Transition to Comprehensive Health Care Reform

by AliceRivlin on April 5, 2009

Digg This
Reddit This
Stumble Now!
Buzz This
Vote on DZone
Share on Facebook
Bookmark this on Delicious
Kick It on DotNetKicks.com
Shout it
Share on LinkedIn
Bookmark this on Technorati
Post on Twitter
Google Buzz (aka. Google Reader)

I. Introduction–what this paper is about

I hope this paper will spark serious, constructive dialogue about decisions that must be made to move the United States toward comprehensive health care reform.  That reform should have three objectives:  

1. It should set the country on a clear path to universal coverage, even if coverage is broadened incrementally.

Universal coverage means that everyone has a health plan that offers a standard benefit package at an affordable price. Coverage could be broadened in stages, but there should be a clear plan to get to universality within a few years. The contents of the standard benefit package will be controversial, but can be adjusted over time. What we need first is a process for deciding what should be in the initial benefit package, a mechanism for adjusting the package over time, a revenue source to pay for the necessary subsidies, and a structure for accomplishing the universal coverage over a defined period.

2. It should move the system toward more efficient and cost-effective care.

Meeting this objective will require aggressive efforts to deliver more health care value per dollar expended. It will require investment in health information technology and use of that technology to collect and analyze evidence on the cost and comparative effectiveness of alternative treatments. It will require use of that evidence to improve quality of care, and implementation of payment and pricing incentives to encourage cost effective decisions by payers, providers and consumers.

3. It should slow the currently projected growth in total health spending.

Broadening coverage will increase total spending, while reducing waste and moving to more efficient care will tend to reduce it. On balance, health spending will likely continue to grow faster than other spending. However, the recent and currently projected excess growth in health spending over other spending (both in the federal budget and in other public and private budgets) is unsustainable. Explicit mechanisms for controlling overall spending, such as global budgets or protocols for rationing expensive procedures, will be needed.  The new system should build in mechanisms for controlling over-all cost growth.

Coverage, efficiency, and cost growth must be addressed simultaneously–not separated or pursued sequentially. Covering the currently uninsured without moving toward greater effectiveness and slower cost growth will only add to spending in a costly and wasteful system.  Improving effectiveness and controlling the growth of spending will be more difficult as long as a large and growing fraction of the population remains uncovered. Hence, transition to an improved system must encompass all three elements and move toward them at the same time

Why 2009 is the right time to launch comprehensive health reform

The recession should be seen by policy-makers as a reason to accelerate comprehensive health reform, not delay it.  Health reform was high on the electorate’s agenda a year ago, but has been eclipsed by the rapidly deteriorating economy. Conventional wisdom now holds that economic recovery will prove so expensive that it must push all but minor health reforms off the agenda for several years.  However, in this case, conventional wisdom is wrong.  For both economic and political reasons 2009 is the optimum moment to launch comprehensive health reform–far more propitious than 1993-4 or some future year after economic recovery is underway. The Administration and Congress should start now, so they have a good chance of getting comprehensive reform enacted in the first term. (Passing the legislation would put a firm plan and schedule in place, but the elements would not all need to be implemented right away).

Fiscal conditions are actually favorable because health reform requires large up-front expenditures that will have long run payoff. We need to invest now in the skills and infrastructure (such as, health information technology and creation of new institutions to collect and analyze information on effectiveness of treatments and delivery mechanisms) that will allow us to make health delivery more efficient and effective in the future.  This infrastructure will cost money right away and (we have reason to hope) slow the growth of health spending down the road. Broadening coverage to the uninsured will also cost money, but is an important component of an economic recovery plan.  

In 1993-4, the Clinton Administration proposed comprehensive health reform at the same time it embarked on aggressive action to reduce the budget deficit immediately. The Clinton team needed to constrain near-term spending and could not afford significant upfront investment. They needed to construct a health reform proposal that was arguably budget-neutral.  This imperative led to downplaying health infrastructure and required complex mechanisms for cost control, which fed the perception that the plan was too complicated to be workable. It also led to less-than-credible estimates of the rapidity with which efficiency savings–needed to offset the increased cost of broader coverage–could be accomplished. By contrast, the Obama Administration faces a deteriorating economy for which appropriate fiscal policy involves increasing near-term deficits coupled with longer-run reductions in spending growth–just the ticket for comprehensive health reform if done right.  Precisely because of the probable severity of the recession, the current obstacle to reform is not scarcity of resources in the near-term; it is competition for the attention of the policy leadership, both executive and congressional.

