Click for next page ( 94


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 93
Impact of the Changing Medical Payment System on Technological Innovation and Utilization STUART H. ALTMAN When Medicare legislation was drafted in 1965, legislators were determined to control health care costs. Not unreasonably, they decided that the best way to control federal spending on health care was to pay only for the cost of the care that was provided. Legislators reasoned that there are only a few ways in which a particular illness can be treated; if hospitals are paid only for services rendered, they will not make profits from Medicare and costs will be contained. Since 1965, however, hundreds of analyses and documents have been prepared demonstrating that cost-based reimbursement had the opposite effect. It created a set of economic incentives that rewarded spending and penalized attempts by hospital managers to provide medical care at lower cost. Annual health care spending rose from about 5.5 percent of our gross national product before Medicare was enacted to more than 11 percent in 1987. Some of this additional spending may indeed be beneficial, but there are also many examples of questionable health care expenditures. For the past 10 years, health care economists and others have warned policymakers, medical practitioners, and the public that the U.S. health care system could collapse if such spending increases continued. Well, the system has not collapsed, but it continues to prosper at the expense of other needed federal services, such as mental health care and guaranteed incomes for the poor. Our gross national product represents the total capacity of our country to purchase needed goods and services. When we double the amount spent on health care, there are fewer resources available for other needs. Although there is 93

OCR for page 93
94 CURRENT TRENDS no magic number that represents an appropriate amount for the United States to spend for health care services, the question of economic and social trade-offs remains important. PROSPECTIVE PAYMENT SYSTEM Congress debated the issue of how to control spending on health care throughout the 1970s without a significant change in the way hospitals were reimbursed by the federal government. In 1982, how- ever, Congress made a fundamental change in the hospital payment system for Medicare. It created the Prospective Payment System (PPS). PPS was heralded by its supporters as a system that would put in place incentives that would make hospitals more cost efficient, a change that was long overdue. On the other hand, some individuals were concerned that there would be problems with PPS. For example, critics feared that many hospitals would go bankrupt and patients would be asked to leave hospitals while they were still in need of care. Before 1982, Medicare paid for hospital services based on the actual cost of providing that care to Medicare patients. As such, costs varied from hospital to hospital, even for patients being treated for the same illness. PPS changed this by removing the direct link between cost and payment. Under PPS, the payment for each hospital patient is based on the categorization of the patient's illness into one of 468 diagnosis-related groups (DRGs). After a 1-year phase-in, the payment amount for each DRG reflects the average cost of treating patients in that category throughout the United States. Hospital payments do vary, however, to reflect their wage area, the extent to which they are a teaching hospital, whether they are in an urban or rural area, and whether they provided a disproportionate amount of care to low- income patients. As part of the PPS legislation Congress created the Prospective Payment Commission (ProPAC) to advise it and the executive branch on how to make PPS responsive to changing health care technologies and procedures. The Office of Technology Assessment is responsible for selecting 17 individuals representing various groups involved in the health care system to serve on the commission. I was selected to be its chairman. ProPAC began operation in 1984. Since its inception, it has devoted a major portion of its time and all the time of its 25 staff members to developing an appropriate pricing system for about a dozen new technologies and medical procedures which have become accepted in our medical system. In this paper, I will briefly review the issues surrounding three of these technologies.

OCR for page 93
IMPACT OF THE CHANGING MEDICAL PAYMENT SYSTEM 95 Economists have been very successful in teaching those involved in public policy decision-making processes two fundamental lessons: (1) economic resources are finite and (2) economic incentives matter. It is time that policymakers learn two more economic principles. The first is the concept of elasticity of supply and demand. For example, if the revenue of a product to the manufacturer is reduced (holding the cost of production constant), there is an incentive for manufacturers to produce less of that product but how much less? The larger the elasticity of supply, the greater will be the reduction in output. Similarly, we know that if the price of a product to the consumer increases, there is a tendency to buy less of that product- but how much less? What is the elasticity of demand? When we suggest that there are incentives under PPS to use less health care technology, such a statement says nothing about how strong such incentives are and whether they operate across all hospitals. Unfortunately, we almost never talk about the strength of the incentives or the degree of response, which is the elasticity of supply. The second economic principle concerns income and substitution effects. We know that when the price of a good changes, there is likely to be a change in the quantity consumed. Where the price goes up, people usually use less of the good, and when it goes down, people, on average, use more. This is the substitution effect. On the other hand, when the price of a good goes up but one's income also goes up, one might ultimately use more of that good. Despite an incentive to use less of a more expensive good, increased wealth generates an opposing incentive to buy more. This is the income effect. These principles apply to the hospital industry under PPS. Under the previous cost-based system, a hospital was reimbursed for whatever services it used on a patient. PPS, however, increased the cost of using new medical technology because hospitals faced a fixed budget per patient. Once a DRG category has been established and per-patient reimbursement decided, any extra service or use of an additional test or procedure adds to the cost of treatment but not to the hospital's revenue. Many analysts initially believed that the negative impact of the changed reimbursement system on the purchase and use of medical technologies would be severe. That is, the substitution effect would operate to reduce the acquisition and use of new medical technologies in hospitals. However, during the first few years of PPS, there have been substantial increases in hospital revenues, which may have made it possible for hospitals to purchase more technology not less. It is as yet uncertain what the cumulative impact of PPS will be on the

OCR for page 93
96 CURRENT TRENDS purchase and use of medical technologies. In economic terms, will the positive income effect overpower the negative substitution effect? New Incentives Under the Prospective Payment System Under PPS, hospitals are paid a fixed payment per diagnosis, and that payment is known beforehand. Therefore, there is an incentive for hospitals to minimize the costs of treatment, including the use of expensive new medical technologies. But the existence of such an incentive does not mean that patients will not receive needed services or that new medical technologies will not be used. Instead, it means that there is some tendency for hospitals to act in that direction. Often, we hear concerns that the health care system should not cut back too much on those technologies that improve quality of life or the outcome of a medical intervention while adding to health care costs. Congress, too, had similar concerns and instructed ProPAC to advise it on how to introduce new medical technologies into the PPS pricing system. But the technology issue is more complicated and needs more sophisticated analysis than simply saying PPS increases the price of new technology and, therefore, much less technology will be used. For example, there is another incentive that operates in the opposite direction. PPS provides a powerful incentive for hospitals to seek more admissions. The best way to get more admissions is to have doctors who are willing to admit patients to that hospital. The best way to encourage doctors to admit patients to a particular hospital is to make sure that the hospital is up-to-date and has the latest technologies and equipment. MEDICAL TECHNOLOGIES, HEALTH CARE COSTS, AND THE ROLE OF ProPAC In 1977, under the auspices of the Robert Wood Johnson Foundation, a conference was held in Sun Valley, Idaho, to look at the impact of medical technology on health care costs. Participants concluded that no one knew. We could document several areas where technology had clearly increased costs. But among the cost-increasing technologies, it was unclear whether technologies with the largest acquisition and operating costs were really the biggest culprits. There were also examples of technologies that had led to substantial reductions in medical costs. When the positive and negative examples were summed, the total impact was unclear. This finding went against the prevailing wisdom that said we could

OCR for page 93
IMPACT OF THE CHANGING MEDICAL PAYMENT SYSTEM 97 control escalating health care costs by limiting the use of expensive technologies. Although the evidence did not support it, the general impression in most public policy forums remained that medical tech- nology was overused in the United States and that tough planning and certificate-of-need legislation was needed to control its use. This assumption led to considerable efforts by state and federal planners to limit the availability of a then emerging, new, and expensive technology, computerized tomographic scanners (CTS), and other expensive technologies as well. It is possible to classify medical technologies into one of four categories. The first type is one that reduces costs and improves the quality of medical care. The pressure to expand the use of these technologies under PPS is substantial. The second type of technology is one that increases costs to the hospital but reduces costs to the health care system. These technologies are troubling under PPS because the DRG system applies only to hospital care. For example, rehabilitative technologies can result in patients feeling better quicker and, therefore, getting back to work faster. But hospitals which use these technologies only see more costs and no added revenue. The third type of technology is one that increases costs to the hospital and the health care system, but also increases the quality of medical care. Here the question is whether the added quality justifies the added costs. Finally, there are those technologies that increase cost but affect quality in only a limited way. Some critics would further argue that there are medical technologies that increase costs and decrease quality. I would like to believe that, over time, such technologies are eliminated by the medical profession itself. But there are knowledgeable people who believe that this is not the case and that a significant amount of money is spent each year on useless and harmful technologies. In any case, there can be no serious objection to a substantial reduction in the use of such technologies. For ProPAC, the major concern is to focus on the second two types of technologies: those that increase costs to the hospital but decrease costs to the system, and those that increase both costs and quality. ProPAC commissioners spend much time worrying about these technologies. In general, the commission wants the price system under PPS to be neutral: The payment amount should neither retard the introduction of beneficial technologies nor promote their overuse. Three new medical technologies that ProPAC has reviewed or is currently studying are cardiac pacemakers, magnetic resonance im- aging, and penile prostheses.

OCR for page 93
98 CURRENT TRENDS Cardiac Pacemakers There are several types of cardiac pacemakers. Some have one chamber and are relatively simple to operate. In previous years, these single-chamber devices were not as expensive as the dual-chamber, programmable pacemakers that are now available. Unfortunately, the DRG system does not recognize differences in price among the various types of pacemakers; reimbursement is the same regardless of the type of device implanted. Led by the pacemaker industry, ProPAC was educated to this issue and reviewed PPS reimbursement for this technology. The alternatives are few. If each device is priced differentially by the reimbursement system, then Medicare is returning to cost-based reimbursement, and that is what DRGs were supposed to stop. The current system, however, is blind to the differences in costs associated with different types of devices. Further, although it costs the hospital more to implant a more expensive device, it does not cost the physician more. Quite the opposite. The physician receives more money for implanting the more complicated device; the physician's incentives may, therefore, be opposed to those of the hospital. What kinds of incentives does that set up? To the extent that the hospitals face an incentive to implant the least expensive device that will produce a desired medical result, that is to be preferred. But what if the incentive is so strong that hospitals implant devices which produce inferior results? The commission has been very concerned about this potential and decided that the PPS reimbursement approach for pacemakers was not correct. In 1986 ProPAC recommended that there be different payments for single-chamber and dual-chamber devices since these were two distinct technologies and PPS should reimburse them at different rates. The Health Care Financing Administration (HCFA), however, did not go along with this recommendation. In 1987 the commission again looked at the evidence and found that reimbursement for pacemakers under PPS has become even more complicated. The distinction between single- and dual-chamber devices now appears to be less meaningful: There are more expensive single- chamber devices and cheaper dual-chamber devices. There is actually more price variation within the two pacemaker types than between them. Also, evidence suggests that the current reimbursement system has not yet had a major impact on the type of pacemaker chosen by physicians. Therefore, the commission rescinded its 1986 recommen- dation and decided to continue to study the issue of appropriate reimbursement for pacemakers.

OCR for page 93
IMPACT OF THE CHANGING MEDICAL PAYMENT SYSTEM 99 Magnetic Resonance Imaging Another issue ProPAC is currently examining is how to pay for the operating costs of magnetic resonance imaging (MRI). After consid- erable study, HCFA decided that MRI is a useful medical technology. The next question HCFA faced was how to reimburse MRI use under PPS. Under the current system, the purchase price of this new technology is paid for by a capital-cost pass-through. But there is no reimbursement for the higher operating costs of this procedure. MRI scans are used as a diagnostic test for many different diseases, so it is not possible to establish a separate DRG category for MRI. Fortunately for manufacturers of MRI equipment and for many patients, an MRI scan is often performed as an outpatient procedure and is reimbursable under different rules. But there are Medicare patients who must be hospitalized and who need an MRI scan. For those patients, the hospital receives no additional payment above the appli- cable DRG rate if an MRI scan is performed. Again, ProPAC believes that it is inappropriate to introduce a valuable new technology and not develop a mechanism to reimburse hospitals for that technology's operating expenses. ProPAC's goal is neither to encourage the use of MRI nor to discourage its availability and use. To achieve their goal and to develop a better understanding of the uses and economics of this technology, ProPAC recommended that there be a temporary add-on payment each time an MRI imager is used. The add-on amount ProPAC proposed was modest and was based on what the reimbursement level should be if imagers are being used at their most efficient level. ProPAC's proposal violated some basic tenets of the DRG system: ProPAC recommended that the payment system be related to the use of a medical device and not a fixed amount per diagnosis. But the commission faced a difficult choice. If it did nothing, this technology would not get paid for and may become underused, particularly if in the future capital expenditures also are included in the DRG payment. Although ProPAC recommended in 1986 and again in 1987 that HCFA develop an MRI add-on for a 3-year period, HCFA and the Congress have not yet accepted this recommendation. Penile Prostheses The third technology or device I will discuss is an appropriate payment system for a new type of penile prosthesis. Again, the implanting of penile prostheses is a procedure for which there is one

OCR for page 93
100 CURRENT TRENDS payment amount regardless of the type of device used on the patient. A new type of prosthesis is now available that apparently works much better but costs more than the old one. The prospective payment system does not fully recognize that change in the technology and pays a weighted average of the costs of the various devices that are available. A number of urologists have petitioned ProPAC claiming that the present PPS payment is preventing their patients from receiving what they (the physicians) and their patients believe is the correct treatment. Just as with cardiac pacemakers, a Medicare patient cannot go to a hospital and request the more expensive device and offer to supplement the government's reimbursement. Only patients who do not accept any Medicare reimbursement can select the technology of their choice. However, they must pay the total hospital bill. Patients who are not wealthy must accept the technology that the hospital and their doctor determine is appropriate. There are those who say that if PPS becomes even more restrictive, we will see increasing limitations on what Medicare will pay. This will be followed by added pressure to allow patients to supplement the basic reimbursement. Others voice concern about creating a two-class Medicare program in which poorer patients must accept the treatment dictated by Medicare and wealthier patients can choose to supplement that treatment. If no supplementation is allowed and PPS does become more restrictive, then the technological issues reviewed in these three examples will become increasingly important. IMPACT OF PPS ON M EDICAL TECH N OLOGY ProPAC believes that the DRG system needs to be more responsive to new medical technology or there will be inappropriate reductions in the use of these services. But to make this case, the commission needs to show the negative effects of the existing DRG system on development and use of medical technologies. What impact or leverage has medical care reimbursement under PPS had on the use, and ultimate manufacture, of new medical technologies? Unfortunately, we still know very little about what has happened. Prospective payment for Medicare was introduced in 1983, and by 1987 we had just begun to understand what happened in 1983-1984. Although there have been financial problems for some hospitals, particularly rural hospitals, the overall American hospital system appears to have adjusted well to the introduction of PPS. To determine how well hospitals have fared under PPS, ProPAC reviewed changes in hospital revenues and expenses since DRGs were put in place. From

OCR for page 93
IMPACT OF THE CHANGING MEDICAL PAYMENT SYSTEM TABLE 1 Change (in percent) in Hospital Expenses, 1976-1986, Adjusted by the Consumer Price Index (in percent) 101 Total Inpatient Expenses Total Expenses Inpatient per: Year Expenses per Capita Expenses Capita Admission 1976 12.6 11.5 12.2 11.1 8.6 1977 8.6 7.5 8.2 7.1 5.5 1978 4.8 3.7 4.3 3.2 3.9 1979 1.9 0.8 1.8 0.7 - 0.8 1980 3.0 1.8 2.9 1.6 0.0 1981 7.6 6.5 7.2 6.2 6.3 1982 9.1 8.1 8.9 7.8 8.8 Average increase, 1976-1982 6.8 5.7 6.5 5.4 4.6 1983 6.8 5.8 6.1 5.1 6.7 1984 0.6 - 0.4 - 0.7 - 1.6 3.1 1985 2.7 1.8 0.5 - 0.4 5.7 1986a 7.0 6.2 5.1 4.2 7.7 Average increase, 1983-1986 3.5 2.5 1.6 0.7 5.5 aEstimate based on the first 8 months of 1986 compared to the first 8 months of 1985. SOURCE: American Hospital Association National Panel Survey. 1976 to 1982, inpatient expenses, after adjusting for overall inflation, went up 6.5 percent (Table 1) and hospital inpatient revenues went up by an average annual rate of 7.1 percent (Table 21. Under cost-based reimbursement before PPS, therefore, hospital revenues per admission were increasing about 9 percent faster than expenses. During 1983, the first year of PPS, both revenues and expenses grew by approximately 7.0 percent. But in 1984, inpatient expenses increased 3.1 percent and inpatient revenues grew by 4.2 percent. Two facts about 1984 are important. First, there was a substantial drop in both hospital revenues and expenses. But, more important, hospital reve- nues increased more rapidly than expenses. That was the year when hospital administrators refrained from purchasing new medical tech- nologies and forced price concessions from hospital suppliers, ex- pecting a tight limit on their revenues. But hospitals found that reimbursement revenues from Medicare did not fall nearly as fast as they had expected. In fact, during the first year of PPS, actual per-admission costs per Medicare patient were 14.8 percent lower than Medicare revenues. This is a substantial difference, considering that each 1 percent difference is equal to $400 million. Before PPS, Medicare revenues and costs were increasing at

OCR for page 93
102 CURRENT TRENDS TABLE 2 Change (in percent) in Hospital Revenues, 197~1986, Adjusted by the Consumer Price Index Revenues per: Year Total Inpatient Outpatient Other Inpatient Outpatient Revenues Revenues Revenues Revenues Admission Visit 1976 13.6 13.9 17.2 - 0.1 10.2 13.5 1977 9.2 9.2 12.8 0.3 6.4 6.3 1978 4.7 4.0 8.2 8.9 3.6 7.7 1979 2.3 2.1 3.0 5.4 - 0.6 3.4 1980 3.7 3.7 5.1 - 0.4 0.9 2.0 1981 7.7 7.1 9.5 13.3 6.3 8.0 1982 9.6 9.5 11.3 6.7 9.5 10.1 Average increase, 1976-1982 7.3 7.1 9.6 4.9 5.2 7.3 1983 6.8 6.4 11.0 2.2 7.0 8.0 1984 1.7 0.3 9.3 3.8 4.2 7.8 1985 2.4 0.0 14.3 8.5 5.1 9.3 1986a 5.8 3.7 15.4 7.9 6.4 6.7 Average increase, 1983-1986 4.2 2.6 12.5 5.6 5.7 8.0 aEstimate based on the first 8 months of 1986 compared to the first 8 months of 1985. SOURCE: American Hospital Association National Panel Survey. about the same rate. According to the Congressional Budget Office, these higher-than-expected "margins" i.e., revenues minus costs- will generate $27 billion in excess hospital revenues over 5 years, or approximately $5 billion in excess revenues per year. When we examine the effect of PPS on specific medical technologies, we find that the use of higher-cost pacemakers has been growing faster than the use of their less expensive counterparts by a margin of four to one, a phenomenon many would not have expected when PPS was introduced. When we look at the newer, more expensive penile prosthesis, the same thing is happening. And MRIs are being purchased by hospitals almost as rapidly as one would have expected before PPS. Even though Medicare is not providing larger reimbursements to hospitals for use of higher-cost devices and incentives therefore remain to use lower-cost technologies the introduction and use of expensive new medical technologies appears to be continuing to grow. Why has this happened? Is it because economic incentives do not matter? I do not think so. I believe it is because of the high Medicare per patient margins earned. For hospitals, the income eject (i.e., larger than expected Medicare revenues) appears to have swamped

OCR for page 93
IMPACT OF THE ClIANGING MEDICAL PAYMENTSYSTEM 103 the substitution effect. Can we be sure this will continue to happen in the future? I am not optimistic. Congress will not allow such high margins as profits to continue to accrue; our nation's budget problems are too pressing. CONCLUSIONS I believe that the PPS has much merit and that for most illnesses an average payment amount per DRG is appropriate. But there are situations such as the three technologies I discussed in this paper where an average national rate per diagnosis should not be followed. Instead, I believe we should use a blended reimbursement rate- somewhere between a patient-specific amount and one which relies on a national rate. In so doing, we will take into account somewhat the special needs of each patient. How fiscally hard do you have to hit a hospital to make the point that they ought to think twice about purchasing and using an expensive new technology? Does the penalty have to be 100 percent, or can it be 25 percent? Economists deal on the margin. We complained for years that cost-based reimbursement generated the wrong incentives by paying for all the added costs of treating a patient with a more expensive procedure. Now, I believe that PPS also generates the wrong incentives by paying for none of the extra expense of using an expensive procedure. We may have to wait several years before finding out whether the incentives under PPS are as strong as I suggest. My own view is that they ultimately will be shown to be too strong.