This Article Abstract Submit a response to this article Services E-mail this article to a friend Alert me to new issues of the journal

F O R E W O R D 

26 September 2005  
  

Foreword

This special collection of Health Affairs papers focuses on a topic central to both health policy and public finance: the evolution of Medicare spending as the U.S. population over age sixty-five grows rapidly during the next quarter-century. The papers focus on a set of issues that are key to understanding this evolution: the likely course of disability in successive cohorts of Medicare beneficiaries and the relationships between technological progress and health spending.

Three decades ago Lewis Thomas proposed a useful taxonomy, differentiating “halfway technology”—efforts to compensate for the incapacitating effects of certain diseases whose course one is unable to do very much about—from the “genuinely decisive technology of modern medicine,” such as immunization and antibiotics for bacterial infections.¹ Others argued for an inverted U-shaped curve of costs, with the highest cost for halfway technology and the lowest costs for palliation at one end and high technology at the other end. For years the hope has been that molecular medicine would result in more low-cost, basic solutions to prevent or cure disease and that information technology would target procedures efficiently. The findings from the papers in this collection converge on a conclusion that foreseeable improvements in health will cost more money, rather than saving any. This is not, of course, the same as arguing that forecast medical expenditures are not “affordable,” still less that the technological changes that largely drive expenditure growth are not valuable.² Robert Fogel has argued that health and leisure are the next consumption frontiers; with most material needs satisfied, spending on health will likely continue to expand greatly as a fraction of gross domestic product (GDP).³

Microsimulation models such as the one underlying these papers have proved their worth in many policy fields. Social Security and pension policy (in related work by Dana Goldman and colleagues) and income support and training are examples. Their main function is not to produce precise forecasts but to explore the implications of current trends and foreseeable demographic shifts. These exercises are very useful in that they bring together a large set of interacting variables and point to needs for further data and more focused questions. These models are also useful tools for comparing different strategies for improving health and allocating research resources. However, these “what-if” simulations do not produce precise forecasts of expenditures, as a long trail of forecasts from the past can show. The papers in this collection are, appropriately, as concerned with the uncertainty surrounding the central expenditure paths as with the central paths themselves.

One of the most difficult puzzles is to be able to forecast the impact on health care costs of longer lives under various scenarios that include improved or worsened functioning. There is much evidence that disability in the elderly population has declined during the past two decades, but the future course of the trend is in doubt, especially given rising levels of obesity.⁴ There are few analyses of the economic consequences of the disability decline, including savings, if any, for long-term care. Ideally, such analyses would be carried out using rich panel data that include all diagnostic procedures, medical and behavioral interventions, prescriptions, and insurance and out-of-pocket costs, along with measures of health, functioning, and well-being. Panel data that continue until the end of life are needed to be able to estimate the impact of interventions on the downstream consumption of medical and other services as well as on well-being. It would be ideal if future economic analyses were also accompanied by parallel analyses of the impact of the interventions on the level of well-being and quality of life for patients and their family and friends. Characterizing the long-term impact of interventions on well-being is even more necessary but is now at an even more primitive level than estimates of the impact on medical spending, although Daniel Kahneman and colleagues are developing promising new approaches.⁵

With some exceptions, Medicare claims data have been silent on use of prescription drugs, central to understanding both increasing expenditures and the improvements in disability of the elderly in recent decades. With implementation of the Medicare prescription drug benefit in 2006, the most significant change in the history of Medicare, it becomes even more urgent to include pharmaceutical use in panel data.

The simulation model used by the authors of these papers is built on a nationally representative data set. National averages mask profound, persistent variations in spending for apparently similar Medicare beneficiaries in different regions—even different hospitals. John Wennberg has argued that much of this variation is attributable to differences in “preference-based care” that is motivated by tradition and local practice, not by patients’ informed preferences, nor by the simple economic interests of providers.⁶ Models such as those used here could help illuminate the expected monetary costs of unwarranted variation—for example, showing how spending forecasts would change if Minneapolis medicine replaced Miami medicine on a gradual timetable. Cross-national research, such as the series of studies done for the Organization for Economic Cooperation and Development (OECD) comparing how health systems in different countries treat specific diseases and at what cost, also can suggest what is possible.⁷

But at any one time, care actually being delivered is an amalgam of care at various levels of effectiveness; the incentives in the system do not push providers and patients toward using technology whose expected benefits outweigh its costs in the particular setting. The policy problems are to accelerate the trend toward more effective medicine—moving the production possibility curve outward, in the model of textbook economics—while providing incentives that discourage delivery of ineffective (often expensive, and often not much wanted) care.

An even more ambitious accounting system would involve the construction of a set of National Health Accounts, comparable to the National Income and Product Accounts that measure gross national product. A satellite system of National Health Accounts would need to measure both spending on health and its impact on a summary measure of health, incorporating a valuation of the quality of life associated with different health states. The National Institute on Aging (NIA) has begun an initiative to develop a prototype set of accounts for the older population.

Goldman and his coauthors show that a strong tide is running against our efforts to continue the disability decline of recent decades as new cohorts entering Medicare eligibility in the next twenty-five years are showing early antecedents of late-life disability. But we can also ask the models to set intermediate targets for us. We can use them to set intermediate goals such as reducing disparities among subpopulations to keep old-age disability rates on a sustained downward path. We can refuse to accept the disability trends toward which the models show us drifting. Indeed, as the NIA plans for population-level interventions to reduce disability in the older population, we expect that these microsimulation models will play a key role in identifying sets of interventions that will most efficiently move the production frontier out as far as possible.

Richard Suzman and John Haaga
National Institute on Aging

NOTES

1. L. Thomas, “The Technology of Medicine,” in The Lives of a Cell: Notes of a Biology Watcher (New York: Bantam Books, 1974), 37–38.
2. M. Chernew, R. Hirth, and D. Cutler, “Increased Spending on Health Care: How Much Can the United States Afford?” Health Affairs 22, no. 4 (2003): 15–25.
3. R.W. Fogel, The Escape from Hunger and Premature Death, 1700–2100: Europe, America, and the Third World (Cambridge: Cambridge University Press, 2004), 66–95.
4. K. Manton and X. Gu, “Changes in the Prevalence of Chronic Disability in the United States Black and Nonblack Population above Age Sixty-five from 1982 to 1999,” Proceedings of the National Academy of Sciences (U.S.) 98, no. 11 (2001): 6354–6359.
5. D. Kahneman et al., “A Survey Method for Characterizing Daily Life Experience: The Day Reconstruction Method,” Science 306, no. 5702 (2004): 1776–1780.
6. J. Wennberg, “Variation in Use of Medicare Services among Regions and Selected Academic Medical Centers: Is More Better?” Duncan W. Clark Lecture to New York Academy of Medicine, 24 January 2005, www.dartmouthatlas.org/lectures/NYAM_Lecture_FINAL.pdf (31 August 2005).
7. Organization for Economic Cooperation and Development, A Disease-based Comparison of Health Systems: What Is Best and at What Cost? (Paris: OECD, 2003).

Access the table of contents for this package.

DOI: 10.1377/hlthaff.W5.R1
©2005 Project HOPE–The People-to-People Health Foundation, Inc.