Appendix A


Learning What Works Best: The Nation’s
Need for Evidence on Comparative
Effectiveness in Health Care




AN ISSUE OVERVIEW

image

IOM ROUNDTABLE ON EVIDENCE-BASED MEDICINE






September 2007 version. This Issue Overview was prepared at the request of the IOM Roundtable Working Group on Sustainable Capacity by J. Michael McGinnis, LeighAnne Olsen, Katharine Bothner, Daniel O’Neill, and Dara Aisner.



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



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

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

OCR for page 333
Appendix A Learning What Works Best: The Nation’s Need for Evidence on Comparative Effectiveness in Health Care AN ISSUE OVERVIEW IOM ROUNDTABLE ON EVIDENCE-BASED MEDICINE September 2007 version. This Issue Overview was prepared at the request of the IOM Roundtable Working Group on Sustainable Capacity by J. Michael McGinnis, LeighAnne Olsen, Katharine Bothner, Daniel O’Neill, and Dara Aisner. 

OCR for page 333
 LEARNING WHAT WORKS MARCH 2009 UPDATE The American Recovery and Reinvestment Act of 2009 In the time since the preparation of this white paper, $1.1 billion of fed- eral funds have been provided by Congress, through the American Recovery and Reinvestment Act of 2009 (ARRA), to increase national capacity for clinical effectiveness research. AHRQ (Agency for Healthcare Research and Quality) has received $700 million of these funds, of which $400 million will be transferred to the Office of the Director of NIH (National Institutes of Health) to conduct or support comparative effectiveness research (CER) activities. An additional $400 million will be allocated at the discretion of the Secretary of HHS (Department of Health and Human Services) to: “…accelerate the development and dissemination of research assessing the comparative effectiveness of health care treatments and strategies, through efforts that: (1) conduct, support, or synthesize research that compares the clinical outcomes, effectiveness, and appropri- ateness of items, services, and procedures that are used to prevent, diagnose, or treat diseases, disorders, and other health conditions; and (2) encourage the development and use of clinical registries, clinical data networks, and other forms of electronic health data that can be used to generate or obtain outcomes data.” The recommendations from an Institute of Medicine consensus com- mittee report and from a newly established Federal Coordinating Council on CER within HHS will be considered by the secretary’s office in desig- nating activities to receive funds. Members of the 15-member council will be federal employees or officers appointed by the President, at least half of which will have clinical expertise.

OCR for page 333
 APPENDIX A LEARNING WHAT WORKS BEST THE NATION’S NEED FOR EVIDENCE ON COMPARATIVE EFFECTIVENESS IN HEALTH CARE Contents Introduction Implications for Stakeholders Current Activities in Clinical Effectiveness Research Activities and Needs Related to Comparative Effectiveness Research Models for a Stronger Approach to Comparative Effectiveness Research Decision and Implementation Considerations Concluding Observations APPENDICES 1. Current National Capacity for Clinical Effectiveness Research 2. International Activities in Clinical Effectiveness Research 3. Potential Model: Federally Funded Research and Development Centers 4. Potential Model: NIH Public-Private Partnership Program

OCR for page 333
 LEARNING WHAT WORKS 5. Potential Model: National Academies’ Transportation Research Board 6. Potential Model: Federal Reserve 7. The Business Case for Comparative Effectiveness Research: A Commissioned Analysis

OCR for page 333
 APPENDIX A Acknowledgments This Issue Overview on current and needed capacity for comparative effectiveness research was developed by staff of the IOM Roundtable on Evidence-Based Medicine, initially as background material for the activities of the Roundtable’s Sustainable Capacity Working Group, and to inform discussion at the March 19, 2007 Roundtable meeting. Guidance on structure and content were provided by members of an advisory group including: Carmella Bocchino (AHIP), Queta Bond (Burroughs Wellcome Fund), Kathy Buto (Johnson & Johnson), Steve Galson (FDA), Mark McClellan (AEI-Brookings Joint Center for Regula- tory Studies), Lisa Payne-Simon (Blue Shield of California Foundation), Diana Petitti (University of Southern California), Jean Slutsky (AHRQ), Sean Tunis (Center for Medical Technology Policy), and Gail Wilensky (Project Hope). We also extend special thanks to those who took the time to review and comment on various sections or draft versions of this paper, including: Wade Aubry (HealthTech), Tanisha Carino (Avalere Health), Nancy Featherstone (AstraZeneca), Mark Gibson (OHSU), Carmen Hooker Odom (North Car- olina HHS), Michael Johns (Emory), Peter Juhn (Johnson & Johnson), Doug Owens (Stanford), Steve Pearson (AHIP), Steve Phurrough (CMS), Eugene Rich (House Committee on Ways and Means, Health Subcommit- tee), Wayne Rosenkrans, Jr. (AstraZeneca), and Jeff Shuren (FDA). IOM staff who contributed include: Dara Aisner, Katharine Bothner, Michael McGinnis, LeighAnne Olsen, and Daniel O’Neill. Roundtable sponsors: The work of the Roundtable is supported by the

OCR for page 333
 LEARNING WHAT WORKS Agency for Healthcare Research and Quality, AHIP (America’s Health Insurance Plans), AstraZeneca, Blue Shield of California Foundation, Bur- roughs Wellcome Fund, California Healthcare Foundation, Centers for Medicare and Medicaid Services, Department of Veterans Affairs, Food and Drug Administration, the HWG Fund, Johnson & Johnson, sanofi-aventis, and Stryker. Suggested citation Institute of Medicine. 2007. Learning what works best: The nation’s need for evidence on comparative effectiveness in health care. http://www.iom.edu/ebm-effectiveness.

