Wednesday, April 29, 2009


I’ve been reading some of the testimony on delivery system reforms from the House Ways and Means Committee meeting earlier this month, in particular the lengthy statements from MedPAC Chairman Glenn Hackbarth and Urban Institute Senior Fellow Dr Robert Berenson.

Hackbarth and Berenson are each distinguished health care figures, and their remarks are worth careful study. Together, they paint an all too familiar gloomy picture of a system whose costs are out of control, in which quality is often poor, and where there is little correlation between expenditures and outcomes. Few would disagree with the causes that they identify: payment structures that reward volume, lack of coordination among providers, an overemphasis on specialty care, and a system that seems more often driven by supply than demand.

The two sets of testimony include several very important recommendations, like more emphasis on public health, dissemination of comparative effectiveness information, and higher payments for primary care (although several years will elapse before this makes a real impact on physician career choices).

Other testimony proposals, however, especially those focused on Medicare, carry the risk of distracting us from more important changes. Chronic care coordination (including the medical home model) has not yet convincingly been demonstrated to cut costs. Accountable care organizations (this year’s buzz-phrase) require more willingness to cooperate than many providers have so far shown. Bundled hospitalization payments make good sense but require the same kind of willingness to cooperate. Tying payments to quality introduces questions of data interpretation and validity of guidelines.

Aside from their uncertainty of success, these proposals share two problems. Each offers potentially inadequate incentives for changing entrenched provider behavior, and each represents yet another government attempt to make an inherently inefficient and ineffective system work just a little better.

Rather than trying to breathe life into a system that might be better left to wither and die, perhaps we should focus on the potential of market Darwinism—in which the most cost-effective survive, and the less fit fall.

In the context of Medicare, Darwinism means redesigning Medicare Advantage so that private plans compete with the FFS program on the basis of price alone—without subsidies. While Hackbarth’s testimony noted the need for financial neutrality between MA and FFS, he stopped short of the recommendation that would make survival of the fittest a reality: allowing beneficiaries to offset any MA savings against their Part B premiums, and requiring that they assume the full additional burden when MA premiums exceed FFS costs.

In the context of national health care reform, Darwinism means setting a standard benefit package that insurers must offer through an insurance exchange. Insurers should be allowed to offer separately priced supplemental coverage, but again—except for low-income individuals—there should be no counter-productive government subsidies that disguise differences in premium cost for the basic benefit.

What might be the results? Assuming a functioning risk adjustment mechanism to discourage “cherry picking,” the least cost-effective insurers should rapidly disappear, the better HMOs should thrive, while the best managed other insurers should apply the kind of cost pressures and pricing innovations that Hackbarth and Berenson have presented (and should rapidly abandon them if they are ineffective).

This doesn’t mean that Medicare FFS demonstration projects aren’t worthwhile. It would be wonderful if the costs of traditional Medicare could be controlled. However, we should be focusing more on the kind of health care system that will work long-term, rather than expending too much of our energies on the application of band-aids and baling wire to something that is inherently faulty.

Monday, April 27, 2009


Wall Street Journal editorial writers and other folk with touching faith in classic economic theory wonder from time to time why competition doesn’t work better in our health care system. (Actually, the WSJ people are sure that it could, if it were not for government bureaucrats and their spendthrift liberal friends).

It does seem as if Adam Smith’s “invisible hand” is affected by a strange palsy as it nears the realm of health care. But why, given the legions of insurers all apparently eager to edge each other out in the race for our dollars?

Theoretically, employer-sponsored insurance—more than 90 percent of non-government coverage—provides two competition opportunities: when an employer selects plans to be offered, and when employees choose from these plans.

Unfortunately for the reputation of the invisible hand, employers picking insurance options don’t necessarily choose the best value plans. Cost issues aside, they are influenced by the need to avoid the disruption and employee unhappiness that could result from changing plans. Employees won’t want to travel further to a provider, or switch specialists in the middle of treatment, or be forced to leave a well-liked family physician. Large employers may try to sidestep employee concerns by offering a choice of plan types (like a PPO, an HMO, and a HDHP/HSA), but rarely competing plans within a type; smaller employers may be unable to provide any choices at all, unless provided by the same carrier. The result: incumbency may allow insurers to bid higher premiums, even though experience should mean lower risk, while employers may prefer to shift costs to employees rather than switch carriers.

Not only are employers likely to lean towards an incumbent, they may find competing proposals impossible to compare. Large employers have the clout to demand that insurers bid against the same benefit specifications, but others will struggle to compare whatever off-the-shelf packages insurers believe fit their needs. Small groups—even those using a broker—may not have that luxury; insurers will offer take-it-or-leave-it choices of their standard coverage. And, as in other business arenas, volume counts for a lot. Smaller employers face a double cost whammy: their business isn’t important enough to attract aggressive pricing, and individual risks can’t be pooled among thousands of other employees.

The invisible hand doesn’t shake off its palsy at the employee level, either. Although most workers will have some choice, it will be limited, and “best value” may be impossible to determine. Even more than at the employer level, employees—assuming the employer is paying most of the premium—will pick their coverage to maintain existing provider relationships; only those without such relationships or who must pay a substantial part of the premium themselves are likely to put cost first. If it’s available, HDHP/HSA coverage may appeal to the youngest and healthiest, but—alas again for classic economics—at the expense of increasing premiums for other options.

Current government programs do no better than employers. Medicare Advantage, far from allowing health plans to go head-to-head with traditional Medicare, subsidizes plans by several percent while encouraging add-on benefits that make comparison impossible. Medicare’s drug program is so popular (read profitable) with insurers that in urban areas there can be fifty or more plan choices, with different combinations of formulary, pharmacy network, deductible, and monthly premium, proof that more competitors doesn’t mean more effective competition. (Pity the eighty-year-old who must navigate this muddle to pick the best value plan.)

Outside of Medicare, state Medicaid programs have tried to create competition among contracted health plans, but such attempts are undermined by inability to pass cost differences on to patients. FEHBP—suggested as a model for a national insurance exchange—does rather better, offering a shorter list of choices, but unfortunately one in which there is no actuarial equivalence to help determine best value.

So what might make health insurance more price competitive? Reform proponents’ suggestions to help the invisible hand include:

1. Move the primary responsibility for insurance choice to individuals (for example, via an insurance exchange) and require that—except for low-income persons—they bear some meaningful part of the premium cost.

Not everyone in this scenario will make a best value choice, but it should result in greater price sensitivity and greater awareness of the implications of personal health decisions. However, while this would lower costs for small groups and individuals, premiums for large employers (or their employees, depending on the financing structure) would increase if they were forced to switch to the exchange.

2. Establish a standard set of benefits, so that price differences between insurers are apparent.

This doesn’t have to mean “one-size-fits-all;” supplementary benefits (perhaps including alternate deductible levels) could be priced separately, as they are in other nations with guaranteed coverage. The standard benefits could be adjusted periodically to reflect insurance costs and funding availability, although this would shift consumer dissatisfaction from insurers to the government entity administering the exchange function.

3. Guarantee portability, without pre-existing condition limitations.

Essential to universal coverage, this would eliminate the medical shackles that tie workers to their employers (and their insurers) solely to maintain coverage, but would increase insurers’ risk.

4. Provide a risk-adjustment mechanism for insurers participating in the exchange.

Any major change to the insurance system will increase actuarial risk. Insurers participating for the first time in an exchange may have little experience to help set premiums, and will build in a risk factor—unless they receive some protection, either though a risk-adjustment formula (as in the Dutch system) or through reinsurance (a simpler mechanism).

The difficulties facing employers and employees are not the only failures of price competition in our current system. Elsewhere in the health care jungle, insurers face their own competition issues, which will be discussed in Part 2 later this week.

(An earlier version of this piece appeared in The Health Care Blog.)


Today’s Politico reports that Conservatives for Patient Rights is launching a million dollar television advertising campaign focusing on the pitfalls of government-run health care systems.

The ads are reported to feature British and Canadian physicians talking about the problems that patients experience due to their nations’ systems.

Conservatives for Patient Rights is headed by former Columbia/HCA CEO Rick Scott, who featured prominently in the federal investigation of improper Medicare claims by Columbia that led to a $1.7 billion settlement. Scott presented his views on reform in a controversial March 2009 article in The Health Care Blog—one that generated scores of comments ranging from enthusiastic support to virulent opposition.

Saturday, April 25, 2009


Reconciliation. It’s an odd word for something that could precipitate a knock-down, drag-out fight in Congress, but the process that Senate Democrats agreed this week to adopt if health care reform legislation isn’t passed by October 15 was originally intended to reconcile differences among House and Senate budget bills. What the process does is to replace the usual Senate requirement of a three-fifths majority—needed to end a filibuster, but also consistent with Senate traditions of compromise—by a simple majority.

