Evidence-Based Medicine and the Changing Nature of Health Care: 2007 IOM Annual Meeting Summary (2008)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Summary INTRODUCTION AND OVERVIEW The rapid pace of scientific discovery and technological innovation over the last several decades is unprecedented and raises the prospect of achieving dramatic improvements in the nation’s health and well-being. Yet stakeholders from across the healthcare system, from patients to prac- titioners to payers, are demanding fundamental improvements to a system that is seen as costly, fragmented, and ineffective. Because of its emphasis on integrating the best available external evidence with clinical experience, evidence-based medicine (EBM) provides a guiding framework for the de- velopment of systems and approaches necessary to deliver the promise of 21st century health care—in which knowledge is both applied and gener- ated as a natural outgrowth of the care process, to ensure delivery of the care most appropriate for each individual patient. The nation relies on the Institute of Medicine (IOM) for indepen- dent, science-based advice on matters of biomedical science, medicine, and health. In part, the IOM’s strength as an advisory organization lies in its ability to draw upon its membership, which includes distinguished health professionals as well as researchers and leadership from the fields of medicine and health care; and, the IOM annual meeting provides the op- portunity for IOM members and other guests to discuss timely topics that are central to the nation’s healthcare policies. In recent years, the annual meeting has focused on stem cells, tissue engineering, longevity and health, and pharmaceuticals in the 21st century. With healthcare reform at the top   This publication has been prepared by the authors to document the proceedings of the 2007 IOM Annual Meeting.  

 EVIDENCE-BASED MEDICINE of the national domestic agenda, the choice of EBM by the IOM Council as the theme of the 2007 IOM Annual Meeting underscores its centrality and importance to healthcare improvement and offers the opportunity to bring this topic into a broader context as a transforming national priority. Evidence-Based Medicine and the Institute of Medicine The IOM has throughout its existence been dedicated to improving the health and health care of Americans. Its seminal contributions to drawing attention to issues and policies important to better quality health care have recently included To Err Is Human (2000) and Crossing the Quality Chasm (2001). In the face of the growing awareness generated about the nature and implications of the gap between healthcare practice and the evidence base, the IOM Roundtable on Evidence-Based Medicine was established in 2006 to provide a neutral forum for discussions and collective action by healthcare stakeholders to help transform the way evidence on clinical effective­ness is generated and used to improve health and health care. Underscoring the challenges faced by healthcare decision makers, the Roundtable has defined EBM to mean that “to the greatest extent pos- sible, the decisions that shape the health and health care of Americans—by p ­ atients, providers, payers, and policy makers alike—will be grounded on a reliable evidence base, will account appropriately for individual variation in patient needs, and will support the generation of new insights on clinical effectiveness” (IOM’s Roundtable on Evidence-Based Medicine, 2006). To support EBM in practice, the Roundtable seeks the development of a learn- ing healthcare system that “draws on the best evidence to provide the care most appropriate to each patient, emphasizes prevention and health promo- tion, delivers the most value, adds to learning throughout the delivery of care, and leads to improvements in the nation’s health” (IOM’s Roundtable on Evidence-Based Medicine, 2006). With the guidance of members and expert panels, the Roundtable has conducted a series of meetings and workshops aimed at fostering progress toward the “learning healthcare system”—a system in which both evidence development and application flow naturally from the care process. In addi- tion to the development of the 2007 IOM Annual Meeting on the topic, the series of workshops exploring the barriers, challenges, and opportunities for this vision include • The Learning Healthcare System (2006); • Judging the Evidence: Standards for Determining Clinical Effective- ness (2007); • Leadership Commitments to Improve Value in Health Care: Find- ing Common Ground (2007); 

SUMMARY  • Redesigning the Clinical Effectiveness Research Paradigm: Innova- tion and Practice-Based Approaches (2007); • Clinical Data as the Basic Staple of Health Learning: Creating and Protecting a Public Good (2008); • Engineering a Learning Healthcare System: A Look at the Future (2008); • Learning What Works: Infrastructure and Clinical Priorities to Learn Which Care Is Best (2008); and • Making Better Choices: Getting the Value We Deserve from Health Care (2008). COMMON THEMES: ISSUES, CHALLENGES, AND OPPORTUNITIES The 2007 IOM Annual Meeting provided the opportunity to juxtapose the potential for health care in the 21st century with current shortfalls and to present to important leaders in the field some emerging resources and key policy opportunities that could help transform health care. Discus- sions focused on four themes: the forces driving the need for better medical evidence; the challenges with which patients and providers must contend; the need to transform the speed and reliability of new medical evidence; and the legislative and policy changes that would enable an evidence-based healthcare system. During the course of meeting presentations, a num- ber of ­ common themes were identified (Box S-1). Among them were the following: • Increasing complexity of health care. New pharmaceuticals, medi- cal devices, technologies, and predictive data offer much promise for improving health care, but they also introduce high levels of complexity, requiring changes on the parts of both caregivers and their patients. • Unjustified discrepancies in care patterns. The intensity of health- care services delivered for similar conditions varies significantly across geographic regions, particularly in areas that require dis- cretionary decision making. However, the higher-spending regions often do not deliver better-quality care, hence offering substan- tial opportunity for reduced spending without sacrificing health outcomes. • Importance of better value from health care. The current healthcare system is not designed to deliver value, and the nation’s long-term fiscal challenges are serious and are being driven predominately by excessive medical spending, often on interventions of no clinical benefit. Opportunities exist to eliminate wasteful spending with no 

 EVIDENCE-BASED MEDICINE BOX S-1 The Changing Nature of Health Care Common Themes • Increasing complexity of health care • Unjustified discrepancies in care patterns • Importance of better value from health care • Uncertainty exposed by the information environment • Pressing need for evidence development • Promise of health information technology • Need for more practice-based research • Shift to a culture of care that learns • New model of patient-provider partnership • Leadership that stems from every quarter reduction in health care, as well as to improve the overall health outcomes, but agreement is needed both on what constitutes best care and on what constitutes value in health care. • Uncertainty exposed by the information environment. An irony of the information-rich environment is that information important to clinical decision making is often not available, or is provided in forms that are not relevant to the broad spectrum of patients— with differing levels of health, socioeconomic circumstances, and preferences—and the issues encountered in clinical practice. This is due to too little clinical effectiveness research, to poor dissemina- tion of the evidence that is available, and to too few incentives and decision supports for evidence-based care. • Pressing need for evidence development. More and better evidence— including comparative and longitudinal data—is needed to deter- mine the effectiveness and usefulness of new medical interventions, treatments, drugs, devices, and genetic information. There is an untapped potential to reduce healthcare costs and improve quality by developing evidence not only for specific medical interventions, but also for the way health care is delivered. • Promise of health information technology. Electronic medical r ­ ecords (EMRs) and clinical data registries offer tremendous po- tential both to generate new evidence and to augment randomized clinical trials. Addressing privacy and proprietary issues that limit data access and sharing would help to support a system in which 

SUMMARY  EMRs, clinical registries, and other types of electronic data could contribute to building a more robust evidence base. • Need for more practice-based research. How might the system better support the notion of a “living textbook of medicine” in which the experience of healthcare diagnosis and treatment is rou- tinely captured in order to better care for those in the future. To develop best evidence for the delivery of medicine that is geared toward the needs of individual patients, investment is needed into infrastructure for the gathering and analysis of healthcare data and information, and standards and protocols to ensure their accuracy and reliability. • Shift to a culture of care that learns. This changing role will require healthcare providers and patients to adopt a culture that supports the generation and application of evidence. Effective culture change must also be accompanied by insurance and reimbursement system reform that encourages development and application of the systems necessary. • New model of patient-provider partnership. With the increasing complexity of care, and the need and demand for more patient involvement, the traditional “physician-as-sole-authority” model will need to adapt to support patients as integral partners in medi- cal decisions. • Leadership that stems from every quarter. Adapting to and taking advantage of the changes in the healthcare environment will take broad leadership. A strategic focus on the development and appli- cation of evidence will require the involvement of both the public and private sectors working together, and with policy makers, providers, patients, insurers, and other stakeholders in the steps toward change. PRESENTATION SUMMARIES The Changing Nature of Health Care (Chapter 1) Meeting co-chairs Mark B. McClellan, director of the Engelberg Center for Health Care Reform and Leonard D. Schaeffer Chair in Health Policy Studies and Economic Studies of the Brookings Institution, and Elizabeth G. Nabel, director of the National Heart, Lung, and Blood Institute of the National Institutes of Health, opened the meeting with introductions and overviews of key issues. Comments were also provided as context for each session by the moderators: Denis A. Cortese, chief executive officer of Mayo Clinic; Michael M. E. Johns, chancellor of Emory University; John W. 

 EVIDENCE-BASED MEDICINE Rowe, professor, Columbia University; and John K. Iglehart, editor emeri- tus of Health Affairs. A summary of their comments follows. The content of the session presentations also follows, with more detail provided by the authors in Chapters 2 through 7. Mark B. McClellan and Elizabeth G. Nabel, Meeting Co-Chairs In his remarks, McClellan focused on two core challenges facing health care in the 21st century: rising and unsustainable costs and the untapped potential presented by biomedical and technological advances. Creating an evidence-based system is essential to achieving both the promise of per- sonalized medicine and the value needed from health care. In the current medical system, practices vary widely, which can affect health outcomes and costs. Treatments effective for some may not actually be beneficial for oth- ers, and may even carry significant risks. Needed evaluation of interventions and delivery system issues in real-world practice will require that we look beyond the traditional research focus to new methods and new opportuni- ties presented by emerging data sources, including electronic health records (EHRs) and clinical registries. For sustained change, the system also must be better oriented around value and outcomes. A systems-level approach is needed to align policy and reimbursement to support and reward better quality at lower costs. Nabel noted that while there is much consensus about the problems with the current healthcare system, agreement and collaborative work are needed to develop solutions. The roles and responsibilities of all stake­ holders in the healthcare system are undergoing tremendous transforma- tion; patients, providers, payers, industry, and policy makers alike must work together to contend with these changes and make needed reforms. P ­ roviders are operating in an increasingly complex system with an ever- growing amount of medical information and treatment options to consider. The ­ physician-patient relationship is changing as patients have access to more Web-based health information and are more empowered to make decisions about their own health care. Research methodologies will need to be adapted to take better advantage of the increased amount of real-time and real-world data from EHRs and clinical registries. Beyond the evalua- tion of interventions, methodologies are needed to model and analyze work processes and decision management. This type of research may require restructuring of funding to include research on how to attain changes in work processes and set value in the system. Both McClellan and Nabel emphasized the many technical and policy challenges that must be overcome to fulfill the vision of the Roundtable. While EBM is increasingly at the forefront of policy making and is driving 

SUMMARY  various reform initiatives at government agencies, greater collaboration by all stakeholders is essential for progress. An increased role of healthcare product companies and other healthcare industries is particularly ­important. Public-private partnerships are needed to identify new approaches to the generation and application of evidence in medical practice in order to i ­mprove care and reduce costs. Denis A. Cortese, Chief Executive Officer, Mayo Clinic Cortese noted that the current system provides no incentives for pa- tients to seek—or for practitioners to provide—high-quality, cost-effective health care. If the value of the healthcare system is measured in outcomes, safety, and service in relation to the cost over time to provide services, the return on our investment is falling far short of its potential. Current dissatisfaction with health care provides an opportunity to develop a vision, create a strategy, and specify goals for a different kind of healthcare system. The ideal system would reward prevention of the onset of illness, tailor specific treatments for possible cures, control chronic illnesses, and improve patients’ quality of life. The system should be affordable for both individuals and the country. The overarching vision, as conceived by the IOM’s Roundtable, is one of a learning healthcare system “designed to generate and apply the best evidence for the collaborative choices of each person and provider.” The system would support innovation and discovery as a natural outgrowth of care and ensure quality, safety, and value. Essential to progress toward this shared vision is a better understand- ing of the forces driving the need for better medical evidence. Chapters 2 and 3 summarize presentations on several of the key issues introduced by Cortese: the tremendous negative economic consequences of the rising, unsustainable costs of health care; implications of the variations in health- care spending and health outcomes in different regions of the country; the quality and quantity of evidence needed to guide clinical decisions; the increasing diversity of new health products; and the complexity of insights generated by genetic research. Michael M. E. Johns, Chancellor, Emory University Chapters 4 and 5 examine how patients and providers might begin to contend with the healthcare system’s increasing complexity and the role of information technology (IT) in the process. Johns’s introductory com- ments outlined the challenge presented by the vast amount of healthcare information available to consumers—information of varying quality and complexity. This is daunting even to providers with years of education 

 EVIDENCE-BASED MEDICINE and experience, but it is far more challenging to the non-expert. Tools are becoming available that will assist navigation of the information needed to better support healthcare providers, foster improved patient-physician partnerships, and empower patients in the “Information Age.” However, attention is also needed on educational efforts to promote a better under- standing by patients and providers of what constitutes good evidence for effective care. As more and better evidence is developed, effective processes and IT systems will be necessary to ensure that healthcare practices actually apply the evidence. Methods for accelerating the dissemination and incorpora- tion of new knowledge into practice should be identified so that it does not take years for providers to catch up on new knowledge and skills. Effective systems must include proper rewards and incentives for providers to imple- ment best practices, as well as mechanisms to pay for information systems and process innovations. Education is a key component to ensure needed fluency with new systems and capabilities, as well as to increase the appreciation of and demand by the general public for evidence-based care. Practitioners need to work with the public to identify and implement the right healthcare solutions. Needed from healthcare leadership, therefore, is initiative for policies that empower patients and providers to catalyze and implement needed change. John W. Rowe, Professor, Columbia University In introducing the presentation summarized in Chapter 6, Rowe em- phasized that a key characteristic of the Roundtable’s vision of a learning healthcare system is continual evidence development and refinement. Infor- mation is acutely needed on the effectiveness of individual drugs, devices, and procedures and systems of care and care delivery and, in particular, how these competing therapies stack up against each other. A major strat- egy proposed to hasten the development of the needed evidence base is to take advantage of a broader range of methodologies, beyond randomized controlled trials (RCTs). Concerns have been raised that such approaches are not as reliable as RCTs and that time saved employing different research approaches might not be worth the cost in quality and reliability. However, rather than looking at these issues as absolute, Rowe urges researchers and practitioners to identify strategies to enhance the types of data being used to increase the speed and reliability of evidence generated. Chapter 6 includes presenters’ views of the opportunities presented by EMRs and clinical data registries for generating real-time evidence. In addition, the potential for developing and delivering increasingly tailored therapies is examined. 

SUMMARY  John K. Iglehart, Editor Emeritus, Health Affairs Iglehart sets the stage for the discussion of policy issues and consider- ations which are included in Chapter 7. Key issues include placing a sharper focus on value in health care, as well as understanding the current political environment that must be navigated to drive change. Iglehart cautions that despite the potential for reform, the overall process of policy making is badly in need of repair. Perversely influenced by a campaign finance system that makes candidates beholden to narrowly focused special interest groups, change will require savvy political strategy, as well as sound policy. The Need for Better Medical Evidence (Chapter 2) A close examination of national and regional healthcare spending trends and health outcomes reveals much about where the system is falling short and identifies opportunities for improvement. Presentations at the meeting made the case for more and better medical evidence, which could simultaneously reduce unnecessary healthcare expenditures and improve health outcomes for patients. Health Care and the Evidence Base Elliott S. Fisher, professor of medicine and community and family medicine, Center for the Evaluative Clinical Sciences, Dartmouth Medical School, drew upon small-area analyses to underscore the scope of the chal- lenge faced in bringing evidence to bear on current practice and to point out opportunities to improve costs and quality of care. Two categories of treat- ment are considered: (1) discrete, biologically targeted interventions, which have been the traditional focus of a narrow definition of evidence-based practice; and (2) care delivery strategies that look at how these therapies should be delivered—by whom, where, and with what intensity. For both categories, the evidence base is often limited, and the relative magnitude of uncertainty is often reflected in regional variations in the rate of service use among the Medicare population. Interestingly, these categories have distinct relationships to variations in spending, with higher spending largely due to differences in care delivery: how frequently patients are seen, how much time they spend in the hospital, and the intensity with which they are monitored. When viewed in terms of health outcomes, regardless of organizational level or region, higher spending or intensity of care delivery consistently returns no improvement (and in some cases worse quality or outcomes) (Table S-1). If the organizational structures and practice patterns of the lowest-spending regions of the United States could be adopted across the country, Medicare spending would decline by about 30 percent. 

10 EVIDENCE-BASED MEDICINE TABLE S-1  Relationship Between Regional Differences in Spending and the Content, Quality, and Outcomes of Care Higher-Spending Compared to Lower-Spending Regionsa Healthcare • Per capita supply of hospital beds 32% higher (Fisher et al., 2003) resources •  er capita supply of physicians 31% higher overall: 65% more P medical specialists, 75% more general internists, 29% more surgeons, and 26% fewer family practitioners (Fisher et al., 2003) Content and •  dherence to process-based measures of quality lower (quality worse) A quality of care •  ittle difference in rates of major elective surgery (Fisher et al., 2003; L Wennberg et al., 2002) •  ore hospital stays, physician visits, specialist referrals, imaging, and M minor tests and procedures (Fisher et al., 2003) Health •  ortality up to 5 years slightly higher following acute myocardial M outcomes infarction, hip fracture, and colorectal cancer diagnosis (Fisher et al., 2003) • No difference in functional status (Fisher et al., 2003) Physician •  ore likely to report poor communication among physicians (Sirovich M perceptions of et al., 2006) quality •  ore likely to report inadequate continuity of patient care (Sirovich M et al., 2006) •  reater difficulty obtaining inpatient admissions or high-quality G specialist referrals (Sirovich et al., 2006) Patient-reported • Worse access to care and longer waiting times (Fisher et al., 2003) quality of care •  o difference in patient-reported satisfaction with care (Fisher et al., N 2003) Trends over •  lthough all U.S. regions experienced improvements in acute A time myocardial infarction survival between 1986 and 2002, regions with greater growth in spending had smaller gains in survival than those with less growth in spending (Skinner et al., 2006) aHigh- and low-spending regions were defined as the U.S. Hospital Referral Regions in the highest and lowest quintiles of per capita Medicare spending, as in Fisher et al. (2003). SOURCE: Elliott S. Fisher, 2007. When there is strong medical evidence, physicians tend to agree on courses of treatment across regions of different spending levels, but they differ widely in areas that require discretionary decision making, such as how often to see a patient with well-controlled hypertension and when to hospitalize a patient with heart failure. For these decisions, neither patient preference nor the malpractice environment is associated with variations in practice; however, the local organizational and policy environment pro- foundly influences provider decision making. For example, hospitals and physicians operate in a system that rewards the expansion of capacity and 

SUMMARY 11 the recruitment of procedure-oriented specialists. Because so many clinical decisions are in the “gray areas,” such as how often to see a patient, when to refer to a specialist, or when to admit to a hospital, any expansion of capacity results in a shift in clinical judgment toward greater intensity of treatment. These findings point to the need for much better evidence about the risks and benefits of discrete biologically targeted interventions, and offer insights to the decisions of clinicians, administrators, and policy makers about approaches to care delivery. Not only do these differences in the way care is delivered explain almost all of the geographical variations in spending; but in many cases, the effectiveness of targeted interventions will also depend on the delivery strategy. Therefore developing the capacity to better evaluate the effectiveness of both categories of treatments offers an immediate opportunity for improving the costs and the quality of care. Fortunately, the information systems and analytic approaches required to improve the evidence base for both biologically targeted interventions and care delivery are fundamentally the same. For both types of evidence, certain variables such as patients’ health outcomes and factors such as age, race, sex, and severity of illness are needed. Improved information systems and electronic health records allow for assessments of both short- and long- term health outcomes and effective patient follow-up. In fact, the capacity to evaluate both care delivery and biologically targeted interventions will be critical, since the lack of information about how the interventions were administered and monitored would sharply limit the ability to interpret studies of biologically targeted interventions. Finally, Fisher extended a challenge to the field of academic medicine and the government agencies that fund its research. Because of the traditional focus on understanding disease biology, the dramatic variations in care delivery among academic medical centers have been largely ignored despite their substantial potential to lower costs and improve care. If all U.S. delivery systems could achieve the same efficiency as some of the top-performing academic medical cen- ters, more resources would be available to expand healthcare coverage and access. Academic medicine has the opportunity to lead the development of a learning healthcare system by taking up this obvious opportunity to improve care quality. The High Price of a Lack of Evidence Improving the quality of health care and reducing unnecessary spending is not only a priority for the healthcare system, it has become an important national economic imperative. Peter R. Orszag, director of the Congres- sional Budget Office, noted that if healthcare costs continue to grow at the same rate as they have over the past four decades, by 2050 Medicare and 

12 EVIDENCE-BASED MEDICINE Medicaid spending alone would account for 20 percent of the total U.S. economy—an amount roughly equivalent to the entire size of the federal budget in 2007. Although upward-spiraling costs are often misdiagnosed as the consequence primarily of an aging population, lower fertility rates, and longer life expectancy, spending increases actually result mostly from the rising cost per individual beneficiary. In short, the rate at which healthcare costs grow is the central longtime fiscal challenge facing the United States. Simply reducing payment rates of Medicare and Medicaid (and therefore access) may not be consistent with the nature of these public programs. To be sustainable, reforms may therefore need to trigger changes that will have an impact on the overall healthcare system. Cost increases are being driven by a lack of information on the effec- tiveness of medical interventions and healthcare delivery and a payment system that accommodates the delivery of low- or negative-value care. Be- cause lower cost sharing increases health costs overall, many have argued for more cost sharing and health savings account approaches. However, while more cost sharing among consumers could help reduce costs, the overall effect would be limited because a significant amount of cost sharing already exists in health plans and because costs are concentrated among the very sick. For example, 25 percent of Medicare beneficiaries account for 85 percent of total costs. Therefore, instituting more cost sharing probably would not be able to reduce costs significantly without impairing health quality. On the provider side, better evidence and changes in payment incentives could help reduce costs. Evidence on comparative effectiveness is needed for a variety of clinical interventions, but careful consideration of what is meant by comparative effectiveness research and how it would be imple- mented is necessary. Building out the evidence base across the spectrum of clinical interventions and practice norms by relying exclusively on random- ized trials is impractical, but reliance on nonrandomized evidence comes with well-known limitations. Using statistical techniques on panel data sets from EHRs, insurance claims, and other medical data seems to be the only cost-effective and feasible mechanism for significantly expanding the evidence base, but separating correlation and causation is difficult with such an approach. Simply making comparative effectiveness information avail- able is insufficient to impact costs. Releasing information from systematic reviews will have some effects, but they are likely to be modest. Utilization of a broader range of studies—including those using clinical registry or EHR data—will have greater effects, but real gains in improving the quality of health care and reducing costs will come when the results of research are built into incentive payments for healthcare providers. In sum, there is a pressing need for better medical evidence on what works best for whom. When combined with increased consumer informa- 

SUMMARY 13 tion and provider incentives that reward higher-value care, comparative effectiveness research offers an opportunity to reduce healthcare costs and improve health outcomes. Circumstances Accelerating the Need (Chapter 3) Scientific and technological advances sometimes offer dramatic op- portunities to improve treatments, boost the efficiency of care delivery, and provide more options for patients and providers. However, efforts to adopt and integrate these advances into health care confront complex challenges. Advances in the field of genetics in particular are providing a wealth of information on factors influencing the development of disease. The com- plexity and costs introduced by these developments and other innovations in medical products also impact the healthcare system as a whole. Providers will be challenged to find ways to evaluate the effectiveness of these new technologies while also ensuring that innovation is continuously supported, captured, and applied to health care. New Healthcare Product Introduction Molly J. Coye, founder and chief executive officer, Health Technology Center, discussed the challenge presented by the increased complexity and diversity of new medical devices and interventions to the development of adequate evidence to support healthcare decision making (Figure S-1). The clear benefits of medical technologies have underscored the importance of supporting innovation in healthcare product development, and perhaps most pressing in this respect is the development of new approaches to ac- celerate the evaluation of new technologies. The complexity introduced by many new technologies has had two major consequences: (1) waste and inefficiency in the processes of evidence generation, due to poorly defined evidence needs with respect to coverage and reimbursement decisions; and (2) delayed action on the potential of transformative technologies that might enable disruptive and positive changes in clinical care and adminis- trative processes. Key challenges for health care are the development of ap- proaches for the early identification of beneficial technologies, as well as for generating the evidence needed to evaluate these emerging technologies. Recent Food and Drug Administration (FDA) approvals of new prod- ucts have actually been relatively level. Medium- and low-risk products have been approved each year in far greater numbers than novel products, and off-label extensions of drugs and devices are increasing without any kind of formal evaluation. In addition, providers have been contending with the expansion of IT products, most of which do not require FDA ap- proval. However, Coye argued that emerging technologies, including new 

14 EVIDENCE-BASED MEDICINE Diversity, complexity, and cost rather than quantity � Pharmaceuticals biotech drugs � Diagnostics pharmacogenomics and biomarkers � Information technology � Hybrid technologies: devices + pharmaceuticals + information technology � “Follow-on biologics” and biologic equivalency Rapid evolution of technologies � Ongoing, rapid-cycle innovation � Need for guidance documents for evidence development Challenge = current evidence paradigms FIGURE S-1  Key challenges to the current evidence paradigm. SOURCE: Molly J. Coye, 2007. S-1 new.eps biologics and hybrid pharmaceutical and device products, present the most pressing burden on current capacities to develop evidence for regulatory and coverage decisions. For example, the field of biotechnology, including genetic diagnostics and therapies, is growing at twice the rate of chemical pharmaceuticals. Several factors complicate efforts to generate appropriate evidence on the effectiveness of biotech drugs. Often costly and targeting rare or life- threatening diseases, these drugs are in urgent need of assessment, but high patient demand often limits trial participation. Off-label uses for biotech drugs often target unrelated diseases and may quickly become accepted in practice, again limiting opportunities for clinical trials. For “follow-on biologics,” the evidence basis for regulatory decisions will likely be fraught with controversy and legal actions until scientific and legal issues relevant to regulatory policy are resolved. Similarly, the number of biomarkers—which combine the knowledge of genetics, proteomics, and bioinformatics—is expanding rapidly in hopes of providing better indicators of normal biologi- cal or pathogenic processes or of pharmacological responses to therapeutic interventions. Although they are potentially important for developing di- agnostics, appropriate research approaches and regulatory criteria do not yet exist for biomarkers to ensure their appropriate use and relevance to clinical care. Finally, biotech drugs, devices, and information technologies are also converging to produce entirely new classes of technology that 

SUMMARY 15 pose substantial challenges to the established evidence paradigm. Devices, information technologies, and hybrid products evolve more rapidly than they can be evaluated in controlled trials, necessitating new evaluation methodologies and approaches. The complexity of the current regulatory environment is leading to waste and inefficiency because critical research targets have not been iden- tified for technology developers before they initiate clinical studies. With the exception of the largest, most sophisticated firms, the understanding of what needs to be demonstrated in order to win coverage, reimbursement, and support from payers and providers varies widely. Coordinated public and private efforts are needed to identify important research targets, in- cluding priority populations, side-by-side comparisons of effectiveness with competing technologies, and cost effectiveness. This regulatory complexity is also obscuring the contributions of cer- tain technologies that have the potential to significantly transform medi- cal care. These transformative technologies—which include telemedicine, remote monitoring of chronic disease, the tele-ICU (remote monitoring of intensive care units), pharmacogenomics, hemofiltration for congestive heart failure, and remote video interpretation—enable a wide range of positive changes in clinical care and administrative processes, reducing net expenditures and improving the value of health care. These technologies provide important opportunities for progress toward national goals of im- proved quality and efficiency in health care, and—in contrast to biologics and hybrid devices—they present only modest challenges to capacities for evaluation. Extracting the full value of emerging technologies will advance n ­ ational goals of quality, efficiency, and improved patient experience and will ­require new approaches to generating and evaluating evidence. Few transformative technologies are seen as lifesaving treatments, despite the fact that some actually do save lives and lessen the burden of illness. It falls to purchasers, payers, providers, and policy makers to craft new a ­ pproaches to evaluate and simultaneously foster the development of transformative technologies. Fast Adoption of Significant Technologies is one such effort being pilot tested by the New England Healthcare Institute, in collaboration with the Health Technology Center and the Massachusetts Technology Collaborative. This iterative, coordinated, and proactive approach focuses on accelerating the adoption of classes of technology that lower costs and improve quality and emphasizes rapid design evolution, testing across a variety of care settings and applications, clear definitions of research targets that correspond to value, and a com- mitment to support the coverage and reimbursement for technologies of demonstrated benefit. 

16 EVIDENCE-BASED MEDICINE Rapidly Developing Insights into Genetic Variation In just the past few years, researchers have made striking progress in the study of genomics and genetics (Figure S-2). In 2007 alone, the system- atic study of genetic variation across large patient samples revealed more than 75 genetic risk factors for common diseases, tripling the number pre- viously identified. The implications of a vastly expanded amount of new knowledge about human genetics offers great promise for better diagnosis and treatment of disease, but also pose challenges to the healthcare sys- tem, according to David M. Altshuler, director of medical and population g ­ enetics for the Broad Institute. Progress in identifying the genetic causes of common diseases holds great promise to catalyze the development of new insights into pathophysi- ology. However, testing for genetic variations is of uncertain value for the individual patient and the healthcare system. In vivo biological relevance does not necessarily imply clinical utility. The role of inheritance is only one factor in the development of common diseases. In addition, it has become increasingly clear that the inherited contribution itself is divided across many genes. The polygenic nature of common diseases made it difficult to identify any single genetic variant that was reproducibly contributing to FIGURE S-2  Progress in the identification of gene variants for common diseases. SOURCE: David M. Altshuler, 2007. 

SUMMARY 17 risk. Recently, a variety of tools have been developed to enable a simple but comprehensive association study approach for the role of common ­genetic variants in diseases, and findings have greatly contributed to a better under­ standing of underlying physiological processes. For example, a common single-letter change in one complement factor influences the risk of develop- ing macular degeneration fivefold or more. Research prior to these genetic studies had not explored the effects of complement factors, and this added biological insight suggests the possibility that targeting the complement pathway might be a key in preventing the disease. Similar discoveries have been made for Crohn’s disease and Type II diabetes. Each of the newly localized genetic variants is common, so they are present in a substantial proportion of the population. Although it is simple to test a patient to determine whether he or she carries a genetic risk factor, whether this is actually useful information to individuals is much less clear. Unlike rare genetic diseases such as Huntington’s disease, the risks attrib- utable to these newly found genetic variations are typically very modest. A key challenge is determining whether and how clinical testing for such genetic variations can improve patient care. To evaluate this question for Type II diabetes, researchers from Mas- sachusetts General Hospital and the Diabetes Prevention Program (DPP) conducted a landmark study of diabetes prevention involving 5,000 people with impaired glucose tolerance. As part of that effort, the DPP exam- ined a study on the gene TCF7L2, which has the largest effect of any single common variant yet described in Type II diabetes. In that study, patients homozygous for the gene variant—between 5 and 10 percent of the participants—had double the risk of contracting Type II diabetes as identical patients who did not have this high-risk genotype. The study re- sults validate that measurement of TCF7L2 conveys predictive information above and beyond the clinical standard. However, an even more interesting finding was that the lifestyle intervention was equally effective in preventing the onset of diabetes in the high-risk genotype group as in the population as a whole. While these results suggest little value in testing for this gene, a diagnostic is available and is being actively marketed for use. Clearly, to develop an evidence-based approach to genetic testing, clinical research is needed to determine how such information might influence individual behavior, health outcomes, and healthcare utilization. Performing such re- search will be difficult due to a lack of incentives and the rapidly changing nature of genetic information. While the long-term value of identifying genes and DNA variations that influence diseases could be significant advances in prevention and treatment, the marketing of genetic information is a much more uncertain enterprise that currently lacks evidence for improvement of people’s health and well-being. 

18 EVIDENCE-BASED MEDICINE Contending with the Changes (Chapter 4) As new technologies and scientific advances continue to revolutionize what is possible in health care, providers and patients at the front lines of care must contend with an increased number of medical options, as well as an overwhelming amount of information to guide their healthcare choices. Adopting new technologies and medical interventions and embed- ding emerging information into practice will require a cultural shift in the behaviors, beliefs, and practices of individual healthcare providers and delivery organizations. Evaluating the quality and usefulness of medical evidence, putting the evidence into practice, and continuing to adjust and monitor patients’ treatment and health outcomes are tasks that both pro- viders and patients will have to perform. Beyond Expert-Based Practice William W. Stead, McKesson Foundation Professor of Medicine and Biomedical Informatics and associate vice chancellor for strategy and trans- formation at Vanderbilt University, postulates that in order to keep pace with new developments and make informed decisions in a timely man- ner, the healthcare system will shift from expert-based practice, which is built around the extensive knowledge and experience of the physician, to systems-supported practice. In expert-based practice, the individual expert provides extensive knowledge and technical skill based on his or her education and experience. He or she is expected to remember facts, assimilate data, recognize patterns, judge, and make decisions wisely. Stead suggests the demise of expert-based practice is inevitable given the rapidly increasing gap between human cognitive capacity and the number of facts to consider in a single clinical decision (Figure S-3). However, in systems-supported practice, the focus is on the system’s performance. Teams of people, well-defined processes, and IT tools work in concert to produce the desired result consistently. People provide compassion, pattern recognition, and judgment and are supported by well-defined processes that standardize and simplify work flow. IT tools decrease dependence on memory and force action when needed. While both expert-based and systems-supported practices rely on evi- dence, the difference is in how they translate evidence into action. Stead described the rigorous, painstaking steps taken by Vanderbilt University Medical Center to develop a systems-based approach to caring for patients on ventilators. Although there is an abundance of evidence surrounding the care of ventilator patients, much of it is not translated into a form that can be used by practitioners. Vanderbilt’s team built a set of standard practices 

SUMMARY 19 1000 Facts per Decision Proteomics and other effector molecules 100 Functional Genetics: Gene expression profiles 10 Structural Genetics: Human e.g., SNPs, haplotypes Cognitive 5 Capacity Decisions by Clinical Phenotype 1990 2000 2010 2020 FIGURE S-3  Schematic depicting the increase in number of facts per clinical deci- sion with new sources of biological data. SOURCE: Daniel R. Masys and William W. Stead, 2007. new S-3.eps with specific process steps for implementing each practice and measures for assessing them. Moving from an expert-based to a systems-based practice is complicated by the ever-changing nature of health care. The constantly evolving biological systems that healthcare providers work with, combined with the variability of individuals and conditions, make it impossible to standardize treatments. Other industries are able to isolate change in their systems and adjust accord- ingly, but the rate of discovery in the biological sciences and the introduction of new technologies require rapid experimentation and iterative change. To achieve consistent performance and accommodate a number of clinical problems, the variability in biology and values, and the rate of change in biomedical knowledge, standardization should not focus on specific practices, but on a systems approach to practice. Continuous system development and refinement through iterative cycles of development will yield local standard practices consistent with global knowledge, yet adapted to local resources and capabilities and responsive to changing evidence and system performance. In such a scenario, rather than focusing on managing individual patients as an experiment with an N of 1, the expert applies judgment to develop and iteratively refine the system of practice for their organization. 

20 EVIDENCE-BASED MEDICINE To make these changes, policy makers and payers need to understand the characteristics of health care that make moving beyond expert-based practice challenging. Without this understanding they will continue to ask for and pay for changes that are unlikely to produce the desired result. Healthcare providers need to appreciate the expert’s role in the systems approach and understand that the systems approach does not replace or devalue the expert. The Partnership Imperative in an Evidence-Driven Environment Underscoring that each individual patient has differing life circum- stances, cultural needs, preferences, and socioeconomic status, Marc Boutin of the National Health Council discussed the importance of developing an evidence base that takes into account the unique needs of each patient to deliver and ensure the “right” health care for each person. When used in a strong provider-patient relationship, EBM can help to close the quality chasm in patient care as well as target resources to where they are most effective. However, current efforts to increase the use of evidence in health care have not yet delivered on the promise of EBM. Patients with chronic conditions, for example, require ongoing treatment to maintain their qual- ity of life and enable them to remain productive members of society. Often however, treatments for Medicaid patients with asthma, epilepsy, and de- pression have been denied based on the “evidence.” This may save money for the payer at the time, but it often later results in costly emergency room visits and hospitalizations, in addition to physical or emotional suffering for the patient and financial loss. If EBM is to be taken up broadly and implemented systematically, its benefits must be better communicated to patients—many of whom assume that the care they are receiving is evidence-based. EBM must also be struc- tured with the realization that what works for 80 percent of patients may actually cause harm to, or be inappropriate for, the other 20 percent. In other words, as public health decision models and epidemiology are incor- porated into practice, individual patient data in the hands of an individual health professional should be given equal standing to aggregated public health data. More incentives are needed for providers to promote health, wellness, and prevention, and above all, the patient-provider relationship should be protected to ensure that the physician’s experience and patient preferences are considered along with the best evidence. The Promise of Information Technology (Chapter 5) Although the widespread use of the Internet brings information to the fingertips of healthcare providers and patients, providers often have just a 

SUMMARY 21 few minutes to choose which data are most relevant, evaluate the quality of the information, and find ways to incorporate it into clinical practice. Also, while more information is empowering consumers to become full partners in their healthcare decisions, they must find a way to keep track of this information and decide how it will be useful to them. Information Technology Tools to Support Best Practices of Healthcare Providers Robert Hayward, associate professor of the Departments of Medicine and Public Health Sciences and director of the Centre for Health Evidence (CHE) at the University of Alberta, noted that in the messy informational environment of front-line care, the availability of evidence alone will not lead to improved health. IT will become an increasingly important vehicle for linking evidence to improved outcomes by providing information that is supported with convenience and capacities for discrimination and integra- tion. Examples of IT that serve these needs of clinicians are described. Key characteristics of a convenient decision support environment are outlined by the “Rule of Fives”: it must be “responsive,” with evidence sources immediately accessible and available for searching within five sec- onds; “proximate” to practitioners on the front lines and ready to be searched within no more than five mouse clicks; “guessable”—taking no more than five minutes for sufficient orientation; “comprehensive”—serving at least five distinct information needs (e.g., communication, collaboration, evidence access, decision support, documentation, news); and “rewarding” in that users experience five practice-changing rewards per week of system use. In addition to external evidence, convenience also requires access to “internal evidence” derived from organizational data repositories and rel- evant to specific organizational patient populations and settings. Evidence discrimination requires functions to support recognizing, gathering, and reflecting upon internal and personal evidence. For example, the Nemours Foundation, which provides pediatric care in multiple states, is supplying information environments specialized to individual practi- tioners, such as emergency room physicians, surgeons, or nurses. These environments provide at least five evidence services: evidence selection, evidence synthesis, evidence in context, evidence management, and evidence literacy training. Collaborative evidence management tools led to the emer- gence of “communities of inquiry” in clinician groups and changes in the organization that over time increased capacity for use of the best external evidence. Finally CHE has developed integration systems that monitor evidence behaviors in practice. The systems analyze patterns of information use in practice to provide feedback on the quality of the information environment, 

22 EVIDENCE-BASED MEDICINE use of internal and external evidence, level of information seeking and re- flection, and support for organizational change. Collectively these tools provide an information culture that rewards explicit approaches to uncertainty and the use of just-in-time knowledge by making it easier for decision makers to find, collaborate around, and use information. Information Technology Tools that Inform and Empower Patients The past two decades have seen an explosion of data and information relevant to medical care, which is projected to grow by a factor of thou- sands in the coming years. Yet the information will not be useful unless it is accessible to the right people at the right time (Figure S-4). Peter M. Neupert of Microsoft’s Health Solutions Group suggests that consumers are taking a more active role in navigating the health system out of neces- sity, and that these engaged consumers will increasingly become disruptive agents of change in the health ecosystem—demanding better and more in- tegrated information as well as tools to support their healthcare decisions. The availability, portability, sharing, and use of health information will be key to achieving this transformation. This shift in responsibility is inevitable because the consumer cares more about health outcomes than other stakeholders in the system. With TOMORROW • Continuous location tracking • Constant vital signs monitoring • Transcutaneous monitoring of glucose, ETOH, etc. • Noninvasive testing • Exhaled breath analysis • Testing to maintain wellness • Increased image utilization • Genomic data • Proteomics NOW • Personalized pharmacy • Radiology • Radiology • EKG • EKG • Labs • Labs THEN • Meds • Meds • Monitoring • Monitoring • Monitoring • MD exam • MD exam • MD exam • Nurse observations • Nurse observations • Nurse observations THEN NOW TOMORROW FIGURE S-4  Data advances in medicine. SOURCE: Peter M. Neupert, 2007. new S-4.eps 

SUMMARY 23 consumers driving change, competition for healthcare dollars of individu- als will likely result in a system oriented around quality, transparency, and accountability. In short, EBM will no longer be purely a clinical term as consumers increasingly recognize the importance of evidence to guide deci- sions about health care. To facilitate this change, public and private sectors should work to support consumers with best evidence and information management tools, as well as to reverse the fragmentation of the healthcare system and health data. Microsoft recognized that a truly consumer-focused healthcare infor- mation system would also need a private and secure data storage and sharing platform enabling the exchange of data between thousands of health applications and devices. Along these lines, the company released in beta HealthVault™, a free Web-based platform designed to put people in control of their health data. It helps them collect, store, and share health information with family members and participating healthcare providers, and it provides people with a choice of third-party applications, services, and devices to help them manage things such as fitness, diet, and health. HealthVault also provides a privacy- and security-enhanced foundation on which a broad ecosystem of players—from medical providers and health and wellness device manufactures to health associations—can build innova- tive new health and wellness management solutions to help put people in control of their and their family’s health. Collectively these tools are an important step toward achieving the vi- sion of a patient-centered healthcare system. Transforming the Speed and Reliability of New Evidence (Chapter 6) Keeping pace with scientific advances and medical innovations will re- quire an evidence base that adapts and builds over time to provide informa- tion that is timely and up-to-date. A reliance on clinical trials is impractical due to limitations such as cost, amount of time to complete studies, and inability to generalize results to broader populations. Moreover, RCTs can- not answer many important questions about medical interventions and care delivery. EMRs and clinical registries offer the opportunity to capture im- portant data and information at the point of care and speed the generation of evidence to inform clinical practice. New tools such as biomarkers, mod- eling, adaptive trial designs, and patient enrichment approaches are helping to accelerate the development and evaluation of tailored therapies. EMRs and the Prospect of Real-Time Evidence Development George Halvorson, CEO of Kaiser Permanente, discussed the potential of EMRs, if well designed and adequately compiled and supported, to revo- lutionize medical research. Advantages include instant access to a wealth of 

24 EVIDENCE-BASED MEDICINE data; provision of comprehensive and longitudinal data that can span de- cades; access to massive data sets from millions of patients rather than the narrower populations of traditional studies; and greater flexibility in data utilization. These data could be used to support highly structured clinical trials, track progress and care results in the post-market environment well into the future, and analyze population health data in new ways. Electronic data will allow researchers to search for unforeseen cor- relations in ways that were previously impossible. For instance, files could be searched to determine if there is a relationship between specific patient populations and various diseases and comorbidities. Caregivers can gain up-to-the-minute information on which treatments are working for which patients with just-in-time learning and searches within EHRs. In the future, providers could even examine genetic correlations. Some examples of important research facilitated by EMRs at Kaiser Permanente include the recognition of adverse effects of Vioxx in certain patients and the identification of adverse reactions in patients a number of years after treatment with heart stents. Medical records provide great potential in terms of follow-up studies and longitudinal data. As more organizations utilize EHRs for research, they should be careful to design records with research goals in mind. For example, careful consideration is needed to ensure that EHRs facilitate outcome analyses, support clini- cal trials, have data approaches that incorporate genetic information, and contain data sets that can be sorted by relevant demographics such as race, ethnicity, gender, or economic status. Five medical conditions drive more than half of healthcare costs in the United States (mood disorders [depressive and manic depressive disorders], diabetes, heart disease, hypertension, and asthma) (Druss et al., 2001) and present the opportunity to dramatically improve care, as well as re- duce costs. Payment models, delivery structure, data reporting, community priorities, and education should all be aligned toward the generation and application of evidence on how these five diseases are best treated. Build- ing the right electronic data sets, which enable these types of analyses, will transform medical research into a direct tool of medical reform. Research Methods to Speed the Development of Better Evidence— The Registries Example Eric D. Peterson, professor of medicine at the Duke Clinical Research Institute, outlined how clinical registries provide infrastructure and re- sources that help to address current shortfalls in the cycle of evidence development and adoption. Clinical trials tend to involve younger, healthy patients, treated in ideal conditions, and to measure short-term treatment efficacy. Yet the full measure of an intervention’s safety and effectiveness 

SUMMARY 25 can be determined only when it reaches the real-world market of patients and caregivers. Registry data are collected at the point of care and offer the opportunity not only to augment existing evidence development but to accelerate evidence adoption in practice. Clinical registries are clinician-organized networks for collecting de- tailed patient information for a given population, often defined by a par- ticular disease or treatment. Ideally, registries would accurately capture detailed clinical information at key points and events in a patient’s life. These data would also be linkable with other data sources and enable the user to construct a long-term record of a patient’s care and health out- comes. In addition, the registry could be accessible to health services and discovery researchers, as well as to clinicians. The majority of these features exist or are being planned by the major cardiovascular provider-led registries, such as the Society of Thoracic Sur- geons’ National Cardiac Database, the American College of Cardiology’s National Cardiac Data Registries, and the American Heart Association’s programs. The size and scope of these programs are substantial, and al- though participation is voluntary, a growing number of external forces are providing strong incentives for their engagement. For example, a large healthcare insurer encourages registry involvement as a condition for ob- taining “Premium Provider Status,” and some states are requiring participa- tion as part of state-based programs. In addition, more registries are being launched or planned for carotid stenting and acute coronary syndromes, congenital heart disease, and cardiovascular imaging. As registries enter the electronic age, progress in several areas—such as standardization of data elements; clarification of patient privacy rules; development of new data harvesting tools; connection of longitudinal da- tabases; and growing collaboration among professional societies, insurers, and government regulators—is allowing for more integrated and cross- purpose clinical registries. The data and infrastructure provided by clinical registries can aid evi- dence generation in several ways—for example, providing data for national epidemiological and health services research. The Surveillance, Epidemiol- ogy, and End Results Program of the National Cancer Institute provides information on cancer incidence and survival in the United States, and cardiovascular registries have been used to determine national variability in disease treatment, disparities in care among patient subgroups, and trends in treatments over time. Data can also be used to provide larger patient samples for genomic research, such as genome-wide association studies that attempt to link a given genetic variation to a disease state, offering incredible potential to better predict patients’ susceptibility to the disease and their response to treatments. Clinical registries offer opportunities to have detailed pheno- 

26 EVIDENCE-BASED MEDICINE typic and longitudinal outcomes for a very large cohort of patients. These longitudinal data are also useful for post-market surveillance studies that track long-term outcomes of therapies used in diverse patient populations and under different clinical conditions and settings. Registries could improve efficiency in the design and conduct of RCTs, and practical clinical trials might eventually be embedded within clinical registries. In situations where randomized treatment comparisons are not possible, observational comparative effectiveness studies using registries provide another source for evidence development. Finally, as a tool for quality assessment and improvement, registries can also ensure that evidence is fully and appropriately translated into clinical practice. Historically, registries have been useful in uncovering issues of overuse, underuse, and misuse of proven therapies in clinical practice. This information can provide specific guidance on what is not working and how to fix it and, ultimately, help practitioners deliver better care. Even with all of their advantages for evidence generation and applica- tion, participation is often voluntary and resources for clinical registries are shrinking in light of demands from government and insurers for alternative performance assessment data. Physicians also worry that clinical informa- tion might be used against them in a malpractice lawsuit. Clinicians need to make a strong case that registries are best run and most valuable when they remain in the hands of clinicians. Product Innovation—The Tailored Therapies Example Steven M. Paul, president of Lilly Research Laboratories, discussed how the dual challenges of rising costs and realizing the potential of bio- medical research have been reflected in the recent experience of biophar- maceutical companies. Stakeholders are demanding more information on the effectiveness of therapies, as well as more predictable and demonstrable health outcomes. Despite these heightened expectations, there is an efficacy and safety gap for today’s drugs. Paul noted that only about 50 percent of patients respond to any given therapy, and many of these do not respond in the same way. This increased focus on outcomes has put a burden on biopharmaceutical research development in terms of both longer develop- ment timelines and overall costs. To contend with this changing environment, biopharmaceutical com- panies are exploring the concept of tailored drug therapies: the right drug, for the right patient, at the right dose and the right time. For years, medi- cines have been tailored using biomarkers such as blood pressure, LDL (low-density lipoprotein) cholesterol levels, and hemoglobin A1c, but new tools for discovery and development are accelerating movement toward personalized medicine. Tools such as imaging modalities are increasingly 

SUMMARY 27 becoming more sophisticated in certain areas of drug therapies. However, the root causes and factors contributing to the progression of disease are often very complex and the routine development of more personalized drugs is a distant prospect. Tailored therapies represent a paradigm shift for drug development away from traditional approaches such as phased clinical trials to what is termed “value-based” drug development, which identifies the subpopula- tions of patients that would most benefit. This allows drug companies to stratify clinical development by identifying which agents should move to phase II or phase III clinical trials, and also assists in terms of marketing drugs. This shift may provide an important advantage given the current regulatory and market pressures on drug development. According to Paul, drug patents are shorter than they were previously—ranging from about 10 to 12 years today, as opposed to a previous life span of 17 to 20 years, making it difficult for drug companies to get new drugs to the market be- fore patents expire. With increased pricing pressures along with regulatory requirements, there is a paucity of new medicines from big pharmaceutical and biotech companies. The ability to stratify drug development using biomarkers and the tailoring approach offers potential for reducing costs and approval times. Many companies also have tied their drug development efforts to an ac- companying biomarker strategy to accelerate the identification of safety and efficacy issues. However, such stratification poses financial challenges for pharmaceutical companies. Based on the investment necessary for develop- ing a drug, if only a small subset of patients benefit, companies might have a difficult time generating returns. However, this is often offset by a higher value proposition, longer days/duration of therapy, and better reimburse- ment. The big challenge for drug companies in the coming decades will be to develop drugs that can be tailored to patients, while balancing the formidable risks and costs of drug development. Policy Changes to Improve the Value We Need from Health Care (Chapter 7) Many of the most important advances needed to bring more evidence into the healthcare system will face political hurdles. Proposals such as establishing universal EHRs, determining how to measure the quality of care services, and revamping the payment system will require political and legislative muscle and continuing oversight. Therefore, making a strong case to policy makers and the public for the importance of supporting the devel- opment and application of evidence in health care is crucial for facilitating positive change. Two presentations advanced some key considerations and opportunities for progress. 

28 EVIDENCE-BASED MEDICINE Regulatory and Healthcare Financing Reforms Donna E. Shalala, president of the University of Miami and former U.S. Secretary of Health and Human Services, raised several important points for healthcare leaders to consider as they incorporate EBM in healthcare re- forms. First, there is a suspicion among healthcare professionals that EBM is primarily about cost-cutting. Attention is needed to how these efforts are framed, and a primary lesson from the 1990s underscores the fact that the power of words must be recognized. Second, the political system powerfully affects the quality and delivery of health care. Often, large, single-payer sys- tems of care are viewed as the key to improve evidence-driven transforma- tion of care. However even within single-payer or large-purchaser systems such as the Department of Veterans Affairs and Medicare, and even where the evidence base is strong, implementing change has been challenging. Therefore, the expectation that Medicare or Medicaid can lead efforts to reform healthcare delivery through introducing more evidence should be questioned. Despite a strong evidence base for many proposals, each step toward implementing change affects a constituency and the members of Congress supported by those constituencies. In looking at successful public health efforts to apply evidence to healthcare policy, such as the efforts of the U.S. Department of Health and Human Services to increase childhood immunizations in the 1990s, three conditions are important for change: a very specific goal, consensus on the problem, and a public-private partnership on how to provide the solution. However, unlike the politically uncontroversial initiative for childhood im- munization, clear answers on how to bring about change do not exist for many current problems in health care. Although all healthcare professionals want the best care for their patients, the structure of the healthcare system, payment methods, and expectations of patients for an evidence-based sys- tem can lead to many complications. Complicating these issues is the fragmentation of the healthcare system and research institutions. Centralization of these efforts has been suggested by some but healthcare leaders and the public need to consider whether a central research agency will be able to withstand political fallout for u ­ npopular evidence. The experience of the Agency for Healthcare Research and Quality’s predecessor, the Agency for Healthcare Policy and Research, suggests that establishing a base of support for research is important, but political protection is also necessary for those who conduct and translate potentially unpopular research findings. Healthcare leaders must also consider whether there is enough political will and pressure to make big changes in health care. To make grand social policy changes, there must be consensus on both the problem and the solu- tion, as there was when Social Security was enacted and in the development 

SUMMARY 29 of Medicare and Medicaid. Although there is interest by presidential can- didates to move toward an evidence-based system of health care, making this a reality will require healthcare professionals to make a strong case as shrewdly as possible. Defining and Introducing Value in Health Care Michael E. Porter, Bishop William Lawrence University Professor at the Harvard Business School, argued that for any meaningful reform to occur, the healthcare system must organize care and delivery around the value it provides patients, as measured by patient health outcomes per dollar spent. Much of the confusion around improving health care comes from different definitions of value for the various actors in the system (e.g., healthcare pro- viders, insurers, device manufacturers, patients) working at cross-purposes toward undefined goals. Organizing reform efforts around improving value will unite the interests of all parties and is fundamental to achieving many other goals, such as equity in health care and cost containment. Outcomes and costs should be measured separately and compared to determine value. This is especially important in health care because one of the most powerful ways to contain costs is to improve outcomes, such as through early detection or less invasive treatment. Treating cost contain- ment itself as a goal, instead of a patient value, has been a major stumbling block to improving the value of health care. Value in health care is largely unmeasured in the United States. Mea- suring value depends on properly measuring health outcomes and then comparing the total costs in achieving them. Currently, there is much confu- sion about measuring processes instead of outcomes. Measuring structural factors such as adherence to best practices, protocols, or guidelines is an imperfect indicator of health outcomes, although studies have shown that deviation from them can lead to poor outcomes. Guidelines are often incomplete and fail to adapt care to individual patient circumstances. In addition, process guidelines can slow innovation because they often need to be refined as new evidence becomes available. Similarly, focusing solely on health indicators as outcomes provides an incomplete picture. Indicators such as hemoglobin A1c levels used as a marker of blood sugar control in diabetes care should be closely correlated with acute episodes and complica- tions. They are predictors of results, not results themselves. Patient value is found in the integrated care of a patient’s medical condition rather than care from a single specialist or discrete intervention. Care for a medical condition such as breast cancer, diabetes, asthma, or congestive heart failure usually requires multiple specialists. Value for the patient is created by the combined efforts to care for that medical condi- tion and any other comorbidities patients may have. Therefore, the value of 

30 EVIDENCE-BASED MEDICINE health care is often revealed only over an extended period of time, through measuring long-term outcomes such as sustainability of recovery, the need for more interventions, or the occurrence of treatment-induced illnesses. Yet because health care is often fragmented by facility or specialty, outcomes and processes tend to be mismeasured. Providers tend to measure only their own interventions or services, even if this is not what determines overall value. Gathering long-term longitudinal data on outcomes is challenging, and even more so because of current organizational processes and practices. These same obstacles also hinder accurate measurements of costs. For every medical condition, multiple outcomes collectively define pa- tient value. Measuring the entire hierarchy of outcomes will be essential to improving value. Doing so will enable progress to be made at different rates and different levels of care. As survival rates get very high, for example, research could be focused on the speed of treatment or reducing discomfort. Advancements in medical science have led to the development of therapies to address a great majority of medical conditions in some way—including organ transplantation, new cancer therapies, and joint replacement. Today, there is the opportunity not only to develop new therapies but to improve and reduce the cost of existing therapies. To determine value, the full costs of care must be measured. Like outcomes, costs should be measured for medical conditions over the cycle of care. Providers and health plans need to work together to measure the cumulative costs by activity for each individual patient over time. Health plans will play an important role in measuring costs, and the focus of re- imbursement for health care should shift to bundled models for medical conditions. Health plans, providers, employers, and government policy can all contribute to making measurement of value in health care a reality. If all of the parties in health care could truly measure and embrace value as the central goal, opportunities for improvements in healthcare delivery would be almost limitless. REFERENCES Druss, B. G., S. C. Marcus, M. Olfson, T. Tanielian, L. Elinson, and H. A. Pincus. 2001. Comparing the national economic burden of five chronic conditions. Health Affairs 20(6):233-241. Fisher, E. S., D. E. Wennberg, T. A. Stukel, D. J. Gottlieb, F. L. Lucas, and E. L. Pinder. 2003. The implications of regional variations in Medicare spending. Part 1: The content, qual- ity, and accessibility of care. Annals of Internal Medicine 138(4):273-287. Institute of Medicine’s Roundtable on Evidence-Based Medicine. 2006. Charter. http://www. iom.edu/CMS/28312/RT-EBM/33544.aspx (accessed May 15, 2008). Sirovich, B. E., D. J. Gottlieb, H. G. Welch, and E. S. Fisher. 2006. Regional variations in health care intensity and physician perceptions of quality of care. Annals Internal Medi- cine 144(9):641-649. 

SUMMARY 31 Skinner, J. S., D. O. Staiger, and E. S. Fisher. 2006. Is technological change in medicine always worth it? The case of acute myocardial infarction. Health Affairs (Millwood) 25(2): w34-w47. Wennberg, J. E., E. S. Fisher, and J. S. Skinner. 2002. Geography and the debate over Medicare reform. Health Affairs (Millwood) Supp Web Exclusives:W96-W114.