7

Reimbursement and the Dynamics of Surgical Procedure Innovation

SOPHIA W. CHANG * and HAROLD S. LUFT

While the economic environment has important effects on medical practice, we focus here on a single part of the picture: the interaction between payment and the diffusion of surgical procedure innovation. In doing so, we make the distinction between reimbursement (i.e., the repayment of costs incurred) and payment (i.e., compensation that may be more or less than incurred costs). The former entails few cost-saving incentives, whereas the latter may have powerful cost containment effects. It is also important to distinguish surgical procedure innovation —the development of a new procedure— from its application and dissemination. Although we discuss the former, most of our focus will be on the latter. We will also attempt to contrast the characteristics of surgical procedure innovation with those of pharmaceuticals and medical devices.

Surgical innovation often has a significant relationship with new drugs and devices. For example, the expansion of transplantation surgery was facilitated by the development of better chronic immunosuppressive drugs (e.g., cyclosporine). New devices, such as fiberoptic endoscopes, have made possible a wide range of surgical procedures, from arthroscopies to laser atherectomies. The development of total hip replacement relied on such new materials as methylmethacrylate, a plastic cementing mixture allowing better fixation of the prosthetic hip joint. The complexities of these relationships are deep and far reaching. In a study by Comroe and Dripps (1),

* Sophia W. Chang, M.D., was supported by a NCHSR National Research Service Award HS0026-04 postdoctoral fellowship.



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 96
The Changing Economics of Medical Technology 7 Reimbursement and the Dynamics of Surgical Procedure Innovation SOPHIA W. CHANG * and HAROLD S. LUFT While the economic environment has important effects on medical practice, we focus here on a single part of the picture: the interaction between payment and the diffusion of surgical procedure innovation. In doing so, we make the distinction between reimbursement (i.e., the repayment of costs incurred) and payment (i.e., compensation that may be more or less than incurred costs). The former entails few cost-saving incentives, whereas the latter may have powerful cost containment effects. It is also important to distinguish surgical procedure innovation —the development of a new procedure— from its application and dissemination. Although we discuss the former, most of our focus will be on the latter. We will also attempt to contrast the characteristics of surgical procedure innovation with those of pharmaceuticals and medical devices. Surgical innovation often has a significant relationship with new drugs and devices. For example, the expansion of transplantation surgery was facilitated by the development of better chronic immunosuppressive drugs (e.g., cyclosporine). New devices, such as fiberoptic endoscopes, have made possible a wide range of surgical procedures, from arthroscopies to laser atherectomies. The development of total hip replacement relied on such new materials as methylmethacrylate, a plastic cementing mixture allowing better fixation of the prosthetic hip joint. The complexities of these relationships are deep and far reaching. In a study by Comroe and Dripps (1), * Sophia W. Chang, M.D., was supported by a NCHSR National Research Service Award HS0026-04 postdoctoral fellowship.

OCR for page 96
The Changing Economics of Medical Technology the development of cardiac surgery was traced to a series of 25 “essential bodies of knowledge” that enabled Gibbons to perform the first successful open-heart surgery with bypass. The gamut of technologies ranged from antibiotics to electrocardiography to pump oxygenators. Thus, if new payment systems affect innovation in the drug and device arenas, there may be indirect and unanticipated effects on surgical innovation. Unlike pharmaceuticals and devices, which are often developed by industry with specific marketing goals or strategies, surgical procedure development occurs in a more diffuse and less “market-oriented” fashion. Similarly, the entry of drugs and devices into the market is regulated heavily by formal and often costly and time-consuming procedures. Surgical procedure development and diffusion are subject to little or no regulation and appear to be more influenced by payment strategies. Our discussion focuses on how payment affects the development and diffusion of new surgical procedures and secondarily on how these innovations affect medical costs. To understand this dynamic better, we first discuss different types of surgical procedure innovation. Still focusing on the innovative process, we look at how procedures are developed and the incentives inherent in their development. We then examine the diffusion process and discuss how procedures are paid for, with an emphasis on payer recognition and coding practices. This is followed by a consideration of the interaction between competition and diffusion of surgical innovations. Finally, we speculate about the possible effects of the new payment environment on innovation in surgery. DEVELOPMENT OF NEW SURGICAL PROCEDURES Types of New Surgical Procedures Surgical procedures generally are based on a “vocabulary” of basic surgical techniques. For the individual surgeon, mastering surgical procedures requires facility with an array of techniques that, when combined in different ways and in different settings, result in different operative procedures. Some procedures, such as appendectomies, remain relatively simple and have undergone little or no change in the past decade. There is little variation in how the procedure is performed throughout the United States. Greater expertise, as well as greater support from a team of providers, is required to execute the larger array of techniques that comprise more complex surgical procedures. For example, whereas simple procedures such as suturing of lacerations can be performed by any surgeon, such operations as cholecystectomies require the help of anesthesiologists and nurses, and cardiac surgery requires specialized nurses, anesthesiologists, pump technicians, and intensive care specialists. New procedures generally fall into one of two categories: new themes

OCR for page 96
The Changing Economics of Medical Technology and variations on a theme. New themes are considered to be a leap in technological innovation. They result from the invention of new techniques or application of existing techniques in a new context. One example is coronary artery bypass surgery, which when first performed using saphenous vein grafts was a new theme; a related technique using the internal mammary artery can be considered a variation on a theme. Because the complex development of new technologies relies on previous discovery, the distinction between new themes and variations can be blurred somewhat. However, costly changes are more likely to be considered new, whereas less costly changes are usually considered to be variations. The process of surgical innovation includes a broad spectrum of activities that range from relatively simple changes, such as using a new type of suture material, to modifying a surgical technique and to developing a transplantation program. Although some changes require little or no financial investment and might actually decrease costs (by improving outcome, decreasing hospital stay, or decreasing anesthesia time), others require significant investment. Thus, procedures can be classified as either “little-ticket” or “big-ticket” items. Big-ticket technologies usually require institutional (rather than individual professional) support for development and adoption. A liver transplantation service, for example, requires a team of specialists, both medical and surgical (2). Such expensive big-ticket technologies often are used when a disease is life threatening and there is no effective alternative treatment. Some big-ticket procedures, however, can also be cost saving in the long run. In the case of end-stage renal disease, for instance, the costs of renal transplantation generally are less than those of chronic hemodialysis, especially when social costs are included (3). Not all new theme procedures are big-ticket technologies. Certain new procedures, such as the arterial switch procedure for repair of congenital transposition of the great vessels, require no new devices or personnel. This particular procedure does not increase operative time and has better patient outcomes than its previously used alternative (4). Similarly, biliary lithotripsy, a noninvasive therapy for gallstones, is being introduced at centers that already own renal lithotripsy units (5). Although some changes in adjunct equipment are required, most renal lithotripters are adapted easily to biliary treatment and allow their more efficient utilization. Variations are often little-ticket technologies that may increase surgeon efficiency and/or improve patient outcome. Different vagotomy techniques for treatment of peptic ulcer disease, which ligate specific portions of the vagus nerve, can be considered variations. Selective approaches are more time consuming than truncal vagotomy, and there is mixed evidence concerning their relative efficacy (6). Blood cardioplegia, which adds some of the patient's own blood to the “paralyzing” solution used during open-heart surgery, is an example of a variation that improves patient outcome without incurring much expense (7).

OCR for page 96
The Changing Economics of Medical Technology The Development Process for Surgical Procedures New theme procedures generally begin with an experimental phase that is carried out primarily at academic centers with animal models. Research at this stage focuses on the new procedure's technical feasibility and evidence of efficacy without undue risk. Once these surgical procedures are improved or perfected, they are performed on selected human subjects who might benefit from them. Results of animal and human surgery are reported in peer-review journals and/or at professional meetings, usually as case reports. Other forms of peer review occur internally through institutional review boards (required prior to human experimentation) and at morbidity and mortality conferences, where individual cases with poor or unexpected outcomes are discussed among colleagues. At this stage of development, with relatively small numbers of patients involved, it is possible to detect only large differences in outcomes. Furthermore, ethical precepts generally preclude the use of sham operations or the randomization of patients to a new procedure unless there is some evidence that it is at least as good as the standard treatment. The acceptance of a procedure as standard or nonexperimental occurs more readily when the disease is life threatening and there is no effective treatment. The first cardiac transplant at Stanford, for example, did not undergo institutional review for clinical experiments; Shumway maintained that he was trying to save someone's life—not conduct an experiment (8). Indeed, in such critical situations it is difficult for surgeons (and the public) to withhold experimental therapies. In the past, once a series of patients underwent a procedure with some perception of “good” outcome, the procedure became accepted as standard practice. Clinical acceptance of a procedure generally occurred informally through its adoption by colleagues. In the case of coronary artery bypass graft (CABG) surgery, the procedure was first reported in 1969 and disseminated rapidly. By 1973 to 1974, before publication of any clinical trial results, it had become the established treatment for angina pectoris with known two- or three-vessel disease (9). In many cases major innovations have become accepted and paid for without formal evaluation, as is required for new drugs and devices. Some of these practices, such as gastric freezing for the treatment of peptic ulcer disease and gastric balloon placement for the treatment of morbid obesity, have since been shown to be ineffective or even potentially dangerous (10,11). The importance of surgical procedure evaluations should not be underestimated. For example, the $9 million federally funded, randomized extracranial/intracranial arterial bypass trial found that a procedure deemed efficacious and widely practiced for 16 years actually had worse outcomes than nonsurgical treatment. It is estimated that the study saved the United States $11 million in charges for professional and hospital services not provided to those patients randomized

OCR for page 96
The Changing Economics of Medical Technology to medical care during the trial (12). The potential reductions in cost and morbidity of applying these results to the general population are far greater. Recently, increased concern and emphasis on more rigorous methods of technology assessment has helped preclude the rapid diffusion of certain procedures based only on case reporting. A recent case in point is that of neurosurgical transplantation of adrenal medulla tissue into the brain (caudate) of patients with severe Parkinsonism. The inability of investigators to replicate the early reported results essentially has stopped this practice (13). On the other hand, there are occasional egregious examples of highly questionable procedures being undertaken for years by selected surgeons without any formal review (14). New procedures that are variations often undergo an even less formal development process. Certain new procedures might be performed by necessity in patients undergoing reoperations or with slightly different underlying anatomy. Decisions to perform a variation of a procedure often are made during surgery. The resulting new procedures might be disseminated to peers as case reports in professional journals and then used by other practicing surgeons. Other changes are more explicit attempts to improve surgical efficiency and patient outcome (15). In either case, formal or systematic review of the procedure is uncommon. Untoward outcomes might be discussed with peers at morbidity and mortality conferences; otherwise, results are simply added to the participating surgeon's anecdotal experience with that particular procedure. Comparison of the Development of Drugs, Devices, and Procedures In order to better understand the development process of surgical procedures, it may be useful to examine the apparent differences among surgical, pharmaceutical, and medical device innovation.Table 7.1 presents some observations on the characteristics of innovation in drugs, devices, and procedures. The cost of developing a new drug is often enormous as measured in both money and time. Many surgical procedures are developed without external funding. Those with outside sources of support typically are small in scale compared with the cost of pharmaceutical development. New medical device innovation is in an intermediate category of cost. In part, development costs for new drugs and devices reflect regulatory requirements for testing, whereas little formal testing is required of a new procedure. However, even without Food and Drug Administration (FDA) regulations, drug and device development would likely still be orders of magnitude more expensive than that of most surgical procedures. It is not surprising that these cost differences are balanced by differences in the ability of the developer to capture the rewards. The patent process allows the developer of a new pharmaceutical or medical device to recap

OCR for page 96
The Changing Economics of Medical Technology TABLE 7.1 Aspects of Innovation in Drugs, Devices, and Surgical Procedures Characteristics Drugs Devices Procedures Cost of development (regulatory hurdle) High Moderate Low Patentability Yes Maybe No Uniformity High Moderate Low Evaluation Double Blind Yes No Rare Controllable Yes Yes No Link between evaluation and diffusion Yes Yes No Diffusion Corporate Corporate Professional Local monopolies No Sometimes Frequent Coverage Once approved (usually) Once approved May not even seek approval ture the investment over a period of time. Surgical procedures are not patentable. Although the surgical innovator may gain national renown, higher fees, and more patients, it is impossible to earn licensing fees for procedures performed by others. The drug manufacturing process is designed to assure a uniform product. The same is true for devices, although problems of calibration and maintenance may reduce uniformity. Surgical procedures, in contrast, depend on the technique and the skill of the surgeon, patient factors, and sometimes the skill and cooperation of the rest of the care team. Thus, while it is reasonable to assume that a given drug would be equally effective in different hospitals, the same cannot be assumed of surgical procedures. Evaluation of new surgical procedures remains much less formal than evaluation of drugs and devices. Drugs are classically evaluated through a double-blind, randomized, controlled trial. New devices may be evaluated in the same fashion, although it is much more difficult to blind the investigators and use placebos. With surgical procedures, randomization and blindedness are very difficult under reasonable ethical guidelines. The number of cases evaluated also is typically far smaller than for drug and device evaluations. Correspondingly, the detectable differences in outcomes must be far greater in magnitude. Other outcome research methods, such as meta-analysis, are being developed to synthesize formally the disparate results of clinical reports and trials that often differ in design, size, and findings. Most importantly, the timing of the evaluation process in relation to product diffusion differs. FDA guidelines for drugs regulate market entry by mandating product evaluation before dissemination. Although less standardized, the evaluation of medical devices has a similar timing. Post-marketing surveillance is becoming more

OCR for page 96
The Changing Economics of Medical Technology common (especially for devices), however, regulation requires that some known efficacy be demonstrated prior to diffusion. Surgical procedures, on the other hand, generally are not evaluated prior to diffusion, making their evaluation more difficult once they have been accepted professionally as “standard of care.” Surgical procedures are evaluted only with post-diffusion clinical trials, if at all. The diffusion process for various innovations also differs. New pharmaceuticals and devices are marketed actively by manufacturers at professional meetings and through sales representatives. New surgical procedures generally are first discussed in the professional literature. Then, if necessary, the techniques are demonstrated in continuing medical education courses. In general, new drugs are available to all the relevant physicians at roughly the same time. Thus, with the exception of experimental drugs, there are no local monopolies. Some devices, by virtue of their high capital cost or regulatory constraints, may be available only at selected hospitals, but most centers are unlikely to be local monopolies. Sites of new surgical procedures, such as transplants, may be local monopolies for similar reasons, but even when the procedures are widely diffused, local surgeons can develop reputations for being more skilled (or having better outcomes) and thus may attract a disproportionate share of patients. Once a new drug has been approved by the FDA, it is usually covered by most third-party payers. There are some important exceptions for payers that use formularies to select drugs. The same situation generally holds for devices, although explicit formularies are less likely. For surgical procedures recognition and approval for funding are even more closely linked. However, many innovative surgical techniques are thought to be billed with the use of preexisting codes, making them unrecognizable to payers as innovations. Incentives in the Innovation Process The allocation of research funds can have a direct impact on the nature of innovation. This is most clear in the development of new drugs and devices, which is influenced directly by the priorities of the companies involved. At the other extreme, the classic National Institutes of Health (NIH) model of investigator-initiated research allows the pursuit of science, rather than external priorities, to determine the focus and funding of new research. Academic surgical departments generally have relied on several funding sources, including federal agencies (especially the NIH), industry, private foundations, and clinical revenues/teaching funds (16). The level of funding for surgical innovation is difficult to estimate, for most information concerning biomedical research funds does not specify the amounts allocated to surgery departments nor to research focusing on surgical procedures. Funds gener

OCR for page 96
The Changing Economics of Medical Technology ated within academic surgical departments include institution-wide funds for “teaching cases” and clinical revenue. A shrinking share of NIH research funds is being spent on clinical trials and clinical research, with a growing share spent on molecular research (17). Private foundation research funds often are disease or organ specific in orientation and have generally echoed the NIH trends in supporting basic science rather than clinical pursuits. Industry support of surgical research has been restricted primarily to procedures related to devices, such as cardiac pacemakers or new artificial graft materials. There is a longstanding tradition of surgical research and development (R&D) in university medical centers. Innovation is often the route to academic prestige (18). For the academic surgeon this translates into tenure evaluation criteria that emphasize journal publications and the development of new procedures. Although there is little literature on the personal incentives for surgical innovation, it appears that surgical innovators, like those in other areas, are driven more by a desire for knowledge and improvement of patient care than a desire for money. Issues of payment and competition become more prominent in the adoption and diffusion of surgical innovations. Outside of academic centers, practicing surgeons continue to be pushed by the need to remain proficient with old skills and to keep up to date with newer developments. These community-based surgeons rely on a handful of mechanisms to receive knowledge of new procedures. The most common way to stay current is to read surgical journals. However, a large number of newly reported procedures are published in journal supplements that generally are not peer reviewed. Specialty society meetings, grand rounds at individual hospitals, and short courses at innovating centers are other means of disseminating new procedures through “continuing medical education.” New surgical procedures involving medical devices often are taught by manufacturer representatives on artificial models and in the operating room. For some of the less complicated, smaller-ticket changes in surgical procedures, the individual surgeon can try to implement a new procedure in his or her own practice without formal instruction. The pressure on surgeons to innovate and adopt innovations has been exacerbated by the loss of surgical “turf” to nonsurgical specialists who have developed less-invasive substitute procedures. For example, radiologic computerized tomography (CT)-guided percutaneous biopsy and drainage procedures have replaced some surgical open-lung biopsy and intraabdominal abscess drainage procedures (19,20). Endoscopic papillotomies performed by gastroenterologists have replaced open common bile duct exploration for the most part (21). Other therapies, such as ulcer treatment with oral H2 blockers, have further decreased the rate of surgery for peptic ulcer disease (22). Thus, with traditional surgeries becoming less common, the surgeon is pressured to adopt newer procedures in order to maintain a busy practice and to compete successfully with nonsurgical specialties.

OCR for page 96
The Changing Economics of Medical Technology Incentives for hospitals to remain competitive parallel those for surgeons. In general, the public focus on new life-saving procedures has meant that the world-renowned doctor is more often a surgeon than an internist. This is based in part on the surgeon's more clearcut and dramatic intervention and the public interest in curative treatments. To some extent the reputation of an individual surgeon or surgical team casts a “halo” on the whole hospital and is seen as beneficial to the institution. From the hospital or institutional standpoint, surgical innovation is highly prestigious. Public recognition as an innovator in one field often is equated (justifiably or unjustifiably) with institution-wide high-quality care. A reputation for providing the latest in technology can often provide the type of institutional advertising that can increase the hospital's competitive edge. A favorable reputation boosts referral patterns benefiting the institution 's providers. More patients will use hospital services, not only for the innovative procedure, but also for more routine services. A perhaps apocryphal example of this phenomenon is a story of a survey of Louisville, Kentucky, residents. In it they identified Humana Hospital Audubon, the site of Humana's artificial heart program, as having the best obstetrical service. In fact, the hospital had no obstetrical service. Even if the story were not true, it nonetheless reflects the belief that high technology brings significant rewards. A recent example of the symbiotic relationship between hospitals and surgeons is the report of a new twist to outpatient herniorrhaphy using laparoscopically guided internal stapling. The innovative surgeon claimed that the benefits for the practitioner of the new, yet not well-evaluated, procedure included shorter operating time, improved productivity, and less risk of complication. The projected benefits to the hospital included more efficient use of hospital operating rooms, less need for operating room support staff, increased hospital revenue with shorter lengths of stay, and the “marketing advantage of a much less invasive procedure” (15). Reported at a professional meeting, knowledge of this new procedure spread rapidly when it was noted in the Journal of the American Medical Association, with a circulation of over 350,000 in the United States. PAYMENT FOR SURGICAL PROCEDURES A surgical innovator desiring payment for his or her new procedure can follow one of two general approaches. One is to record the new procedure as a minor variation of a standard and accepted procedure and thus be paid at the same level. Under this approach, no new name is given to the procedure; for payment purposes, nothing is recognized as having changed. The other approach is to identify the procedure as new or as a significant variation and seek a different payment (usually a more generous one). In general, this requires a third-party payer to make an explicit coverage decision. Historically, coverage evaluation criteria focused simply on whether the

OCR for page 96
The Changing Economics of Medical Technology new procedure was a “reasonable” and “necessary” form of treatment. Concern over rising medical costs has since led to the development and integration of more rigorous technology assessment criteria into coverage decisions. The Health Care Financing Administration (HCFA), the federal agency responsible for the administration of Medicare, claims that its consideration of a procedure as “reasonable” and “necessary” must be based on evaluation in the literature showing it to be “safe, effective, and not experimental” or “generally accepted in the medical community as safe and effective for the condition for which it is used” (23). The implementation of this relatively vague terminology is complicated further by the fact that coverage decisions are made on a case-by-case basis. Over 150 entities contract with HCFA to review and adjudicate claims for Medicare services. The contractors, known as intermediaries (for Medicare Part A or hospital services) or carriers (for Medicare Part B or physician services), include Peer Review Organizations, health maintenance organizations (HMOs), commercial insurers, and Blue Cross/Blue Shield plans. The distinction between payments for hospital services versus physician services is most clear under the Medicare program, in which different payer authorities review different aspects of the claims generated by the same clinical episode. However, even when the same payer is involved as a private insurer, hospital service claims are handled differently from physician service claims. The hospital claim focuses on the patient's diagnosis and the specific charges for services rendered, such as room charges, operating room time, and medications. The specific procedure performed, however, is of secondary importance and may not even be coded. On the other hand, professional fee claims focus primarily on specific procedures and their coding. Submitted code numbers and charges for procedures are scrutinized, since there are screens for reasonable fees for each procedure and different rates for similar yet distinctly coded procedures. Determining Coverage for New Procedures The first step in the coverage decision process is procedure identification. New procedures are sometimes recognized only after a claim is filed. However, carriers are now being approached by some physician and hospital providers (as well as manufacturers) to consider a new procedure and make an explicit ruling prior to billing. This generally occurs when the new procedure is likely to incur substantially higher costs. Reviews of provider claims can also detect new procedures based on the absence of codes or the presence of unrecognized codes and through excess charges for an accepted service (24). The ability of this process to identify new operations is limited by the sheer volume of individual claims processed. In fiscal year 1987, the HCFA estimates that 400 million claims were processed by its contractors (23); we have no estimate of the number of claims processed by other

OCR for page 96
The Changing Economics of Medical Technology payers outside the Medicare insurance program, but we can assume that it is even larger. Most coverage decisions are made by local fiscal intermediaries or carriers. However, if a contractor cannot resolve a coverage question satisfactorily or believes a national coverage decision may be necessary, the issue is referred to the central HCFA office, specifically the Bureau of Eligibility, Reimbursement and Coverage (BERC). BERC formally reviews 20 to 30 services each year, most of which are referred by contractors. “In general, the more expensive a service is or promises to become, either in an individual case or in the aggregate, . . . the more likely it is to be referred” (23). BERC will review procedures meeting any of the criteria, which include being a significant scientific advancement, a new or costly product, a procedure having potential for rapid diffusion, or a procedure considered to be outmoded or under question concerning its safety and effectiveness. Generally, HCFA will postpone making a national policy decision if the service involves a new or emerging technology or practice for which there are limited clinical data. Individual contractors maintain significant autonomy in deciding what procedures are reimbursed. They generally focus on the appropriate use of a technology in a specific setting (e.g., does this patient require inpatient care for an elective herniorrhaphy?) and do not carry the burden of following strict technology assessment criteria. Coverage of those procedures that are truly novel and are referred to BERC can be left unresolved if there is inadequate information for a national coverage policy decision. Furthermore, there are no formal checks on the compliance of contractors in following national guidelines; most decision making remains at the local level. If a procedure does not appear to be markedly new or expensive, it might easily be considered eligible for reimbursement by some local contractors. In the past, the HCFA led the way in the recognition of new surgical procedures. Once Medicare accepted a procedure for payment, commercial payers generally followed suit. In general, private payers are contractually bound to cover accepted procedures, and the Medicare stamp of approval signified a new technology as legitimate. However, large private insurers recently have led the way in the coverage of new surgical procedures (e.g., cardiac and liver transplants). This acceptance has put greater pressure on HCFA to provide Medicare payment for these procedures (25). All third-party payers must rely primarily on claims reporting to identify new surgical procedures. Evaluations of procedures generally are performed in house by private insurers, with direct consultation from physicians (e.g., the Blue Cross/Blue Shield National Association Medical Advisory Panel) and specialty groups. Although not necessarily less rigorous, the process of private insurer technology assessment is generally less formal and can be implemented more quickly than national Medicare evaluations. Other public programs, such as state-administered Medicaid programs, remain highly divergent in

OCR for page 96
The Changing Economics of Medical Technology Interaction of Diffusion and Competition One manifestation of hospital competition is avid adoption of high-technology procedures to remain competitive and attract patients (24). This impetus for the diffusion of new surgical techniques has come from both the hospital and the individual practitioner (33). With the recognition of technology as a competitive tool, nonacademic institutions are rapidly adopting new surgical procedures. The spread of new procedures to nonacademic settings also reflects the success of training programs, which have produced larger numbers of surgeons to compete as providers (34). Without the financial commitment to support research and teaching costs that academic centers must make, many community-based institutions are able to attract well-trained surgeons and offer better prices for specific operations. This can, in turn, reduce the patient load at innovative centers. At Stanford University, for example, which is world renowned for its cardiac surgery service, the number of CABG procedures fell from 623 to 378 between 1983 and 1986 with the proliferation of neighboring competitors (data derived from discharge abstracts made available through the California Office of Statewide Health Planning and Development). Furthermore, surgical case loads are decreasing on average, making it harder for surgeons to remain proficient at routine procedures. From 1982 to 1985 the average number of operations performed annually by general surgeons decreased by 25 percent (35). The development of a major new program, such as a liver transplant service, requires a substantial institutional investment of space, faculty positions, and other resources in addition to funds. Such decisions are clearly made at the highest levels in the institution. Not all decisions are in favor of new technology adoption; after 2 years of in-house debate, the trustees of Massachusetts General Hospital chose not to adopt a cardiac transplantation program. Reasons for their decision included costs (with one transplant consuming the resources of six to eight open-heart surgery cases) and the paucity of new knowledge the proposed clinical program was thought to add. It was a clear (and perhaps unusual) case of an institution overriding professional desire to implement a new surgical innovation (36). CABG surgery is a commonly cited example of a rapidly diffused technology. Although substantial investment in equipment and personnel is required to perform this operation, hospitals project a positive cash flow in several months and profitable operation of an open-heart surgery unit within a year or two. These rosy projections do not even take into account the fact that a cardiac surgery unit enables a hospital to provide the required backup for the performance of percutaneous transluminal coronary angioplasties, another major source of hospital revenue. Having a cardiac surgery unit also projects the image of being a state-of-the-art hospital. These competitive pressures are so great that the number of heart surgery units continues to grow despite controversies about the appropriateness of CABG surgery

OCR for page 96
The Changing Economics of Medical Technology relative to aggressive medical management for many patients (37) and evidence that hospitals with low volumes of CABG surgery tend to have worse patient outcomes (38). Given the concerns about the potentially excessive diffusion of CABG surgery, it is instructive to examine the roles of market competition and regulatory forces. Hospitals with more local neighbors (who are likely to be competitors) are much more likely to offer open-heart surgery. Furthermore, the more neighboring hospitals with open-heart units, the higher the probability that a competing hospital will offer CABG surgery. Thus, there appears to be a “medical arms race ” in which the presence of cardiac surgery services in one hospital results in an escalating availability of the service among local competitors (39). In some states, however, regulatory agencies have sought to constrain this type of behavior. For example, New York State has a stringent certificate of need approval process for new beds and open-heart surgery units. In 1986, with a population of 17.8 million, New York had 27 hospitals offering CABG, with an average volume of 362 operations per hospital per year. In contrast, California has a very lenient approval process. This state, with a population of 27.7 million (1.5 times that of New York), had 89 hospitals offering CABG (three times that of New York), with an average volume of 201 procedures per hospital per year. Los Angeles alone had over 40 hospitals offering the procedure, none more than 7 miles from another (40). Selective Contracting for Surgical Procedures With interest in containing costs while maintaining quality care, one payer response has been selective contracting.4 Although the concept of selective contracting has been in existence for decades (e.g., coverage for specific services at designated hospitals by Crippled Children's Services), there is a recent national proliferation in this approach to payment. It is useful to distinguish two types of selective contracting. In one approach a payer establishes certain standards that must be met by a provider before payment will be made. The second approach combines standards with the notion that even among providers meeting the standards, only certain ones will be selected for contracts. The primary example of selective payment for a high-cost surgical service is cardiac transplantation. In 1983 California Blue Shield began providing payment for cardiac transplantations on a case-by-case basis at Stanford University, based on the view that the procedure was standard accepted care for that community.5 With the diffusion of the procedure to other centers, pressure for reimbursement by Medicare increased, and by 1985, the HCFA made a concerted effort to evaluate the procedure, including issues of cost effectiveness (41). The evaluation of cardiac transplantation made it clear that the institutional program, with its interdisciplinary team and ancillary

OCR for page 96
The Changing Economics of Medical Technology support services, was just as important as the performance of the procedure itself. Furthermore, the costs and outcomes of the surgical procedure were highly dependent on the postoperative management of the transplant recipient (41). Medicare announced coverage of cardiac transplantation in 1986 and, for the first time, included institutional criteria for coverage. Although it has been argued that these criteria are not strict enough nor sufficiently sensitive to truly discriminate among centers (42), it was nonetheless a policy move that acknowledged a concern over too rapid diffusion of high-cost technology. A policy of selective contracting sought to limit rapid growth by setting limits on whom HCFA would pay for specific services. The shortage of available organs in the case of cardiac transplantation also concerned policy makers, who feared that the spread of the procedure to a large number of institutions could lead to a less than optimal matching of organs and recipients. These selective payment programs have placed a greater onus on the hospital or institution to make significant investments to develop experience with a new surgical procedure prior to receiving coverage approval. Despite efforts to control their growth, transplantation programs have continued to diffuse at a rapid rate. Community hospitals see transplantation programs as potential big moneymakers and have made the initial investments to start their own programs. These hospitals generally are able to attract academically trained transplantation surgeons to a nonuniversity setting by offering better remuneration, less teaching responsibility, and, generally, less pressure to pursue surgical research. This is causing increased tensions as academic centers are competing not just with each other but with community hospitals that do not invest in research, development, and training. These rivalries are exacerbated further by the fact that many of the community surgeons trained at or were previously on the faculty of academic centers. Supporters of “standards-based” selective contracting have been somewhat disappointed in its results. The initial hope was to significantly limit the diffusion of high-cost surgical technologies and improve patient outcomes by restricting new procedures to sites meeting specific criteria with respect to staffing levels, support, and patient outcomes. However, there has been continued growth in the number of transplantation centers, and institutions have carefully selected patients to ensure a good outcome rate for reimbursement (RW Schaffarzick, personal communication, November 2, 1989). In essence, the standards-based approach does not necessarily limit the number of institutions that are eligible; it merely specifies what any center must do in order to qualify. By rewarding programs that achieve favorable results through selection of healthier patients, this policy provides incentives for providers to avoid difficult patients who might have the most to gain from the new procedure. The second type of selective contracting sets minimum criteria for entering the “game” and then selects certain institutions from among those eligible.

OCR for page 96
The Changing Economics of Medical Technology These more restrictive modes of payment have begun to introduce cost consciousness and occasionally price competition in lieu of non-price competition (43-45). With the current emphasis on competitive bidding and an increased focus on price and possibly on quality or outcome, the next wave of selective contracting may be for new surgical procedures. We are already witnessing competing price bids by large institutions for such accepted procedures as CABGs. For example, the Texas Heart Institute established a subsidiary corporation in 1984 to offer CABG surgery at a package rate that included the hospital, surgeon, cardiologist, and other associated costs (including transportation) for $13,800, in contrast to the Medicare rate of $24,588 (46). The HCFA is now developing a program of selective contracting with “centers of excellence” to provide big-ticket surgical services for Medicare beneficiaries. Similar programs have been developed by commercial insurers, such as Prudential's program of centers of excellence and Metropolitan Life's centers of quality. Self-insured employers, such as Honeywell, have made similar arrangements for certain high-cost, schedulable procedures. Larger payers generally are able to bargain for better quality and price because they purchase a large volume of services for a large number of beneficiaries. This buying power has enabled HMOs to contract selectively for specific surgical services. HMOs select the providers and often specify their own criteria for deciding whether the procedure should be done. In this way they are able to control not only the cost of the procedure but also the frequency of its use. Limitations of Selective Contracting The criteria used to evaluate centers that provide new surgical procedures are relatively limited. Mortality and morbidity rates, while informative, may not provide sufficient information to assess a program's quality. Factors such as patient selection become highly important; certain institutions may choose only patients who are likely to have good outcomes but who may not be most in need of the procedure. With low mortality rate procedures, evaluation of outcomes becomes more difficult because enormous volumes are needed to calculate reliable mortality rates. In this situation the volume of services provided is often used as a proxy. Although there is a relationship between volume and outcomes for procedures such as CABG surgery, the evidence of such a relationship for other procedures is mixed (47). Thus, there is the danger that selective contracting for such procedures may focus on price simply because there are no readily available measures of quality. In addition, just as surgical procedures continue to evolve, an individual surgeon's proficiency may continue to improve. Thus, the time frame in which a provider is assessed may be important. Overall improvements in proficiency across settings might require more stringent criteria with time if the purpose is

OCR for page 96
The Changing Economics of Medical Technology to limit availability. Increased facility with a surgical procedure will often enlarge the patient population considered eligible for the procedure. This may mean the inclusion of sicker patients (in the case of CABG surgery) or healthier patients (in the case of abdominal aortic aneurysm repair). The constantly evolving nature of surgical procedures requires periodic reevaluation of hospital costs (with the development of some economies of scale) and of professional fees (with increased efficiencies) (48). Some have voiced concern that current payer strategies may squelch surgical innovation, especially if initial costs appear to be high. Blue Shield of California has developed a system of “modified selective contracting” to encourage the development of new procedures, especially those deemed to be potential cost-saving substitutes (46). The institutional and professional fees for such emerging or investigational procedures are based on successful outcome; if the outcome is unsuccessful, the institution agrees to absorb the cost of the procedure. An example is catheter ablation of abnormal cardiac conduction foci. This percutaneous method uses accepted electrophysiologic mapping (EPS) techniques to locate the focus and then ablates it with electroshock to the abnormal area. The older approach to this problem uses EPS to identify the lesion and then surgically destroys the abnormal heart tissue via open-heart surgery (49). The contingent payment approach essentially means that as the success rate approaches 100 percent, providers will be paid in full, but if the rate approaches zero, providers will not be reimbursed at all. The emphasis on controlling cost can also slow the dissemination of more expensive procedures that might have better outcomes and even potentially lower costs in the long run. Current studies are examining the use of slowly inflatable breast implants with breast reconstruction at the time of mastectomy. Although it increases intraoperative time, this procedure can eliminate the need for a second operation (mastectomy, then reconstruction). Use of the slowly enlarging prosthesis also causes less discomfort and permits better wound healing (50). Changes in practice induced by measures to reduce costs may also affect outcomes negatively. One study of the impact of the PPS on the results in elderly hip fracture patients found that while the mean length of stay fell as expected from 21.9 to 12.6 days, the proportion of patients discharged to nursing homes rose from 38 percent to 60 percent. Unfortunately, the numbers remaining in nursing home care 1 year after hospitalization also rose from 9 percent to 33 percent (51). Although this change is not due to a new surgical procedure, it nonetheless reflects payment effects on surgical outcomes. Outpatient Procedures and Payment Incentives Most of the discussion so far has emphasized hospital costs and hospital-based surgical procedures. In 1980 Medicare changed its reimbursement

OCR for page 96
The Changing Economics of Medical Technology policies to encourage the growth of ambulatory surgery. The 1980 Omnibus Budget Reconciliation Act waived the Part B deductible and copayment for the facility fee and paid 100 percent of “reasonable ” professional charges once a surgeon accepted assignment.6 Hospitals also received payment on a cost rather than DRG basis for outpatient surgery. With these changes in payment policy, the volume of ambulatory surgery rose explosively; by 1983 over two-thirds of ambulatory surgery for Medicare enrollees was performed in physician offices (52). Although arguments have been made that some specific procedures are cost saving when performed on an outpatient basis (53), those savings have not necessarily been recouped by the payers. Certain procedures, such as ophthalmic intraocular lens implantation, have continued to receive high fees in spite of the volume of services provided and presumed cost decreases (48). Although there are some efforts to monitor the quality of care in the hospital setting and although it is possible to imagine selective contracting based on rigorous evaluation of the quality of institutional bidders, it is almost impossible to imagine such approaches applied routinely to the outpatient setting. For example, Blue Shield of California has 36,000 individual participating physicians, or 80 percent of the physicians in the state (RW Schaffarzick, personal communication, November 2, 1989). Utilization review and peer group norms linked with quality assessment have been used as evaluation tools in programs such as the Professional Review Organizations; but these efforts are likely to capture only the most extreme outliers among providers. More importantly, routinely collected data from outpatient claims files include little information about diagnoses and are far less reliable than the much-maligned hospital data. In addition to problems with existing data systems for outpatient procedures, the current system essentially divides the medical care world into inpatient (institutional) claims and outpatient (physician) claims. Although there has been an enormous growth in outpatient procedures, they still comprise only a small fraction of total physician claims, which are dominated by office visits and routine tests. One alternative approach would be to segregate claims for “substantial ” procedures and require more data in these cases. Review organizations would then be able to compare the outcomes of all procedures such as herniorrhaphies or lens implants done on an inpatient or outpatient basis. Eligibility for payment could be made conditional on more intensive review of quality and appropriateness. OBSERVATIONS, IMPRESSIONS, AND SPECULATIONS One of the most important observations of our review is that there is very little empirical research on surgical innovation in general, let alone about the influence of payment policies on innovation. Thus, our review must be considered tentative and exploratory, with substantial gaps filled by hypoth

OCR for page 96
The Changing Economics of Medical Technology esis and argument rather than evidence. Given the importance of surgical innovation for both the health of the population and the cost of medical care, this absence of research is unfortunate. Describing an area with so little research is difficult; therefore, predicting the effects of different types of payment on surgical innovation may be foolhardy. Nonetheless, we offer some speculation to encourage further examination, discussion, and, perhaps most important, important research. The changes in hospital payment in the last few years may have important consequences for surgical innovation. It appears that in the past many innovations were financed internally, through professional fees generated within departments of surgery and by hospitals that were able to pass unreimbursed development costs to other patients and payers. Prospective payment and selective contracts for hospital care have made it far more difficult for hospitals to undertake this type of subsidy. There are already concerns in some teaching hospitals about the sources of financing for new procedures, and it is likely that these concerns will reduce innovation. The revision of physician fees under a resource-based relative value schedule (RBRVS) (48) will likely reduce surgical fees significantly and will certainly eliminate some of the surplus funds used by departments to support innovation. The adoption of an RBRVS also is likely to have other important effects. First, it will focus much more attention on issues of procedure coding. There may be a formal recognition of experimental or developmental procedures in the coding system, and this could lead to a more formal evaluation of such innovations before their diffusion. While a simpler and more generalized professional fee coding system has been proposed to decrease the rate of miscoding, attention must also be focused on the system's ability to identify new procedures (30). The change in physician payment is intended to be more than just a coding reform—it is supposed to redistribute payments among providers. Although the effects will vary by specialty, the changes generally will reduce payments for procedures and increase them for evaluation and management services. This may lead to financial disincentives for specializing in surgery (or at least reduce the rewards for the choice of such specialties) and perhaps reduce surgical innovation. Unlike the old cost-based reimbursement system for hospitals, which had no incentives to reduce costs, the old physician payment system did have some incentives for developing cost-saving innovations that could make accepted procedures faster and less expensive. An RBRVS approach would allow payers to recapture some of these savings by lowering standard fees as surgeons became more efficient. However, since fees would be based on average performance, strong incentives for individual surgeons to become more efficient would remain, since they would be able to capture the difference between the fee and their own costs. On the other hand, the new system would make it less attractive for less efficient surgeons to continue to do the procedure since they could not charge above the

OCR for page 96
The Changing Economics of Medical Technology average fee. The net effect might be a somewhat slower diffusion of new procedures. The growth of selective contracting for specialized procedures likely will play a stronger role than the RBRVS in affecting innovation and diffusion. Assuming that such contracting is based on quality as well as price and that the price will tend toward an all-inclusive fee, there are several implications. First, there will be strong pressures for cost-reducing innovations to reduce not only operative time but also all other surgery-associated fees, such as hospital costs, laboratory tests, and other physician services. Second, methods of modified selective contracting may help limit the premature diffusion of costly surgical innovations and provide selective support of research in the investigational stage. Third, since there will be more attention to the measurement of quality of care and clinical outcomes, such contracting may also encourage quality-enhancing innovations, particularly if the enhancements are measurable. Fourth, the potential to obtain such contracts means that research institutions might be able to recapture some of their development costs by establishing a reputation for outstanding, state-of-the-art care at volume prices. It probably will be easier for academic centers to do this in fields in which procedures are continuing to develop, while selected community hospital centers may be limited to performing procedures once they become routine. Furthermore, the conventional division between hospital and professional charges will become less prominent. Payer interest in cost containment and quality assurance will create greater pressure for the development of all-inclusive fees that cover all services related to a surgical procedure. Such package fees will impel providers (both physicians and hospitals) to determine the more cost-effective methods of treatment and the appropriate balance of services required for better patient outcome. If one looks back to the projections made in 1983 of the likely effects of prospective payment for hospital care, it is apparent that some were correct and others were entirely wrong. Our speculations are based on far less evidence and research. However, it is reasonable to expect future changes in the payment system to have some, and perhaps substantial, effects on the innovation and diffusion of surgical procedures. Better predictions about the nature of those effects and their desirability are important policy questions that should be given careful consideration. NOTES 1. Surgical CPT codes are translated directly into the HCFA Common Procedure Coding System (HCPCS). 2. This is similar to the phenomenon of “DRG (diagnosis-related group) creep,” described by health policy analysts as a method used by hospitals to maximize revenue by classifying an acute care episode under more generously reimbursed diagnoses.

OCR for page 96
The Changing Economics of Medical Technology 3. “Unbundling” is the process of individually charging for techniques that comprise a multistep procedure, resulting in a total fee that is greater than the traditional fee for the overall procedure. 4. Selective contracting is a mechanism used by insurers to contract for the provision of certain services with specific providers. By representing large numbers of patients, the insurer can bargain for better prices from providers (both hospitals and physicians). In general, to encourage patients to use contracted services, financial incentives are offered if the contracted provider is chosen. 5. At that time Blue Shield felt that Stanford had shown an adequate “track record” with the procedure, with favorable outcomes and specific eligibility criteria. 6. “Assignment” is the term used to designate provider acceptance of Medicare reimbursement rates as full payment for the services provided. The provider agrees to not “balance bill” the patient for any difference between physician charge and Medicare payment, except for the normal 20 percent copayment, which is waived in this instance of 100 percent coverage. REFERENCES 1. Comroe JH , Dripps RD. Scientific basis for the support of biomedical science . Science 1976 ; 192 : 105-111. 2. Maddrey WC , Van Thiel DH . Liver transplantation: an overview . Hepatology 1988 ; 8 : 948-959 . 3. Garner TI , Dardis R . Cost-effectiveness analysis of end-stage renal disease treatments . Medical Care 1987 ; 25 : 25-34 . 4. Castaneda AR , Mayer JE , Jonas RA , Wernovsky G , Di Donato R. Transposition of the great arteries: the arterial switch operation . Cardiology Clinics 1989 ; 7(2) : 369-376 . 5. Freiherr G , Weiss M. Romancing the market for stones . Healthweek, December 4, 1989 ; 3 : 18-24 . 6. Hoffmann J , Jensen HE , Christiansen J , Olesen A , Loud FB , Hauch O. Prospective controlled vagotomy trial for duodenal ulcer: results after 11-15 years . Annals of Surgery 1989 ; 209 : 40-45 . 7. Cunningham JN , Adams PX , Knopp EA , et al. Preservation of ATP, ultra-structure and ventricular function after aortic cross clamping and reperfusion . Journal of Thoracic and Cardiovascular Surgery 1979 ; 78 : 708-711 . 8. Bunker JP . The Anesthesiologist and the Surgeon: Partners in the Operating Room . Boston : Little Brown , 1972 . 9. Bunker JP , Hinkley D , McDermott WV . Surgical innovation and its evaluation. Science 1978 ; 200 : 937-941 . 10. Ruffin JM , Grizzle JE , Hightower NC , McHardy G , Shull H , Kirsner JB. A co-operative double-blind evaluation of gastric “freezing” in the treatment of duodenal ulcer . New England Journal of Medicine 1969 ; 281 : 16-19 . 11 . Kramer FM , Stunkard AJ , Spiegel TA , Deren JJ , Velchik MG , Wadden TA , Marshall KA. Limited weight losses with a gastric balloon . Archives of Internal Medicine 1989 ; 149 : 411-413 . 12. Merz B. Neurosurgeons address EC/IC study; question controlled surgicaltrials . The Journal of the American Medical Association 1986 ; 256 : 165-167 . 13. Sladek JR , Shoulson I. Neural transplantation: a call for patience rather than patients . Science 1988 ; 240 : 1386-1388 .

OCR for page 96
The Changing Economics of Medical Technology 14. Holoweiko M. Why was the love surgeon allowed to keep cutting? Medical Economics 1989 ; July 17 : 125-141 . 15. Goldsmith MF. Some new twists to one of most common procedures in US general surgery . Journal of the American Medical Association 1989 ; 262 : 3248-3249 . 16. Rikkers LF , Bland KI , Kinder BK , Lawrence PF , Lynch TG , Modlin IM , etal. Funding of surgical research: the roles of government and industry. Journal of Surgical Research 1985 ; 39 : 209-215 . 17. Ginzberg E , Dutka AB. The Financing of Biomedical Research . Baltimore : The Johns Hopkins University Press , 1989 . 18. Bahnson HT. Presidental address: education of a surgical chairman . Annals of Surgery 1988 ; 208(3) : 247-253 . 19. Gleich S , Wolin DA , Herbsman H. A review of percutaneous drainage in splenic abscess . Surgery, Gynecology and Obstetrics 1988 ; 167 : 211-216 . 20. Chaffey MH. The role of percutaneous lung biopsy in the workup of a solitary pulmonary nodule . Western Journal of Medicine 1988 ; 148 : 176-181 . 21. Silvis SE. Current status of endoscopic sphincterotomy . American Journal of Gastroenterology 1984 ; 79 : 731-733 . 22. Jamieson GG. Surgery for peptic ulcer disease in the era of H2 receptor blockers . Digestive Diseases 1989 ; 7 : 76-85 . 23. Health Care Financing Administration . Medicare program: criteria and procedures for making medical services coverage decisions that relate to healthcare technology . Federal Register, January 30, 1989 ; 54 :4302-4318 . 24. Ruby G , Banta HD , Burns AK. Medicare coverage, Medicare costs, and medical technology . Journal of Health Politics, Policy and Law 1985 ; 10 : 141-155 . 25. Rettig RA. The politics of organ transplantation: a parable of our time . Journal of Health Politics, Policy and Law 1989 ; 14 : 191-227 . 26. Wagner JL. Reimbursement shapes market for technology . Hospitals 1979 : 91-94 . 27. Bunker J , Fowles J , Schaffarzick R. Evaluation of medical-technology strategies . New England Journal of Medicine 1982 ; 306 : 620-624 . 28. Beahrs OH. Reimbursement in the future: the physician payment review commission . In Rutkow IM (ed). Socioeconomics of Surgery . St. Louis : C.V.Mosby Company , 1989. 29. AMA editorial staff . Physicians' Current Procedural Terminology . Chicago : American Medical Association , 1988. 30. Physician Payment Review Commission . Annual Report to Congress—1989 . Washington, D.C. : PPRC , 1989 . 31. Kane NM , Manoukian PD. The effect of the Medicare prospective payment system on the adoption of new technology: the case of cochlear implants . New England Journal of Medicine 1989 ; 321 : 1378-1383 . 32. Roe BB. The UCR boondoggle: a death knell for private practice? New England Journal of Medicine 1981 ; 305 : 41-45 . 33. , Kronenfeld JJ , Baker SL , Amidon RL. An appraisal of organizational response to fiscally constraining regulation: the case of hospitals and DRGs . Journal of Health and Social Behavior 1989 ; 30 : 41-55 . 34. Schroeder SA , Zones JS , Showstack JA. Academic medicine as a public trust . The Journal of the American Medical Association 1989 ; 262 : 803-812 . 35. Rutkow IM. Surgical operations and manpower: can technical proficiency be

OCR for page 96
The Changing Economics of Medical Technology maintained? In Rutkow IM (ed). Socioeconomics of Surgery . St. Louis : C.V. Mosby Company, 1989 . 36. Knox RA. Heart transplants: to pay or not to pay . Science 1980 ; 209 : 570-575 . 37. Jones EL. Evolving status of operative versus nonoperative management of coronary arterial obstruction . Cardiovascular Clinics 1987 ; 17 : 73-89 . 38. Showstack JA , Rosenfeld KE , Garnick DW , Luft HS , Schaffarzick RW , Fowles J. Association of volume with outcome of coronary artery bypass graft surgery, scheduled vs. nonscheduled operations . Journal of the American Medical Association 1987 ; 257 : 785-789 . 39. Robinson JC , Garnick DW , McPhee SJ. Market and regulatory influences on the availability of coronary angioplasty and bypass surgery in U.S. hospitals . New England Journal of Medicine 1987 ; 317 : 85-90 . 40. Hannan EL , O'Donnel JF , Kilburn H , et al. Investigation of the relationshipbetween volume and mortality for surgical procedures performed in New York state hospitals . Journal of the American Medical Association 1989 ; 262 : 503-510 , and case abstract data provided by the California Office of Statewide Health Planning and Development . 41. Evans RW.The economics of heart transplantation . Circulation 1987 ; 75 : 63-76 . 42. Renlund DG , Bristow MR , Lybbert MR , O'Connell JB , Gay WA.. New England Journal of Medicine 1987 ; 316(14) : 873-6 . 43. Luft HS. Competition and regulation . Medical Care 1985 ; 23 : 383-400 . 44. Robinson JC , Luft HS. Competition and the cost of hospital care, 1972 to1982 . Journal of the American Medical Association 1987 ; 257 : 3241-3245 . 45. Robinson JC , Luft HS . Competition, regulation, and hospital costs, 1982 to 1986 . Journal of the American Medical Association 1988 ; 260 : 2676-2681 . 46. Schaffarzick RW , Bunker JP. Regionalized surgical health care . In Rutkow IM (ed). Socioeconomics of Surgery . St. Louis : The C.V. Mosby Company, 1989 . 47. Luft HS , Garnick DW , Mark D , McPhee SJ. Volume of services in hospitals or performed by physicians . In U.S. Congress Office of Technology Assessment, The Quality of Medical Care: Information for Consumers, OTA-H-386, Washington, D.C. : US Government Printing Office , 1988 . 48. Hsiao WC , Braun P , Yntema D , Becker ER. Estimating physicians' work for a resource-based relative-value scale . New England Journal of Medicine 1988 ; 319 : 835-841 . 49. Newman D , Evans GT , Scheinman MM. Catheter ablation of cardiac arrhythmias . Current Problems in Cardiology 1989 ; 14 : 117-164. 50. Artz JS , Dinner MI , Sampliner U. Breast reconstruction with a subcutaneoustissue expander followed with a polyurethane-covered silicone breast implant . Annals of Plastic Surgery 1988 ; 20 : 517-521 . 51. Fitzgerald JF , Moore PS , Dittus RS. The care of elderly patients with hip fracture: changes since implementation of the prospective payment system . New England Journal of Medicine 1988 ; 319 : 1392-1397 . 52. Leader S , Moon M. Medicare trends in ambulatory surgery . Health Affairs 1989 ; 158-170 . 53. Kambouris AA. Ambulatory surgery: its impact on general surgical practice . The American Surgeon 1986 ; 52 : 347-350 .