So, with the Democrats having decided on an aggressive approach (Republican Senator Michael Enzi has called it “like a declaration of war”), what are the implications for the reform legislative process (beyond making Congressional Republicans mad)?

First, is October 15 an absolute drop dead date?

The answer is, not quite. Not only does the reconciliation process provide for up to twenty hours of debate (which could move the deadline out by just two or three days), but Senate Democratic leaders might prefer to continue negotiations on a reform bill if they felt they were close to the magic sixty votes. This would require the vote of at least one Republican, as well as the only Independent (Joe Liebermann), but would allow Democrats to claim bi-partisan support—even if only a little.

Second, will the threat of reconciliation result in the refusal of Republicans to compromise on reform language?

This is the big gamble that the White House and Senate Democrats are taking. No one likes to negotiate with a threat hanging over their head, and most Senate Republicans are already strongly opposed to much that is being proposed from the other side of the aisle. The risk is that Republicans who might be willing to support a reform bill in some form—for example, the handful who have signed on to the Wyden-Bennett bill—could be so alienated that they too will be unwilling to compromise.

On the other hand—and presumably this is Democrats’ hope—at least some Republican senators may prefer to negotiate in order to try to protect the interests of their business, insurance, and medical backers. Grudging support of a reform bill that they have made more tolerable for their constituents—and that will get fifty-one votes in any case—may seem like a rational strategy.

Third, will the reconciliation process do what Democrats hope it will?

Maybe, but there’s more to the process than just changing the majority vote rules. A key part of the reconciliation structure is the so-called “Byrd rule,” named for long-time (fifty years) Senator Robert Byrd. The Byrd rule, part of the Budget Act, provides a point of order in the Senate against extraneous matter in reconciliation bills. Determining what is extraneous can be a procedural quagmire requiring resolution by the Senate Parliamentarian.

To make matters even more tricky, the Byrd Rule and other Budget Act rules, including a prohibition on provisions that would cause the deficit to increase, may only be waived by a three-fifths vote—potentially bringing the issue full circle. Thus, the result of stretching the reconciliation process too far beyond its original intent can be passage of the legislation shielded against filibustering—but with the original bill emasculated by the deletion of the “extraneous” material.


The New York Times and New Republic each report that Democratic leaders—albeit in some cases unwillingly—and the White House have agreed on a “reconciliation instruction” that would potentially protect health care reform legislation from the Senate filibuster process.

Under the agreement reached after a multi-hour negotiating session in Majority Leader Harry Reid’s office, the Congress will have until October 15 to pass reform legislation under the usual rules. After October 15, the Senate will be able to pass a reform bill by a simple majority, rather than the requirement of a three-fifths vote.

The agreement was apparently reached only after considerable pressure from the White House. Senate Budget Committee chair Kent Conrad was quoted in the New York Times as stating “Virtually everyone who has been part of these discussions recognizes that reconciliation is not the preferred way to write this legislation, but the Administration wants to have a reconciliation instruction as an insurance policy.” Senate Finance Committee chair Max Baucus also was quoted as preferring not to pursue health care reform legislation through the reconciliation process, and hoping instead to produce a bill that would “get significantly more than sixty votes.”

Friday, April 24, 2009


Senate Finance Committee chair Max Baucus offered some intriguing clues in his breakfast meeting today (see previous post) about the direction that reform legislation might take.

Baucus’ comments, mostly relating to the roles of insurers, seemed carefully calibrated to gain support from—or at least neutralize opposition from—business and the insurance industry. In general, he avoided discussion of provider payment and system costs, financing, and individual or employer mandates. However, he did point to his November 2008 White Paper as a foundation for his proposals.

Baucus made three key points: coverage should be available to individuals through an insurance exchange but a public plan option might not be essential (but he reserved the right to revisit the issue); the present structure for self-insured employers should be preserved; and the insurance market should be reformed to eliminate pre-existing condition limitations and to guarantee the right to coverage. Each of these points has important implications.

Proposals for a public plan option have aroused fierce resistance from insurers and have been a focus of Republican attacks on Democrats’ reform efforts. Insurers perceive a public plan, particularly one based on Medicare (as suggested in Baucus’ 2008 White Paper) as a double threat. First, it would create government-sponsored competition in a government-regulated market. Second, it would further increase cost-shifting as individuals moved from private insurance to government coverage. Together, these factors could have a huge and destructive impact on the insurance industry. (In its February 2009 report on possible costs and savings due to reform, the Commonwealth Fund estimated that a Medicare-based government program would drive provider payments down by thirty percent and attract up to two-thirds of the individual and group markets.)

Not having a public plan has its own implications. While the insurance exchange concept favored by Baucus would presumably put price pressure on competing insurers, the pressure would likely be less without a public plan than with one—at least in any region in which there were not efficient HMOs or PPOs, the two private structures that under Medicare Advantage have costs closest to the Medicare FFS program. (With less than one percent of Massachusetts residents covered by its Connector, there is probably insufficient current evidence to determine the price effects of insurance exchange competition.) On the other hand, avoiding undermining the traditional insurance industry, for all its faults, would reduce the risk involved in transition to a reform structure.

Preserving the present structure for self-insuring employers would also minimize the transition risks of reform. As presented at the breakfast meeting, the intent seems to be to insulate large employers—those most likely to self-insure—from the potential impacts of reform. Self-insurance is regulated by the terms of the Employee Income Retirement Security Act (ERISA), which presumably would remain unchanged. One feature of ERISA is that it reserves to the Federal government the regulation of self-insured plans, something that has been a concern of health reformers at the state level, who feared that employers not wishing to comply with state reform requirements could decide to self-insure, in effect hiding behind ERISA. In Massachusetts, the only state to have fully implemented reform, this has not so far been a problem, perhaps because the Commonwealth’s business community has generally been supportive of reform. However, if Federal reform allows any employer to self-insure and so avoid other reform requirements, this could become an issue.

The final key point, insurance market reform, also involves risk, but here it is an actuarial concern. Guaranteeing issuance and eliminating pre-existing condition exclusions obviously increase insurance risk. Assuming that market reform will include provisions for minimizing “cherry-picking,” large insurers will have a considerable advantage over their smaller rivals, unless the insurance exchange system provides for either very extensive risk-adjustment or very generous reinsurance. All these options imply increased costs and/or endless argument with government regulators.


Senate Finance Committee chair Max Baucus offered several clues today about features he hopes to include in his committee’s version of a health care reform bill.

Backing off somewhat from earlier strong backing of a public plan option to be included among insurance exchange offerings, Baucus told reporters at a breakfast meeting hosted by Health Affairs that he wanted to focus first on preserving the insurance structure for larger employers while expanding private insurance and public programs like Medicaid and SCHIP.

Baucus did express considerable support for the insurance exchange structure, citing Massachusetts’ Connector as a model, but emphasized that the benefits to be offered would have to meet (or, presumably, exceed) national—rather than state—standards.

Baucus emphasized his desire to retain much of the present system, and especially to minimize the impact of reform on employers that currently have insurance. However, his remarks focused almost entirely on self-insured companies covered by ERISA. It was not clear if he intended to exclude only current ERISA employers from the potential requirements of reform, or would also exclude any employer wishing at any future time to self-insure.

Among other comments, Baucus noted his intent that reform legislation should provide for guaranteed issuance and elimination of pre-existing conditions. He also remarked that he had not abandoned the public plan concept and might return to it.

A much more detailed presentation of Baucus’ proposals is expected to be provided when he meets in closed-door session with other Finance Committee members on April 29.


No, it’s not one of those Chinese operas from the Chairman Mao years, but rather my reaction to a recent report from the prestigious Commonwealth Fund. “The Path to a High Performance US Health System,” and its accompanying technical documentation, forecast savings for a “comprehensive set of insurance, payment, and system reforms that could guarantee affordable coverage for all by 2012, improve health outcomes, and slow health spending growth by $3 trillion by 2020.

On a positive note, both the report and the technical documentation are well worth reading. The report assembles in a single “system” most of the proposals currently being talked about by HHS Secretary-designate Kathleen Sebelius and senior staff in the White House Office of Health Reform, while the technical documentation provides a comprehensive analysis of costs and savings that might result from these changes.

So, should we have confidence that the proposed “system” can get us close to universal coverage and make a $3 trillion dent in health care costs?

Unfortunately not. While the Commonwealth Fund report contains many sensible ideas, the conclusions are undermined by five major fallacies.

Fallacy Number One: Small businesses will accept a “play-or-pay” proposal that forces them to pay a minimum of seven percent of payroll for health care.

There are practical reasons why play-or-pay won’t be effective, but the biggest obstacle is political feasibility. While a seven-percent levy might seem modest to businesses that currently pay much more for coverage, it’s inconceivable that such a proposal in the middle of a recession would produce other than fierce opposition from NFIB and its allies. Unless health care reform is incorporated in a budget reconciliation bill—which would upend the Senate tradition of compromise—it will require sixty yea votes, something that small businesses can pretty much guarantee to prevent. (The Commonwealth Fund seems to have forgotten that business lobbyists helped defeat California’s reform bill that called for just a four percent levy.)