The recession also strengthens the political case for comprehensive health reform. Massive job loss will worsen the plight of the uninsured in 2009, increase their numbers, and dramatize the need for stable, portable health coverage. Health reform that assures workers they will not lose their coverage if they lose their jobs will contribute to restoring consumer confidence and spending, especially among those who are not actually unemployed, but fear they might be. Moreover, escalating benefit costs will continue to squeeze both employers and employees and accelerate the erosion of employer-provided benefits. The drivers of higher health spending–an aging population and continued innovation in drugs, treatment and equipment–will not pause for recession. Absent policy intervention, there currently appears to be little prospect of offsetting savings, such as those associated with the widespread shift to managed care, which slowed the growth of health spending in the 1990s.  The plight of the automobile companies has also dramatized the cost of health care to businesses competing in world markets.  On balance, the recession seems likely to strengthen the political forces favorable to comprehensive health reform, much as the Depression strengthened the case for Social Security because it worsened the already serious plight of the elderly and underlined the need for retirement security. 

However, while the 2008 election was a mandate for change, it did not show that people were looking for major disruption of familiar patterns or ideological purism–on the contrary.  President Obama must be a pragmatic problem-solver who governs from the center, especially in an emotion-laden area like health care, and he must be careful not to add to the widespread anxiety.  During the campaign, he repeatedly stressed that people who were satisfied with their health insurance would not have to change it and that those without adequate coverage could buy into a system with familiar characteristics, such as the system that covers federal employees. The severity of the recession has made it even more necessary to accomplish health reform without plunging people into complex and unfamiliar territory.    

The challenge for the Obama Administration and the Congress is three-fold:

  • agree on a strategy for moving to universal coverage in the least disruptive way possible and for implementing the strategy on a defined schedule;  
  • simultaneously put in place infrastructure for improving efficiency and effectiveness of health delivery;
  • ensure that the evolving structure of health care delivery and financing is conducive to controlling the growth of health spending over time (this will be the hardest part!).

Difficult choices must be made to move forward on all three tracks.

II. Transition to what? 

The short-comings of the U.S. health system are well known. American health care is more expensive than that of other industrial countries (per capita and as a percent of GDP), but Americans are not healthier as a result. Some 47 million people do not have insurance coverage and must pay out of pocket or depend on public or private charity when they need care. They are often treated belatedly and expensively when earlier and cheaper interventions would have been more beneficial. The number of uninsured has been rising and the increase is likely to accelerate as the economy worsens. Research has shown wide ranges in quality of care, a distressing frequency of medical errors resulting in adverse consequences ranging from unnecessary deaths to added cost, and big disparities in health outcomes by race and income. Research has also revealed enormous differences in intensity of care among areas and providers without corresponding differences in outcomes. Clearly there is room for reducing wasteful treatment and moving toward cost-effective practices. The question is how to do it.

Moreover, the health spending increases of recent decades are simply not sustainable for long.  Health care spending has been rising about two and a half percent faster that the economy has been growing for the last 40 years.  At almost 17 percent of GDP, health spending is already squeezing out other spending federal, state, corporate, non-profit, family, and individual budgets and is on track to squeeze more.  If current trends do not change, the federal government will have to raise taxes and/or cut other spending drastically just to pay for Medicare and Medicaid. But cutting federal health benefits without increasing efficiency and reducing overall health spending growth will only shift the burden to state, local and private payers. Maintenance of the status quo into the future will doom Americans to spending an increasing share of their total income for an inefficient health system that covers fewer and fewer people.

In addition,  much of what needs to be done to make Americans healthier involves behavioral changes–in diet, exercise, smoking, and other substance abuse–that are not easily impacted by the health care system.  Fixing health care delivery by itself will not make Americans as healthy as they could be, but even the daunting task of restructuring health care may prove more tractable than changing unhealthy lifestyles.