OCR for page 333
9 APPENDIX A IOM ROUNDTABLE ON EVIDENCE-BASED MEDICINE Working Group on Sustainable Capacity∗ LEARNING WHAT WORKS BEST The Nation’s Need for Evidence on Comparative Effectiveness in Health Care INTRODUCTION A core objective for the nation is achieving the best health outcome for every patient. This objective simply cannot be accomplished until we have better evidence on which to base healthcare decisions, as well as more effective application of the knowledge we have. Each is vitally important. We know, for example, that failure to deliver proven interventions is a substantial challenge to the quality of health care for Americans—and is a key concern of the IOM Roundtable on Evidence-Based Medicine. Yet, with the current pace of change, the most rapidly growing problem is our inability to produce the needed evidence in a timely fashion. This paper provides background for discussion about the evidence gap—the fact that the nation’s capacity has fallen far short of the need to produce reliable and practical information about the care that works best. Medical care decision- making is now strained, at both the level of the individual patient and the level of the population as a whole, by the growing number of diagnostic and therapeutic options for which evidence is insufficient to make a clear ∗ Jack Rowe, Columbia University (Chair); Adam Bosworth, Google; Helen Darling, Na- tional Business Group on Health; Michael Johns, Emory University; Steve MacMillan, Stryker Corporation; Mark McClellan, AEI-Brookings; Richard Platt, Harvard University; Steve Ud- varhelyi, Independence Blue Cross; Bill Weldon, Johnson & Johnson; Janet Woodcock, FDA. The material here is a staff paper prepared at the request of the working group. Information on the IOM Roundtable on Evidence-Based Medicine may be obtained at www.iom.edu/ebm.

OCR for page 333
0 LEARNING WHAT WORKS choice. Biomedical insights and medical innovation continue to advance opportunities to increase the health and life-span of the American public, yet to capitalize fully on this potential requires enhanced capacity to ensure that decisions, in the face of increasing complexity, can be supported and guided by the best available scientific information. Health care in the United States underperforms on many dimensions. At the macro level, with per capita expenditures more than 20 percent higher than any other country in the world and more than twice the average expenditure for European countries[1], the nation ranks well below others on key health indices—28th in overall life expectancy at birth and 23rd in infant survival [2, 3]. In part this is because people often do not receive the care they need. One study found that, where evidence exists, only about 55 percent of recommended services were actually delivered [4]. In part it is also because the services people receive are not always necessary or the right ones for them. The intensity of services for similar conditions with similar results—in particular, for procedures such as lumbar surgery, hysterectomy, and bypass surgery, where discretion plays a stronger role—can vary by as much as a factor of 20 depending simply on where one lives. In Idaho Falls, Idaho, 4.6 lumbar fusions were reported per 1,000 Medicare enrollees annually, as compared with 0.2 in Bangor, Maine, with no difference in the outcomes [5]. Similarly, wide geographic variations have been reported for conditions such as hip fracture, colorectal cancer, and acute myocardial infarction as well as end-of-life care [6], with a nearly fourfold difference in cardiac bypass surgery rates, a phenomenon primarily related to the region’s number of cardiac catheterization labs per capita rather than ill- ness rates [7]. One estimate suggested that only 27 percent of the weighted discrepancy in Medicare spending across regions could be explained by population illness levels [7], and if all regional spending levels matched those of the lowest decile, Medicare could see savings of up to $40 billion (1996 dollars) without compromising health status [8]. Clearly, more does not by itself equate to better—and the variation is greater for conditions in which the evidence is more limited. Ultimately, the central challenge is not primarily a matter of overuse or underuse of services but instead is related to the lack of available evidence to achieve the right care for any given patient. Information on which to compare the results from drugs with the same purpose is often not avail- able. For example, both Lucentis and Avastin are promising new drugs for treatment of macular degeneration, but head-to-head information on the relative outcomes is not available—and one costs about 20 times as much as the other. Similarly, different approaches to radiation therapy—intensity- modulated radiotherapy and conformal radiotherapy—have very different costs but currently inadequate information on which to base clinical judg- ments. And the pace of introduction of new genetic prognostic tests is on

OCR for page 333
 APPENDIX A an exponential course without the necessary evidence about the results of clinical decisions and outcomes. Estimates range widely concerning the proportion of medical care in the United States that is based on, or supported by, adequate evidence [9-14]. However, given concerns about the extent to which this information may be generalized and the quality of the evidence that is used, some place this figure at well below half. Regardless of the precise level, there is no question about the need for improvement. Part of the challenge is the appropriate delivery of what has already been proven effective. Medical care is becom- ing more complex with the increase in multifaceted chronic diseases, the development of new interventions, and the pressures to reduce the time of patient-provider interaction in the face of greater administrative burdens. New care management approaches, decision support systems, and incen- tives will be required to help providers and patients work together to ensure that the care delivered is the care that is known to be most effective. The most rapidly growing problem may soon relate not so much to shortfalls in applying what is known—a clearly significant problem—as to the inability for evidence of comparative clinical advantage to keep pace with innovation. It is both a capacity investment and a resource allocation problem. Of the nation’s more than $2 trillion annual health expenditure—nearly half of it borne by government—currently less than 0.1 percent [15, 16] is invested in assessing the comparative effectiveness of available interventions. Although about 5 percent of overall health expen- ditures is devoted to research, most is devoted either to basic research or to product development [17], as opposed to assessing how well medical treatment options perform. If trend data were kept, it would likely reveal that the proportion of expenditures devoted to this assessment “budget” was actually shrinking every year, yet the complexity of clinical decisions continues to compound. A testament to the power of innovation is the fact that new pharmaceu- ticals, medical devices, biologics, and procedures are introduced constantly, and the pace is quickening. From 1991 to 2003 the number of medical device patents per year doubled, from 4,500 to nearly 9,100 [18]. From 1992 to 2001 the total biotechnology patents granted per year tripled, from more than 2,500 to nearly 7,800, and the number of biopharmaceutical patents granted in the United States increased nearly four-fold, from about 1400 to nearly 5,200 [19, 20]. In the same period, annual sales of biolog- ics and pharmaceuticals more than doubled [21]. Between 1993 and 2004 there was a more than 80 percent increase in the number of prescriptions received by Americans [22]. Data for the growth in procedures are more difficult to obtain; however, as one example, between 1989 and 1995 spe- cialized procedures in major teaching hospitals nearly tripled [23]. A recent review by the Kaiser Family Foundation suggests that half or more of the

OCR for page 333
2 LEARNING WHAT WORKS growth in medical spending in recent years is attributable to technological change [24]. Much, but certainly not all, of this change has resulted in better care. Diagnostic imaging services, for example, grew more rapidly than any other type of physician service under Medicare. Between 2000 and 2005 spending on radionuclide imaging (RNI) doubled from $6.6 billion to $13.7 billion [25]. Yet an American College of Cardiology Foundation technical panel convened in 2005 to assess the appropriateness of cardiovascular RNI imaging for 52 indications [26] found that the lack of clear evidence on the best and most effective uses yielded strong disagreement on the appropriate circumstances. In addition to the growth in use of drugs, devices, biologics, and proce- dures, the world of health care is about to experience dramatic new insights concerning the variation in individual response to different diagnostic and treatment interventions. The 3 billion base pairs of the human genome have now been sequenced, revealing the 99.9 percent of the genetic code that is common to all humans. Additional differences, such as the gain or loss of regions of the genetic code, increase the variation between two random individuals by five- to ten-fold. Cataloging and characterizing these differ- ences by haplotype mapping and other initiatives is currently in progress and will begin to reveal how people vary in their susceptibilities to diseases and their responses to diagnosis and treatment. The age of personalized medicine will soon be a reality, if the capacity can be developed to contend with its insights. Today the average clinical encounter already requires a health provider to manage more variables than would be considered reasonable given what is known about the capabili- ties of the human mind, and over the next decade that same encounter will require contending with perhaps an order of magnitude more [27]. The traditions of developing evidence through one-at-a-time studies and relying for quality assurance on the recall capacity of an individual provider are no longer practical. Over the long term, substantial changes will emerge in the way the nation goes about generating and applying evidence for clinical decision making. A learning healthcare system is one in which the clinical research paradigm depends more judiciously on the serial conduct of randomized controlled trials—important, but often too expensive, untimely, and of limited applicability—and draws more heavily on electronic health records (EHRs) to generate evidence as a natural by-product of the clinical experi- ence. But while these longer-term capacities emerge, substantial near-term improvement will be necessary in our capacity to assess the relative effec- tiveness of different interventions—to understand what works best for whom under what circumstances. We need better understanding, agree- ment, and focus on the value we get from our health care—including what

OCR for page 333
 APPENDIX A constitutes value and how to measure it. Without this capability, it is likely that the inefficiencies that currently characterize the U.S. healthcare system will be compounded, perhaps considerably. Conversely, a more systematic and sustained effort to develop evidence on comparative and “real-world” effectiveness should stimulate more investment in research on innovation that will deliver better outcomes and greater value. Engaging the immediate need for a much stronger and sustained capac- ity to meet the need for guidance on the clinical effectiveness of medical interventions is the subject of this paper. Discussion follows on the perspec- tives of the various stakeholders, the current capacity and activities on clini- cal effectiveness research, the key functional needs to be met, and, finally, some possible approaches to addressing the issues, including consideration of decision principles, governance, funding, and public support. IMPLICATIONS FOR STAKEHOLDERS Better evidence is essential to securing trust in our healthcare system. In the face of uncertainty borne of insufficient evidence, patients, provid- ers, insurers, and health product companies frequently find themselves at odds and distrustful of each others’ motives. Concern about the shortfall in the national capacity to determine what medical care is actually best for whom is shared among many stakeholders. Most important, of course, are the patients who receive medical care and the health providers who deliver it, but large stakes are also held by healthcare delivery organizations, insur- ers, manufacturers, and others engaged in various aspects of health care, with the shared goals of improving patient health and delivering the best value. Increasing the level of investment in clinical effectiveness research, and doing so in a comparative fashion, is key to facilitating significant gains toward these shared goals. Roundtable teams are currently reviewing the perspectives and action prospects as the various sectoral stakeholders work to improve the prospects for the development of a learning healthcare system. Following are some of the more important implications of accruing substantially better information on clinical effectiveness. Consumers-Patients Each patient should be able to feel confident that there is solid evidence that the care received is the appropriate care for his or her circumstances. Yet, increasingly, this notion is strained. The American public has tradition- ally expressed strong support for medical care, research, and technology development, while also expressing a strong interest in both individual patient prerogative and better information to aid decision making. But with the increasing complexity of care and an increasing awareness of its

