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

P E R S P E C T I V E
F U T U R E  E L D E R L Y 

26 September 2005  
  

The Potential For Cost Savings 
In Medicare’s Future

Examination of five scenarios to project spending on cancer
among the elderly through 2030.

By David M. Cutler

ABSTRACT:

The papers in this special Health Affairs collection offer a fresh and very important look at Medicare’s future. Taken as a whole, they suggest that financing problems in Medicare will be particularly acute in coming decades. Other pieces of evidence suggest that this outlook is not as problematic, however. The technological changes that the RAND authors consider will likely come to pass, and they will drive up Medicare spending (often with good value). But there is enormous potential for cost savings as well, which we have the capacity to realize.

Papers in this special Health Affairs collection contain a remarkable series of papers by Dana Goldman and his colleagues at RAND, using a detailed microsimulation model to examine different scenarios for the future of the elderly, and the Medicare program in particular. Goldman and colleagues have built a model of the aged population, termed the Future Elderly Model (FEM). The details of the model are complex, but it is conceptually quite simple. The model takes the current health status of the elderly; estimates what medical services people will use; and then simulates the change in health status, including deaths, over the course of the year. For good measure, a new set of sixty-five-year-olds enters every year, and their health is forecast as well. Rolling the process forward, the model can predict medical costs and health status far into the future.

Like a new luxury car, the FEM is fun to drive. It is without a doubt the shiniest, most sophisticated car on the road; the engine is state of the art. The FEM is the standard we should use for Medicare policy evaluations, and it should be integrated into the work of the Medicare actuaries and outside researchers.

But how is the ride overall? What do we know about the future of Medicare that we didn’t know before this model was introduced? The ride is good but still a bit bumpy. Perhaps the central question facing the government is how to afford Medicare in the future. The papers here could well lead to despair. For example, Goldman and colleagues and Jay Bhattacharya and colleagues conclude that technological innovation in medical care will lead to much higher medical spending, with uncertain cost-effectiveness.¹ Geoffrey Joyce and colleagues find that the healthy elderly spend only a little less over their lifetimes than the less healthy elderly. Thus, having a healthier group of sixty-five-year-olds will not save Medicare much money.² Further, Michael Chernew and colleagues show that medical spending is increasing more rapidly for the less disabled than for the more disabled, so whatever advantage there is to keeping people healthy is declining.³ And Darius Lakdawalla and colleagues estimate that rising rates of obesity among the elderly will lead to much higher medical costs.⁴ As assessed in these papers featuring the FEM, the future of Medicare is dire, and we don’t know what to do about it.

To some extent, dire predictions about Medicare are unavoidable. Under current projections, Medicare spending is expected to rise from 2.6 percent of U.S. gross domestic product (GDP) today to 9.2 percent in 2050. About half of this increase is a result of the demographic trends toward more elderly people; the other half results from medical cost growth above rates of GDP growth.⁵ At current levels of general revenue financing, the income available for public medical spending will increase to only 3.8 percent of GDP by 2050. So one would need to cut medical spending for the elderly in 2050 more than in half to balance spending with revenue forecasts. No one has a way to do that.

Cause for pessimism. But can’t we make some progress? To some extent, these authors understate the magnitude of some of their findings. Joyce and colleagues show that the healthy elderly do spend less over their lifetime than the elderly with comorbidities. At age sixty-five, the difference between those with no comorbidities and those with three or more comorbidities is 29 percent. Put another way, if all incoming Medicare beneficiaries had no comorbidities, lifetime medical spending for the average person would fall by 17 percent. That is below the 50 percent goal, but it is still sizeable.⁶

Whether these savings can be realized—or whether the situation will be even worse—is a matter of debate. Lakdawalla and colleagues argue that increases in obesity will lower the average health status of the young elderly population.⁷ On the other hand, smoking rates have been falling for decades, and control of chronic conditions such as hypertension and high cholesterol has improved. It is not impossible to envision a reduction in obesity. An improved medical and public health system could lead to substantial improvements in the health of the elderly and thus to some relief for Medicare.

Still, even these savings are not enough. Joyce and colleagues estimate that even a healthy sixty-five-year-old will still spend $88,000 on medical care over his or her remaining lifetime.⁸ Why so much? There are two important reasons. First, virtually everyone winds up with some chronic disease ultimately, and chronic disease is very costly. Even healthy people at age sixty-five develop arthritis, suffer broken bones, or have heart attacks. About half of people age eighty-five and older, for example, have a limitation in either activities of daily living (ADLs) or instrumental activities of daily living (IADLs).⁹ People with one or two ADL limitations spend twice what the nondisabled do annually. Second, death is very expensive—more accurately, people get very sick before they die, and being very sick is quite costly. Joyce and colleagues show that about one-quarter of lifetime medical costs for the elderly occur in the last year of life.¹⁰ About 30 percent of Medicare spending each year is for people who will die within a year.¹¹

Addressing the issue of high end-of-life costs is an important issue that the FEM does not address, but could. In the FEM, the costs of death are independent of age, based on data suggesting that health status matters more than age. But the high cost of death is not immutable. Medical spending in the last year of life falls from $32,000 for elderly who die young (ages 65–69) to half that amount for those who die above age 90.¹² The diseases that older people die from are different, and the amount of medical care provided differs as well. If we were able to keep people healthy until age eighty-five or ninety, when people would then succumb to pneumonia or other less costly illnesses, one could envision ample savings in end-of-life costs. The types of interventions that would accomplish this are not modeled in the FEM, but they could be.

Cause for optimism. The FEM misses two other types of innovations that could be important in helping Medicare. The first is spending for innovations that do what we do now, but at lower cost. Laparoscopic surgery instead of open surgery is a clear example; laparoscopic operations take less time and cost less. In cardiac disease, angioplasty often substitutes for more expensive bypass surgery. And some pharmaceuticals save money in reduced inpatient stays or fewer physician visits. In the past, such innovations have lowered unit costs but led to corresponding increases in volume that more than offset the unit cost savings. Thus, we spent more on gallbladder removal after laparoscopic surgery was developed, since so many more people had their gallbladders removed.¹³ The same is true about the diffusion of angioplasty and many pharmaceuticals such as antidepressants.¹⁴ “Treatment expansion” proved to be quantitatively more important than “treatment substitution.”

But there is reason to suspect that this might change in the future. As more people are treated with existing technologies, the share of untreated people in the population will fall. The number of people left to expand treatment to would thus naturally diminish. As a specific example, suppose we developed a newer form of cardiac intervention that involves less operating time than angioplasty, and thus lower cost. A good part of the use might be to replace existing angioplasties, rather than increasing the total number of people receiving coronary interventions.

The technologies in the FEM generally are not this type of innovation. Most of the technologies these RAND authors consider are incremental technologies—with use that is mostly expansionary rather than substitutable. The one exception is the analysis of cancer by Bhattacharya and colleagues, which considers therapies such as a cancer cure.¹⁵ That cure does save money, since people with cancer are virtually always treated now. Following the logic, the cure does save money, albeit modest amounts ($5 billion annually). It could be that the panel of experts RAND consulted did not think that substituting innovations would occur, but in truth we do not know. There is a question that these authors could ask of their model that would be relevant: For the typical innovation expected in the future, what share of the use would have to substitute for care already received for the net cost impact to be zero or negative?

The final type of innovation not included in the FEM is information technology (IT) or other process improvements. Aficionados of electronic medical records and other IT innovations hope that they will improve health and save money. It is certainly true that there is enormous waste in medical care, and it is generally believed that IT can help reduce that waste.

The omission of this type of technology from the FEM is understandable, and it was the right decision. The FEM is not a model of medical care interactions; it is a model of the medical system overall. Still, users should be aware of this limitation. Estimates are all over the map, but they suggest that 20–50 percent of medical care spending could be eliminated without skimping on valuable services.¹⁶

Accounting for these factors, my forecast about medical spending is rosier than the FEM model suggests. The technological changes that the RAND authors consider will likely come to pass, and they will drive up Medicare spending (often with good value). But there is enormous potential for cost savings as well, which we have the capacity to realize. One can be an optimist even when the storm clouds are gathering.

The author is grateful to the National Institute on Aging for research support.

NOTES

1. D.P. Goldman et al., “Consequences of Health Trends and Medical Innovation for the Elderly of the Future,” Health Affairs, 26 September 2005, content.healthaffairs.org/cgi/content/abstract/hlthaff.w5.r5; and J. Bhattacharya et al., “Technological Advances in Cancer and Future Spending by the Elderly,” Health Affairs, 26 September 2005, content.healthaffairs.org/cgi/content/abstract/ hlthaff.w5.r53.
2. G.F. Joyce et al., “The Lifetime Burden of Chronic Disease among the Elderly,” Health Affairs, 26 September 2005, content.healthaffairs.org/cgi/content/abstract/hlthaff.w5.r18.
3. M.E. Chernew et al., “Disability and Health Care Spending among Medicare Beneficiaries,” Health Affairs, 26 September 2005, content.healthaffairs.org/cgi/content/abstract/hlthaff.w5.r42.
4. D.N. Lakdawalla, D.P. Goldman, and B. Shang, “The Health and Cost Consequences of Obesity among the Future Elderly,” Health Affairs, 26 September 2005, content.healthaffairs.org/cgi/content/abstract/hlthaff.w5.r30.
5. R. Lee and T. Miller, “An Approach to Forecasting Health Expenditures, with Application to the U.S. Medicare System,” Health Services Research 37, no. 5 (2002): 1365–1386.
6. Joyce et al., “The Lifetime Burden.” In this case, the analysis by Joyce and colleagues is actually more optimistic than that of James Lubitz and colleagues, who found that at age seventy, the healthy elderly spent only 6 percent less over their lifetime than the community-dwelling elderly with ADL limitations. See J. Lubitz et al., “Health, Life Expectancy, and Health Care Spending among the Elderly,” New England Journal of Medicine 349, no. 11 (2003): 1048–1055.
7. Lakdawalla et al., “The Health and Cost Consequences.”
8. Joyce et al., “The Lifetime Burden.”
9. K.G. 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.
10. Joyce et al., “The Lifetime Burden.”
11. S. Calfo, J. Smith, and M. Zezza, “Last Year of Life Study,” 17 September 2004,
www.cms.hhs.gov/statistics/lyol/default.asp (3 August 2005).
12. Ibid.
13. A.P. Legorreta et al., “Increased Cholecystectomy Rate after the Introduction of Laparoscopic Cholecystectomy,” Journal of the American Medical Association 270, no. 12 (1993): 1429–1432.
14. D.M. Cutler and R.S. Huckman, “Technological Development and Medical Productivity: The Diffusion of Angioplasty in New York State,” Journal of Health Economics 22, no. 2 (2003): 187–217; and D.M. Cutler, Your Money or Your Life: Strong Medicine for America’s Health Care System (New York: Oxford University Press, 2004).
15. Bhattacharya et al., “Technological Advances.”
16. J. Skinner and J. Wennberg, “Regional Inequality in Medical Care: The Key to Medicare Reform,” in Frontiers in Health Policy Research, vol. 3, ed. A. Garber (Cambridge, Mass.: MIT Press, 2000), 69-90; and D.U. Himmelstein and S. Woolhandler, “Cost without Benefit: Administrative Waste in U.S. Health Care,” New England Journal of Medicine 314, no. 7 (1986): 441–445.

David Cutler (david_cutler{at}harvard.edu) is the Otto Eckstein Professor of Applied Economics and dean for the social sciences at Harvard University in Cambridge, Massachusetts.

Access the table of contents for this package

DOI: 10.1377/hlthaff.W5.R77
©2005 Project HOPE–The People-to-People Health Foundation, Inc.