Rejecting Extreme Solutions

In view of these short-comings it is tempting to say, “Let’s blow it up and start over!” Surely, we can design a system that provides better care to more people for less money. Advocates of a single-payer system favor scrapping our complex mixed public-private system and covering the whole population with uniform health insurance designed and financed by the federal government.  They point to single-payer systems in other countries, which, while not without disadvantages, cover all residents at lower cost than the US sustains and have no worse (often better) health outcomes. These systems are simpler than ours and have lower administrative expenses. While single-payer systems in other countries are experiencing rising costs as medical knowledge advances and their populations age, they may be more effective in restraining spending growth than our fragmented system.  By definition a single payer system has a global budget that can be controlled by government decision. Spending restraint takes political will, however, especially when the government is the payer, The big unanswered question for Americans is whether our political system would be able to control the costs of a single payer system. Medicare, after all, is a single payer system, which has not effectively restrained spending growth, in part because of political opposition of providers, suppliers, and insurance companies to making Medicare more cost effective.  Creation of a single payer that covered the whole population would not make those strong political forces disappear.

In any case, blowing up the current system in favor of a single payer approach does not appear politically realistic. The United States is a huge, diverse country whose political institutions resist revolutionary change. Much of the public is deeply suspicious of government, especially expansions of federal power. Most people have confidence in the quality of care provided by American doctors, hospitals and other providers and are reluctant to disrupt a system they perceive works well for them. The current patchwork of public and private health insurance covers a large fraction of the population reasonably well and has strongly-entrenched stakeholders with substantial political power. It is hard to imagine the circumstances in which the public would countenance scrapping the existing system in favor of unfamiliar single payer system run by the federal government. 

Nor are the drastic proposals of the right, such as abandoning regulation in favor of a free market in health care and informed consumer choice, any more realistic or feasible. Some free market advocates favor replacing the current system with one in which consumers paid for their own health care directly aided by individual health savings accounts. Only high-deductible health insurance, covering truly catastrophic health events,  would be permitted. Consumers would have access to improved information about the cost of various treatments and the track records of doctors and hospitals and would be able to make well-informed choices about how to spend their health care dollars. These proposals put a heavy burden on consumers to chose wisely even when they are ill.  They risk discouraging needed routine and preventive care, while failing to reduce the high spending for serious illness or end-of-life care that would exceed any realistic level of deductible.  While better information about cost and success of treatment would help consumers make better informed choices, health care cannot be treated as just another consumer good to be bought and sold in the market place.

In the 2008 election campaign the leadership of both political parties apparently bought into this premise: only reforms that build on the main features of the current system and accomplish needed change without whole-sale disruption stand a reasonable chance of being successfully undertaken and implemented.  

Options for Expanding Coverage

But which features should we build on?  At least four possibilities have received serious consideration: (1) Build on state initiatives; (2) Build on (and improve) the individual insurance market by switching to health care vouchers; (3) Build on existing federal programs; (4) Build on existing employer-based insurance, but provide an alternative. I believe that reform in 2009 should build on employer-based insurance (because that is what most people have), but that movement to a voucher system could be anticipated over time.  

Build on state initiatives. This approach would take advantage of the diversity of the states and the opportunity it presents for state by state experimentation and evaluation of different approaches.  Massachusetts and Vermont have moved toward universal coverage and other states were considering doing so–at least until the recession hit them.  In a 2004 paper Henry Aaron and Stuart Butler (Aaron and Butler 2004) proposed a major effort to use federal financial incentives to encourage all states to move toward universal coverage in their preferred way.  Under the plan states would be encouraged to take different routes, including single payer systems, and to evaluate the results against specific criteria, including the percentage increase in residents covered by insurance. The authors did not make clear whether they thought that eventually a single national model should be chosen or whether the objective was to tie the most successful models together in a national network. However, it would be hard to close down a state approach that had succeeded in reaching universal coverage and was popular in the state, so a state-by-state approach would tend toward institutionalizing state variation, rather than creating a uniform national system. 