OCR for page 333
2 LEARNING WHAT WORKS In another study, mortality rates and quality-of-life measures were compared for patients undergoing coronary angiography in Texas, where the utilization of the procedure is high (45 percent), and for similar patients in New York, where utilization is low (30 percent). After adjusting for case mix differences, the researchers found no health advantages associated with greater utilization, suggesting that savings associated with reduced utilization of the procedure in Texas could be achieved with no deleterious clinical consequences [7]. One estimate suggests that, in aggregate, only 27 percent of the weighted variation in Medicare spending across regions can be explained by popula- tion illness levels [6]. If spending levels in all regions were made to match those in the lowest decile (age-, sex-, and race-adjusted), then Medicare could see savings of up to $40 billion in 1996 dollars [8]. ii. Inappropriate use The second body of research that addresses waste in the system attempts to directly measure how frequently certain medi- cal services are delivered for medically inappropriate indications. Results from this literature often demonstrate high levels of inappropriate use. For example, a 1993 study of members of seven managed care organizations found that about 16 percent of hysterectomies performed were judged to have been clinically inappropriate, and 25 percent of the patients under- went hysterectomy for uncertain indications [9]. A more recent study (in 2000) on hysterectomies found more dramatic results. Among hysterec- tomies performed in a capitated medical group in Southern California, 70 percent of cases were judged to have been inappropriate, according to RAND appropriateness criteria. Of the 497 women studied, 71 had hys- terectomies for conditions covered by three recent ACOG criteria sets. The recommendation for hysterectomy was judged inappropriate for 53 percent of that subset by the RAND criteria and for 76 percent according to the ACOG criteria [10]. In other cases, the rates of inappropriate use are relatively low, but there is a wide range of situations in which appropriateness is uncertain, which demonstrates the need for a stronger evidence base. For example, in one study, 4 percent of coronary angiographies performed at 15 hospitals in New York State were rated inappropriate; another 20 percent were rated uncertain. The rate of inappropriate use varied from 0 percent to 9 percent among hospitals, but the difference was not significant [11]. In another study, 4 percent of percutaneous transluminal coronary angioplasty (PTCA) performed at 15 hospitals in New York State were rated inappropriate; another 38 percent were rated uncertain. The inappropriate rate varied from 1 percent to 9 percent by hospital, the uncertain rate from 26 percent to 50 percent [12]. Trends toward inappropriate and uncertain use appear in other clini-

OCR for page 333
29 APPENDIX A cal areas as well. Reviewing cases of new-onset chest pain not due to myocardial infarction at one of five Los Angeles-area hospital emergency departments revealed that 7 percent of those who received some form of diagnostic cardiac testing had tests that were judged to be inappropriate. A literature review on cases of metastatic renal cell cancer (MRCC) rated 46.9 percent of treatments as inappropriate and 25.8 percent as uncertain [13]. A review of Medicare patients in three geographic areas revealed that 32 per- cent of the sample had carotid endarterectomy for inappropriate reasons, and 32 percent for uncertain reasons [14]. Seventeen percent of diagnostic upper gastrointestinal endoscopy procedures for Medicare patients were performed for inappropriate indications, and 11 percent were performed for uncertain indications [15]. In cases of hospital use, 23 percent of admis- sions were judged to be inappropriate and an additional 17 percent could have been avoided by the use of ambulatory surgery [16]. These studies often examine a specific intervention (e.g., upper gastro- intestinal endoscopy or percutaneous coronary angioplasty) and evaluate the usefulness in a number of clinical indications. Most of the appropriate- ness research focuses on high unit cost services. However, significant expen- ditures associated with overuse may accrue from inappropriate utilization of low unit cost services if they are used in sufficient volume (e.g., routine blood testing, imaging procedures). Moreover, many of the studies cited above are based on data from the 1980s. The more recent small area varia- tions literature suggests that substantial inappropriateness likely still exists, but much more work is needed in the area if we are to better understand, and address, the inefficiencies in the system. These findings of substantial variation in practice patterns and often large rates of inappropriate use highlight the fact that the merit of a specific medical intervention depends on the precise reason for use. Thus, in most situations, detailed patient-specific information is required before reporting whether the use of a drug, test, or device is worthwhile. It is important to recognize that one cannot say that a particular medical service is always appropriate or always inappropriate. Consider an example in the area of diagnostic imaging: radionuclide cardiovascular imaging (RNI). This is but one type of diagnostic imaging, but understand- ing the appropriateness of imaging as a whole is crucially important. Diag- nostic imaging services reimbursed under Medicare’s physician fee schedule have grown more rapidly than any other type of physician service. Between 2000 and 2005, spending doubled from $6.6 billion to $13.7 billion [17]. In 2005 the American College of Cardiology Foundation convened a tech- nical panel to assess the appropriateness of RNI for 52 indications [18]. Of the 52 indications, 13 were deemed inappropriate, 27 appropriate, and for 12 the appropriateness was uncertain. Moreover, there was not even consensus on all of the indications for which RNI was deemed appropriate.

OCR for page 333
0 LEARNING WHAT WORKS For example, for 6 of the 27 indications deemed appropriate, there was strong disagreement among the panelists about that designation. Much more research is needed to reduce the level of clinical uncertainty and move the system toward efficient practice patterns. However, CER will not be sufficient to eliminate overuse. Even when identified, system factors and the complexities of care limit the ability of the system to eliminate the waste. Research on these system factors, includ- ing patient- and system-oriented interventions such as benefit design and clinician/hospital reimbursement, will be needed to complement CER and to allow development of the systems needed to realize the potential offered by CER. b. Underuse Paradoxically, while overuse in the healthcare system is com- mon, underuse of medical services rigorously determined to provide sub- stantial clinical benefit is also widespread. While the small area variation discussion commonly focuses on overuse, similar aggregate-level outcomes in high-expenditure areas and low-expenditure areas imply that some of the small area variation may be due to underuse. For example, among patients with heart attacks who were considered “ideal candidates” for beta-block- ers, those who actually received the needed drug ranged from 5 percent to 92 percent of patients among the 306 Dartmouth Atlas Hospital Referral Regions (HRRs) [6]. A substantial portion of underuse reflects the failure of individuals or their physicians to use preventive services or to manage their chronic illnesses as the scientific evidence would recommend. CER is needed to improve our ability to identify when variation represents underuse and when it represents overuse so that the system can respond appropriately. However, as with overuse, CER will not be sufficient to eliminate underuse. While the clinician-patient relationship plays a critical role in this shortcom- ing, systemic effects such as access to care, benefit design, and ability to pay are also likely contributors, and more research examining these factors will be needed to improve the ability of the system to integrate CER findings into practice. 2. Effects of Inefficiency on Key Stakeholders Inefficiency in the healthcare system, particularly that which leads to unnecessary expenditures, affects all stakeholders. Both overuse and unde- ruse reduce the value of the resources devoted to the healthcare system. The enormous incremental costs associated with this inefficiency are borne throughout the system. a. Individuals Individuals, whether they use the system or not, pay for these inefficiencies in several ways and are unmistakably worse off. First, in some

OCR for page 333
 APPENDIX A cases, individuals pay out of pocket for services (e.g., total body imaging scans) that provide little value in terms of clinical outcomes. Second, the financial costs associated with waste are reflected in higher healthcare premiums. These are paid by workers either directly or, because higher healthcare costs lead employers to pay lower wages, indirectly [19]. Third, higher costs for public programs are financed by taxpayers. The costs of the largest public program, Medicare, rose 8.9 percent to $342.0 billion between 2004 and 2005 [20]. Furthermore, projections suggest that Medi- care will grow at an annual rate of over 9 percent between 2005 and 2015 [21]. The growth of Medicare spending will represent a serious burden for taxpayers and a significant challenge for policy makers. It is well established that the tradeoff between access to medical care and how to pay for it is a complex and extremely political issue. Fourth, high healthcare costs are also associated with declining rates of health insurance coverage [22]. To the extent that greater waste leads to fewer covered individuals, those that are un- and under-insured must bear greater financial risk and suffer the consequences of diminished access to valuable care in the event that such care is needed. Finally, inefficiency generates additional adverse consequences for patients already engaged in the system. Specifically, the over-consumption of care often entails clinical risk as well as financial costs. Over the past decade, the “patient safety” movement has brought to light the extent of the clinical and economic ramifications of avoidable medical errors. For example, hospital-acquired infections are estimated to be responsible for between $3.5 billion and $5.7 billion in excess healthcare costs each year [23, 24]. Under-utilization also generates suboptimal clinical outcomes as patients forego utilization of important services. b. Employers The clinical and financial effects of inefficient care delivery on other stakeholders are more complex. To the extent that employers bear a large fraction of the costs associated with inefficiency, they are adversely affected. As mentioned above, standard economic models supported by empirical evidence suggest that, over time, employers shift the costs of higher healthcare spending to workers in the form of lower wages. How- ever, in the short run, employers (or the shareholders) may bear some of the costs of inefficiency. Moreover, the ability to shift cost to workers is limited for retiree expenses, suggesting that shareholders will bear the costs of inef- ficiency for this population of workers. Employers may also bear some of the administrative costs associated with managing healthcare benefits in an environment of rising costs and considerable inefficiency. c. Health insurers The fiscal implications of inefficiency for insurers are also complicated. To the extent that cost increases can be anticipated, they may

OCR for page 333
2 LEARNING WHAT WORKS be included in premiums. However, as premiums escalate, the demand for coverage may be dampened, suggesting that, on balance, insurers will find it challenging to remain profitable in a rising cost environment over the long run. Yet with the challenge comes opportunity. If insurers can develop ways to address the problems of inefficiency in the healthcare system, substantial profit opportunities may arise. d. Providers of healthcare services Providers of healthcare services—espe- cially those whose income is related to productivity, not quality of care—may be one stakeholder group that benefits from inefficiency. Since one group’s expense is another’s revenue, the payments for unnecessary interventions are income for the providers of those services. Thus, while no physician or hospital may intentionally, or even knowingly, provide unnecessary services, they likely reap some financial gain from the services delivered, necessary or not. The magnitude of this effect for particular providers depends on the extent to which they deliver unnecessary care. Providers of necessary care would not be adversely affected by reductions in the use of unneces- sary services. Reductions in the use of unnecessary care may offer indirect benefit to providers in the long term. Specifically, higher costs lead to fewer people with coverage. This may place a burden on providers who are increasingly called on to provide uncompensated care. Providers may also benefit from any reductions in inefficient care because they may find this type of cost containment preferable to other approaches (such as fee reductions). e. Manufacturers The impact of inefficiency (and efforts to reduce inef- ficiency) on manufacturers is much the same as on the providers of those services. Any reduction in utilization may be a reduction in revenue, but the effects will target low-value or unnecessary services. Manufacturers that have the potential to make important clinical advances can thrive in a low- waste environment. Moreover, relative to other cost containment efforts that may impact manufacturers, efforts to reduce unnecessary use of certain medical products may be preferable. . Uses of CER Discussions about CER frequently focus on the use of these evalua- tions to assist in development of practice guidelines or in coverage and payment decisions. While CER could be used in these specific endeavors, CER is needed for more far-reaching efforts to improve the efficiency of the healthcare system. The critical nature of a comprehensive CER agenda arises because of the lack of controlled assessments of available therapeutic options and the substantial amount of patient heterogeneity that exists. Waste generally arises when services that are valuable in some clinical situ-

OCR for page 333
 APPENDIX A ations are applied to other indications. CER is an essential tool to determine which intervention should be delivered to which person and in what clinical circumstance. By facilitating improved targeting of both the clinical intervention and the specific patient population, the information provided by CER can benefit key stakeholders, particularly patients and payers. Specifically, by reducing the uncertainty about which treatment course is most appropri- ate, CER can decrease the frequency that patients receive inappropriate care, reducing costs and the potential for harmful medical errors. Similarly, CER can facilitate efforts to develop coverage policy and design value- based insurance packages, which should enhance the return on healthcare expenditures made by payers—private or public [25]. Taking the perspec- tive of the provider, the effects of CER on utilization will depend on both the nature of the product and the incentives in place to use the service. If coverage and reimbursement levels reflect the findings of CER (i.e., payment based on clinical effects, not exclusively on production costs), providers and manufactures of high-value services should find that the CER increases their market share. However, the demand for low-value services will likely (appropriately) decline. Given that the burden of proof necessary to demon- strate value in the marketplace may intensify, so might the costs to perform the requisite CER studies. A particular concern for providers is that cost containment efforts designed to eliminate the use of unnecessary services often inadvertently lead to restrictions on the provision of needed care. In almost all of the studies that report the appropriate indications for the use of a specified intervention, the appropriateness is “uncertain” in a significant portion of situations. Recall that there are few instances where the use of a specific drug, diagnostic test, or procedure is always appropriate or inappropriate. This underscores the need for a CER agenda that is able to measure health and economic impact on a granular level that will ultimately target those specific circumstances when certain interventions should and should not be used. While the evidence examines both under- and overuse of selected medi- cal services, one cannot accurately predict the net effect of a more efficient system on expenditures. This is related to the tradeoffs of how a better evidence base drives the increased use of more valuable services (and likely increases expenditures) and slows the utilization of low-value interventions (and decreases spending). Individual CER studies may not always suggest that the least expensive course of action is the appropriate course of action—recall that “the most expensive therapy is the one that doesn’t work.” However, medical culture tends to be driven toward the adoption of new, expensive services, and cost growth has widely been attributed to the development and diffusion of new

OCR for page 333
 LEARNING WHAT WORKS medical services [25-27]. Therefore, on balance, we would expect that CER would tend to dampen spending to a level below that which would other- wise occur, because the ”adopt everything for everyone” mentality would be replaced with an “adopt when appropriate” paradigm. For example, a 2006 study examined whether some stable, high-risk patients with persis- tent total occlusion of the infarct-related coronary artery should undergo percutaneous coronary intervention (PCI) in addition to receiving optimal medical therapy [28]. Although use of this procedure in such cases was not universal, the authors reported an inclination among physicians toward its use. In this case, a randomized trial demonstrated that PCI did not improve clinical outcomes, suggesting that resources could be saved by foregoing the procedure. The trend would likely have been toward greater use, and the CER-suggested lower use was medically appropriate. Since the literature on diffusion of medical technology clearly shows a preference among U.S. clinicians to use new interventions before definitive clinical data are available, one can safely assume that the clinical data pro- vided by a CER agenda will improve the quality of care. However, it should not be assumed that the completion and implementation of a CER agenda will save money in the short term. The short-term financial consequences will depend on how CER is used and on whether the savings incurred to lower rates of use of low-value interventions will offset the added expenses of the increased use of higher-value services. While enhancing the health of Americans is a noble goal, we acknowl- edge that cost containment is an integral and inevitable part of the future healthcare policy. Without a strong investment in CER, patients and provid- ers are more likely to face unintended “across the board” restrictions on the provision of valuable care because of the fiscal pressures that are being imposed on public and private health care payers. Whether these are mani- fested by fewer insured individuals or by the underinsurance of those with some type of benefits, CER provides the knowledge base by which providers of high-value services can advocate their continued use, using accepted sci- entific approaches to make their case. The findings of research that directly compares the pros and cons of available treatment options from numerous perspectives will be important for clinical practice, regardless of the cost containment/benefit reform approaches being considered. Cost containment efforts that rely on an improved evidence base are likely preferable to cur- rent efforts to drive all practice toward those of the lowest cost. Findings from CER should be used to better target, not to limit, care. The exact mechanisms by which CER will lead to enhanced efficiency will vary based on the level of detail of the data generated by the studies and the ability of the system to implement the findings in everyday practice. On the quality improvement side, similar challenges have been identified in studies examining the suboptimal uptake of evidence-based practice

OCR for page 333
 APPENDIX A guidelines. From the financial perspective, cost-sharing approaches aim to control spending by making patients pay more at the point of service. Most efforts to raise patient out-of-pocket costs have resulted in higher costs across all services (with the possible exception of some preventive health services). It has been demonstrated that financial disincentives are often placed at the patient level, making adherence with evidence-based care difficult. Yet when faced with higher costs, patients often make poor clinical decisions, which in fact could, in some cases, lead to greater overall costs. Thus, the alignment of clinical and financial incentives is a necessary component to ensure the attainment of an efficient delivery system. The status quo has been unable to align quality improvement and cost contain- ment initiatives. In fact, in some instances they actually compete with one another, contributing directly to ineffiency [29]. Such an alignment of incentives is possible in the setting of improved clinical evidence—driven by CER—and health benefit reform. Value-based insurance design (VBID) represents a “clinically sensitive, fiscally respon- sible” approach that advocates keeping patient out-of-pocket payments low on high-value services and raising them on services of no or marginal clinical value. Similar processes can be developed for clinician payment (e.g., payment based on quality of care delivered, not productivity). Imple- mentation of such a scheme, in any form, would require greater CER since the relative value of services would be based directly on the findings. The advantages of such an alignment of clinical and economic incentives are obvious when compared to the current approach of untargeted “across the board” cost-sharing schemes, where the rates of both non-essential and essential services are negatively affected by higher out-of-pocket rates. By using incentives to encourage the use of high-value services and discourag- ing low-value ones, VBID has the potential to achieve marked increases in the efficiency of the healthcare system. Supply-side-oriented healthcare reform approaches could also benefit from added investment in and coordination of CER. Certainly, coverage policy and clinical guidelines require such knowledge. But other initiatives, such as provider education, disease management, or pay-for-performance programs, all require an understanding of which services provide value in which settings and how quality and cost metrics can be designed in a clini- cally meaningful way. 4. Conclusions Healthcare cost growth has placed a growing strain on our healthcare financing system. Although there is no consensus about how we can address the healthcare cost issue, most stakeholders would probably agree that the resources devoted to health care must be allocated more efficiently. This will entail being able to identify situations when more resources are necessary to

OCR for page 333
 LEARNING WHAT WORKS overcome the problem of underuse of highly valued services that improve health, as well as when money is being wasted on interventions that do not improve health, or worse, actually produce adverse consequences. Regardless of the reform approach considered—market-based health savings accounts or a system administered through a single source— enhanced efficiency will require more detailed knowledge about the relative effectiveness of different interventions in specific clinical indications. All vested stakeholders should encourage investment in an infrastructure that prioritizes and undertakes investigations that yield practical information on which services to provide to which patients and when. Our healthcare system is too complex and too large to be guided without an appropriate knowledge base. Moreover, because innovation in the healthcare sector is substantial, investment in an infrastructure that would allow the assessment of the clinical and economic impact of new and existing diagnostic and treatment modalities is essential. Creating this infrastructure will require a substantial investment. For those who consider the upfront investment necessary to create such an infrastructure to be unaffordable, it is imperative to contemplate the costs of the status quo that propagate tremendous inefficiency. REFERENCES 1. Schoen, C., et al. 2006. US health system performance: A national scorecard. Health Affairs Web Exclusive, W457-475. 2. Kaufman, J. S., et al. 1998, Subcutaneous compared with intravenous epoetin in pa- tients receiving hemodialysis. Department of Veterans Affairs Cooperative Study Group on Erythropoietin in Hemodialysis Patients. New England Journal of Medicine 339(9):578-583. 3. Fisher, E. S., and J. E. Wennberg. 2003. Health care quality, geographic variations, and the challenge of supply-sensitive care. Perspectives in Biology and Medicine 46(1):69-79. 4. Fisher, E. S., et al. 2003. The implications of regional variations in Medicare spend- 4. spend- ing. Part 1: The content, quality, and accessibility of care. Annals of Internal Medicine 138(4):273-287. 5. Fisher, E. S., et al. 2003. The implications of regional variations in Medicare spend- 5. spend- ing. Part 2: health outcomes and satisfaction with care. Annals of Internal Medicine 138(4):288-298. 6. Wennberg, J. E., E. S. Fisher, and J. S. Skinner. 2002. Geography and the debate over 6. Medicare reform. Health Affairs Suppl Web Exclusives:W96-114. 7. Guadagnoli, E., et al. 1995 Variation in the use of cardiac procedures after acute myo- cardial infarction. New England Journal of Medicine 333(9):573-578. 8. Wennberg, J. E., and M. M. Cooper, eds. 1999. The Dartmouth Atlas of Health Care. Chicago: American Hospital Association. 9. Bernstein, S. J., et al. 1993. The appropriateness of hysterectomy. A comparison of care in seven health plans. Health Maintenance Organization Quality of Care Consortium. JAMA 269(18):2398-2402. 10. Broder, M. S., et al. 2000. The appropriateness of recommendations for hysterectomy. Obstetrics and Gynecology 95(2):199-205.

OCR for page 333
 APPENDIX A 11. Bernstein, S. J., et al. 1993. The appropriateness of use of coronary angiography in New York State. JAMA 269(6):766-769. 12. Hilborne, L. H., et al. 1993. The appropriateness of use of percutaneous transluminal coronary angioplasty in New York State. JAMA 269(6):761-765. 13. Halbert, R. J., et al. 2006. Treatment of patients with metastatic renal cell cancer: A RAND Appropriateness Panel. Cancer 107(10):2375-2383. 14. Winslow, C. M., et al. 1988. The appropriateness of carotid endarterectomy. New Eng- land Journal of Medicine 318(12):721-727. 15. Kahn, K. L., et al. 1988. The use and misuse of upper gastrointestinal endoscopy. Annals of Internal Medicine 109(8):664-670. 16. Siu, A. L., et al. 1986. Inappropriate use of hospitals in a randomized trial of health insurance plans. New England Journal of Medicine 315(20):1259-1266. 17. Kuhn, H. 2006. Testimony on Payment for Imaging Services under the Medicare Physi- cian Fee Schedule. July 18. Online. Available at http://www.hhs.gov/asl/testify/t060718. html. Accessed January 22, 2007. 18. Brindis, R. G., et al. 2005. ACCF/ASNC appropriateness criteria for single-photon emis- emis- sion computed tomography myocardial perfusion imaging (SPECT MPI): A report of the American College of Cardiology Foundation Quality Strategic Directions Committee Appropriateness Criteria Working Group and the American Society of Nuclear Cardiol- ogy endorsed by the American Heart Association. Journal of the American College of Cardiology 46(8):1587-1605. 19. Gruber, J. 1994. The incidence of mandated maternity benefits. American Economic Review 84(3):622-641. 20. CMS Office of Public Affairs. 2007. Press Release: CMS Releases U.S. Health Spending Estimates Through 200. January 9. Online. Available at http://www.cms.hhs.gov/apps/ media/press/release.asp?Counter=2069&intNumPerPage=10&checkDate=&checkKey= &srchType=&numDays=3500&srchOpt=0&srchData=&keywordType=All&chkNews Type=1%2C+2%2C+3%2C+4%2C+5&intPage=&showAll=&pYear=&year=&desc=& cboOrder=date. Accessed January 22, 2007. 21. Borger, C., et al. 2006. Health spending projections through 2015: Changes on the ho- rizon. Health Affairs 25(2):w61-73. 22. Chernew, M., D. M. Cutler, and P. S. Keenan. 2005. Increasing health insurance costs and the decline in insurance coverage. Health Services Research 40(4):1021-1039. 23. Burke, J. P. 2003. Infection control—A problem for patient safety. New England Journal of Medicine 348(7):651-656. 24. Hollenbeak, C. S. 006. Factors associated with risk of surgical wound infections. Ameri- can Journal of Medical Quality 21(6):29S-34S. 25. Chernew, M. E., et al. 1998. Managed care, medical technology, and health care cost growth: A review of the evidence. Medical Care Research and Review 55(3):259-288; discussion 89-97. 26. Cutler, D. M. 1995. Technology, Health Costs, and the NIH. Cambridge, MA: Harvard University and the National Bureau of Economic Research. 27. Newhouse, J. P. 1992. Medical care costs: How much welfare loss? Journal of Economic Perspectives 6(3):3-21. 28. Hochman, J. S., et al. 2006. Coronary intervention for persistent occlusion after myocar- myocar- dial infarction. New England Journal of Medicine 355(23):2395-2407. 29. Chernew, M. E., A. B. Rosen, and A. M. Fendrick. 2006. Rising out-of-pocket costs in disease management programs. American Journal of Managed Care 12(3):150-154.

OCR for page 333