Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
APPENDIX THE IMPACT OF STATE REGULATION p ON THE ADOPTION AND DIFFUSION OF NEW MEDICAL TECHNOLOGY \ Jack Needleman and Lawrence S. Lewin In l950, acute care hospitals spent $l0 per patient day; in l975, they spent $l50.! Changes of comparable orders of magni- tude have occurred for other medical services. Part of these increases is simply the result of inflation. A significant part of the increase, however, has been the result of changes in the product. These changes, especially those that have resulted from the development of new equipmentâ"machine introduced technology"âhave recently been the focus of con- siderable public attention. Issues have been raised regarding the magnitude of investment in this technology, the payoffs of such investment in terms of patient care and comfort, and the factors that influence the distribution of new technology within the- medical system. This paper focuses on one aspect of these issues, the current and potential impact of state regulatory programs on the adoption and diffusion of equipment-embodied technology. This discussion is presented in three sections: I. Background A discussion of the scope and limitations of this analysis, the nature of relationships between regulation and technology policy, and the importance of this issue in formulating policies regarding medical technology. II. Analysis of the Impact of State and Local Regulation Six principal modes of state regulation are identified and ana- lyzed: â¢ Certificate of need â¢ Reimbursement controls â¢ Medical devices regulation 240
24l â¢ Utilization review and quality assurance programs â¢ Institutional licensure and accreditation â¢ Personnel licensure Following a brief description of the nature of the regulatory program and how it operates, empirical evidence of impact is presented, followed by a discussion of factors likely to determine its effect and the probable implication of each regulatory mode for technology policy. III. Policy Issues and Implications The concluding section synthesizes the policy implications discussed in Part II and offers some caveats and guidelines for those involved in making both regulatory and technology policy. I. BACKGROUND Scope of this Paper While the most visible direct form of regulating equipment- embodied medical technology is the federal Medical Devices Legislation of l976, P.L. 94-295, considerable influence over the adaption and utilization of medical technology is exerted by regulatory programs at the state level. Because the direct regulation of equipment is relatively new and almost nonexistent at the state level, the greatest influence at this level appears to result from the secondary effect of such activities as control over,reimbursement for services. This paper discusses a wide range of regulatory activities at the state level, including direct control modes such as certificate of need, medical device regulation, and institutional licensure and the indirect modes such as reimbursement controls, utilization review and quality assurance programs, and personnel licensure. The analysis of these modes in Part II draws, where possible, on existing evidence of the impact of technological change; but with the exception of certificate of need, prospective reimbursement, and the Economic Stabilization Program there is little empirical evidence available. In the absence of such evidence the authors have presented a logical analysis of the probable effect of regulatory practices both to date and in the future. A review of the literature failed to produce a model that could be used to test the impact of interactive regulatory modes on the adoption and utilization of medical technology. While this paper does not presume to present such a model, much of the discussion in Part II is devoted to defining factors
242 that influence regulatory impact and will hopefully facilitate subsequent efforts to develop such a model. In conceptualizing regulatory impact, a major difficulty arises from the fact that many of the state regulatory practices that affect medical technology are designed primarily to meet other objectives and the effect on adoption or utilization of technology is in many of these cases a side effect. The modeling effort is further complicated by the absence of a clear objective function or a statement of desired outcomes for technological change. Relationship Between Regulatory and Technology Policy To make qualitative judgments about the desirable or undesirable effect of state regulation requires an understanding of the goals of technology policy. While in general the goal of medical technology can be to improve the effectiveness of medical diagnostic, preventive, treatment, and rehabilitative services and the efficiency with which these services are provided, such statements are too imprecise to provide practical guidance. It is not within the scope of this paper to set forth specific policies; however, the type of policy statements that are needed would be in response to such questions as: â¢ In what areas is more sophisticated, efficacious, and efficient technology most/least needed? â¢ How can decision makers deal with the often conflicting strains of cost and effectiveness? How, for example, can decisions be made regarding the cost level at which specific improvements in medical technology are, from a societal stand- point and in the face of constrained resource, no longer cost- effective? â¢ Since many high-cost technology utilization decisions involve judgments about the value and quality of life, how are these judgments to be made and by whom? â¢ Where cost containment is a major concern, to what extent should regulators make service- or equipment-specific decisions as compared with more general decisions that produce constraints and that in turn force providers to make specific trade-off decisions? For example, with respect to CAT Scanners regulators could (l) prohibit a hospital from purchasing one, (2) limit reimbursement to a level that assumed full utilization, or (3) place a ceiling on each provider's capital expenditures for new services and let individual providers decide whether to buy a CAT Scanner or some other piece of equipment. â¢ Since medical services once established tend to remain in use even at low levels of utilization and despite high unit
243 costs, what constitutes optimum rates of diffusion, particu- larly before efficacy and cost-effectiveness are fully established? Put another way, how can regulators most intelli- gently decide how widely and rapidly untested technologies should be installed? In making these decisions, how should they consider the opportunity cost of foregoing treatment or diagnostic benefits, and the need to test the technology's efficacy and efficiency? The purposes and goals of regulatory policy in the health care services field are somewhat clearer. For our purposes, regulatory goals can be classified into three broad categories: â¢ Protection of the consumer from unsafe or high-risk situations. This purpose may be achieved by requiring a product to be tested before being marketed or to meet preestablished quality standards with regulators empowered to either prohibit or limit use or to require that specific product information be provided to the consumer. Another approach is for regula- tors to assure protection by requiring providers of services to meet certain competency standards through professional licensure. â¢ Assuring the accessibility of services in a rational system. While there is vigorous debate as to whether this goal in practice works for the ultimate protection of the consumer or for the industry, the notion of stable and reliable provision of services accessible to all has become a conerstone of public utility regulation. With respect to accessibility, it takes the form of readiness-to-serve and community service requirements. With respect to stability and the rationality of the system it takes the form of franchising of exclusive service areas and of "gatekeeping" functions like certificates of necessity. Since this latter requirement erects serious barriers to entry, it often is limited in its application to so-called "natural monopolies" like power generating public utilities where economies of scale are significant. The health field, which uses certificates of need extensively (now virtu- ally universally) relies on this form because of the unique characteristics of its financing mechanisms and its character- istics of supply-induced demand. â¢ Constraining the prices of services and total system expenditures. Economic regulation usually draws its rationale from structural deficiencies in the marketplace that in the absence of regulation would result in either destructive competition or monopolistic pricing. In the case of the health care services industry, economic regulation can also be designed to constrain total expenditures in response to what appears
244 to be supply-induced, price-inelastic demand characteristics. This relative inelasticity of demand reflects in part the unconstrained nature of the financing system wherein the demand for health care services has many of the characteristics of a free good. The basic human values at stake in the provision of health care services and the growing preoccupation with new technology as a means of delivering a greater intensity of medical services further contribute to the problems of con- straining total expenditures. This regulatory goal is achieved in a variety of ways, including establishing prices or tariffs either across the board or differentially by service or product, by constraining total institutional revenues as a function of actual or prospective costs, or by rationing or otherwise in- fluencing the utilization of services. For purposes of our anal- ysis, an important distinguishing characteristic is whether the focus of the regulation is on overall costs or revenues or whether it becomes involved in decisions about the utilization of specific services or products. For example, Phase IV of the Economic Stabilization Program for hospital services (and the Carter hospital cost control proposal as well) sought to con- strain average revenues per admission, leaving it up to the hospital to determine the mix of services, new technology, etc., that could permit needed care to be provided without costs ex- ceeding allowable revenues. In British Columbia and other Canadian provinces, however, provincial officials approve hospi- tal budgets on a line-by-line basis that requires them, and not just the hospital administrators or medical staff, to share in the trade-off and investment decisions. Why the Impact of Regulation is a Timely Issue The rapid growth of medical technology is widely acknowledged as a major contribution to the inflation in health care costs whether measured in terms of cost per day, cost per admission, cost per encounter, charge per unit of service, expenditures per capita, or total personal health care service expenditures. Gaus and Cooper have estimated that the "technology factor" (changes in intensity from new equipment, supplies, and person- nel) represents 47.3 percent of the 9-year increase in the expense per patient day in community hospitals from $49 to $l47.2 Although not all new technology contributes to higher costs or higher intensity, there is little doubt that new tech- nology presents increases in both unit and system costs. The role of new medical technology in the health care cost spiral, coupled with a growing consensus for increased regulation of the health care for other reasons, almost certainly means
245 that technological innovation, adoption, and diffusion will be increasingly under the influence of public regulation, much of which is at the state and regional level. While regulatory initiatives are occurring in a number of areas, regulators and students of the subject are looking more and more to a strategy of controlling supply and capacity a,s a way of constraining demand and, thereby, total expenditures. This strategy can be carried out through direct controls such as certificate of need strengthened by the application of regional caps, and indirectly through reimbursement controls. If this tack is pursured, policymakers will need to understand more clearly than they now do how such strategies will affect the adoption and utilization of medical technology as well as other capital investments. The capacity of regulatory bodies to be sensitive to sound technology policy is very much at issue. First, most regula- tory modes do not incorporate specific goals with respect to technology, nor have they even considered the need for such goals. Second, there are few guidelines available to help planners and regulators make the trade-offs among alternative service/technology options. Finally, there are serious ques- tions as to who should be making what can become triage decisions. As the Congress appears reluctant to extend open- ended coverage like that afforded under the End Stage Renal Dialysis Program to other diseases, the capabilities of medical technology may outstrip the public's willingness to finance its application and "who shall live" decisions may become increas- ingly acute and politically uncomfortable. For these reasons, the likely impact of state regulation on medical technology and the need for a more conscious linkage between regulatory goals and technology policy must be better understood by policymakers. II. ANALYSIS OF THE IMPACT OF STATE AND LOCAL REGULATION This section reviews six types of state or local regulations that have the potential to influence the introduction and diffusion of medical technology. The six regulatory programs reviewed are: Certificate of need Reimbursement controls Medical devices regulation Utilization review and quality assurance programs Institutional licensure and accreditation Personnel licensure
246 For each, the purpose and administrative mechanisms are described, the impact on the introduction of new medical tech- nology examined conceptually and, where possible, empirically, and the policy implications of this analysis presented. The section following this one addresses policy issues that cut across individual types of regulation. Certificate of Need Certificate of need is a program of public review and approval of the capital expenditures and service changes of health care providers. Review in most states is conducted by both state agencies and local health systems agencies, with local review advisory and state review final. As part of this review process, judgments are made on the need and desirability of proposed expenditures. In reaching these judgments, reviewers can consider a wide range of factors, including the efficacy or usefulness of the proposed project, the appropriate distribution pattern for the service or equipment sought (both geographically and with respect to the preferred provider or setting), the reasonableness of the financing arrangements, cost and proposed charges and financial feasibility, the adequacy of the proposed staffing and physical layout, and the competence of the provider and overall quality of the services to be offered. Because of the scope of review, certi- ficate-of-need programs directly confront states with the full range of issues affecting the introduction of new medical technology. These programs are widespread. Currently, 33 states have certification programs. Thirty-seven states have contracted with the federal government to do similar reviews under the authority of Section ll22 of the Social Security Act. Only Missouri has neither program.3 The National Health Planning and Resource Development Act of l974, P.L. 93-64l, virtually mandates all states to enact certificate-of-need programs. The programs do not cover all medical equipment purchases, but they are extensive. The federal regulations for state programs require hospitals, nursing facilities, health main- tenance organizations, kidney treatment centers, and ambulatory surgical facilities to be subject to review. For these pro- viders, all capital expenditures over $l50,000, bed changes, and all new services must be reviewed. The services offered and equipment purchased by private practitioners are not included in most laws, but two states have introduced some controls on physicians' purchases, and more are considering this step.
247 These programs are intended to serve a number of purposes. There is a concern with cost containment, although it is a very specific one. The impetus for these programs has been a belief that there has been uncessary overinvestment, leading to inefficient use of capital and additional operational expenses, and encouraging inappropriate utilization of facili- ties and services. The concern is not with absolute levels of costs, nor with the desirability or impact of expanding medical services per se. Most programs have retained a commitment to assuring access to care that is at least as strong as their cost containment goal. There have been several analyses of the effects of certifi- cate of need on the investments of health care institutions. Hellinger, based on early data, concluded that certificate- of-need legislation had not lowered hospital investment and that hospitals had increased their investment in the period immediately preceding the enactment of the legislation in anticipation of the program. Salkever and Bice concluded that certificate-of-need pro- grams had constrained the increase in hospital bed supply but did not reduce total hospital investment. Instead, assets per bed had increased faster in certificate-of-need states, suggesting that investment had been redirected into moderni- zation and special equipment. They also found that certificate of need was associated with decreases in patient days but increases in the costs per day. Since these have opposite effects on the overall costs of health care and the relative magnitude could not be determined, the ultimate impact of certificate of need on total costs was not estimated.5 In an update of their analyses with data through l974,6 the authors found continued support for their earlier conclu- sions. An analysis was made of the experience of the five states that had certificate-of-need programs prior to l97l, and for these states the programs were found to be associated with control of assets per bed, although at levels the authors judged not statistically significant. (The significance level was 0.07.) Four of these five states, it should be noted, have institutional rate review programs. In its l975 study of the impact of the Section ll22 and other programs to control capital expenditures, Lewin and Associates7 concluded that these programs had not been parti- cularly successful in controlling hospital investment. Forty- six percent of the 20 states and 40 areawide agencies included in the sample had approved hospital beds in excess of their published 5-year need projection. While approved increases in assets per hospital bed between l972 and l974 were less than the estimated amount of inflation plus economic growth
248 in most surveyed states with these programs, this could not be compared with the growth of assets in states without this form of regulation. Those states that were most successful in controlling per bed assets generally had some form of institutional reimbursement control. The Lewin study also found that proposals to purchase equipment and add new hospital services are almost always approved. The principal exceptions are X-ray and other scanner equipment, where the still infrequent denials are largely the result of several institutions applying to purchase the same equipment and only one being approved. A table of approval rates is presented in Table l. In sum, based on evidence drawn from the first years of operation of these programs, there is little evidence that operating independently they have limited the ability of institutional providers to invest in new technology. There is also some evidence that the process, by defining its coverage in terms of types of providers rather than types of TABLE l Approval Rates by State Agencies under Certificate of Need and ll22 Reviews for Equipment and Service Projects in Sample States Applications Percent Received Approved Equipment CAT scanners 9l 96.7 Radiological l06 93.4 Renal dialysis 24 l00.0 Cardiac catheter 9 l00.0 Other equipment 5 l00.0 All equipment 235 95.7 Services ICU-CCU l4 92.9 Renal dialysis l4 78.6 Other specialty 69 94.2 Emergency medical services l9 89.5 Other outpatient 73 97.3 Rehabilitation 28 l00.0 Other ancillary 8 l00.0 All services 224 94.6
249 services that require regulation, has encouraged the intro- duction of secondary and tertiary level technology into primary care settings. For example, the Office of Technology Assess- ment reported that l5 percent of CAT scanners were in private Q physician offices or clinics. Several factors have influenced this current level of performance by these agencies. The programs are relatively new, as are many of the agencies involved. The ideal situation would be one in which the standards for project review, in- cluding definitions of need, were established prior to project submission, so that they could be immediately applied. In most agencies, however, planning activities and standards development have lagged. A frequent pattern has been for agencies to approve the first project (sometimes the first several) of a given type and begin developing a standard for a new service or equipment after the first application has been received. The standards development process itself has a number of weaknesses. There is now available a considerable body of literature with suggested standards that have been developed at both the national and local levels. These materials cannot simply be adopted by a health planning agency. Rather, they are reviewed to address local circumstances and patterns of treatment. These reviews often blend local health politics with technical adjustments. In addition, the development of standards for new services, particularly standards regarding the appropriate number of service units, can often only be done once there has been experience with the service. Requests for new technology, however, will often precede the development of information on efficacy or levels of utilization, placing pressure on review agencies to act before standards can be developed. There has been a tendency in the agencies to grant approval when uncer- tain. (This is not inevitably the case; it is a function of agency orientation. In Quebec, under the province's hospital insurance program, which includes both certificate of need and reimbursement authority, the province's announced policy is to limit technological diffusion to services of demonstrated efficacy. As part of this effort, the province had limited the introduction of in vivo radioisotope facilities to only a few hospitals and required these to develop research protocols to evaluate the efficacy and cost impact of these services. ) There also appear to be some areas in which prior standards development is infeasible, or at least very difficult. Some of the most subtle decisions agencies must make involve deter- mining the appropriate level of technical sophistication in existing services. This includes questions such as: Does a
250 l50-bed community hospital need tomographic or radioisotope capacity in its radiology department? Or, is laboratory volume sufficient to justify an automated analysis and, if so, with what capacity? Standards have not been developed because the decisions involve a great many situational factors that limit general answers. Since these issues do not involve large projects, they have received lower priority in the standards development process than bed planning or plans for entire services. Yet, collectively, these decisions have a major impact on the technological level of routine care. Finally, there is an issue of whether the entire standards developmentâand indeed certificate-of-needâprocess is mis- oriented. Currently, these efforts are directed at assessing the need and appropriateness of services, beds, and equipment on an absolute standard of whether the service is productively and efficiently used. If so, it is approved. The process does not address the basic issue of establishing priorities to allocate limited resources. No certificate-of-need program has incorporated into its planning process judgments of the rate at which resources should be added to the health care system or medical technology diffused. It is therefore not surprising that these programs seem most able to limit invest- ment in states where reimbursement control programs have introduced separate limits on available resources. It is within this context that the proposal to limit hospital investment incorporated into the administration's cost con- tainment program should be noted. If adopted, this limit would introduce a fundamental change into the certificate-of-need process. Agencies will be forced to assess projects on the basis of relative as well as absolute need. This has occurred on a limited scale in Rhode Island under the reimbursement control program for hospitals, but for other agencies this would be a new experience. Among its other consequences, it is likely to place basic service expansion into competition with new technology. The potential impact of certificate-of-need programs on the adoption and diffusion of technology is great. This includes the potential to severly limit the total level of investment. Without the introduction of financial pressures into the certificate-of-need environment, however, the relative ease with which new services and equipment have been approved is likely to continue.
25l .Reimbursement Controls While certificate of need can directly regulate the introduc- tion of new technology, the effect of reimbursement controls on service decisions is potentially greater. Ultimately, reimbursement policy influences the resources available to invest in services and equipment and the incentives for their use. Furthermore, the effects of reimbursement may be unin- tentional, inadvertent results of policies directed toward other goals or priorities. In certificate of need, there is at least explicit consideration of the value of the tech- nology. All payment programs reimburse providers within specific rules. In this sense, all have reimbursement controls. The term is generally meant to distinguish a payment system from those widely in use that give providers considerable discretion to determine their reimbursement. Physicians For physicians and invidual practitioners, there are two common payment approaches used by third-party payers. One is the payment of the provider-established usual and customary fee. The other is the payment of a flat amount under a fee schedule established by the payer (sometimes with provider participation). The provider may ask patients to supplement the fee schedule payment or, for those public programs where he cannot, may refuse to treat the patient. Under fee sche- dules, payment for procedures that are not included is at the provider-established level or based on a comparison of com- parable procedures. The main controls that are introduced into these systems are to limit the ability of the provider to charge an additi- onal amount and to refuse to reimburse for procedures not initially included in the schedule. This last has particular relevance to the development of new techniques and procedures. The U.S. experience with controls on physicians' fees is limited. Medicare only reimburses for approved procedures, and DHEW has established a system in which the Bureau of Quality Assurance, Food and Drug Administration, and National Institutes of Health will be consulted to determine whether a new procedure should be added, based on demonstrated safety and efficacy. Some Blue Shield plans negotiate fee schedules that are accepted as payment in full by participating physi- cians. These types of activities are not common at the state level. There has been limited systematic examination of the impact
252 of fee levels on practitioner behavior. There is limited evidence from Canada that physicians modify their patterns of practice in response to fee levels. The provincial govern- ments have considered adjusting the fees for specific services in an effort to use this effect (e.g., increase vaccinations, decrease routine physical examinations).10 In the United States, it has been claimed that the development of proprietary hemo- dialysis centers was fueled by high levels of reimbursement and that some procedures, such as cardiac bypass, are used in marginal cases because they are well reimbursed. Conceptually, it can be argued that manipulating fee schedules can provide one mechanism for modifying physician use of specific treatment methods or technologies. The extent to which fees would influence choice is not known, however. Changing physician practice patterns would have an indirect effect on the medical technology investment decisions of hospitals, since these are made largely in response to staff requests. There are two current practical limits on the use of this technique. First, not enough is known about how physicians respond to fee levels. Specifically, it not clear whether most physicians respond to fees for individual procedures or whether the response is to the total pattern of fees in a field that, taken together, defines the potential earnings of practicing in a specific technical style. Until this is better understood, a comprehensive regulatory strategy could not be constructed. Efforts to manipulate the use of specific technologies might nontheless be attempted. The second practical limit is the likely inability of states to enact regulatory programs to control physician fees and the inability of individual purchasers to impose the stringent fee limits necessary for this policy. Physicians have been in the forefront of opposition to the financial regulation of institutional providers, largely because of their fear that this is an initial step to the regulation of their incomes. In many states, this opposition has been a key factor in limiting the scope of the programs developed. It is unlikely that in any state their opposition could be overcome and general regulation introduced. The regulation through individual third parties is flawed by the general effects differentials in fee levels haveâre- fusal by physicians to treat patients, requiring additional payments of them (in the case of Medicaid and Medicare, ille- gally) , and encouragement of fraud. Whether this would occur if general fee levels were comparable but specific technology- related fees had differentials is uncertain. In all likeli- hood, it would not. But unless a given technology encompasses
253 a significant portion of a physician's practice (such as in the dialysis situation), or unless no reimbursement at all is available, adjusting fees should only have a marginal effect. Hospitals The incentives that can be created at the institutional level through reimbursement are much greater than at the physician level, primarily because regulation can be introduced for institutional providers with greater ease (although not easily) and because they are more sensitive to the impact of changes to reimbursement patterns from individual third par- ties.11 The two most common forms of reimbursement for hospitals are payment of hospital-established charges or payment of an allocated portion of actually incurred costs. In both these cases, it is the hospital's decision that determines the level of costs and reimbursement for services. Coupled with the growth of third-party payment and the concommitant reduction in bad debt and charity loads, these payment mechanisms have improved the financial conditions of hospitals and generated the funds used for service expansion and upgrading technology. Nine states have introduced programs to regulate the pay- ments to hospitals and establish payment rates on a prospective basis. These programs vary in their coverage. Some only cover Medicaid and/or Blue Cross rates (e.g., New York and New Jersey). Some cover all payers (e.g., Connecticut, Mary- land, and Washington). One is a program of nonbinding mandatory state review of rates (Arizona). In some cases the programs regulate per diem or individual service charges; in others, total institutional budgets or revenues. There is some evidence that controlling reimbursement has an impact on capital expenditure plans and, consequently, the introduction of new technology. A survey of hospital admini- strators on the effects of the Economic Stabilization Program at the end of the first year of the program indicated that approximately one-fifth had reduced their capital expenditure plans for that year, and that the average reductions among those was 43 percent. Approximately l3 percent indicated that they had foregone the introduction of a new service.13 Cromwell examined the impact of Economic Stabilization Programs and concluded that it had lowered the rate of adoption of complexity-expanding and community services, although this effect declined over time. A comparison of a sample of New York hospitals to a sample from a state without prospective reimbursement indicated that the adoption of complexity- expanding services in New York was lower, although because of
254 the existence of the certificate-of-need program, he hesitated to attribute this solely to the reimbursement controls. For smaller hospitals, contraints were associated with changes in net revenues, although with significant lag. No similar constraints were found for larger hospitals, which the author speculated reflects their better access to external funds.14 Thus, the potential for reimbursement controls to influence the investment decisions of institutional providers appears to be confirmed. Not examined in these studies is the impact of alternative methods of introducing controls over the capital and operating budgets of providers. There are two broad alternative approaches to the imposition of reimbursement controls. One is to exercise close scrutiny over the individual expenditures or charges of the provider. Under these systems, the efficiency of individual departments or appropriateness of specific activities is examined. The results of these reviews can be implemented in two ways: by prohibiting the institution from making an expenditure or by establishing reimbursement at levels that will not allow the expenditure to be recovered if it is made. The latter method is more common and is assessed in the discussion that follows. To take an example, under the detailed review or line-item approach, if a radiology department were judged overstaffed, an amount equal to the estimated savings would be subtracted from the budget base. If a service were judged unnecessary, the estimated operating expenses would be subtracted from the budget. If a service or new piece of equipment is to be added to the hospital, its estimated operating expenses would be added to the budget base. Alternatively, a standard reim- bursement rate can be established for a service, such as a brain scan, based on estimates of the per-unit expense of an efficiently organized institution providing an appropriate volume of services. For those services judged inappropriate for a given institution, the rate might be zero, that is, the service would not be reimbursed. Establishing standard reimbursement rates allows the institution to calculate whether it can break even on the service. In either case, whether the adjustment is to the budget base or the reimbursement rate, the institution is free to continue the service if it can finance it from other sources, such as gifts or research grants. With respect to the introduction of new technology, the line-item approach requires reimbursement agency concurrence in each investment decision. This allows considerable control by the state, but also requires greater sophistication in the reimbursement agency. One element in this sophistication is a health planning capacity similar to that required in
255 certificate of need to evaluate the need and appropriateness of the proposal. In fact, this is usually obtained by accep- ting the certificate-of-need decision as a commitment by the state to allow reimbursement for the service. The level of reimbursement may be disputed, however. There have been instances where the reimbursement agency has not agreed with the level of capital or operating cost approved by the certi- fication agency and renegotiated the scope of the project. To limit the possibilities for this occurring and to approve the level of cost and financial analysis in certification decisions, reimbursement programs are increasingly partici- pating in the analysis of certificate-of-need applications. The alternative to line-item review is establishing overall limits to the growth of hospital charges, budgets, or revenues. The hospital is free to allocate its resources at its discretion. This was the approach taken in the Economic Stabilization Program and in the administration's cost-containment proposal. It is essentially the method used in New York and as a screen- ing device in New Jersey. Under an overall increase approach, the hospital retains its discretion to invest in new technology (subject to other regulatory constraints). The control's main impact is on the level of funds available to support these investment decisions. Programs can directly manipulate the amount they make available. In the final Phase IV regulations under Economic Stabilization, for example, an estimate of the base inflation rate for unchanged operations was adjusted to reflect estimated improv- ments in productivity (l percent) and an allowance for increase intensity, i.e., an increase in the complexity and number of services provided per patient of 2.5 percent. Both these factors reflect assumptions about technological change. The New Jersey reimbursement methodology adopted in l976 assumed a l percent increase in intensity to be offset by a l percent increase in productivity. This was a statement of public policy regarding new investment in sophisticated services rather than a history-based projection. The New York system includes neither a productivity nor an intensity factor, implicitly assuming that the production function does not change from year to year. Because rates have been exclusively cost-based until this year, if a hospital could finance ex- pansions in services through working capital or charges to patients not covered by the control program, these would ultimately be built into the rate base. It is clear that other policy goals, such as contraction of the system, could be incorporated into overall increase limits. It should be noted that no system in use is a pure repre- sentative of either approach. The Economic Stabilization
256 Program and New York systems both allowed providers to request changes in their rate bases to reflect large capital expendi- tures and their associated operating expenses. These reviews were quite similar to those described in the line-item approach above. In addition, specific ceilings, penalities, or incentives can be incorporated into the general increase approach. For example, in New York, hospitals are expected to main- tain minimum volumes of certain complex services (open-heart surgery, cardiac angiography, transplantation) or rates are adjusted down by formula. Similarly, in line-item systems, review is simplified through the use of analytic screens. If technological changes can be introduced within these limits, they may not be subject to detailed review. (Some programs, however, focus their review on operational and technical changes by asking for separate line-item capital budgets and breakdowns of new costs to be incurred because of changes in service.) The Rhode Island system of project prioritization mentioned in the preceding section on certificate of need is another illustration of the blending of these two approaches. Rhode Island engaged in line-item budget negotiation within a fixed total budget ceiling for the state. The inclusion of new services into budgets was subject to agreement on the level of operating costs of existing services. This examination of institutional reimbursement controls makes it apparent that any control program includes an explicit or implicit policy toward capital expenditure in general and new technology in particular. Such a policy has several com- ponents : â¢ It sets limits on the overall level of allowable invest- ment, although these may be fluid. Line-item systems without priorities or targets for total increase limits can be prepared to add any amount of new services to the system, so long as they have been certified as needed by the appropriate planning agency. â¢ It determines who is to be involved in establishing priorities among new technology, specifically defining the relative roles of regulators and institutional personnel. Regulatory personnel are not a homogeneous group, but can be divided broadly into service planners and financing categories. State policy determines not only the degree of institutional autonomy but also the relative participation of the two sets of regulators. â¢ It distributes the costs of new technology. This was not discussed in detail above, but one of the critical decisions
257 made in reimbursement regulation is the distribution of costs among different patients and third parties and between patients and other sources of revenue such as gifts, tuition, or research grants. All these are inherent in rate regulation. The regulatory mechanism is flexible enough to accommodate virtually any policy desired within these components. Medical Devices Regulation Under the l938 Food, Drug and Cosmetic Act, the federal govern- ment assumed a number of regulatory roles with respect to drugs and medical devices. The drug role was an extensive one, involving premarket approval of new drugs, as well as authority to require reports of manufacturers and monitor the manufacturing process and files of the producer. Approval to market a drug could be withdrawn. With respect to medical devices, however, the authority was limited. The government was given the authority to seize a sample of a misbranded or adulterated device and begin judicial proceedings to halt distribution. The burden of proof was on the government, and sales could continue while the case was being adjudicated. Most states followed the federal lead and adopted laws allowing them to proceed against devices that were adulterated or misbranded. In general, the guidelines adopted for assess- ing this followed the federal guidelines, so the net effect was to create a national approach to regulating medical devices with both federal and state enforcement mechanisms. In l976, the Medical Devices Amendments to the Food, Drug and Cosmetic Act were signed into law. These strengthened the regulatory authority of the Food and Drug Administration to control medical devices, increasing the information device manufacturers were required to provide the government and shifting the burden of demonstrating safety and effectiveness to the manufacturer. The act requires the government to classify all devices into one of three groups defined by the appropriate level of regulatory control the device requires. The groups are: premarket approval, establishment of perform- ance standards, and general controls regarding adulteration and misbranding. To date, the Food and Drug Administration has not published its proposed regulations classifying devices. The agency is not likely to quickly follow the final classi- fying devices. The agency is not likely to quickly follow the final classification regulations with its performance standards.15
258 While the amendments preempt state activity in this area, they contain provisions that allow states to request an exemp- tion to administer their own medical devices program. It is unlikely that this provision will lead to intensive state regulation or that the introduction of state programs would limit the introduction of new medical technology more than the federal program operating alone. The judgment that states are not likely to adopt this form of regulation is based on four factors: â¢ California is the only state currently operating a similar program that has expressed its interest in receiving a waiver. The low level of activity by other states in this area is probably an accurate reflection of interest. â¢ The resources, especially the technical resources, needed to develop a program encompassing performance standards and premarket clearance are extensive and only a few states are capable of developing a program in this area. â¢ Unlike some forms of regulation in which state actions to complement or reinforce federal programs are encouraged (e.g., fair trade), the intent of this act is federal pre- emption and states are less likely to move into a new area of regulation under these circumstances. â¢ States are likely to be preoccupied with regulation in the health planning and finance areas. The resource issue also makes it unlikely that state pro- grams will operate so at variance with federal efforts that it will impose additional barriers. It is useful to distin- guish between the enforcement aspects of these programs and the technical evaluation aspects. California's interest is sparked primarily by concern over enforcement. The state began its program of premarket approval because it felt federal efforts were inadequate and that stronger licensure and surveil- lance were required. A primary reason for seeking the exemption under the Medical Devices Amendments is concern over the time the federal government will take to establish standards and begin operating the program. These are enforcement issues. On a technical level, however, the state is unlikely to duplicate the standards development process undertaken by the federal government. It is, among other things, very expensive. Currently, the state is essentially a consumer of standards, using those developed by national organizations. It will probably adopt federal standards, and currently a bill is pending in the legislature that would require this. Similarly, the California requirements for premarketing demonstrations of safety and efficacy are also likely to be comparable to
259 the requirements adopted by the federal government, at least in terms of basic approaches to be used. The state may examine the evidence more or less critically, but is not likely to require a different research protocol. Marginal modifications rather than new systems should be expected from the states in this area. The parallel operation of this form of regulation by the federal government and a small number of states might have a negative impact on the diffusion of efficacious technology in several ways. First, since the main reason for California's interest in the program is to bring devices under regulation sooner than the federal government is prepared to, it is likely that the state will establish standards before the national standards are established. If the federal requirements are more stringent than state standards, it is likely that the standards will be upgraded. If federal requirements are less stringent, the state may not downgrade, producing a situation comparable to that with respect to automobile emission standards. With a tenth of the medical market in California, it is unlikely manufacturers would fail to modify equipment and leave the state unsupplied with the technology, but responding to two separate standards might increase the pro- duction and distribution costs. Costs might also be increased by the need for manufacturers to deal with two separate sets of regulators. This is an additional, marginal cost, which may be high enough in some cases to make the introduction of a new device infeasible. If our judgment of the extent to which state standards of demonstration are likely to be consistent with federal stan- dards, and if the state and federal governments are prepared to coordinate their requests of manufacturers even while reserving the right to make independent judgments, the impact of this factor should not be great. Utilization Review and Quality Assurance Programs Another form of regulation that might have an impact on the adoption and diffusion of medical technology are programs to review the appropriateness of the utilization of health care services and the quality with which services are provided. For the most part, these are not state regulatory activities but are locally and regionally organized. The concept of reviewing medical care practices to assess the quality of care and appropriateness of the services provided has evolved over a period of 40 years. One of the first formal organizational settings for such review was
260 created in l954 when the first private medical foundation was established to review patterns of care within the context of fee-for-service independent practices. The development of Medicare in l965 increased the demands on the medical system to develop mechanisms to assess utili- zation by requiring all institutional providers to have utilization review programs for their Medicare patients. In l967, this requirement was extended to Medicaid. These pro- grams were institutionally organized, although requirements were established for state medical audits of nursing home patients under Medicaid. The Social Security Amendments of l972 created the Pro- fessional Standards Review Organization (PSRO) program. This program was a distinct break from previous utilization review efforts in that it called for the establishment of a network of review organizations external to the individual institutions and separate from government to assume responsibility for assuring that health services reimbursed by federal or feder- ally assisted programs under the Social Security Act are medically necessary, meet professionally recognized standards of care, and are provided in the most economical and appro- priate setting. To meet this responsibility, PSRO's will review admissions to hospitals and other health facilities, certify the need for continued treatment, review extended or costly treatment, and review the patterns of practice of individual providers, institutions, and the community as a whole. The PSRO's will, as part of their activities, conduct medical care evaluation studies and develop standards and protocols for care of specific conditions. PSRO's have initially focused on reviewing the length of stay of patients, and the few evaluations done to date have focused on this aspect of their performance. The results, while preliminary and needing more analysis and verification, suggest that most programs have not affected length of stay.16 It is in the development of the standards for appropriate patterns of treatment and the application of these standards to review of individual cases or provider profiles that these programs might most influence the adoption of new medical technology. There is a strong likelihood that the standards they develop will become norms for medical practice. Aside from the pressures inherent in the peer review process, the current malpractice crisis creates additional incentives for physicians to practice consistently with one another. State actions to address the malpractice problem by establishing new insurance pools or changing the liability laws might
26l therefore have an influence on the impact of the PSRO program on provider practice patterns. If a PSRO is conservative, it can create pressures to main- tain current practice patterns and retard the introduction to new methods and technology. If a PSRO adopts standards and treatment protocols that call for the routine use of the best available technology, this might create pressures on institutions for new investment to meet the recognized commun- ity standard. While there is considerable conjecture on the likely direction this standard-setting process will take, there are no current studies of the impact of PSRO's on the level of technology used in medical treatment. Nor is there any reason to assume that the impact of PSRO's will be con- sistent, given local governance. Variations, reflecting differences in the perspectives of the medical community in different regions, are rather to be expected. There are several vehicles to coordinate policy among PSRO's. These include the National Professional Standards Review Councils and the DHEW staff in the Bureau of Quality Assurance. Efforts to introduce common approaches for incorporating judgments on new technology into local standards of practice will have to be made through these coordinative bodies. There is another role PSRO's might play in the introduction of new technology. It is to serve as the locus of efforts to evaluate the impact of new technology on the costs and outcomes of medical care. Such a role is envisaged in the scope of medical care evaluation studies that these agencies are obligated to conduct. The PSRO program manual, in giving examples of possible studies, includes: â¢ Studies of the outcomes of hospitalization for given diagnoses. â¢ A study of the use of combination antibiotics with the criteria specifying the indications and contraindications for their use. â¢ Detailed analysis of the process of care for a particular problem, the criteria used in the study to be based on scientifically derived evidence of the efficacy of given diagnostic or therapeutic procedures. For these last studies, the manual recommends that if no evidence is available, the best judgment of experts be used. An alternative approach would be for the PSRO to initiate or supervise the clinical studies needed to make these judgments or to cooperate with other efforts to do this through the use of their review records to assess outcomes. This would involve the PSRO's not only in the standards-setting process, but also the research process necessary to establish standards.
262 Institutional Licensure and Accreditation Institutional licensure is one of the most widespread forms of state health regulation. Virtually all states license hospitals, although it was not until l946 under the impetus of the Hill-Burton Act that most states developed comprehensive licensure laws. Other health facilities, such as nursing homes and clinical laboratories, are also frequently subject to licensure.17 Licensure is a mandatory program in which a provider must meet minimum standards in order to receive permission from the state to operate. The licensure programs focus on fire and life safety considerations, water and sanitation standards, minimum service standards, and guidelines for staffing and staff qualifications. There are generally three aspects to the licensing program: a review of architectural plans of proposed facilities for consistency with the construction code, an initial licensure inspection, and periodic inspections once the facility is opened. One of the few studies of licensing programs that has been conducted concluded that regulations were stronger, less ambig- uous, and more stringently enforced in matters relating to construction, maintenance, and sanitation of the physical plant than for matters relating more directly to patient care.18 It has been asserted that institutional licensure programs have intended to over specify the techniques and inputs that institutions must use. Closely related to licensure are two voluntary programs for reviewing institutional qualityâaccreditation and certification. Accreditation is conducted under private auspices and goes beyond minimum standards. The main accrediting organization in the United States is the Joint Commission on the Accreditation of Hospitals (JCAH), operating under the sponsorship of the American College of Surgeons, American College of Physicians, American Hospital Association, and American Medical Association. The program is intended to assess the quality of institutions, but, like licensure programs, has been accused of focusing too much on inputs and not enough on process or outcome measures. Certification is a process by which third-party payers, most notably the federal government but also Blue Cross, assess facilities as meeting the standards required to receive reim- bursement from the program. Under Medicare, JCAH accreditation is one basis for certification. If an institution is not accredited," it may be surveyed and assessed against Medicare standards. The federal government has contracted with the states to make this survey, and in most states it is conducted by the licensing agency using the federal standards.
263 Little is known about the impact of licensure or accredi- tation on the operations of hospitals and specifically on the introduction of new technology. The standards established focus on physical layout and personnel, and in the accreditation process on training and the maintenance of appropriate policies and procedures for each department. In determining the appropriateness of equipment and technology, both processes rely on the professional judgment of surveyors. The New York State standards for laboratory services, for example, require the extent and complexity of service to be "commensurate with the size, scope and nature of the hospital."19 The standard for radiology is comparable, calling for facilities to be available "according to the needs of the hospital." If radiotherapy services are provided at the hospital, they must "meet professionally approved standards for safety and person- nel qualifications."20 JCAH standards similarly call for professional judgment, requiring, for example, in laboratories, that "equipment and instruments be appropriate for the services required,"21 or "that radiation therapy facilities shall be adequate for modern and appropriate treatment according to the needs and size of the institution."22 The character of these programs may have several influences on the introduction and diffusion of new medical technology, although analysis of these influences is currently only con- jecture. Their direct effect on the introduction of new technology is likely to be slight, except in those cases where it affects basic construction methods or the layout of the facility, areas in which the standards are most specific. The attention these programs pay to the adequacy of personnel and in particular formal training and qualifications may slow the diffusion of technology if specialized training is required. Reliance on professional judgments by the review teams of the availability of appropriate equipment may, because of profes- sional orientations, encourage the introduction of new, "best available" technology. Personnel Licensure Personnel licensure laws are intended to assure the quality of health care by: â¢ Reserving certain functions to only those capable of doing them well and prohibiting others from performing them. â¢ Identifying the individuals judged able to perform the function.
264 In recent years, several complaints have been raised about the operation of these lawsâthat they often use the wrong criteria to assess competence; that the multitude of categories and variations among states has reduced occupational mobility of health workers; that they do not distribute tasks among different classes of personnel based on their ability to perform them, and, as a result, prevent the efficient delivery of health services; that the proliferation of narrow licensing categories has created not only inefficiencies but obstacles to innovation. In understanding the potential impact of these laws on the introduction of new technology, it is this last issue that is critical, and it is relevant primarily with respect to allied health professionals. Specifically, as new technology has been developed in the last decade, there has been a proliferation of personnel trained in the technical support of specific technology. The specificity of the technical knowledge required of the allied health personnel has encouraged movements for licensure of each specialty. The potential effect of too many narrow licensing categories on health care institutions may be to reduce their ability or increase the costs of introducing technological change/ by fixing staffing patterns and assign- ments, by requiring larger investments in manpower when introducing new services, and by increasing training costs. It is not clear that licensing regulation has had any of these effects, and interviews with individuals knowledgeable about licensing programs suggest that the impact has not been great. Indeed, impact appears to have been in the other direction, with rapid changes in technology leading to a multiplication of professional and occupational categories. The growth in new state licensure categories was raised as a federal concern in the Health Training Improvement Act of l970, and the federal government prepared a report on licensure and related programs in l97l. That study identified l7 cate- gories of allied health personnel which were licensed in at least some states. These included: Clinical laboratory directors Clinical laboratory medical technologists Dental hygienists Inhalation therapists Midwives Opticians Optical technicians Optometrists Pharmacists Physical therapists Physical therapy assis- tants Physician assistants Psychiatric attendants Psychologists Radiologic technologists Sanitarians Sanitarian technicians
265 As a result of the study, the federal government called for a 2-year moratorium on state adoption of new licensing categories, while additional study was made of a number of issues including: the potential of national volutary certification programs to serve as a substitute for state licensure, the development of national standards for health worker licensure, and the need for continuing education activities and periodic relicensure.23 In l973, DHEW recommended a 2-year extension of the moratorium. 1> The moratorium appears to have had only a limited effect on state actions in this area. A recent report by DHEW on the issue, Credentialing Health Manpower, calls for a restraint in the introduction of state licensing for new categories of workers. It advises states to "entertain proposals to license additional categories of health personnel with caution and deliberation." As an alternative, the report stresses the 'role that can be played by private certification programs. These programs are also intended to identify individuals able to perform specific functions. Unlike licensure, the category is not defined by law and the functions are not reserved to the indi- viduals certified. Certification programs are often related to the completion of formal education programs in the field and the range of such programs gives an indication of the scope of certification activities. The American Medical Association Council on Medical Education has established accreditation standards for 26 programs. These include (with the date the standards were established): Occupational Therapist (l935) Medical Technologist (l936) Physical Therapist (l936) Medical Record Administrator (l943) Radiologic Technologist (l944) Medical Record Technician (l953) Cytotechnologist (l962) Respiratory Therapist (l962) Certified Laboratory Assistant (l967) Radiation Therapy Technologist (l968) Medical Assistant (l969) Nuclear Medicine Technician (l969) Nuclear Medicine Technologist (l969) Orthopaedic Physician's Assistant (l969) Histologic Technician (l970) Assistant to the Primary Care Physician (l97l) Medical Laboratory Technician (l97l) Specialist in Blood Bank Technology (l97l)
266 â¢ Medical Assistant in Pediatrics (l972) â¢ Operating Room Technician (l972) â¢ Respiratory Therapy Technician (l972) â¢ Urologic Physician's Assistant (l972) â¢ Electroencephalographic Technician (l973) â¢ Electroencephalographic Technologist (l973) â¢ Surgeon's Assistant (l974) â¢ Opthalmic Medical Assistant (l975) The advantage certification may have over licensure in this area is that it is not tied to legal limits on the scope of activities. Hospitals are encouraged to hire certified per- sonnel because of the institutional accreditation process, but they are not barred from using uncertified personnel or shifting them to new functions. The certification process appears also to be better able to shift standards in the light of changes in technology and, for both these reasons, to be more responsive to an environment of rapid technological change. III. POLICY IMPLICATIONS Even though the impact of state regulation on medical technol- ogy cannot be fully assured, it is clear that the potential exists for the various forms of regulation to have considerable impact. This effect will likely be felt at both the level of aggregate capital investment and the level of individual equipment and service decisions. The regulatory modes most likely to influence the adoption and diffusion of medical technology are certificate of need and reimbursement controls. The growing emphasis on cost control, which guides their policies, places a premium on technological innovations that result in more efficient use of resources. However, these two programs, and reimbursement controls in particular, do not focus on technology policy per se, and their impact could be inadvertently and unintention- ally adverse. If the appropriate development of new technology is to become a matter of public activity, then a productive and rational relationship between regulatory and technology policies will have to be developed. For this to occur, the scope of these policies, the individuals to be involved in making policy, and the technical capacity to make decisions consistent with policy will all have to be established. Currently, there is little formal consideration of technol- ogy policy in regulation. But it is not clear what the nature of such a policy would be. There is no agreement on the current
267 need for new technology or its priority relative to other goals for the medical system, such as improved access to primary care. Without specific goals and objectives with respect to technology, and the integration of these goals with others, the explicit consideration of these issues in regulatory settings is impossible. Critical in the process will be the questions of who should be involved and how should technology policy be set. This is likely to remain a decentralized process, in which indivi- dual regulatory programs preserve their right to establish their own approaches. Conflicts will occur among agencies that will be negotiated much as other current conflicts are. The processes of local policymaking and conflict resolution can be improved, however. There is a critical need for the potential decision makers to obtain much more information than they have on the nature and scope of these issues, the effect their decisions can have, the critical issues or factors that should be considered as they establish policy in their programs, and the best available objective information and analyses by advocates of various positions on the current technological orientation and needs of the American medical system. Finally, moving from the level of policy to individual decisions, it is not at all clear that the technical capacity to make good planning, review, or investment decisions regarding the diffusion of medical technology exists within the regulatory agencies or, for that matter, elsewhere. Considerable effort will be required so that technical questions and issues of human values can be addressed by regulatory agencies, where appropriate, in logical and informed ways. NOTES AND REFERENCES l. American Hospital Association. Hospital Statistics: l976 Edition, p. 4. Chicago: AHA, l976. 2. Gaus, Clifton R., and Cooper, Barbara S. "Technology and Medicare: Alternatives for Change." Presented at the Conference on Health Care Technology and Quality of Care, Boston University Health Policy Center, November l9-20, l976. 3. Department of Health, Education, and Welfare. Bureau of Health Planning and Resources Development. "Status of Certificate of Need and Section ll22 Programs in the States (as of June 30, l977)." n.d. 4. Hellinger, Fred J. "The Effect of Certificate-of-Need Legislation on Hospital Investment." Inquiry l3(June l976)(2):l87-93.
268 5. Salkever, David S., and Bice, Thomas W. Impact of State Certificate of Need Laws on Health Care Costs and Utili- zation. Washington, D.C.: National Center for Health Services Research, l976. 6. Memorandum to Ira Raskin, Bureau of Health Planning and Resources Development, from Thomas Bice and David Salkever, on the Implications of CON Findings, March 30, l977. 7. Lewin & Associates, Inc. Evaluation of the Efficiency and Effectiveness of the Section ll22 Review Process. Springfield, Va.: National Technical Information Service, l976. 8. U.S. Congress, Office of Technology Assessment. "The Computed Tomography (CT or CAT) Scanner and Its Impli- cations for Health Policy." Cited in Institute of Medicine. Computed Tomographic Scanning: A Policy State- ment, p. 52. Washington, D.C.: National Academy of Sciences, April l977. 9. Lewin & Associates, Inc. Government Controls on the Health Care System: The Canadian Experience, p. 4.30. Springfield, Va.: National Technical Information Service, l976. l0. Ibid., p. 2.26f. ll. For example, the penalties for failure to conform to a decision under the Section ll22 program is limited to Medicare and Medicaid refusal to pay for their allocated portion of capital costs. Operating costs would still be reimbursed. In the Lewin study of the ll22 program, no cases were identified of institutions accepting this pen- alty in order to pursue an unapproved project. l2. See Lewin & Associates, Inc., An Analysis of State and Regional Health Regulation (Springfield, Va.: National Technical Information Service, l975) and Nationwide Survey of State Health Regulation (Springfield, Va. : National Technical Information Service, l974) for descrip- tions of these programs and programs operated by Blue Cross. Since those reports, Washington has introduced comprehensive rate regulation, Massachusetts has shifted from regulating only Medicaid rates to regulating rates for all payers, and Oregon has started to introduce a program. l3. American Hospital Association Bureau of Research Services. Findings of a Special Survey of Hospital Opinion and Behavior Regarding the Economic Stabilization Program. l4. Jerry Cromwell. The Impact of Rate Regulation on the Diffusion of New Technologies in Hospitals. Cambridge, Mass.: Abt Associates, October l976. l5. Interviews with FDA personnel, May l977.
269 l6. This conclusion is based on conversations with contractors developing and testing a PSRO evaluation protocol for the Bureau of Quality Assurance and a comparison of review determinations made independently by PSRO's and New York State utilization review employees. l7. Department of Health, Education, and Welfare. Licensing of Health Facilities, l968. Washington, D.C.: U.S. Government Printing Office, l972. l8. Fry, Hillary G. The Operation of State Hospital Planning and Licensing Programs. Hospital Monograph Series No. l5. Chicago: American Hospital Association, l967. l9. New York State Code of Rules and Regulations, Title l0, Chapter 5A, Article 2, Section 405.l028. 20. Ibid., Section 405.l029. 2l. Joint Commission on Accreditation of Hospitals. Accredi- tation Manual for Hospitals, l976 Edition, p. l36. Chicago: JCAH, l976. 22. Ibid., p. l59. 23. Department of Health, Education, and Welfare. Report on Licensure and Related Health Personnel Credentialling. Washington, D.C.: U.S. Government Printing Office, June l97l. 24. Department of Health, Education, and Welfare. Develop- ments in Health Manpower Licensure. Washington, D.C.: U.S. Government Printing Office, l973.