simply because they may appear less cost-effective for the “average” patient.

  • Cost-effectiveness analysis will not address underlying drivers of costs: Comparative effectiveness research will undoubtedly improve the evidence base and lead to better use of health resources, but it will not have a major impact on costs, as the Congressional Budget Office (CBO), RAND, and others have found.15 One misconception is that pharmaceuticals and medical devices, one focus of interest in applying cost-effectiveness, make up a relatively large percentage of health care costs, when in fact the percentages are quite small.16 In focusing on including cost-effectiveness as part of comparative effectiveness, the committee’s recommendations do not address some of the major drivers of cost including the fact that 75 percent of costs are driven by chronic disease and the role of the fee-for-service payment system in Medicare, where doing more generates more reimbursement. The committee does recommend outcomes-based reimbursement, which is a good thing—but this alone will not significantly change either the management of chronic disease or the incentives that reward doing more.


P. Orszag, September 5, 2007. Letter to the Honorable Pete Stark. Washington, DC. In the letter to Chairman Stark estimating the impact of enacting a comparative effectiveness entity, Orszag states, “CBO estimates that the information produced by enacting section 904 would reduce total spending for health care services. Specifically total spending—by public and private purchasers—would be reduced by about $.5 billion over the 2008–2012 period and by about $6 billion over the 2008–2017 period. Direct spending by the federal government—mostly for Medicare, Medicaid, and the Federal Employees Health Benefits program—would be reduced by $.1 billion over the 2008–2012 period and $1.3 billion over the 2008–2017 period.” A RAND COMPARE analysis reaches a similar conclusion, that comparative effectiveness research will not result in significant savings in the near term. See


For prescription drugs, the Centers for Medicare and Medicaid Services estimates spending in 2006 was about 11 percent of total health care spending. A. Caitlin, C. Cowan, M. Hartman, S. Heffler, and the National Health Expenditure Accounts Team. January/February 2008. National health spending in 2006: A year of change for prescription drugs. Health Affairs 27:14-29. Recently, CBO issued a report on the impact of technological change on the growth in health care spending, attributing about half the growth to “changes in medical care made possible by advances in technology.” CBO is able to “count” as contributors to growth in costs such factors as growth in personal income, aging of the population, and rising personal income. Everything else is counted as “technological change,” including changes in physicians’ practices, price increases in technologies and treatments, and other hard-to-quantify cost increases. Congressional Budget Office. January 2008. Technological change and the growth of health care spending. Washington, DC: CBO.

The National Academies of Sciences, Engineering, and Medicine
500 Fifth St. N.W. | Washington, D.C. 20001

Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement