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 173
The Artificial Heart: Prototypes, Policies, and Patients 10 Conclusions and Recommendations AS THE PRECEDING CHAPTERS HAVE RELATED, the artificial heart program of the National Heart, Lung, and Blood Institute (NHLBI) is nearing one of its goals, that of developing a fully implantable device that offers indefinite maintenance of cardiac function to individuals suffering from end-stage heart disease. Surgeons will soon implant the first long-term ventricular assist device (VAD) that needs no skin-penetrating air hose or wire. It is therefore time, despite many uncertainties, to begin examining how appropriately these devices will be used in health care. Because of their life-or-death implications and high cost, their use poses profound economic, ethical, and clinical questions. At the request of NHLBI, an Institute of Medicine (IOM) study committee examined the agency's artificial heart program and the devices being developed for long-term use.1 This document is the committee's report. Nine questions set forth below, formulated in 1989 by another IOM committee, have guided this evaluation. A number of closely related issues have also been addressed, including government and private-sector roles in research and in the development of new technologies. To succeed in improving patient care, the goals of the NHLBI artificial heart program and the outputs of its R&D efforts must be compatible with the U.S. health care system, particularly because: 1 This discussion and the report as a whole do not consider devices that are intended only for temporary cardiac support, as they were not a focus of the study.
OCR for page 174
The Artificial Heart: Prototypes, Policies, and Patients on a per-patient basis, the mechanical circulatory support systems (MCSSs) being developed will be among the most costly therapeutic devices to date; MCSSs may be coming into general use as the health care system is under growing scrutiny for the quality and cost of care and patients ' access to it; and the heart, an organ that has special cultural symbolism and life-sustaining importance, is involved. FOCUS OF THIS STUDY A 1989 IOM planning committee formulated nine questions to be addressed in this evaluation although, as seen in the preceding chapters, the committee's deliberations went beyond answering the questions. The committee's responses to the questions are very briefly summarized here. What are the nature and magnitude of the target populations for which MCSSs may be applied?Once a fully implantable MCSS is established to be clinically useful for long-term cardiac support, many more patients will receive one than the current annual volume of 2,000 heart transplant recipients. The number of potential MCSS candidates in the primary group (those most urgently in need of cardiac support) is between 35,000 and 70,000 annually, but practical limits on the growth of this technology's use will hold patient volumes below this range, for perhaps 10 years. In the 2010-2020 period, if device and transplantation outcomes are then similar, potential use may grow substantially beyond this range. Most of the devices implanted during the first decade of MCSS use will be VADs. As many as 10,000 to 20,000 of each year's primary group of patients have impairment of both of the heart's ventricles, however, and will be candidates to receive a total artificial heart (TAH) after these devices become available, likely between 2005 and 2010. What are the alternative technologies for preventing and treating end-stage heart disease that may affect the need for MCSSs?For the foreseeable future, transplantation will remain the treatment of choice for end-stage heart disease. Current forms of treatment other than transplantation offer most patients only limited benefits, such as relieving symptoms but not significantly prolonging life. Early in the next century, advances in drug therapy may at least be able to delay the onset of end-stage disease for some patients, and perhaps prevent it. What is the potential for MCSSs to prevent and treat end-stage heart disease, and what are the current technological and other barriers to their development?Almost three decades of research and development have overcome many technological barriers to developing a fully implantable
OCR for page 175
The Artificial Heart: Prototypes, Policies, and Patients long-term device and have identified areas susceptible to solution by further technological development. Forthcoming clinical trials may clarify the relative importance of those areas needing additional research. Areas that have already been identified for further R& D include potential thromboembolic problems, through-the-skin power transmission, valves, and biomaterials. What is the clinical effectiveness of MCSSs?The experience to date with temporary-use devices provides reasonable assurance that one or more long-term MCSSs can be proven clinically effective for specific patient groups, probably before the turn of the century. As R&D efforts yield more technological advances, device performance and effectiveness levels (e.g., survival probability) are likely to improve steadily. What are the projected costs of research and development of MCSSs? NHLBI currently spends about $7 million each year through the contracts that provide most of the support for MCSS development and trials. Another $8 million is spent annually for investigator-initiated grants related to cardiovascular technology. The level of past funding of MCSS contracts has been constraining for MCSS developers; the annual total may need to be slightly greater if NHLBI accepts the committee 's recommendations to continue R&D with TAHs, as well as with more than one model of VAD, for an interim period. Not including the costs of the Novacor VAD trial, for which funds are already committed, approximately $2 million per year will be needed for additional VAD development for at least the next two fiscal years, and thereafter a possible additional amount to support further clinical trials. For TAH development, the total amount needed will depend upon findings during the recommended extensions of current contracts. The extensions themselves will likely cost a total of $3 to $6 million per year for a 2- or 3-year period, perhaps to be followed by a total cost of $30 to $90 million (over 10 years) for the two final stages of R&D, as discussed further in Chapter 6. (All of the foregoing amounts are expressed in 1991 dollars.) A number of NHLBI decisions in future years will have a substantial impact on these funding requirements. The devices themselves will, in the initial phase of use, cost about $50,000 for a VAD and $100,000 for a TAH, expressed in 1991 dollars. The hospital and physician care required to implant either type will cost about $100,000 in addition. Later costs will depend heavily on the extent to which care is required for complications or for repair or replacement of the device. What is the cost-effectiveness or cost-benefit of research and development of the various MCSSs, and how does this compare with what is known about the cost-effectiveness or cost-benefit of the research and development of alternative technologies for preventing and treating end-stage heart disease?The cost-effectiveness of long-term TAH use depends on the extent to which a device benefits patients, in particular the quality-adjusted life years (QALYs) gained in comparison with outcomes of conventional medi-