The state-based approach had considerable appeal when the chances of federal action appeared dim and some states were willing, especially with federal help, to move more quickly toward universal coverage. But the election of 2008 has greatly increased the chances of federal action on comprehensive health reform, which has advantages over locking in and expanding state variability.  Even with federal inducements to expand coverage, some states would be reluctant to participate and others would have difficulty with implementation. Having 50 different universal coverage systems would add more complexity to an already complicated system and increase administrative cost. Basic benefit packages would differ and coverage would not be portable across state lines. States with generous benefits would tend to attract workers, but might lose individual or business taxpayers to states with lower taxes–a combination that could jeopardize the sustainability of their universal coverage. Perhaps most important, the state-by-state approach would undermine the opportunity to couple broader coverage with implementation of the evidence-based changes in pricing and reimbursement needed to improve the cost effectiveness of the health delivery system. In short, a state-by-state approach to comprehensive health reform is an idea whose time has passed. Lessons learned from Massachusetts and other state efforts, however, should be heeded in constructing comprehensive reform at the federal level. (Gruber 2008)).

In the meantime, however, it is crucial to help the states survive the recession.  Most states have been hard hit by falling revenues and increased needs for state services. States always have a tough time balancing their budgets when the economy turns down, and this recession has already been unusually severe, especially in the industrial states, and will get worse before it gets better.  Aid to states designed to forestall their cutting safety net services and raising taxes is an important part of federal efforts to mitigate the effect of the recession. The recent increasing of the federal match for Medicaid (FMAP) will help. However, pushing the states to move toward universal coverage does not seem like a useful step toward comprehensive health reform in 2009.

 

Health Care Vouchers–building on the individual insurance market.  Ezekiel J. Emanuel and Victor R. Fuchs proposed accomplishing universal coverage by providing everyone with a federally-financed voucher good for guaranteed, renewable enrollment in an approved health plan offering basic insurance coverage (Emanuel and Fuchs, 2007).  The insurer or health plan would get a risk-adjusted payment. The vouchers would be financed by a new federal value added tax (VAT), estimated at 10-12 percent. The exclusion of employer paid health insurance from taxable income would be eliminated, leading to the disappearance of most employer-based insurance and likely an offsetting increase in wages. Health plans would compete on regional health insurance exchanges under the supervision of a National Health Board and regional boards.    

A more limited concept of a voucher plan was floated by Senator McCain in the 2008 campaign. Under the McCain plan everyone would get a refundable tax credit that could be used to purchase a basic health insurance and the employer exclusion would be phased out to pay for it. President Bush made a similar proposal earlier, but his credit was not refundable and, therefore, would have had little impact on lower-income people.

Voucher plans have a lot of appeal. They are easy to understand and extremely equitable, since everyone gets the same benefits. They preserve the choice and competition that are so appealing to conservatives, but under tight controls that do not allow adverse selection to penalize people with health problems. Even if financed by a VAT, a voucher system would be more progressive than the status quo. Moreover, a universal voucher plan would have a built-in top-line cost (a global budget). At least conceptually, the global budget would provide a means of controlling the total cost of health care.  Reluctance to raise the VAT might constitute a brake on aggregate cost growth that would trigger more aggressive efficiency measures or rationing of expensive care.

A voucher system has many attractions, especially if accompanied by aggressive efforts to improve efficiency, but such a drastic change does not appear feasible at present. Replacing other coverage with vouchers would require millions of people to give up their familiar employer plans and shop for health insurance on the new regional exchanges. Moreover, the employer-based system is seen as privately financed, despite the huge tax subsidy. Vouchers by contrast, are clearly government expenditures. Paying for the universal voucher would require a major new revenue source, such as a VAT. Retailers and state governments would oppose the tax. Consumers would resent it and workers might not believe that their wage increases were a by-product of health reform, rather than a reflection of their merits and hard work. Businesses might see the VAT as move to bigger government, even though they were benefiting from lower health benefit costs. Insurers might be unhappy about losing the right to cherry pick and do a “Harry and Louise” number on the proponents. With all these impediments, moving to a voucher plan right now involves too much change for too many people. However, while a voucher plan does not constitute a good starting point for comprehensive reform in 2009, vouchers might still be where we end up down the road as the employer-based system erodes. 

