Michael E. Chernew, Ph.D., Lindsay Sabik, B.A., Amitabh Chandra, Ph.D., and Joseph P. Newhouse, Ph.D.
Much of the recent health care reform debate has focused on achieving budget neutrality over a 10-year period, but this goal is less important than the reform’s long-run fiscal sustainability. If the rate of growth of health care spending continues to exceed the rate of income growth by its historical margin of more than 2 percentage points, the consequences for beneficiaries, federal and state budgets, and the entire economy — given the implied increase in tax rates and forgone consumption — will be dire.1 For this reason, health care reform is an economic issue as much as, if not more than, a health issue.
Growth in health care spending is often a good thing. We want to spend more of our growing income on medical advances whose benefit exceeds their cost. Yet the U.S. health care system is larded with inefficiency. For example, it is well established that some regions spend considerably more than others in the Medicare program without delivering higher-quality care or generating greater patient satisfaction.2,3
Yet low spending is not the same as low spending growth, and even efficient areas can experience considerable spending growth. In fact, many areas that had low spending in 1992 did not have notably lower spending growth between 1992 and 2006 than other areas (see graph). Rochester, Minnesota, and Salt Lake City, for instance, are known for high-quality, integrated providers and have low Medicare spending per beneficiary. Yet between 1992 and 2006, inflation-adjusted Medicare spending per beneficiary rose 4.3% annually in Salt Lake City and 3.8% annually in Rochester, as compared with 3.2% for the country as a whole. In short, areas with exemplary delivery systems do not necessarily have exemplary rates of spending growth.
Level of Medicare Spending in 1992, Plotted against Growth in Spending between 1992 and 2006.
The plot shows little correlation between levels of Medicare spending in 1992 and the growth of such spending from 1992 through 2006, according to hospital-referral region. Thus, areas with exemplary medical delivery systems do not necessarily have exemplary rates of spending growth. Data include all spending for Medicare Parts A and B, with adjustment for age, sex, and race but not for price differences, with spending levels presented in 2005 dollars. Data are from the Dartmouth Atlas of Health Care.
Of course, some of the 300-plus regions have both low levels of spending and low spending growth (e.g., San Francisco). These areas should be studied for any lessons that can be applied elsewhere, but such lessons may be few, because low growth rates often do not persist. For example, the correlation between spending growth from 1992 through 1999 and spending growth from 2000 through 2006 is –0.12. This modest inverse correlation partially reflects regression to the mean, but that is the point: high (or low) spending areas are not systematically high (or low) spending-growth areas, so strategies that reduce costs in high-spending areas will not necessarily “bend the curve” (except transitorily).
The historical flow of valuable but costly medical advances raises a profound question — and not just for Americans: Can that flow be maintained in future years without rates of spending increase that wreak economic havoc? To answer affirmatively, the nation must develop effective strategies to curb spending growth and must address the role of government in health care.
A number of options have been proposed for lowering spending growth. These include delivery-system reform, especially the development of “integrated” organizations to reduce costs and improve patient outcomes; payment reform involving a greater role for bundled payments, including episode-based payments and various forms of capitation; and “promarket” strategies that make individuals more sensitive to costs and rely on individual choices to constrain spending, such as benefits packages with substantial cost sharing, taxing of high-cost plans, and insurance- and provider-market reforms that enhance competition.
Delivery-system reform, undoubtedly important to increased efficiency, may not slow spending growth sufficiently in the long run, even if it transitorily reduces it, because even the most efficient delivery systems must wrestle with the adoption of expensive new technologies. Thus, delivery-system reforms will probably need to be coupled with meaningful payment reforms, promarket reforms, or both if the rate of spending growth is to fall.
Payment reform can slow spending growth but only if the process for setting updates for bundled rates is disciplined. Bundled or capitated payments cannot be routinely increased to cover all new services. Although all payers can adopt more bundled payments, government health insurance programs, because of their market power, can have the largest effect on spending growth. Government restraint in rate setting, however, may be weakened by the political process. On the other hand, if government forces payment updates into a tight budget straitjacket, there could be adverse effects on innovation, access, and patients’ health. Getting reimbursement rates right is a daunting task.
Promarket strategies have their own pitfalls. Left unimpeded, markets will eventually slow spending growth because people will ultimately not want to give up other goods in favor of still more health care. But markets may not lead to efficient, or equitable, outcomes. Market failures stemming from having too few competitors, which results in excessive market power, or a surplus of competitors, which increases the cognitive burden of choosing among them, may impede markets from achieving efficiency even if there are appropriate incentives for selecting efficient plans or providers. Moreover, patients do not always make appropriate decisions about the use of services when faced with cost sharing, so plans with high cost sharing may lead to suboptimal clinical outcomes.4 This concern may be mitigated by more sophisticated insurance products, such as those with a “value-based insurance design,” which lower the out-of-pocket costs for high-value services. Unfortunately, such strategies are in their infancy.5 Reference pricing systems may help, but defining classes for a reference price is often challenging. Even when promarket systems are more fully developed, their success in stemming spending growth will depend on tolerance of the income-related disparities that such strategies will surely generate. Subsidies and vouchers, which can mitigate these disparities, not only are expensive but also inherently dampen the impact of promarket policies on spending growth. Moreover, if they are restricted to low- and middle-income families, subsidies have to be phased out, which creates large increases in marginal tax rates that can reduce the incentive to work.
Our country’s record of making the sorts of tough choices that are required for payment or promarket reforms to restrain spending growth is not encouraging. The experience with Medicare’s “sustainable growth rate” formula for physician payments shows the difficulty of resisting political pressure to increase fees in order to preserve access to care and avert large reductions in providers’ incomes. And private payers, despite pioneering innovations such as disease management and various forms of bundled payment, have not been able to constrain spending growth to sustainable levels.
Yet lamenting such past failures does not provide a way forward. Unfortunately, an obvious, painless solution does not exist. Many observers point to waste, fraud, and abuse that should be eliminated from the current system, but these problems have proved resilient to myriad eradication efforts. Even if they could be reduced to negligible levels, the underlying spending growth would then resume. Current analyses frequently focus on the 10-year savings associated with various reforms, but these are often one-time savings that do little to change the long-term spending trajectory. Instead, the priority now should be to create institutions that will allow future cost containment to be successful. The appropriate balance between the roles of government and markets within these institutions is a function of politics and philosophy as much as economics. Institutions that emphasize markets will struggle with income disparities and market failures, whereas those that give government a more central role will wrestle with political pressures affecting payment updates and inefficiencies generated by administratively set prices. Although assessing the magnitude of the effects of both these strategies is necessarily speculative, failing to move forward at all seems a bigger risk.
Financial and other disclosures provided by the authors are available with the full text of this article at NEJM.org.
Source Information
From the Department of Health Care Policy, Harvard Medical School, Boston (M.E.C., L.S., J.P.N.); and the John F. Kennedy School of Government, Harvard University, Cambridge, MA (A.C.).
This article (10.1056/NEJMp0910194) was published on December 9, 2009, at NEJM.org.
References
Chernew ME, Hirth RA, Cutler DM. Increased spending on health care: long-term implications for the nation. Health Aff (Millwood) 2009;28:1253-1255. [Free Full Text]
Sutherland JM, Fisher ES, Skinner JS. Getting past denial — the high cost of health care in the United States. N Engl J Med 2009;361:1227-1230. [Free Full Text]
Fisher ES, Wennberg DE, Stukel TA, Gottlieb DJ, Lucas FL, Pinder EL. The implications of regional variations in Medicare spending. Part 1: the content, quality, and accessibility of care. Ann Intern Med 2003;138:273-287. [Free Full Text]
Chernew ME, Newhouse JP. What does the RAND Health Insurance Experiment tell us about the impact of cost sharing on health outcomes? Am J Manag Care 2008;14:412-414. [Web of Science][Medline]
Chernew ME, Rosen AB, Fendrick AM. Value-based insurance design. Health Aff (Millwood) 2007;26:w195-w203. [Free Full Text]
Email
Printer Friendly
Permalink
facebook
StumbleUpon
Digg
Twitter
Yahoo! Buzz
Fark
reddit
LinkedIn
del.icio.us
MySpace
Loading ...
Comments [0]