Monetary incentives shape the kind of health care Americans get. Free-spending fee-for-service brought us bountiful, patient-centered medicine and runaway inflation. Soon, if Congress has its way, millions of older Americans will find themselves in an unfamiliar system where a fixed number of dollars must be alloca-ted among many. Understanding how the system works may be one key to a healthier old age.
The failure of President Clinton's effort to restructure American health care seemed to write finis to major changes in that sector of public policy for a long time to come. Yet in the last two years, the health care landscape has been transformed. The traditional system of "fee-for-service" medicine is rapidly giving way to "managed care."
And now Medicare, the final bastion of fee-for-service, is under powerful political and economic pressures to move its 37 million retired and disabled Americans into managed-care arrangements.
Noisy partisan battles in Congress aside, leaders of both parties have seen the writing on the wall and are trying to head off a crisis in Medicare funding. At some point in the not too distant future--estimates vary--Medicare will be broke, overwhelmed by rising costs, the retirement of the baby boomers, and the fact that older Americans are living longer than previous generations and ultimately requiring more--and more expensive--care. Whether the estimates are accurate or off by a decade is beside the point--before long, market forces will have thrust traditional fee-for-service care beyond the reach of all but the wealthiest.
Managed care in some form is the future, and we are heading toward it at breakneck speed along a road that has many blind curves. What does all this mean for aging Americans? Can they count on getting the same excellent health care after retirement that their parents now enjoy? The answers will depend on how skillfully society accomplishes the transition to a system of medicine that is primarily driven not by the preferences of patients and their doctors, as in the past, but by the need to control costs.
Under the private fee-for-service system, doctors and hospitals bill insurers for each service they provide. The more doctors and hospitals do, the more they are paid. Although some cost-control mechanisms now exist in fee-for-service, years of unconstrained spending helped push U.S. health costs from 6 percent of national income in 1965 to 14 percent in 1994.
Managed care, however, seeks to control costs by the efficient management of health care resources. Some of the ways it does this are utilization review ("second opinion" panels), the use of primary-care ("gatekeeper") physicians to limit access to higher-paid specialists, financial incentives that reward doctors for holding down costs, and in general offering patients fewer services than they would get in a fee-for-service setting.
Health maintenance organizations (HMOs)--the most familiar type of managed care--have been around a long time. What is new is the rate of growth in HMO enrollment and the proliferation of many new and untested managed-care arrangements. Almost overnight there has emerged a bewildering assortment of nonprofit and for-profit managed-care plans involving various combinations of employers, insurance companies, hospitals and physicians.
Some of the earliest data on HMOs come from the massive RAND Health Insurance Experiment, launched in 1971. Although HMOs represented only a tiny segment of the health care market at the time, they were beginning to attract notice, and so one large, well-established HMO was included in the study.
The major finding--that large, staff-model HMOs are able to control costs and still provide care as good as that in the fee-for-service system--still stands. How the many new and competing forms of managed care will perform and affect consumers remains to be seen.
Despite the good news about HMOs, researchers in the health insurance study warned that there are problems inherent in any health care system that elevates cost containment to the level of a principle. A bottom-line emphasis could lead to denial of access or erosion in the quality of care for vulnerable populations like the poor, the old and the chronically ill.
The great virtue of managed care in the eyes of policymakers is its promise to curb the growth of health costs. How well has it lived up to its reputation? Definitive answers have been hard to come by, but a recent RAND study of hospital competition in California by economists Glenn A. Melnick and Jack Zwanziger moves the debate several giant strides forward. Their work was funded by grants from The Robert Wood Johnson Foundation and the U.S. Department of Health and Human Services.
In June 1982, California adopted trendsetting legislation designed to maximize competition in the health sector by encouraging the growth of managed-care plans. Over the decade, enrollment in these plans exploded from under 20 percent to over 85 percent of the state's insured population.
Melnick and Zwanziger found that health expenditures in California have grown at a dramatically slower rate than those in the country at large, not only overall but within every major category of health spending.