 

Building on existing federal programs. This approach would move toward universal coverage by expanding Medicaid and SCHIP and providing opportunities for buy-in to Medicare with subsidies for low-income people. Raising the income limits on SCHIP would be a relatively easy way to ensure that virtually all children have health insurance. Allowing people aged 55-65 to buy into Medicare with a sliding scale subsidy, would reach a particularly vulnerable group of adults who now fall between the cracks. There have also been proposals for moving to universal coverage by expanding Medicare to cover everyone (Anderson and Waters 2007)

Expanding Medicare and SCHIP is a tempting way to increase coverage, because it does not require creating new programs, which is both a plus a major minus. Putting more people into Medicare involves expanding a system which is still largely fee for service and has few effective incentives to deliver care efficiently. Since Congress has proved resistant to allowing major moves toward making Medicare more cost effective, adding even more constituents with a stake in the Medicare status quo could slow needed change.  Adding middle income children to SCHIP has the disadvantage of additional reliance on the states with their variable records on efficiency. In short, while expanding existing programs may solve some coverage problems quickly, it is not a winning strategy for accomplishing comprehensive reform.      

Building on employer based coverage. Most people get their health insurance through their employers, so a reform strategy that keeps employer coverage and fills in gaps promises minimal disruption. As a presidential candidate President Obama promised “affordable, comprehensive and portable health coverage for every American.” The plan he sketched out started by assuring everyone with insurance that they were free to keep it.  It promised that those not covered (or not satisfied with their coverage) could participate in a National Health Insurance Exchange similar to the exchange available to federal employees, where they could buy a basic insurance package at an affordable price and not be turned down because of poor health or preexisting conditions. Competition among the plans on the regulated exchange was anticipated to bring greater efficiency and lower premiums (a rash promise). A public plan would be offered to those who wanted it. Employers would have to contribute to health insurance for their employees or pay into a fund that would finance the public alternative. There would be income-related subsidies to help people pay for the public or private plans available on the National Health Insurance Exchange. Details were not extensive, which was appropriate for a campaign pledge, but the concept was “play or pay” for employers, access to a large insurance pool for those not adequately covered by employer-based insurance, and income-related subsidies.  There was no individual mandate, except for children (and it was not clear how that was to be enforced.)  Candidate Obama’s health proposals also included other substantial proposals for enhancing prevention and chronic disease management, subsidizing health information technology, reinsurance for catastrophic expenses, tort reform, etc..

With respect to health care the Obama campaign made two mistakes that may come back to haunt the Obama Administration.  One was criticizing Senator McCain for proposing elimination of the exclusion of employer paid benefits from taxable income (“My opponent would tax your health insurance!”) The Obama Administration would be well-advised to back off this position as they work with Congress on a concrete proposal.  The exclusion is extremely costly and the benefits go disproportionately to upper income people.  Capping the exclusion and/or converting it to a tax credit would be a more progressive stance and could help pay for the additional costs of the income-related subsidies. The other mistake was claiming that his health reform would reduce health premiums by $2500 a year.  If people heard that as, “You will actually pay $2500 less than you paid this year” they will be disappointed. If they heard something like “$2500 less than you would otherwise have paid five years from now,” which is more likely what was meant, they are unusually sophisticated, but will have no way of checking the claim.

The general approach sketched out by President Obama in his campaign seems to me the most politically salable and the least disruptive option for moving to universal coverage that is available in the current circumstances. But it may not be a stable long run solution.  Employer-based insurance has been eroding for some time and the pay or play approach could accelerate the erosion. If the National Health Insurance Exchange operates successfully and offers attractive alternative insurance, employer coverage would become less important as a recruiting and retention tool over time. A growing number of employers, first small and then bigger ones, might decide not to play and to pay the fee instead. Capping the exclusion would accelerate this movement. If this scenario plays out, the situation in a few years could something like this:

  • employer-based insurance is disappearing;
  • the National Health Insurance Exchange functions well and the delivery system is getting more efficient;
  • nevertheless, the cost of the basic plan keeps rising and so do the income related subsidies;
  • additional revenue is clearly needed. 

 At this point it might be advantageous and politically feasible to go for a universal voucher system financed by a dedicated VAT.

