.

Q: I’m talking about the Washington power structure. People will get lost in the technicalities, but who will organize to market that model to Americans across the board so there is a surge in confidence? It’s a marketing of the idea.

GAIL WILENSKY: A major problem is that many in Washington—both in the Congress and in the media—seem to think that only price controls on providers will work to slow spending—an idea that is reinforced by the way the Congressional Budget Office traditionally does its estimating. CBO is much more willing to assume that regulatory changes are implemented as legislated, and very hesitant to include the effects of behavior changes in its estimates. Jim Capretta had a nice column—“The $6,400 Question: The Ongoing Delusion of the Price-Control Solution” in National Review Online on May 19, 2011—explaining that the $6,400 more the President has said it will cost seniors is an artifact of a CBO scoring assumption where all current law is assumed to be implemented as legislated with no disruptive side effects—including, by the way, the 30 percent reduction in fees that is slated for physicians, as well as all the Medicare reimbursement reductions built into the Patient Protection and Affordable Care Act—and that all the shortfall in costs gets shifted to the private sector.

The question is whether people who are both independent and credible will begin to question positions that depend on the implementation of certain portions of current law that are highly unlikely to occur—such as 30 percent reductions to payments to physicians or Medicare reimbursements that fall below Medicaid reimbursements—as the CMS actuary has predicted will eventually happen to Medicare under the ACA.

Q: I have a basic question: Why are we so obsessed with cutting health care costs? After all, we don’t need people to do agriculture anymore; we don’t need people to manufacture things anymore because we’ve made it very efficient with automation; maybe health care is the real growth industry that will create the jobs of the future. That seems to say, if health care is really a place that could employ a lot of people, the question is who is going to pay for it? Right now health care is paid for by business employers and government, and business employers’ main objective, they’ve got to cut costs. The business employer wants to cut costs and doesn’t care much about the rest of it. Taxpayers don’t want to keep paying more tax money; they’ve gotten taxed too much. Doesn’t that say that if health care could be a growth industry, the real question is how do we help the private sector to pay for it? How do we help consumers to pay for it?

GAIL WILENSKY: You raise some good points but it’s a little more complicated than how to help the private sector pay for health care. You are correct that unemployment would be even more dire than it is if the health care sector wasn’t continuing to create jobs. Health care has been one of the few sectors that have been creating jobs since the economic downturn in 2008. Health care spending has also been growing, although it has been growing more slowly these last few years than it had been earlier in the decade. Some people have lost their insurance because they’ve lost their jobs and are therefore using less health care or, even if they have insurance, conserving their funds and not spending more money than necessary because of concerns about the economy. But even with slower spending growth, health care remains one of the few bright spots in terms of job generation.

The problem is that it is unclear whether the amount that is spent on health care is the right amount to spend, reflecting the value that the people who are using health care attach to health care. Normally this isn’t a question that we need to ask, but in health care, because of the pervasiveness of third party payment, it is an important question. Most of health care spending occurs with people spending other people’s money, or at least that is how it is perceived by consumers.

Spending under these circumstances is not necessarily a reflection of the value people attach to its use. As of 2011, spending funded by the government accounts to close to half of all health care spending; employer-sponsored insurance accounts for a lot of the rest. Although employer contributions to premiums consist of the employees’ own money, which could otherwise be provided as wages, many employees don’t look at employer contributions in this way. In addition, once someone has insurance, whoever sponsors it, the cost of health care at the time of its use is much less (or zero) than the cost of the care itself, which distorts its use.

The concentrated nature of health care spending compounds the challenges raised by the pervasiveness of third party payment. Most health care spending goes to a small segment of the population—about 5 percent of the highest users account for 50 percent of the spending, and the top 20 percent account for as much as 80 percent of the spending. Conversely, the bottom 50 percent of the spenders account for only about 3 percent of health care spending. This type of concentration is not only true in the U.S. but everywhere it has been studied, and reflects the nature of spending on health care rather than the institutions that provide health care. This means that the people who are responsible for most of the spending will be using third party payment no matter what kind of insurance they have—assuming they have any insurance at all. High-deductible plans, major medical, MSAs (medical savings accounts) or any other type of plan that moves away from first dollar coverage is not very relevant for the “high spenders” in health care. The potential for being one of these very high spenders is why having health insurance is desirable, but it complicates getting the spending level “right” because it means people aren’t spending their own money. It’s why incentives for clinicians and institutional providers to provide high-value care are important, even with insurance that focuses on the kind of care that insurance was originally intended to cover—care for catastrophic events (which have a low probability of occurring and whose costs are high).

So while various policy changes can improve the link between how individuals value health care and how much they spend—such as changing the tax treatment of employer-sponsored insurance and using defined-contribution models of plan offerings—the concentrated nature of health care spending will complicate achieving better value for those who are in the highest spending categories, which is where a disproportionate amount of the spending on health care occurs.

JIM CAPRETTA: I think that’s all perfectly clear. Just to your point, though, I think health care is a preferred good. When countries enjoy greater wealth, they spend more on health care, and that’s a good thing. That’s true also all over the world. One of the big differences: Why is the us spending more on a per capita basis than other countries? Well, one of the biggest reasons is that our per capita GDP is higher; we devote a higher proportion of the extra on health care than other countries. That’s a big reason why we spend more. That’s a good thing.

I think the question is whether or not we’re getting value. Having this business of third party insurance really distorts that to a great extent. People say, Well, we’re spending a lot, but on some parts of it we’re not getting great value, so there’s a lot of interest in making that more efficient.

The last thing, separate from the things Gail raised: There is an issue of equity in health care. If you have a system that is totally driven by price—which I think lets us allocate resources better and that’s the way we need to go—you then have an issue of, What about people on the low end who can’t pay the premium price for the best health care? You then need a system that brings them up, which will almost inevitably be tax-supported. So you have a tax component of health care, which almost inevitably tries to bring people who can’t afford the premium prices into the same system, or as near as we can to the same system as everybody else. When that happens, tax dollars will be involved and then you do have a question of how much we can afford. It is a complicated question, but I think your basic point, that we shouldn’t worry too much about a wealthy country wanting better health, is right.

Q: You’re making the point that competition among private plans for that Medicare population will drive efficiencies and help keep costs down. With employer-provided health insurance, whether it’s the FEHBP or the larger employer-provided health insurance, there is competition among health plans for that business, but they still also struggle with rising health care costs. What makes this different, or why is this going to be the reason why we get these costs under control now?

GAIL WILENSKY: There are some important differences between the competition that exists among private plans and what we are proposing for Medicare, but whether we will be able to slow spending as much as we would like is an empirical question.

For most employees, their employer negotiates on their behalf and makes available two or three health plans, approximating what the employer thinks the average employee would like to purchase. This model is not true for federal employees who have the FEHB as one of their benefits, which contains many health insurance choices, or for those who work for the state of California and who have CalPERS (California Public Employees’ Retirement System). While 9 million people are included in the FEHB model, it is not large enough to dominate the market for insurance, and the plans offered reflect the dynamics that are going on in the country as a whole.

The closest model that we can consider is Part D of Medicare where seniors have a choice of a variety of private drug plans which they can choose from during each open enrollment. The irony is that the House-passed version of the Medicare Modernization Act that created Part D contained a pilot study that would have allowed direct competition between Medicare and private plans in any area where at least 25 percent of the Medicare population was enrolled in private plans instead of traditional Medicare. Unfortunately, the pilot was dropped during the reconciliation process.

The results from Part D, however, are an interesting reflection of how competition among plans can affect spending, especially when seniors are financially at risk for the premium costs. Plans have to cover at least one drug in each specified drug class, but what else is covered, and at what coinsurance and premium charges, can vary by plan. Although initially there was concern whether private drug plans, which hadn’t previously existed, would form in order to participate, the opposite concern—that there are too many plans in many metropolitan areas—has now been raised. The result has been spending on Part D has run more than 25 percent below the CBO projection and almost 40 percent below the CMS actuary’s projection.

The experience with Part D is very different from what we have seen in other parts of Medicare where attempts to control spending are limited to controlling reimbursements to providers. Not surprisingly, or at least not surprising to me, Medicare providers have traditionally (and successfully) engaged in a variety of strategies to increase their revenues—by changing the volume and mix of services provided or by directly appealing through the political process to increase the rate of reimbursement received.

Q: In the more immediate, you had mentioned a 30 percent cut for physicians under the Medicare sustainable growth rate (SGR) that’s to go into effect January 1, and the House is holding hearings on that. Do you have any thoughts as to a solution for the SGR problem?

GAIL WILENSKY: Removing the SGR is not possible until a different way to reimburse physicians is in place. We know what will happen otherwise because of the experiences from the 1980s. During the 1980s, physicians billed Medicare using a detailed fee schedule with no spending limit, and spending on Part B of Medicare grew faster than the rest of Medicare. It’s the disaggregated billing schedule that required the spending limit, and its removal requires a different reimbursement system.

Reimbursement for physicians must move away from the present reimbursement for 8,000 billing codes to one where reimbursement is for bigger “bundles” of services. This is what has happened elsewhere in Medicare where, for example, reimbursements to hospitals are determined by the patient’s diagnosis at discharge rather than having the hospital bill for each day the patient spends in the hospital or for the services received while in the hospital. Surgery has traditionally used a bundled payment where the amount reimbursed for the surgery typically covers a limited amount of pre- and post-operative care that is provided by the physician. Determining the appropriate “bundles of care” for physician services represents a major undertaking, but one that needs to occur as soon as possible if Medicare is to move away from the current conundrum represented by the SGR and RBRVS (Resource-Based Relative Value Scale).

The Administration has bought itself another two years with the recent budget, but it is urgent that work begins now or the country will continue to find itself in the same morass that it has been for most of the past decade.


Robert E. Moffit
Robert Moffit

Senior Research Fellow, Center for Health and Welfare Policy

Gail Wilensky, Ph.D.


James Capretta