Despite the positive effects of competition on costs, Melnick warns: "There has been almost no research on how managed-care plans are able to achieve their cost savings and whether the savings come from increased efficiency or from reduced access or quality or both."
Moreover, he notes, California has had 13 years to work out the kinks of a transition to managed care. But sweeping millions of elderly overnight into managed-care plans would strain even California's vast network and might overwhelm the managed-care resources of many other states. "The Medicare population," Melnick commented, "has different needs than the working population--for geriatricians, for example, who understand the diseases of aging--so an effective, efficient program would have to be properly researched and structured."
To understand what is happening to health care in this country requires an understanding of how doctors and hospitals are paid, a formidably complex affair.
In a fee-for-service system, very few spending constraints operate upon the consumer because, in the main, the money does not come directly from his pocket. Insurance firms pay, employers pay, and in the case of Medicare and Medicaid, government pays. Until fairly recently, doctors could order tests, referrals to specialists, and hospital services without concern for the cost, and hospitals were free to charge pretty much what the traffic would bear. The result has been a rampant, some say ruinous, inflation in medical costs.
Managed care turns that approach on its head, literally: Health plans are paid the same "per-capita" amount for each enrollee. If the total amount spent on health care for the enrollees exceeds what the health plan collects in capitation payments, the health plan and the providers of care (doctors and hospitals) stand to lose money. Ideally, such payment constraints should ensure that health plans provide only medically necessary and appropriate care to the patient in the least costly setting.
But the ideal may prove elusive. While it is true that a capitation system instills a healthy respect for the bottom line, it also contains incentives that may lead to undesired outcomes. For example, it is in a managed-care plan's economic interest to limit the number of very sick and very old patients (high users of resources) or to scrimp on their care. Health plans that enroll only healthy patients would obviously enjoy a financial advantage over other plans that cover providers (like inner-city hospitals or teaching hospitals) obliged to accept a disproportionate number of very ill patients.
So the challenge is to design a system of payments that will keep financial incentives harnessed to a triad of desired ends: containing costs, to be sure, but also maintaining a high quality of care and protecting access to the system for those, like the elderly, who are in danger of slipping through the cracks.
Here, too, research can play a crucial role. Market forces work to wring cost savings from the health care system, and increased competition among managed-care plans should serve to improve the quality of care. But the market alone will not assure quality. Uniform standards of care must be established and government watchdog agencies put in place to ensure that they are followed. And that is possible only when "quality" can be defined and tools can be devised to measure it.
RAND has been at the forefront of research on quality assessment--developing instruments that guide physicians in determining when a given procedure is appropriate and necessary, as well as devising others to
Using these tools, researchers have learned a great deal about cost-cutting trade-offs in the health care system; by no means is all of the news bad.
In 1983, in an effort both to curb costs and to place a more rational value on services, Medicare introduced a radical reform of its payment system. Instead of reimbursing hospitals after the patient is discharged for whatever costs are incurred, Medicare now pays a predetermined amount for each admission (similar in some respects to a managed-care payment scheme). That sum represents the cost of the services that Medicare expects will be rendered in light of the patient's diagnosis. This new system changed the incentives, putting pressure on hospitals to limit services and to keep hospital stays short. It thus raised concerns that elderly patients might suffer and perhaps die needlessly if they were sent home too soon.
After the reforms had been in place for several years, the Health Care Financing Administration (HCFA), which oversees Medicare, contracted with RAND to evaluate the new system's effect on quality of care for the elderly. The research team found that this form of "prospective" payment saved money by dramatically reducing the number of hospital days charged to Medicare. And although some patients were being discharged before they were stable, the majority received good care and came to no harm as a consequence of the shorter hospitalizations. The study also flagged for HCFA the danger of elderly patients being discharged in unstable condition (they die at a higher rate) and dramatized the need to keep a continued watchful eye on the effects of the reforms.
The larger lesson to be drawn from the study--and one that directly applies to managed-care plans serving increasing numbers of elderly--is that cost-cutting is not necessarily the enemy of quality. It is possible to have both, provided that the adverse effects of the cost-savings are identified early and ways are found to ameliorate them.