If the Obama Administration, working with the Congress,  follows the rough blueprint laid out in the campaign, they will have to decide whether to wrap their proposals up in one big bill or legislate it in pieces.  The important thing is to get the vision clear and agree on a fairly detailed plan and schedule for implementation.  Once this is done, it may be advantageous to break the legislation into several parts. Money for electronic medical records has already been included in the economic recovery package and  coverage expansion might proceed in several steps. But it would be a mistake not to lay out a clear design for how the elements will unfold and fit together.

III. Moving to higher quality and cost effectiveness

No matter how the United States decides to move to universal coverage–and even if it does not–there is an overwhelming case for aggressive action to make health care delivery more efficient and effective, to reduce waste and unnecessary medical errors, and get more health care value for the vast sums currently expended. A lot of effort has gone into documenting the problem and a strong consensus is emerging on what needs to be done.  The questions are:  who should do it and who should pay?

Investing in Health Information Technology (HIT)

The fragmented health care sector has lagged other industries in taking advantage of the on-going revolution in information technology to improve quality and productivity.  Most physicians offices are still paper intensive. Small practices lack the funds, time for training and inputting existing records, and the technical sophistication to join the electronic age. Hospitals and health plans often have heavy investments in IT capacity, but use it mostly for billing and record-keeping rather than configuring it to maximize communication among professionals, reduce errors, or learn from actual experience how best to improve patient care.  Patient records are not easily updated or transferred from one provider to another. Providers tend to bear the cost of upgrading technology, while many of the potential benefits accrue to the payer.
There is widespread agreement on these points:

  • Any version of comprehensive health reform must include substantial incentives (including subsidies) for investment in expanding and upgrading health information systems.
  • The systems must be interoperable so that information can be transferred easily among providers and used for analytical purposes.
  • It is time to set standards (but in ways that allow for innovation and adaptation).
  • There is huge potential for wasting money on IT systems that turn out to be poorly designed, not user friendly, or badly implemented. (This observation is not unique to health care).
  • Just having the technology (even if well designed) will not automatically produce quality improvements, cost savings, or even error reduction. Nothing will happen unless participants have the desire and skills to analyze the information carefully, learn from the evidence and change what they are doing continuously as evidence accumulates.
  • Patients must be confident that their privacy is being protected. There are good ways of ensuring privacy, but it will not take many examples of misused personal health records to destroy confidence.  

Since the potential for wasting money is great, plans should be carefully drawn and executed. It is important to take the time to do it right.     

Comparative analysis of treatments, outcomes, costs, and delivery options
There is also a strong consensus that we need to create a national health care data base (or data bases) which captures the treatments and health outcomes of very large numbers of patients while protecting the privacy of individual records.  We need to create an institution or institutions that can recruit and retain skilled analysts to study the comparative effectiveness and cost of treatments and various ways of delivering care, identify best practices, and develop practice guidelines. They should publish their work in understandable form. They might also be asked to design reimbursement structures with incentives to improve patient outcomes, and test these structures in real situations.  To succeed in improving the efficiency and effective of health care in a big way over time, this comparative effectiveness analysis has to be well-funded, carried out by highly competent researchers, and protected from political interference by interest groups, such as drug companies, equipment manufacturers and groups of health specialists. 

A remarkable number of pundits and policy groups have suggested a national institution modeled on the Federal Reserve Board–a Health Fed. The Fed is perceived as an effective institution–although its halo may be fading in the financial crisis–that takes the technical, but contentious job of setting interest rates out of politics.  Its independence is achieved by appointing highly qualified people to long terms (from which they can be removed only for cause) and giving the institution a generous budget not subject to executive challenge or congressional appropriation.  The Health Fed idea has been proposed by the Commonwealth Fund, the Committee for Economic Development, Emanuel and Fuchs, and former Senator Thomas Daschle. Funding could be from a permanent assessment on all payers, public and private, or from a permanent set-aside out of a new revenue source, such as a health VAT if one were enacted.

The Health Fed concept varies.  In some versions the institution would be in charge of setting the contents of the standard benefit package; in others, only of the comparative effectiveness analysis; in some, of designing reimbursement incentives to be used by public programs; in others, of overseeing the administration of the whole health system and actually paying health plans. 