Fallacy Number Two: The insurance industry will allow the creation of a “public plan” to compete with their own offerings—a plan that the Commonwealth Fund estimates will drive provider payments down by as much as thirty percent compared to traditional FFS insurance, and attract up to two-thirds of the individual and group markets.

Given that for most insurers this is a bigger threat even than the 1993 Clinton bill (where at least insurers had the possibility of turning themselves into managed competition entities), the reality is that the public plan proposal is even less likely to succeed than play-or-pay—and certainly not to gain sixty Senate votes. The assumption that it would be the only FFS plan sold through the proposed insurance exchange is especially likely to leave AHIP leaders foaming at the mouth. Providers are unlikely to be too eager to go along with a proposal that slashes payment rates by thirty percent, either. So, as for play-or-pay, unless Senate Democrats gamble on utilizing the reconciliation process, the public plan—at least as assumed by the Commonwealth Fund and its consultants—is very unlikely to survive Congressional debate.

Fallacy Number Three: Government spending on IT of $120 billion over ten years will yield savings of almost $200 billion.

There are certainly strong arguments for electronic medical records, but the forecast savings are unlikely to be anything but illusory. Integrated health care systems like Kaiser may be able to achieve significant savings (hopefully, given the $4 billion that Kaiser has sunk into its own IT project), but the great majority of US providers have neither the same level of integration nor the same incentives. A more realistic view is found in last year’s Congressional Budget Office report on health care issues: “By itself, the adoption of more health IT offers many benefits, but it is generally not sufficient to produce substantial cost savings because the incentives for many providers to use that technology to control costs is not strong.” (By the way, did anyone in the White House think to ask their own Budget Director, Peter Orszag, who oversaw the preparation of the CBO report, before deciding to spend $19 billion on health care IT?)

Fallacy Number Four: Establishment of a “Center for Comparative Effectiveness and Health Care Decision-Making” will cut expenditures by more than $600 billion over the next decade.

While it’s hard to argue against something that seems so sensible (we’d all prefer our physicians to know what works best), the savings projection seems wildly optimistic. The $600 billion estimate assumes that more intrusive--but unfunded--public program claims processing procedures will dramatically change provider behavior. We know from the Dartmouth Atlas reports that there’s lots of room for improvement, but without the control over resources that the UK’s NICE enjoys, it’s hard to believe that high-cost providers will go along with slashing their incomes (see Fallacy Number Five). And as the CBO report notes: “it would probably take several years before new research on comparative effectiveness could reduce health spending substantially.”

Fallacy Number Five (perhaps the biggest fallacy of all): Providers and patients will behave the way the Commonwealth Fund (and most of the rest of us) would like them to.

Unfortunately, this piece of wishful thinking is at odds with the incentives in our current supply-driven health care system. Outside of entities like Geisinger, Kaiser, and the Mayo Clinic, improvements in provider efficiency are likely to cut incomes, not increase them. It’s no coincidence that areas with the greatest physician and hospital densities have the highest health care costs. In a health care version of Parkinson’s Law (“Work expands so as to fill the time available for its completion”), availability of resources—whether high-tech imaging equipment or physician time—means that the resources will be utilized in patient care. Unless we can change the incentives, or control the introduction or distribution of new resources, we will never solve the health care cost problem.

(An earlier version of this piece appeared in The Health Care Blog.)


Health Care REFORM UPDATE--with the goal of providing "one-stop shopping" for the latest health care reform news, opinions, and analysis--is off and running!

And so is the major Democratic legislative process in the US Senate.

Finance Committee chair Max Baucus and Health, Education, and Labor Committee chair Ted Kennedy this week sent a joint letter to President Obama promising to move reform legislation out of their respective committtees by early June. The letter noted the two key senators' intent to craft similar legislative language that could be merged into a single bill for debate by the full Senate.

A day after sending the letter to the President, Baucus chaired the first of three "roundtable discussions" on health care reform issues, declaring "What we hammer out at these roundtables is going to be the cornerstone of the legislation that we pass in coming months.” This first session focused on cost and quality issues and was attended by Finance Committee members and by thirteen health care industry representatives.

A parallel effort, again with the intent of producing compatible language to be combined into a single bill, is underway in the House, involving three committees (Ways and Means, Energy and Commerce, and Education and Labor).