This brings us back to the power of financial incentives to shape the health care system for good or ill--and the difficulties of steering a course between the Scylla of too few constraints on spending and the Charybdis of too many.
Chronic disease is the scourge of the old. Yet the United States has little experience with an HMO-style payment system for people with these intractable health problems. Because costs are unpredictable and can vary enormously from individual to individual, the chronically ill elderly are not well served by a "one-size-fits-all" payment. An important step toward a solution has been the design by RAND of a flexible capitation strategy for Medicare patients with end-stage renal disease (ESRD), the permanent loss of kidney function.
Kidney failure is a chronic condition that requires expensive in-hospital and outpatient treatment. A research team led by Donna O. Farley has developed a payment scheme for kidney disease that aims to control Medicare costs while maintaining access to treatment and quality of care for these patients. As such, it could serve as a managed-care payment model for the chronic diseases of the elderly. Using analytical modeling techniques and the rich data collected by HCFA, researchers devised a payment structure that combines these elements:
The dialysis payment is risk-adjusted; that is, patients with certain characteristics (diabetes, for example) that are predictive of a greater use of resources command a higher payment. The risk adjustment of payments is intended to reduce health plan incentives to select less-expensive patients. The kidney transplant payment is a fixed monthly amount because these costs are thought to be fairly predictable.
This payment structure was chosen because it recognizes the uniqueness of transplant and graft failures as unusually expensive events by paying for them separately. Kidney transplantation is a desired treatment option for qualified patients. Preset monthly payments would put health plans at high risk for the extraordinary transplant costs, which would discourage plans from using transplantation as a treatment option. A separate fixed payment pays directly for expected transplant expenses while maintaining incentives for cost-effective care. Similar logic applies to graft failure, for which every transplant patient is at risk.
One of the thorniest problems in designing a payment scheme for managed care (or for any medical setting where individual payments are predetermined) is how to mitigate the financial risk to health plans of catastrophically expensive patients. If the predetermined amount is too low to cover these cases--outliers, as economists call them--payers and providers may be tempted to cut back on necessary care or substitute less expensive procedures that may not be appropriate for the patient.
The researchers developed a "fixed-loss" outlier policy to deal with the problem. Outlier payments are paid to all cases whose medical expenses during a calendar year exceed the capitation payment by $50,000. The outlier payment equals 75 percent of the expenses in excess of this threshold.
The kidney disease study graphically illustrates the complexity of balancing reasonable but competing societal goals and suggests just how daunting managed care can be for the elderly, who nearing the end of life may lack the vigor and determination to fight for what they need from the health care system.
Other countries RAND has studied have already made their cost-care trade-offs. Canada, a modern Western nation like our own, spends a great deal less on heart bypass surgery for people over 65. The ESRD example shows that there are alternatives to limiting access simply on the basis of age.
In a well-designed capitated payment system, health plans would be perfect agents that match the use of health care resources to individual patient needs, to achieve the best possible outcomes for any given level of costs and payments. While we may never reach that ideal, the work RAND has done on health care finance and quality shows that if society has the will, research knows the way to bring us much, much closer.
The Effects of the DRG-Based Prospective Payment System on Quality of Care for Hospitalized Medicare Patients: Final Report
Katherine L. Kahn et al., RAND/R-3931-HCFA, 1992, 343 pp., ISBN 0-8330-1220-7, $25.00.
Designing a Capitation Payment Plan for Medicare End Stage Renal Disease Services
Donna O. Farley, Joel D. Kallich, Grace M. Carter, Thomas W. Lucas, Karen L. Spritzer, RAND/MR-391-HCFA, 1994, 122 pp., ISBN 0-8330-1512-5, $13.00.
"Comparison of the Appropriateness of Coronary Angiography and Coronary Artery Bypass Graft Surgery Between Canada and New York State,"
Elizabeth A. McGlynn, C. D. Naylor, G. M. Anderson, et al., Journal of the American Medical Association, Vol. 272, No. 12, September 28, 1994, pp. 934-940 (also available as a RAND reprint, RP-344, no charge).