It is hard to argue against the idea of creating a prestigious institution to preside over a national health data base and ensure that substantial resources are devoted to comparative effectiveness analysis and that these resources are well spent.  But does it have to be exempt from public accountability and oversight?  In particular, should it be outside the normal appropriations process? NIH, for example, is not.  I am wary of handing over too much authority and funding to unelected officials without the discipline of accountability for the use of the taxpayers’ money. If we think special interests have too much influence over elected officials, we fix that problem, rather than giving up on representative democracy.

Reimbursement incentives and competition

Once we have a continuing flow of solid information on comparative effectiveness and cost of treatment, how do we ensure that this information actually changes what how the system operates?  There are basically two ways to go:

  • Change the reimbursement of providers under Medicare (and perhaps other public programs) to reward cost effective treatment and penalize waste and error. The trust that the private sector will keep up.
  • Make the evidence-based guidelines and information on best practice public, and  then  rely on competition among health plans to effectuate the implementation.
  • Or, of course, both.

Medicare is currently the largest payer and has enormous potential to improve the cost-effectiveness of delivery.  However, especially in the last few years Congress has generally shut down efficiency moves that providers or suppliers find threatening. For example, Congress derailed competitive bidding on durable medical equipment for Medicare, which sounded to many of us like an obvious step toward greater efficiency. Still, the public and congressional consciousness of the fiscal importance of improving the efficiency of care may be growing. Indeed, if it isn’t, none of this may be possible. As long as Medicare remains a separate program, shifting to reimbursement mechanisms that favor efficient delivery is imperative.

If the National Health Insurance Exchange (or regional exchanges) becomes a reality, we will have a test of how well competition among plans can translate evidence-base guidelines into more efficient care. It will be important to make the exchange consumer friendly so that buyers can easily the benefits and costs of different plans.  The options presented to federal employees, for example are too numerous and complex to facilitate informed consumer choice.   

IV. Controlling the total cost of health care
Americans may be lucky that our health system is so unnecessarily costly that there is substantial potential for making it more efficient and covering more people at the same time.  I am not optimistic that even aggressive efforts to learn more about cost-effective treatment and delivery organization will actually lower the total cost of health care, but we certainly could hope for slower rates of growth in the future than in the past.  Eventually, however, we will achieve a more efficient system and will have to decide how much we really want to spend on health care as the medical profession keeps finding more ways of extending life.  Is 20 percent of GDP enough? Or 25? Or 30?

If the system evolves toward public financing (either a single payer or a universal voucher system, for example), it will have a built-in decision point, namely, increasing the total public budget for health care. A global budget restraint (e.g., running out of the money raised by the dedicated VAT) does not provide a mechanism for rationing the scarce funds among claimants, but it forces decision-makers to resolve the question.  If our system remains fragmented (as in play-or-pay with subsidies and insurance exchanges) it will be harder to figure out how to control total health spending if it continues to rise significantly faster than GDP. Launching comprehensive health reform in 2009 will add the question of how to control total health spending to the already difficult one of how to control total federal health spending.

References

Gruber, J. (2008, July). Taking Massachusetts National: An Incremental Approach to Universal Health Insurance. Washington, DC: The Brookings Institution, The Hamilton Project.

Aaron, H. J. and Butler, S. M. (2004, March 31). How Federalism Could Spur Bipartisan Action on the Uninsured. Health Affairs.  Retrieved December 28, 2008, from http://content.healthaffairs.org.

Emanuel, E. J. and Fuchs, V. R. (2007, July). A Comprehensive Cure: Universal Health Care Vouchers. Washington, DC: The Brookings Institution, The Hamilton Project.

Anderson, G. F. and Waters, H. R. (2007, July). Achieving Universal Coverage Through Medicare Part E(veryone). Washington, DC: The Brookings Institution, The Hamilton Project.

Senator Daschle, T., Greenberger, S. S., Lambrew, J. M. (2008) Critical: What We Can Do About the Health-Care Crisis. New York: Thomas Dunne Books, St. Martin’s Press.

Committee for Economic Development. (2007). Quality, Affordable Health Care for All
Moving Beyond the Employer-Based Health-Insurance System. Washington, DC: CED.

The writer, a senior fellow at the Brookings Institution, was the founding director of the Congressional Budget Office and was director of the White House Office of Management and Budget from 1994 to 1996.

 

Previous post:

Next post: