This Article Abstract Figures Only Reprint (PDF) Submit a response to this article Alert me when this article is cited Alert me when Comments are posted Alert me if a correction is posted Services E-mail this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to Citation Manager Reprints & Permissions Citing Articles Citing Articles via HighWire Citing Articles via Web of Science (22) Citing Articles via Google Scholar Google Scholar Articles by Lubitz, J. Articles by Gibson, D. Search for Related Content PubMed PubMed Citation Articles by Lubitz, J. Articles by Gibson, D. Related Collections Managed Care - Medicare Medicare

Health Care Trends

Three Decades Of Health Care Use By The Elderly, 1965–1998

Abstract

 
Over the past three decades health spending and hospital use increased more for the elderly than for persons under age sixty-five. Medicare spending for the oldest old (age eighty-five and older) increased faster than for persons ages sixty-five to seventy-four, but that increase was due entirely to greater postacute care use. Health care trends are consistent with the idea that Medicare has improved the health of the elderly. Greater spending increases for the elderly may reflect legislative developments such as the passage of Medicare and its continued fee-for-service nature and the failure to pass universal coverage, as well as changes in the health care delivery system such as the rapid growth in managed care enrollment among persons under age sixty-five.

Policymakers continue to debate the impact of Medicare. After Medicare began, some questioned the value of health spending for the elderly and called for spending limits.² These concerns stemmed from the unexpectedly steep spending growth after Medicare’s inception, the much higher spending in the United States than in other nations, and the temporary slowing of the decline in mortality rates in the 1950s and 1960s.

Today the picture is different. The health of the elderly population has improved, as measured by both longevity and functional status.³ Life expectancy at age sixty-five increased from 14.3 years in 1960 to 17.8 years in 1998.⁴ The chronically disabled elderly population declined from24.9 percent in 1982 to 21.3 percent in 1994.⁵ Some claim that Medicare may be partly responsible for the better survival of the U.S. elderly age eighty and older compared with similar industrialized countries.⁶ It has even been argued that better health of the elderly population will moderate the increase in Medicare spending. ⁷ Furthermore, in contrast to three decades of steady growth, health spending as a percentage of GDP has remained relatively constant since 1992, and, for the first time ever, Medicare per capita expenditures actually declined from one year to the next (fiscal year 1998 to FY 1999). The health care system has changed dramatically since Medicare began. Within Medicare, payment shifted from cost or charge-based reimbursement to prospective payment for hospitals and other providers, a fee schedule for physicians, and capitation for health plans. For the privately insured population, the major change was the growth in managed care coverage.⁸

This paper examines trends in health care use and costs by the elderly over the period 1965–1998, spanning the introduction and evolution of Medicare. By comparing trends between persons over and under age sixty-five, we illustrate Medicare’s effect on health care use and draw lessons for future proposals to change Medicare. We also examine the implications of utilization trends for the health of the elderly. In addition, we compare trends in use and cost for elderly persons under and over age eighty-five. The oldest old (age eighty-five and older) are the fastest-growing segment of the elderly: 7.5 percent of the total aged population in 1970, and projected to be 14.5 percent by 2010.⁹ One of the factors behind the passage of Medicare was concern about access to health care for the oldest old.¹⁰ Interest in this age group comes from the baby-boom generation, which itself hopes for a long and vigorous old age in which disability and illness are concentrated into a short period before death.¹¹

Data And Methods

Top Data And Methods Study Findings Discussion And Policy... NOTES

 
Data to track the elderly’s health care use and spending come from the Health Care Financing Administration (HCFA) and the National Center for Health Statistics (NCHS). HCFA’s Medicare data are used to examine per capita spending by age group for the elderly from 1966 to 1998. Because NCHS data cover the entire population, they are used to examine trends in hospital use by the elderly relative to middle-aged persons (ages 45–64). In addition, NCHS data capture hospital use by Medicare health maintenance organization (HMO) enrollees, for whom Medicare program data are not yet available.¹²

We used published tabulations from HCFA’s Annual Person Summary File for per capita Medicare spending for 1966–1969 and 1973 by age, sex, and type of service. Data by age group and type of service were not available for 1970–1972. Data beginning with 1974 are available from the Continuous Medicare History Sample, a longitudinal, 5 percent, person-based sample of Medicare beneficiaries chosen using their Medicare identification numbers. We used this file for 1974–1998 because it gave us the flexibility to exclude certain groups of Medicare beneficiaries.