OCR for page 176
The Artificial Heart: Prototypes, Policies, and Patients cal treatment. The TAH's estimated incremental cost-effectiveness, $105,000 per added QALY (in 1991 dollars), is substantially less favorable than for heart transplantation and most other forms of heart disease treatment. If technological developments lead to improved TAHs with life-year gains that approach those of transplantation, cost-effectiveness will improve. What should the roles of government and industry be with respect to research and development of MCSSs?The NHLBI artificial heart program is unusual among health care R&D in the extent of its government support, in that the government has funded virtually all work to date, including the device development and testing costs usually borne by industry for similar technologies. As discussed later in this chapter and in Chapter 9, existing circumstances warrant continued NHLBI funding of both VAD and TAH research for at least a two- to three-year period. Further, NHLBI may wish to pursue formal cost-sharing arrangements with industry for some future efforts. Should decisions concerning further investment in the artificial heart program depend on whether the current cost-effectiveness or cost-benefit findings indicate the technology is acceptable or unacceptable, or are there additional factors that should be taken into account? Cost-effectiveness is one of several factors relevant to the committee 's recommendations to NHLBI concerning the artificial heart program. Other factors argue for NHLBI to continue its involvement, at least for an interim period. NHLBI's participation, for instance, may influence the manner in which clinical trials are conducted; also important is studying the quality of life of patients during clinical trials, which might well not occur without NHLBI support. Further, it is important that NHLBI participate in activities such as developing clinical practice guidelines for MCSSs and monitoring posttrial device performance. How can these findings be used to support decisions on allocation of research funds for artificial heart technologies?All of the topics considered by the committee are relevant to the decisions that NHLBI must make. Because of the risk of unforeseen technological problems, if clinical trials are conducted with only one VAD model, NHLBI funding of other scientifically meritorious VAD development efforts should continue for a period of at least two to three years. NHLBI support for the development of TAHs should also continue for the same interim period until the early results of long-term VAD trials are available, because of the potential future benefit to patients from TAHs, and because information from those trials will help in future decisions about TAH development. NHLBI decisions about allocating R&D funds should be aided by the use of cost-effectiveness analysis (CEA) and other explicit criteria, as described in Chapter 3 and Chapter 6. Other topics. The committee has also reviewed a number of other areas related to MCSS development and use. The recommendations in this chap
OCR for page 177
The Artificial Heart: Prototypes, Policies, and Patients ter cover such topics as promoting the appropriateness of MCSS use, deciding about patient access to this technology, and avoiding a possible adverse impact of industry support for academic R&D on open, collegial communication among researchers. PROMOTING APPROPRIATE USE OF MECHANICAL CIRCULATORY SUPPORT DEVICES The committee is very concerned about possible inappropriate MCSS use, similar to concerns expressed over the years about other new technologies. The proper response is not to terminate federal support for R&D in this area. Rather, the decade or more that will be needed to complete MCSS development offers an unusual opportunity for all interested parties to deal positively with this concern. Substantial variations in rates of use of many technologies, as well as in other physician practice patterns, indicate that either overuse or underuse occurs regularly; both possibilities raise questions about the effects of such patterns on the quality of patient care. Further, the overuse of technology continues to be blamed for much of the steady increase in the cost of health care. Many regulatory and educational measures that once seemed promising as means of achieving appropriate technology diffusion and use have proven insufficient for the task. With MCSSs, crucial activities can occur before these devices go into general use, to improve the likelihood that they will be applied appropriately. The 1990s provide an opportunity to improve the prospects for appropriate MCSS use, once their development and clinical trials are completed, and activities with that goal should be undertaken. The committee recommends continued federal support for MCSS development, based on an assumption that this will be done. Among the relevant activities that can begin immediately are the development of provisional clinical practice guidelines (indications for use)2 and of guidelines for the institutional resources and staff expertise necessary for appropriate MCSS use, as well as the initiation of selective coverage programs (e.g., third-party payers approving coverage for MCSS use only when the device is implanted at designated institutions). Participants in developing and implementing measures to enhance the appropriateness of MCSS use should include: the researchers developing 2 In this report, the term “clinical practice guidelines” should be construed as including the closely analogous medical review criteria that will be developed from such guidelines and used by third-party payers in acting on preauthorization and payment requests, as well as in posttreatment reviews of care such as those performed by peer review organizations.
OCR for page 178
The Artificial Heart: Prototypes, Policies, and Patients these devices; cardiothoracic surgeons, cardiologists, and other physicians; allied health professionals; representatives of the Medicare program and other third-party payers; and others such as health services researchers, economists, and ethicists. Advice and consultation should be sought from patients who have experienced MCSS use and from family members and others who have cared for them. Further, because of their in-depth knowledge of this technology, NHLBI personnel should be intimately involved in these activities, in particular to design evaluative studies of MCSS use. Efforts to enhance the appropriateness of MCSS use will fit well with current trends to develop clinical practice guidelines and to tighten the granting of practitioner privileges to use particular technologies. Additional studies of these devices' long-term outcomes will be an important basis for the efforts to enhance appropriateness of use that should continue indefinitely. Because of this technology 's cost and visibility and because clinical studies and technology assessments of MCSS use will be important in these and other efforts, the Public Health Service's Agency for Health Care Policy and Research (AHCPR) might give such studies a special priority. Considering the foregoing, the committee recommends the following: MCSS developers, physicians, and others should work together during the 1990s to establish clinical practice guidelines and other measures to promote the likelihood of appropriate MCSS use. Recognizing the cost of this technology, adequate federal and private research funds should be devoted to clinical studies and technology assessments of MCSS use. Staff of the NHLBI artificial heart program should participate in these appropriateness-enhancing activities and work with AHCPR to develop support for the needed studies. ACCESS BY ALL TO CIRCULATORY SUPPORT DEVICES? The health care community and state and federal governments may or may not take action in the years ahead to assure equitable access, either to health care generally or to specific forms of treatment, for the 30 million or more people in the United States who have no insurance or are inadequately insured. The committee has not considered itself charged to break new ground on the issue of access to health care; see discussions of access to care such as that of the President 's Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research (1983). The committee would, however, be reluctant to recommend continued federal support for MCSS development, if it believed these devices would
OCR for page 179
The Artificial Heart: Prototypes, Policies, and Patients be available only to those with personal resources or adequate insurance coverage. It is also beyond the scope of this committee's charge to decide, or even to recommend, whether an MCSS should be part of a basic level of health care that is applicable nationwide, or whether the federal government or states should provide universal access to MCSSs or to an overall basic level of care. Nonetheless, as with the appropriateness issue, the nation's health care policymakers could and should use the emergence of this new technology during the 1990s as an opportunity to face the access issue squarely and examine it in depth before MCSSs come into general use. The time is ripe for the United States to make clear decisions about access to health care, including costly new technologies. A conscious policy decision about MCSS access would also minimize future claims that products of government-sponsored research should automatically become a public entitlement. Access via Third-Party Payers As an alternative to universal access either to MCSSs for a clinically defined population or to a basic level of services, partial but more uniform access could be achieved via third-party payers. Many state legislatures have mandated specific benefits to be provided by all private health insurers in their states; such measures, although sometimes controversial, have proven to be more feasible at the state than the federal level. If no legislative mandate for MCSS coverage exists, third-party payers have a strong incentive to deny coverage under even the most generous benefit packages, because the high per-patient cost will result in a very high aggregate cost if all those who need MCSSs are able to obtain them. Medicare and state Medicaid programs are likely to be more reluctant, even, than private insurers to approve MCSS coverage. The nature of the health care gains provided by an MCSS presents a strong counter-argument. If these devices are proven clinically effective, it will be difficult for either private or public third-party payers to assert that they are properly a “second tier” technology that can be denied to identified patients. There remains, however, a possibility that long-term MCSSs will, in fact, become available only to those who are able to pay for them through special insurance coverage or with private funds. Clinically Limited Access At least in the early stages of MCSS diffusion, it will be imperative to limit access on clinical grounds, much as the number of heart transplantation candidates is limited by stringent approval criteria for placement on the transplant waiting list. If provisional indications for long-term MCSS use