First, HMO members were excluded. Since HCFA does not receive complete claims information for HMO enrollees (16 percent of all beneficiaries in 1998), it would be incorrect to include them in the denominator. Claims received for them are excluded from the numerator. Second, the study population was limited to persons with both Parts A and B of Medicare (about 95 percent of all beneficiaries). Beginning in January 1983 a series of laws made Medicare the secondary payer for the "working aged"—persons entitled to Medicare who worked, or whose spouse worked, for an employer offering health insurance. As with HMO members, it would be incorrect to include them in the denominator because most of their health care expenses would be paid by employer coverage, not Medicare. Since these persons would have no reason to purchase Part B, we have limited our study population to beneficiaries with both Parts A and B. However, some working elderly have had Part B because the 1984 provision allowing them to delay Part B enrollment without penalty has sometimes been misunderstood.

It is important to remember that Medicare data reflect only payments for Medicare-covered services. Important services not covered by Medicare are prescription drugs and most nursing home care. Medicare pays slightly more than half of the total expenses incurred by Medicare beneficiaries.

Data from the NCHS’s National Hospital Discharge Survey (NHDS) are used to track hospital use by the elderly and compare it to trends among the middle-aged (ages 45–64) from 1965 to 1998. The middle-aged are somewhat similar in their morbidity profile to the elderly, but they differ in insurance coverage. The NHDS is a probability sample of hospital discharges from short-stay, nonfederal hospitals. All discharges, regardless of HMO membership, are included.

Since our focus is on comparing trends by age group, we used the ratio of per capita rates of use and costs for one age group to another. Medicare payment ratios were calculated by dividing the age-specific rate for the older elderly by the rate for the youngest elderly (ages 65–74). The per capita costs are not adjusted for inflation, but since we are interested in the ratio of per capita costs in the same year, the comparisons are not distorted by inflation.

Our Medicare data do not include Medicare disabled enrollees under age sixty-five. In 1998 there were five million such enrollees. Their experience, while not broken out separately, is reflected in the NHDS data on hospital use and in the national health expenditures data used for Exhibit 1.

Top Data And Methods Study Findings Discussion And Policy... NOTES

View larger version (16K): [in this window] [in a new window] EXHIBIT 2 Hospital Discharges Per 1,000 Population, By Age Group, 1965–1998

View larger version (31K): [in this window] [in a new window] EXHIBIT 3 Medicare Per Capita Payments, By Age Group, 1966–1998

View larger version (32K): [in this window] [in a new window] EXHIBIT 4 Medicare Per Capita Payments For Home Health Agency (HHA) Services And Skilled Nursing Facility (SNF) Care, By Age Group, 1966–1998

Trends by inpatient hospital procedure and diagnosis. The PPS had a differential effect on discharge rates by diagnosis among the elderly. Discharge rates for cancer dropped by half after PPS, likely reflecting the increasing use of outpatient settings for cancer care (data not shown). In contrast, the rate for heart disease dropped after PPS but resumed growth after a few years, reflecting, in part, the growth in revascularization procedures among the elderly.

We focus on trends in two groups of procedures by age. The first group includes four procedures designed to restore functioning, which became widespread during the past two decades: coronary angioplasty, coronary artery bypass graft (CABG), carotid endarterectomy, and knee replacement. The second group includes two procedures often performed on ill, often frail persons, to support life: gastrostomy (inserting a feeding tube into the stomach) and continuous mechanical ventilation (assisting persons who are having trouble breathing).

Rates for all four procedures in the first group rose considerably (Exhibit 6). For the two revascularization procedures and for carotid endarterectomy, the rates for the 45–64 age group were initially similar to or greater than the rates for the aged, but the rates for the elderly grew faster, as shown by the increase in ratios.¹⁹ The growth in the rates of knee replacement was no faster for the aged than for the middle-aged. By 1998 a sufficient number of all four restorative procedures were performed on persons age eighty-five and older to begin to appear in the data.

Among the oldest old, discharge rates for four frailty-associated diagnoses—septicemia, dehydration (volume depletion), pneumonia, and urinary tract infection—have risen markedly from a total of forty-five discharges per 1,000 persons in 1979 to 109 per 1,000 in 1998. These conditions are related to infections and are common among those already frail or in poor health.²⁰ They now account for 18 percent of all discharges for the oldest old. There also has been a large increase in the elderly death rate for pneumonia and septicemia in this period.²¹

Discussion And Policy Implications

Top Data And Methods Study Findings Discussion And Policy... NOTES

 
In the past three decades health care spending increased for the elderly population at a greater rate than for the group under age sixty-five. This stands to reason since almost all of the elderly are covered by Medicare, while a considerable proportion of the younger population is uninsured (18 percent in 1998).²² Most of the elderly also have some kind of supplemental coverage to pay for Medicare cost sharing.²³ The steady rise in spending on the aged may reflect the incentives indemnity insurance provides to develop new technologies that increase the costs of care.²⁴ Medicare remains largely fee-for-service, while most of those under age sixty-five are now in managed care plans.²⁵ Additionally, there is simply more illness among the elderly and thus more opportunity to apply new technologies, as shown by the increase in restorative procedures in this age group.²⁶ Even under Medicare’s case-based hospital payment system, there are incentives for more intensive treatments.²⁷

Within Medicare, per capita spending has risen most for the oldest old. This increase was entirely attributable to a much greater rate of growth in Medicare-covered postacute services (home health, SNF, and hospice).²⁸ Spending for acute services actually increased at a lower rate for the oldest old than for the younger elderly. For the past three decades death rates for the oldest old declined less than they did for the younger elderly, which should have resulted in increased growth rates for acute care Medicare services for the oldest old. Perhaps the greater frailty of the oldest old limits the amount of intensive treatments applied to them. Also, there might have been an increase in long-term care costs, which are not covered by Medicare.²⁹

Medicare and improved health. Our findings are consistent with the idea that Medicare-funded services have improved the health of the elderly. A whole range of procedures have developed since Medicare began, including CABG surgery, hip replacement, and cataract surgery. On the other hand, Medicare spending for acute care services by the oldest old increased at a slower rate. This indicates that the oldest old have not been in the past and may not be in the future a disproportionate burden on Medicare.

What do the findings imply for the health of older beneficiaries? We noted an increase in restorative procedures among them. We also noted increases in life-support procedures and in hospital discharges for frailty-related conditions. There may be an increasing number of frail persons, who have survived to more advanced age because of medical advances and who are susceptible to infections. Without further study, it is difficult to know what to make of these trends. Although there is evidence that the functional status of the elderly has been improving, it also has been pointed out that the improvement reflected a decrease only in the percentage with lesser disabilities; the percentage with severe limitations has not fallen.³⁰ Additionally, health is a multidimensional concept that cannot be captured by one measure; different health measures may not trend in the same way.³¹ Our findings are suggestive of a steady or growing percentage of severely disabled persons in a population whose overall average health status is improving.

Shifts in response to policy changes. It is important for policymakers to note the extreme plasticity of the health care system in response to changes in Medicare. We have repeatedly seen how abrupt shifts in Medicare utilization trends follow major policy changes. There was a catch-up period right after Medicare began, then a steady rise in both overall spending and in hospital use until implementation of the PPS. The increase in hospital use met the intent of the framers of Medicare to provide secure health care coverage —especially hospital coverage—to the elderly.³² Policy changes drove SNF and home health spending upward from the late 1980s to 1997. However, after BBA payment changes took effect, home health services use dropped sharply, especially among the oldest old. It will be important to monitor patient outcomes to assure that the tightening of the benefit does not harm patients.

These abrupt shifts in Medicare spending following program changes highlight the power that Medicare exercises as the largest single buyer of health care in the United States. Changes in the private sector are more often the result of cumulative system responses to outside market pressures. If Medicare were restructured to increase reliance on private plans, government’s ability to make such abrupt changes might be tempered by countervailing market power, especially if a few large private plans enroll a large percentage of beneficiaries.³³

Need for consistent data. The study demonstrates the usefulness of consistent data on health care use and costs. Both the NCHS and Medicare data used in this study span a number of major changes in health care delivery and financing, such as the movement from fee-for-service to managed care and the changing role of the hospital. However, neither data set contains a complete picture of total health spending by age group and type of service. The main source of such data is HCFA’s occasional studies of health spending by age group from National Health Accounts data. The most recent data on spending by age group from this source were for 1987.³⁴ Therefore, data on health spending by age group should be produced more routinely by HCFA.

Data limitations. Some limitations of the data need to be pointed out. The decline in the ratio of per capita payments for Medicare acute care for the over-85 age group to the 65–74 age group could be attributable in part to differential enrollment by these age groups in Medicare HMOs. That is, younger beneficiaries are more likely to enroll in HMOs, and if they are also healthier—as many studies of biased selection have shown—then there will be a higher proportion of sicker enrollees left in fee-for-service Medicare. This would lower the ratios of per capita payments as HMO enrollment grows. However, these trends predate the large growth in Medicare HMO enrollment. Also, there are important topics not addressed in this paper, such as trends in out-of-pocket payments and in the use of services not covered by Medicare. As noted, more frequent data on health spending by age group would fill this gap.

The debate continues about how much the nation ought to spend on medical care for the elderly and is bound to sharpen as the aging baby boomers put pressure on the Medicare budget. There has been a shift in thinking from the 1960s and 1970s, when there was doubt about the worth of increased spending on the elderly, to the present, when many believe that medical care may be partly responsible for improved health and longevity. The greater growth of spending on the elderly was the result not of a deliberate policy choice but of an array of developments in the health care system such as the passage of Medicare, the failure to pass universal health insurance, and the rise of managed care. Ironically, in view of the unsolved problem of the financial burden of noncovered services, financial protection—not better health—was the intent of Medicare’s framers. If we believe that Medicare has contributed to improved health of the elderly, then program changes that affect the amount of their care must be monitored carefully.

These findings may provide a lesson as two important issues are debated: the future direction of Medicare, and the growing problem of uninsured persons under age sixty-five. Any change in the scope and nature of insurance coverage of one group is likely to affect the types of service received, the allocation of resources among groups, and perhaps the health not only of the affected group but of the rest of the population.

Editor's Notes

 
James Lubitz is a senior researcher at the Office of Strategic Planning, Health Care Financing Administration (HCFA). He did the work for this paper while at the National Center for Health Statistics (NCHS). Linda Greenberg is a social science research analyst in the Research and Evaluation Group, HCFA Office of Strategic Planning. Yelena Gorina is a statistician in the NCHS Office of Analysis, Epidemiology, and Health Promotion. Lynne Wartzman is a senior computer programmer in the Systems, Technical, and Analytic Resources Group, HCFA Office of Strategic Planning; David Gibson is deputy director of that group.

The authors thank Diane Makuc and Amy Bernstein of the National Center for Health Statistics and two anonymous reviewers for their helpful comments.

NOTES

Top Data And Methods Study Findings Discussion And Policy... NOTES

 

1. Board of Trustees, Federal Hospital Insurance Trust Fund, 1999 Annual Report of the Board of Trustees of the Federal Hospital Insurance Trust Fund (Washington: U.S. Government Printing Office, 1999).
2. W.H. Forbes, "Longevity and Medical Costs," New England Journal of Medicine 277, no. 2 (1967): 71–78[Medline]; V.R. Fuchs, " ‘Though Much Is Taken’: Reflections on Aging, Health, and Medical Care," Milbank Memorial Fund Quarterly 62, no. 2 (1984): 143–166; I. Illich, Medical Nemesis: The Expropriation of Health (New York: Pantheon Books, 1976); and D. Callahan, Setting Limits: Medical Goals in an Aging Society (New York: Simon and Schuster, 1987).
3. K.G. Manton, L. Corder, and E. Stallard, "Chronic Disability Trends in Elderly United States Populations: 1982–1994," Proceedings of the National Academy of Sciences 94 (1997): 2593–2598[Abstract/Free Full Text]; and V.A. Freedman and L.G. Martin, "Understanding Trends in Functional Limitations among Older Americans," American Journal of Public Health (October 1998): 1457–1462.
4. National Center for Health Statistics, Health, United States, 2000, with Adolescent Health Chart book (Hyattsville, Md.: NCHS, 2000).
5. Manton et al., "Chronic Disability Trends."
6. K.G. Manton and J.W. Vaupel, "Survival after the Age of Eighty in the United States, Sweden, France, England, and Japan," New England Journal of Medicine (2 November 1995): 1232–1235.
7. B.H. Singer and K.G. Manton, "The Effects of Health Changes on Projections of Health Service Needs for the Elderly Population of the United States," Proceedings of the National Academy of Sciences 95 (1998): 15618–15622.[Abstract/Free Full Text]
8. J.R. Gabel, "Job-Based Health Insurance, 1977–1998: The Accidental System under Scrutiny," Health Affairs (Nov/Dec 1999): 62–74.
9. NCHS, Health, United States, 1999, with Health and Aging Chartbook (Hyattsville, Md.: NCHS, 1999).
10. S.H. Long and R.F. Settle, "Medicare and the Disadvantaged Elderly: Objectives and Outcomes," Milbank Memorial Fund Quarterly 62, no. 4 (1984): 609–656.
11. J.L. Fries, "Aging, Natural Death, and the Compression of Morbidity," New England Journal of Medicine (17 July 1980): 130–135.
12. As part of the implementation of risk adjustment, HCFA has begun to receive hospital encounter data from health plans beginning with July 1997.
13. National Center for Health Statistics, Health Insurance Coverage, United States, July 1962–June 1963, Vital and Health Statistics, Series 10, no. 11 (Hyattsville, Md.: NCHS, 1964).
14. Social Security Administration, Medicare Annual Statistical Series, Section 1: Health Insurance Summary Data, and Section 2: Health Insurance Enrollment Data, for 1968–1969 and 1973; and HCFA, Continuous Medicare History Sample, 1974–1998.
15. B. Gage, "Impact of the BBA on Post-Acute Utilization," Health Care Financing Review (Summer 1999): 103–126; and Prospective Payment Assessment Commission, Prospective Payment System for Medicare’s Skilled Nursing Facility Payment Reform, Congressional Report C-92-01(Washington: ProPAC, March 1992).
16. Medicare Payment Advisory Commission, Report to the Congress: Context for a Changing Medicare Program (Washington: MedPAC, 1998).
17. H.L. Komisar and J. Feder, "The Balanced Budget Act of 1997: Effects on Medicare’s Home Health Benefit and Beneficiaries Who Need Long-Term Care" (Washington: Institute for Health Care Research and Policy, Georgetown University, 1998); and U.S. General Accounting Office, Medicare: Home Health Utilization Expands while Program Controls Deteriorate, GAO/HEHS-96-16 (Washington: GAO, 1996).
18. Board of Trustees, Federal HI Trust Fund, 2000 Annual Report.
19. Rates for carotid endarterectomy dropped in the late 1980s because of doubts, later resolved, about its effectiveness, and have risen since. See M.R. Chassin, "Appropriate Use of Carotid Endarerectomy," New England Journal of Medicine (12 November 1998): 1468–1471.
20. C.M. Callahan and F.D. Wolinsky, "Hospitalization for Pneumonia among Older Adults," Journal of Gerontology: Medical Sciences 51A, no. 6 (1996): M276–M282; and J.L. Warren, et al., "The Burdens and Outcomes Associated with Dehydration among U.S. Elderly, 1991," American Journal of Public Health 84, no. 8 (1994): 1265–1269.[Abstract/Free Full Text]
21. D.L. Hoyert, K.D. Kochanek, and S.L. Murphy, "Deaths: Final Data for 1997," National Vital Statistics Reports 47, no. 19 (Hyattsville, Md.: NCHS, 1999).
22. J.A. Campbell, Health Insurance Coverage, Current Population Reports (Washington: U.S. Census Bureau, 1999), 60–208. David Cutler and Ellen Meara found similar age trends using survey data for 1963–1987. D. Cutler and E. Meara, "The Medical Costs of the Young and Old: A Forty-Year Perspective," in Frontiers in the Economics of Aging, ed. D.Wise (Chicago: University of Chicago Press, 1998).
23. F.J. Eppig and G. Chulis, "Trends in Medicare Supplementary Insurance: 1992–96," Health Care Financing Review (Fall 1997): 201–206.
24. B.A. Weisbrod, "The Health Care Quadrilemma: An Essay on Technological Change, Insurance, Quality of Care, and Cost Containment," Journal of Economic Literature (June 1991): 523–552.
25. Gabel, "Job-Based Health Insurance."
26. Fuchs, " ‘Though Much Is Taken.’ "
27. M. McClellan, "Hospital Reimbursement Incentives and Expenditure Growth" (Working paper, Stanford University, 1996).
28. A fact also noted by Cutler and Meara for 1985–1995. D. Cutler and E. Meara, "The Concentration of Medical Spending: An Update," NBER Working paper no. W7279 (Cambridge, Mass.: National Bureau of Economic Research, 1999).
29. B.C. Spillman and J. Lubitz, "The Effect of Longevity on Spending for Acute And Long-term care," New England Journal of Medicine (11May 2000): 1409–1415.
30. B.C. Spillman and L.E. Pezzin, "Potential and Active Family Caregivers: Changing Networks and the ‘Sandwich Generation,’ " Milbank Quarterly 78, no. 3 (2000): 347–374.[Medline]
31. E.M. Crimmins, "Mixed Trends in Population Health among Older Adults, "Journal of Gerontology 518, no. 5 (1996): S223–S225.
32. U.S. House of Representatives, Committee on Ways and Means, "Medical Care for the Aged: Hearings before the Committee on Ways and Means," 88th Cong., 1st and 2d sess., on H.R. 3920: A Bill to Provide under the Social Security Program for Payment for Hospital and Related Services to Aged Beneficiaries, Part 1 (18–20 November 1963).
33. H.J. Aaron and R.D. Reischauer, "The Medicare Reform Debate: What Is the Next Step?" Health Affairs (Winter 1995): 8–30.
34. Our 1995 data came from P.A. Hodgson and H.A. Cohen, "Medical Care Expenditures for Major Diseases, 1995," Health Care Financing Review (Winter 1999): 119–164.

                      What's this?

This article has been cited by other articles:

				G. F. Riley Long-Term Trends In The Concentration Of Medicare Spending Health Aff., May 1, 2007; 26(3): 808 - 816. [Abstract] [Full Text] [PDF]

				E. Meara, C. White, and D. M. Cutler Trends In Medical Spending By Age, 1963-2000 Health Aff., July 1, 2004; 23(4): 176 - 183. [Abstract] [Full Text] [PDF]

				K. Im, S. H. Belle, R. Schulz, A. B. Mendelsohn, and L. Chelluri Prevalence and Outcomes of Caregiving After Prolonged ("=" BORDER="0"> 48 Hours) Mechanical Ventilation in the ICU Chest, February 1, 2004; 125(2): 597 - 606. [Abstract] [Full Text] [PDF]

				B. Kisloff Implications of Regional Differences in Spending Ann Intern Med, January 20, 2004; 140(2): 145 - 146. [Full Text] [PDF]

				T Shinoda-Tagawa and D E Clark Trends in hospitalization after injury: older women are displacing young men Inj. Prev., September 1, 2003; 9(3): 214 - 219. [Abstract] [Full Text] [PDF]

				N. G. Castle Searching for and Selecting a Nursing Facility Med Care Res Rev, June 1, 2003; 60(2): 223 - 247. [Abstract] [PDF]

				J. M Wiener and J. Tilly Population ageing in the United States of America: implications for public programmes Int. J. Epidemiol., August 1, 2002; 31(4): 776 - 781. [Full Text] [PDF]