OCR for page 180
The Artificial Heart: Prototypes, Policies, and Patients can appropriately be restricted on specific clinical grounds to a relatively small group of candidates, third-party payers may be more comfortable in approving such a limited degree of MCSS coverage and payment. This approach must recognize, however, that narrow indications for use are likely to be temporary, considering the evolution of indications for technologies such as percutaneous transluminal coronary angioplasty, as additional clinical experience was gained. Limited access to heart transplantation is warranted by the small number of available donors; with a manufactured device, no such curbs will occur on the supply side. Furthermore, those developing any restrictive indications for use and the clinicians who apply them to individual patients must take great care that they do not favor a particular demographic group (e.g., middle-class white males) disproportionately to disease incidence. Limiting Access by Inadequate Payments As discussed in Chapter 9, the Medicare program, in particular, sometimes approves coverage of a new technology but then establishes a payment rate for its use that is substantially below the cost of providing the service. Paying 50 or 75 percent of costs is an unacceptable way to ration access to technology. Medicare and other third-party payers should determine coverage policies on clinical and cost-effectiveness grounds. If coverage is approved for a particular clinically defined group of patients, a payment rate should be established that adequately compensates providers for the costs of the service, both direct and indirect. As discussed more fully in Chapter 8, the committee has four recommendations concerning access: Long-term MCSSs should become an example of a specific technology that, once it qualifies clinically, must be explicitly included or excluded from basic health care coverage as defined for particular insurance programs. Legislators and policymakers at state and federal levels should begin now to decide about equitable access to MCSSs and other technologies, establishing commissions or other broadly based groups to aid in making those decisions. Below-cost payment rates should not be used to restrict access. If access decisions are made on a state-by-state rather than national basis, an adequately funded organization or mechanism should be established to provide information and assistance to such state decision-making processes.
OCR for page 181
The Artificial Heart: Prototypes, Policies, and Patients CLINICAL AND COST-EFFECTIVENESS It is clear to the committee that the years of substantial federal commitments to artificial heart R&D will soon lead to clinically useful long-term devices that are adequate to prolong life and restore function for thousands of end-stage heart disease patients, as discussed more fully in Chapter 2. Clinical trials will first establish whether these devices are efficacious. Later clinical use and patient follow-up studies will be needed to identify more details of their effectiveness, as these terms are differentiated (see Chapter 3). Given the health care environment that now exists and will continue for the foreseeable future, scrutiny of the care that is delivered will become increasingly stringent. Attention to the assessment and assurance of quality in the provision of care is at an all-time high. Costs are receiving an equal degree of review but, in its consideration of the future role of MCSSs, the committee is particularly concerned that increases in aggregate health care costs resulting from MCSS use are matched by patient-care gains. Clinical Effectiveness It is premature to judge precisely when a clinically effective long-term VAD will be approved for general marketing, but such an approval appears likely to occur before the end of the century. A fully implantable TAH will likely follow about 10 years later, depending on future NHLBI decisions about funding of TAH development. The review mechanisms of NHLBI and the Food and Drug Administration (FDA), coupled with the efforts of those developing the devices and conducting their clinical trials, will provide reasonable assurance that devices coming into general use are efficacious, based on evidence from the clinical trials. Data will need to be obtained from a registry and from follow-up studies to establish their long-term clinical effectiveness. Continuing attention to assessing patient gains from implanting these devices will be important as their volume of use grows. Quality of Life No comprehensive assessment of a technology's clinical effectiveness can be complete without carefully examining its impact on patients' quality of life, because therapeutic interventions may have similar physiological effectiveness but still differ in terms of quality-of-life outcomes of patients. Quality-of-life measures can be especially important in assessing the outcomes of patients receiving therapies that carry high risks of negative consequences.
OCR for page 182
The Artificial Heart: Prototypes, Policies, and Patients NHLBI is, perhaps, preeminent among others of the National Institutes of Health (NIH) in focusing attention on this essential component of patient outcomes; the committee is gratified to note NHLBI's broad quality-of-life initiative, not only in the artificial heart program but in other aspects of its research. Assessment of patients' postimplantation quality of life is a component of the current NHLBI-sponsored VAD clinical trial. Irrespective of the degree and direction of change in any one or more quality-of-life domains (e.g., social functioning might improve and self-care might deteriorate for a patient, over time), three general health state outcomes are possible from a treatment. The determination of whether improvement, deterioration, or “no change ” has occurred in health status is subjective, depending on the weights and values that are assigned by the individual making the trade-offs among the various states described. The results or products of trade-offs can be converted to quantitative values, namely preference weights (utilities), for particular health states; in the committee's cost-effectiveness analysis (CEA), these weights form the basis for calculating quality-adjusted life years (QALYs). All NHLBI-sponsored research and clinical trials that include cost-benefit analyses or CEAs should include preference-weighted or utility measures. Further work is needed in quality-of-life instrument development and validation to determine which domains are important to particular patient populations and, within domains, which components are sensitive to burdens perceived by patients. In addition to the more traditional core set of domains used in assessing health states, the study committee identified three health-related quality-of-life domains of potential significance to patients considering or having received an MCSS— machine dependence and societal reaction, the meaning and purpose of life, and spiritual well-being. To the extent that policymakers judge it to be in the public interest to use public resources to develop MCSSs, those decisions should include commitments to support comprehensive assessments of health-related quality of life in patients receiving MCSSs during clinical trials and even thereafter. A core set of quality-of-life domains, similar to those used for this study's utility measures, should be assessed in all MCSS clinical trials. More research is needed to identify and understand support systems and selected other determinants of health-related quality of life that might be helpful in identifying those patients, among the groups meeting certain clinical conditions, who are more likely to benefit from an MCSS. Cost-Effectiveness The cost of health care and its clinical effectiveness, including quality-of-life gains, can be related to one another and assessed in various ways;
OCR for page 183
The Artificial Heart: Prototypes, Policies, and Patients the most highly quantified are CEA and cost-benefit analysis. As described in Chapter 6, the committee has used CEA methods to examine both TAH clinical effectiveness and the funding of TAH development. In clinical studies, CEA yields a cost-effectiveness (C/E) ratio that expresses the incremental cost of each QALY gained from using the technology studied, compared with one or more alternate forms of treatment. The committee's CEA examined the projected cost-effectiveness of TAH use and heart transplantation, comparing both with conventional medical treatment for end-stage heart disease as of 2010, the earliest that TAHs are likely to be in routine use. This required making a number of estimates of such aspects of TAH use as device failure rates and the probability of various TAH-related clinical complications, as well as lengths of hospital stays and outcomes of such events. The estimates developed for the CEA may well be more conservative than occurs when a current technology is examined, because of uncertainty about the outcome of use of these devices 20 years in the future. From the committee's analysis, the estimated relative cost-effectiveness of using a TAH instead of medical treatment is about $105,000 per added QALY (in 1991 dollars), a C/E ratio that is considerably less favorable than those for other generally acceptable forms of heart disease treatment, such as $32,000 per QALY for heart transplantation and $34,000 per QALY for coronary artery bypass surgery for two-vessel disease with severe angina. The C/E ratio for TAH use is also less favorable than C/E ratios for other costly interventions (e.g., renal dialysis). The committee's CEA examined only TAH use, the primary focus of the study. Because of the lower device cost and the fact that VAD failures may cause death less often than do TAH failures, the C/E ratio for long-term VADs may be somewhat more favorable than for TAHs. Further, results of the Novacor VAD clinical trial, other VAD use during the 1990s, and continuing technological gains will likely, over time, form a basis for improved C/E ratios for both types of MCSS. Cost-Effectiveness in Research and Development Decisions Cost-effectiveness analysis can also be applied to the consideration of research and development options. The committee used CEA to examine alternative levels of R&D investment during the next phase of TAH development. Based on assumptions of the impact of increased funding on the length of time required to complete R&D and on the future selling price of TAHs, this CEA reveals that increased R&D funding may be beneficial, assuming that the basic $105,000-per-QALY C/E ratio of TAH use is considered acceptable. If performed with care, CEA can make a useful contribution to decision making about R&D activities.
OCR for page 190
The Artificial Heart: Prototypes, Policies, and Patients are known. Based on the important assumption that the broader concerns previously discussed—appropriate use and access—are being addressed concurrently, the committee has a three-part recommendation: NHLBI should continue to support the four current TAH developers, to the extent warranted by the scientific merit of their work, for a two- to three-year period following the current contract expirations in September 1993 and at a level sufficient to ensure that research teams are not disbanded. TAH R&D should be reconsidered in 1994 or 1995, taking into account the VAD experience and information about other potential long-term TAH benefits generated in the interim. NHLBI should decide at that time whether to move ahead with funding for the next phase of TAH development, after further peer review and a decision-making process that takes into account a then-current cost-effectiveness projection and other relevant criteria. By proceeding in this manner, it may be possible to devote some funds to other uses in the intervening years, additional potential long-term savings and gains can be explored in detail, and the TAH designs that ultimately result can reflect lessons learned from early use of long-term VADs. The committee's cost-effectiveness example, analyzing the impact of increased levels of R&D funding, will be useful to NHLBI in its 1994-1995 TAH decision, as will be the explicit decision-making criteria suggested in Chapter 3. OTHER RECOMMENDATIONS The preceding recommendations and discussions fulfill the committee 's principal obligation to NHLBI, responding to the nine questions set forth in the 1989 report of the IOM planning committee and making recommendations concerning future TAH development. The committee believes that several additional conclusions and recommendations about aspects of MCSS development are also appropriate. Future Development of Ventricular Assist Devices The need for a fully implantable, long-term VAD is substantial, and at least one long-term VAD appears reasonably certain to be approved for routine clinical use during the 1990s. Although it will be perhaps 2010 before the technology's use will diffuse sufficiently to meet the need, an annual primary pool of between 35,000 and 70,000 candidates to receive either a VAD or TAH now exists. If continuing development efforts result
OCR for page 191
The Artificial Heart: Prototypes, Policies, and Patients in a device whose performance approaches transplantation outcomes (now about a 70 percent 5-year survival rate), the size of the candidate pool will increase by perhaps another 200,000 per year, as of about 2020. Taking full advantage of the potential of VADs is thus a vital goal. In this regard, the committee is concerned about the possibility that development of long-term VADs other than Novacor's may not continue. If NHLBI ends its contractual funding of other VAD developers, the economic uncertainties might preclude private funding of their R &D work and they may suspend it. Therefore, a risk exists in relying on a single approach to VAD development. Further, lower device prices may also result, if more than one VAD is available. (This discussion specifically does not refer to VAD development under investigator-initiated [R01] grants; consideration and funding of such grants should continue via the usual peer-review process, under any circumstances.) All current models of VADs were essentially designed in the late 1970s; they do not reflect either more recent R&D results or possible new approaches to VAD design that have not yet reached the prototype stage. Knowledge and a range of approaches gained during the 1980s, but not yet applied, may prevent unnecessary delays in developing a clinically useful VAD, particularly if the Novacor VAD encounters unanticipated problems during the forthcoming trial. An end to federal R&D support may also adversely affect the continuity of research by causing the loss of key research personnel to other fields. Yet another reason for continuing NHLBI involvement in additional VAD research is that it enables the agency to ensure that clinical trials are properly designed and conducted. FDA review of a developer 's investigational device exemption application is theoretically sufficient to ensure that all aspects of a trial are properly designed and conducted, including periodic postimplantation assessments of patients' physiologic status and health-related quality of life. In practice, however, the committee recalls the criticisms of past artificial heart trials at Salt Lake City, Louisville, Phoenix, and other locations as having important shortcomings; by continuing its R&D involvement NHLBI can exert strong influence over all aspects of VAD trials. Finally, because of the relatively large potential volume of long-term VAD use, private-sector interest in this technology will likely increase after one VAD has gone into general use and has received third-party coverage approvals. Commitment of substantial private resources to long-term VADs at their current stage of R&D appears unlikely, however, because of the length of time consumed in the development and regulatory processes and the uncertainty about when regulatory approvals will be granted and general use begin. The committee therefore believes that ending contractual support for all
OCR for page 192
The Artificial Heart: Prototypes, Policies, and Patients but one VAD developer carries substantial risks that NHLBI may wish to avoid and concludes that continued support of other scientifically meritorious VAD development is desirable, for at least a two- to three-year period. This could be done either by implementing a new peer-reviewed initiative such as a request for proposals to design a state-of-the-art implantable VAD or by continuing to support, at least temporarily, one or more VAD developers whose approaches differ from the current Novacor model. After preliminary postdischarge follow-up results for the early patients in the forthcoming VAD trial are available, perhaps in 1993, NHLBI will be able to decide with greater certainty about the need for continuing support of VAD development. The committee therefore recommends that, in addition to the Novacor clinical trial, NHLBI consider the continuation of targeted R&D support for other VAD development for at least a period ending in late 1993 or 1994, with further support to be considered at that time. Clinical Trials and Patient Follow-up The committee has several concerns in the area of MCSS clinical trials and ongoing patient follow-up. Clinical Trials In order to determine the continuing role of MCSSs, three types of data should be routinely collected in clinical trials: (1) physiologic data (e.g., cardiovascular, renal, neurologic, hematologic), obtained periodically throughout the trial period; (2) health-related quality-of-life data, using core sets of test questions that tap both generic and condition- or treatment-specific domains; and (3) data about the clinical outcome and cost of treatment so that reports of the trial can include cost-effectiveness analysis. Funding of all trials, including that of the Novacor VAD, should be adequate to ensure that all aspects of device use, and their implications, can be addressed by those conducting the trial. Trial design and implementation should also ensure that study populations include adequate representation of women and minority populations. It is important that MCSS trials, whether supported by public or private funds, make provision for each patient to provide a fully informed consent, an advance directive concerning termination of treatment, and legal designation of a proxy; see the review of these ethical safeguards in Chapter 8. The committee thus recommends that all NHLBI-supported and private MCSS clinical trials include adequate funding for the
OCR for page 193
The Artificial Heart: Prototypes, Policies, and Patients collection of physiologic, health-related quality-of-life, and cost-effectiveness data and, as well, include provisions protecting patients' ethical rights, such as a requirement for written advance directives and, if allowed by state law, designation of a proxy decision maker. Trial protocols should ensure adequate representation of women and minorities and should provide for comparability with other studies by using, for instance, a standard core set of quality-of-life domains and assessment methods. Postmarketing Surveillance All patients receiving a long-term MCSS should be followed for the remainder of their lives, at least to the extent of obtaining rudimentary survival information on a periodic basis. An MCSS implantation is not a simple, time-limited treatment episode. This type of lifetime follow-up can best be accomplished through a registry that receives information from clinicians who implant devices and patients who receive them. If public and quasi-public funds, the latter provided via health insurers, are to support the development and use of this costly technology, special measures are warranted to oversee its long-term clinical effectiveness. The committee recommends that long-term MCSS patients and their physicians provide lifelong follow-up information to a registry, and that hospitals, stimulated by requirements of third-party payers and their payments for this component of care, fund such a registry as an integral part of providing the care. Posttrial Follow-up The two-year follow-up period that is typical of clinical trials of implanted devices is not long enough to reveal all problems that may occur with MCSS use. In addition to any registry that is established, an adequate sample of patients receiving MCSSs during clinical trials should be followed periodically after the trial period ends, to ascertain and study details of posttrial patient management and outcomes. This could be accomplished by various means. For subjects in clinical trials sponsored by NHLBI, that agency should fund follow-up studies of either all patients or a sample. If such a study is to be undertaken by a device developer, such as to fulfill FDA postmarketing surveillance requirements, provisions should be included to ensure that results of the study will be published promptly, regardless of outcome. For patients receiving an MCSS after it is FDA-approved, AHCPR should fund one or more long-term follow-up studies of an adequate sample of
OCR for page 194
The Artificial Heart: Prototypes, Policies, and Patients patients, to be performed by academic or independent researchers. In all instances, funding should be sufficient for the study to gather and report data about patients' physiologic status, health-related quality of life, and treatment costs. The committee therefore recommends that NHLBI and AHCPR support long-term follow-up studies of an adequate sample of all patients receiving MCSSs and fund such studies adequately. Fostering Collaborative Interdisciplinary Research As previously mentioned, the artificial heart program, with its targeted, largely developmental R&D effort, is unusual not only within NHLBI but also in NIH as a whole. The policies and mechanisms by which the artificial heart program operates were originally developed to review, approve, and manage traditional investigator-initiated, nontargeted research by basic scientists, not the development of complex devices and support of their extensive technical evaluation, redesign, preclinical testing, and clinical trials. Additionally, although traditional scientific peer review is as important to MCSS research as it is to any other field, considering artificial heart program proposals in the existing peer-review committees puts this truly interdisciplinary research involving biomedical engineers, clinicians, and others at a serious competitive disadvantage, as discussed in Chapter 9. Adequate peer review of MCSS R&D requires the involvement of biomedical engineers and professionals with other relevant expertise (e.g., quality-of-life assessment, cost-effectiveness analysis). The committee recommends that all proposed grants and contracts for MCSS development be peer-reviewed by a group or groups including appropriate numbers of biomedical engineers, clinicians, and professionals from other relevant disciplines. The organizational structure within which the artificial heart program operates, designed as it is to administer grants to individual investigators, is not as flexible as would be ideal to make collaborative grants that involve both academe and industry. The committee believes that collaborative multidisciplinary research, structured to encourage industry to share in the R&D costs, is an important aspect of government support of private-sector research, consistent with the examples of R&D support by other government agencies (e.g., Department of Defense, National Science Foundation). The committee recommends that NHLBI amplify its attention to alternative policies and structures for programs such as the
OCR for page 195
The Artificial Heart: Prototypes, Policies, and Patients artificial heart program that allow greater flexibility in operating modes and that might serve as models for future efforts to encourage collaborative R&D on a cost-sharing basis, considering the successful experience of other agencies. Patients' Quality of Life and Treatment Preferences In many sectors of health care, patients' health-related quality of life has not received adequate attention. For a device such as an MCSS, which patients and their families must learn to live with for an indefinite period, as discussed in Chapter 5, the committee recommends that physicians and hospitals involved in implanting long-term MCSSs develop support programs to enhance the quality of life of MCSS patients and their families during the period of hospitalization and after discharge. Attention to explicit patient preferences is a growing aspect of medical care. Fully informed patient preferences, or those of a properly designated surrogate decision maker, should be accorded great weight in treatment decisions about implanting an MCSS in the first place as well as withdrawing support at a later time. Thus, as recommended above, protocols for MCSS clinical trials should provide for advance directives and the naming of surrogate decision makers by prospective patients. Research Needs Heart Failure Research Whatever NHLBI's decisions about continuing to support MCSS development, that support should not be at the expense of basic and clinical research concerning medical treatment of end-stage heart disease. The borderline cost-effectiveness of TAH use, as well as the costliness and other logistical problems of transplantation, support giving a high priority to research that may ultimately prevent heart failure or delay its onset. The committee therefore concludes that NHLBI should not use the development of MCSSs as a reason to reduce its level of funding of basic heart disease research. Instead, the committee recommends that NHLBI continue to support, at an adequate level, scientifically meritorious research into the mechanisms of heart failure and approaches to the prevention and medical treatment of end-stage heart disease.
OCR for page 196
The Artificial Heart: Prototypes, Policies, and Patients Epidemiological Research The committee's projections of MCSS use required reliance on numerous sources of epidemiological data about end-stage heart disease, none of which was completely satisfactory. Little is known from an epidemiological perspective about the course of this disease, as discussed in Chapter 4. Better basic data will make future projections of MCSS use more accurate. The committee therefore applauds NHLBI for supporting the longitudinal studies recently begun; further, the committee recommends that, in addition to current longitudinal studies of heart disease in persons over age 65, NHLBI undertake as a high priority, and possibly with the involvement of the National Center for Health Statistics, the development of additional epidemiological data about the natural history of end-stage heart disease in patients of all ages, with special attention to the inclusion of women and minority populations. Communication Among MCSS Researchers Medical device manufacturers and other private sources of financing are increasingly becoming involved in the support of academic MCSS research and development. The committee encourages this, but it is also concerned that such involvement may have an adverse effect on the open, collegial communication among researchers that has been a valuable aspect of MCSS development to date. The committee recommends that, when undertaking relationships with the private sector, academic researchers work cooperatively to minimize restrictions on their freedom of communication with other researchers, and that their universities develop policies consistent with this goal. It further recommends that this topic be explicitly and thoroughly discussed at a meeting involving all researchers and developers in the field. HOW THIS REPORT SHOULD BE USED This study, following six panels that conducted similar evaluations over the past 22 years, has posed a special challenge for the committee: namely, to review all aspects of a relatively small, but nonetheless important and highly visible, segment of federal support of biomedical research and to make projections and recommendations about the future of this unique technology. Faced with the absence of key data in numerous areas, the committee has made estimates that it considers reasonable based on the current evidence.
OCR for page 197
The Artificial Heart: Prototypes, Policies, and Patients This study has been able to add to the knowledge base of previous NHLBI studies of the artificial heart program because the technologies involved are so much nearer to clinical use than they were in earlier years. It has also reviewed some aspects of the NHLBI research funding process, a topic not discussed in previous studies. Most important, any one of the comprehensive studies of the artificial heart that have been performed could serve as a model of the type of study that should become an integral part of the U.S. health care system. Technologies have gone into widespread use with little understanding of their clinical, quality-of-life, ethical, and economic prospects. By and large, patients have been fortunate; most new technologies prove to have clinical value. Yet technology continues to be blamed for much of the steady rise in health care costs. Comprehensive studies—whether to assess new technologies, guide their diffusion, or measure their impact on costs —have been few in number. It is neither simple nor inexpensive to study complex new technologies, let alone to develop measures that assess and improve the appropriateness of their use. The committee expresses the hope that leaders in key components of the U.S. health care system, as well as policymakers in the administration, Congress, and state capitals, will consider the recommendations of this study, particularly those that are broadly applicable. The research, development, and diffusion of future technologies require attention to methods and oversight mechanisms that reflect the full range of considerations discussed here. These conclusions and recommendations can, if applied both to MCSSs and to other new technologies, improve the ways in which the nation's health care providers and regulators deal with them, as well as the ways in which those who support health care R&D manage the innovative research that yields so many health care gains. REFERENCE President's Commission for the Study of Ethical Problems in Medicine and Bio-medical and Behavioral Research. 1983. Securing Access to Care. Vol. 1. Report on the Ethical Implications of Differences in the Availability of Health Services. Washington, D.C.: U.S. Government Printing Office.
OCR for page 198
The Artificial Heart: Prototypes, Policies, and Patients This page in the original is blank.
OCR for page 199
The Artificial Heart: Prototypes, Policies, and Patients Appendixes
OCR for page 200
The Artificial Heart: Prototypes, Policies, and Patients This page in the original is blank.
Representative terms from entire chapter: