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.
4 Taking the First Steps The committee recognizes the enormity of the change that is required to achieve substantial improvements in the six major aims set forth in Chapter 2â that health care be safe, effective, patient-centered, timely, efficient, and equi- table. The ten simple rules described in Chapter 3, designed to help guide the actions of clinicians, patients, and others in ways that will lead to such improve- ments, also characterize a fundamental cultural transformation taking place today in the health care sector. This nascent cultural transformation embodies a more explicit commitment to evidence-based practice and patient-centered care, and reflects recognition of the importance of well-designed systems of care. To achieve the six aims, there must also be a very strong commitment to redesign. The current system will not work. New information technology should be embraced and new systems of care developed. Methods of payment must be modified to encourage and reward quality care. This chapter provides an ap- proach for achieving better alignment of the evidence base, the organization of care, information, payment methods, and quality measurement around patientsâ health care needs. Common chronic conditions should serve as a starting point for the restruc- turing of health care delivery because, as noted in Chapter 1, chronic conditions are now the leading cause of illness, disability, and death in the United States, affecting almost half of the population and accounting for the majority of health care resources used (Hoffman et al., 1996). Chronic conditions affect people of all ages. Although older people are more likely to have a chronic condition, people over age 65 account for only one-quarter of those living in the community with such a condition (The Robert Wood Johnson Foundation, 1996). 89
90 CROSSING THE QUALITY CHASM Todayâs health care system is not well designed to meet the needs of patients with common chronic conditions. Some patients receive good-quality care that is well coordinated, with good communication among the various clinicians in- volved. For too many others, however, care for even a single condition is frag- mented across many clinicians and settings with little coordination or communi- cation, and some needs remain undetected and/or unmet. Given the magnitude of the change that is required, the committee believes that leadership at the national level is required to initiate the process of change by taking two important steps. First, a short list of priority conditions should be promulgated by the Department of Health and Human Services, and all health care stakeholders should then focus attention on making substantial progress toward the establishment of state-of-the-art processes for these conditions in the next 5 years. Second, resources should be provided to seed innovative projects at the delivery system level, especially those projects that have a high likelihood of producing knowledge and tools that can be applied on a widespread basis through- out the health care sector. The committee believes such a focus on specific common clinical conditions is the best way to achieve the substantial improvement in quality that is required. Such conditions represent the needs around which patients have the greatest interaction with the system and make the most choices about cost and quality (and other issues). This is also the level at which care processes can be designed and refined. Thus, priority conditions offer the best opportunity for undertaking the development of the evidence base for practice; the reorganization of care; the development of supportive information technologies; and the design and refine- ment of quality measures and reporting processes, as well as payment incentives and rewards. Recommendation 5: The Agency for Healthcare Research and Qual- ity should identify not fewer than 15 priority conditions, taking into account frequency of occurrence, health burden, and resource use. In collaboration with the National Quality Forum, the agency should convene stakeholders, including purchasers, consumers, health care organizations, professional groups, and others, to develop strate- gies, goals, and action plans for achieving substantial improvements in quality in the next 5 years for each of the priority conditions. Identifying priority conditions represents a starting point to support the orga- nization of care, bring the evidence base into practice, develop information tech- nology and infrastructure to support care, and develop mechanisms to measure and pay for quality care. Instead of defining care by where it is delivered or who delivers it, the system should be designed to optimize care for patientsâ needs across the entire continuum of care in the most effective and efficient way possible.
TAKING THE FIRST STEPS 91 In identifying priority conditions, the Agency for Healthcare Research and Quality (AHRQ) should consider using the list of conditions identified through the Medical Expenditure Panel Survey (MEPS), a nationally representative house- hold survey of health care use, expenditures, sources of payment, and insurance coverage conducted by AHRQ and the National Center for Health Statistics that includes information on health conditions (Medical Expenditure Panel Survey, 2000). MEPS identifies 15 âpriority conditionsâ based on their prevalence, ex- pense, or policy relevance: cancer, diabetes, emphysema, high cholesterol, HIV/ AIDS, hypertension, ischemic heart disease, stroke, arthritis, asthma, gall bladder disease, stomach ulcers, back problems, Alzheimerâs disease and other dementias, and depression and anxiety disorders. The action plan for each priority condition should include strategies for designing and maintaining evidence-based processes; promoting primary, sec- ondary and tertiary prevention; building the necessary information technology infrastructure to support delivery and coordination of care, system design and ongoing management, payment, and accountability; and aligning the incentives inherent in payment and accountability processes with the goals of quality im- provement. AHRQ should also ensure that each action plan is supported by key stakeholders. In identifying and convening stakeholders, AHRQ should work with the National Quality Forum, a publicâprivate partnership charged with de- velopment of a comprehensive quality measurement and public reporting strat- egy. Input should also be obtained from organizations that have made significant efforts to improve quality, such as the Institute for Healthcare Improvement, the Quality Center at the Bureau of Primary Health Care in the Health Resources Services Administration, the Veterans Health Administration, local delivery sys- tems, and others. Since the identification of priority conditions is a starting point toward achieving the six aims, AHRQ should conduct this work expeditiously. The number of priority conditions identified should grow over time to eventually cover the majority (e.g., 80 percent) of the care provided to patients. Recommendation 6: Congress should establish a Health Care Qual- ity Innovation Fund to support projects targeted at (1) achieving the six aims of safety, effectiveness, patient-centeredness, timeliness, ef- ficiency, and equity; and/or (2) producing substantial improvements in quality for the priority conditions. The fundâs resources should be invested in projects that will produce a public-domain portfolio of programs, tools, and technologies of widespread applicability. Policies, incentives, tools, and technologies will be needed to support the changes required to achieve the six aims and redesign the health care system in accordance with the new rules set forth in Chapter 3. The formation of an innovation fund is one mechanism that can be used to seed projects aimed at redesigning care and developing programs to support the other recommendations
92 CROSSING THE QUALITY CHASM presented in this report. For example, purchasers and delivery systems should work together to develop innovative programs that integrate the new rules for patientâclinician relationships (Chapter 3) and to redesign care processes for the priority conditions, making best use of information technology and engineering design concepts (Chapters 5 and 7). There must be a strong commitment to evaluating the impact and cost-effectiveness of innovative programs, and to the rapid diffusion of programs proven successful throughout the field. Although an innovation fund should support projects related to the priority conditions, it could also support other redesign projects, especially those relating to greater use of information technology. The committee is not recommending a specific dollar amount for the pro- posed fund, but believes a sizable commitment, on the order of $1 billion over 3 to 5 years, is necessary given the magnitude of change needed. Just as a vigorous public commitment and an expenditure of approximately $1 billion over a 4-year period has led to the mapping of the human genome, a similar commitment is needed to retool the health care delivery system, or society will never reap the full value of the trillion dollars spent annually on health care services. THE VALUE OF ORGANIZING AROUND PRIORITY CONDITIONS Identifying priority conditions can benefit all those involved in using, deliv- ering, planning, or paying for health care. Patients and their families today must try to navigate a fragmented, complex health care system with insufficient infor- mation and an unclear understanding of how to find the best-quality care for their specific needs. Similarly, health professionals face pressures to improve quality and measurable outcomes without having systems in place that can help them easily identify the best practices for a given case or means of arranging for follow-up on a patientâs needs across the entire continuum of care. Purchasers appear to focus more on cost than on quality, but have little outcome or quality information available to them. They may be willing to buy on the basis of quality, but see a health system that produces care inefficiently and is character- ized by errors (or âdefectsâ) that would not be tolerated in their own industries. Regulators struggle with how best to provide oversight in a rapidly changing environment. Many people have an interest in ensuring quality care, but do not have a framework or tools for doing so. A focus on priority conditions can align the efforts of diverse participants in the health care system, offer a meaningful level of organization to patients, and provide a starting point on which health care professionals and organizations can focus their efforts. Improved alignment between patient needs and the ways care is organized, delivered, measured, and paid for is important because fragmenta- tion and misalignment in the current system inhibit systematic quality improve- ment. For example, patients need information, support, and reassurance to man-
TAKING THE FIRST STEPS 93 age a chronic condition independently on a day-to-day basis and to recognize when help is needed from a clinician. However, the current health care system is organized around acute care needs. It does not facilitate the flow of information over time; offers little recognition or reward for coordinating care; and pays mainly for face-to-face (office) visits, not for information and/or reassurance that may be needed at other times. The aims described in Chapter 2 cannot be achieved without better alignment of organization, payment, and measurement with patientsâ needs. A focus on specific conditions may also be more meaningful to patients. Prior research suggests that when people have a health care need, they are inter- ested in comprehensive sets of services provided to people with similar condi- tions (Cleary and Edgman-Levitan, 1997; Fennell and Flood, 1998). Instead of describing their health experiences with one service or one provider, patients describe an episode of care (Cleary and Edgman-Levitan, 1997). For example, a patient who has had a heart attack will describe all components of his or her care, including the emergency room, medical service, surgical service, specialty physi- cian office visit, generalist physician office visit, and rehabilitation care. Al- though the performance of each individual unit is important, simply aggregating the individual units is not sufficient for understanding the quality of health care provided when care involves many people and facilities, as is often the case today (Fennell and Flood, 1998). Additionally, people want information about people âlike themselves.â That is, they are seeking information that will tell them how well a health plan or clinical group cares for others with similar conditions (Cleary and Edgman-Levitan, 1997). An approach that facilitates the measure- ment and release of information around specific conditions can provide patients with such information. Also, with appropriately designed incentives and support systems, such an approach can provide an organizing framework for care so that providers have the flexibility to match services along the entire continuum of care to the needs of a specific patient and support continuous care relationships. Defining care processes around specific conditions can also establish a suit- able level of focus for significant quality improvement in health care. To achieve such improvement, it will be necessary to develop information about the pro- cesses and outcomes of care for specific population groups (Friedman, 1995). Meaningful groupings are required because the quality of care for one set of conditions cannot be generalized to patients with different conditions (Brook et al., 1996). The tasks of examining processes of care, linking those processes to outcomes for populations, comparing the effectiveness of alternative approaches, bringing the evidence base into practice, forming the teams that deliver complex care, and accurately adjusting for differences among patients to permit valid comparisons are difficult to accomplish simply by looking at patients in general. Rather, these tasks can best be accomplished for people with comparable needs. Once care has been defined around peopleâs needs, meeting those needs becomes the ultimate target for these basic steps of quality improvement.
94 CROSSING THE QUALITY CHASM It is expected that most priority conditions will be strongly related to chronic conditions. As discussed in Chapter 1, care for people with chronic conditions represents an increasing portion of health care resources in the United States. Four chronic conditions (cardiovascular disease, cancer, chronic obstructive pul- monary disease, and diabetes) account for almost three-quarters of all deaths in the United States (Centers for Disease Control and Prevention, 1999). Compared with people with acute conditions, the annual medical costs per person were more than double for people with one chronic condition and almost six times higher for people with two or more chronic conditions1 (The Robert Wood Johnson Founda- tion, 1996). A study in one health maintenance organization found that 38 percent of enrollees had at least one chronic condition, and their costs averaged twice those of people with no chronic condition (Fishman et al., 1997). A study at another health maintenance organization found that 78 percent of direct medi- cal costs were attributable to just 25 acute and chronic conditions and that three cardiovascular conditions (ischemic heart disease, hypertension, and congestive heart failure) accounted for 17 percent of those costs (Ray et al., 2000). It has been estimated that the top 1 percent of spenders account for 30 percent of health spending, whereas the bottom 50 percent account for only 3 percent of spending (Berk and Monheit, 1992). Given this concentration, the majority of health services utilized can potentially be associated with a definable list of conditions. Yet the health care system is not well designed to meet the needs of the chronically ill. The current delivery system responds primarily to acute and urgent health problems, emphasizing diagnosis, ruling out serious conditions, and relieving symptoms (Wagner et al., 1996b). Those with chronic conditions are better served by a systematic approach that emphasizes self-management, care planning with a multidisciplinary team, and ongoing assessment and follow- up (Wagner et al., 1996a). As noted in Chapter 1, successful chronic disease management programs: â¢ Use a protocol or plan that provides an explicit statement of what needs to be done for patients, at what intervals, and by whom, and that considers the needs of all patients with specific clinical features and how their needs can be met. The care plan is a tool that links the multiple visits and contacts that characterize care for chronic illness. â¢ Redesign practice to incorporate regular patient contact, collection of critical data on health and disease status, and strategies to meet the educational and psychosocial needs of patients who may need to make lifestyle and other changes to manage their disease. Regular follow-up is a hallmark of the design of successful programs. 1Direct medical costs included hospital care, physician services, dental services, other professional services, home health care, prescriptions, medical equipment, emergency services, and nursing home care.
TAKING THE FIRST STEPS 95 â¢ Include a strong focus on patient information and self-management so patients and their families acquire skills in self-management and can make needed lifestyle changes. Structured self-management and behavioral change programs improve patient outcomes. â¢ Ensure the availability of specialized expertise to the primary care prac- tices that frequently have responsibility for managing patients with chronic ill- ness. The traditional mechanism for accessing expertise is through a consultation or referral, which runs the risk of fragmenting care. Alternative approaches for making expertise available include teams with specialized knowledge (e.g., a diabetologist and nurse specialist working with general practitioners who care for diabetic patients); collaborative care arrangements (e.g., where specialists and generalists manage patients together); and, eventually, well-designed computer decision support systems. â¢ Rely on having good information about patients, their care, and outcomes in order to improve outcomes. Registries inform providers which patients have certain conditions to permit proactive clinical management. Use of reminder systems supports patient participation in explicit plans of care. Wagner (2000) also notes the implications of such a model for how teams of clinicians work together. Successful teams should bring in new disciplines in medicine, but also nonmedical personnel. Establishment of care plans, good patient involvement, and a strong information base permit members of a care team to work together beyond organizational and practice boundaries. Although good coordination and communication are essential for all care, they are especially important for chronic care. Patients may move through many settings of care, from home, to clinician office, to hospital, to nursing home, and back. Patients and their families often provide a sizable proportion of routine care, including the administration of medications, performance of some diagnos- tic tests, and compliance with physical therapy and nutritional plans. While the current health care system is built around visits, people with chronic illness need flexible models that provide more time and alternative contacts with the system. Although there are many basic, simple techniques that can be employed today (e.g., telephone follow-up rather than e-mail, or reminder systems that use flags on a chart rather than computerized reminder systems), the committee believes these simple techniques have been available for many years and have not been sufficient to achieve broad-based quality improvements. While any type of progress is welcome, at some point the health care system will need to embrace more automated methods and greater use of information technology to make significant progress. Unfortunately, there are very few well-tested integrated models of chronic care management (Wagner et al., 1996a). While research may focus on specific components of care, it is more difficult to understand the interrelationship be- tween the components and the influence of the organizational environment. For
96 CROSSING THE QUALITY CHASM example, the Diabetes Control and Complications Trial demonstrated how to improve clinical management of patients with insulin-dependent diabetes. But the trial required special skills of both patients and clinicians, services that many insurance policies do not cover, and delivery of care by patient-centered teams (Lasker, 1993). APPLICATIONS OF PRIORITY CONDITIONS Figure 4-1 illustrates the multiple ways in which the priority conditions, once identified, can be applied. First, they can be used to synthesize the evidence base and delineate practice guidelines. This application is closely linked to and should guide the organization of care and coordination of care around patient needs. The priority conditions can also be applied in developing information systems, reduc- ing suboptimization in payment for services, and simplifying the measurement Synthesize the evidence and delineate practice guidelines Organize and Simplify quality coordinate care measurement, around patient Identify priority evaluation of needs conditions performance, (consistent with and feedback the evidence base) Provide a common base Reduce for the suboptimization development of in payment iinformation technology FIGURE 4-1 Applications of priority conditions.
TAKING THE FIRST STEPS 97 and evaluation of care. Each of these applications is described below. Although some applications may occur more quickly than others, Figure 4-1 is not meant to imply a linear order to their accomplishment. Rather, the priority conditions can be used for any of these applications as soon as they have been identified. Synthesize the Evidence Base and Delineate Practice Guidelines The identification of priority conditions provides a framework for synthesiz- ing the evidence, developing practice guidelines, and delineating best practices for clinical care. There is a significant lag between the discovery of better forms of treatment and their incorporation into actual care. The identification of prior- ity conditions supports a well-thought-out organization of information to im- prove its accessibility and utility for both patients and health professionals (see Chapter 6). Identification of these conditions can guide the prioritization of issues for analysis and synthesis of evidence, delineation of practice guidelines, and development and application of automated decision support tools. It can also provide direction for stronger dissemination efforts aimed at communicating this information to clinicians and consumers. Even in clinical areas characterized by strong evidence and general consensus on practice, variability in practice sug- gests that current dissemination efforts could be improved. The Internet offers the opportunity to achieve such improvement by reaching sizable proportions of both consumers and clinicians in a timely manner. One of the strongest examples of synthesizing the evidence base and apply- ing it to clinical care is offered by the Veterans Health Administration (VHA). The VHAâs Quality Enhancement Research Initiative (QUERI) is a quality im- provement program that focuses on eight priority conditions: chronic heart fail- ure, diabetes, HIV/AIDS, ischemic heart disease, mental health (depression and schizophrenia), spinal cord injury, stroke, and substance abuse (Demakis et al., 2000). These conditions were selected on the basis of the number of veterans affected, the burden of illness, and known health risks among the veteran popula- tion. Specific conditions were selected as the focus in the belief that quality improvement is most likely to occur when viewed in the context of the overall care of a patient and population, rather than the individual components of care (Demakis and McQueen, 2000). The process implemented at the VHA involves defining best practices by reviewing currently available information and literature. For some conditions, such as diabetes, a great deal of information is available; for others, the informa- tion must be developed by a planning team. Once the evidence has been re- viewed and best practice defined, the latter is compared with current practice to identify gaps in performance. Policies, procedures, and programs are then devel- oped to organize care around the best practice, which also guides the evaluation of impact and feedback to enable learning from experience and continuously improving care. Thus, best practice affects how care is delivered, but evaluation
98 CROSSING THE QUALITY CHASM of its impact also informs the continued development of best practice. The process used by the VHA also emphasizes the broad dissemination of informa- tion about best practice throughout their system, from large academic centers to smaller, community-based centers. The VHAâs approach is consistent with the concept of focusing on priority conditions in that it provides a framework for organizing and continually updat- ing the evidence base, bringing it to the direct delivery of care, and evaluating its effect on improving care for patients. Synthesis and application of the evidence base, therefore, forces the reexamination of how care is organized to affect qual- ity. By examining where current practice departs from the evidence base and best practice, suggestions for improving care may emerge that can direct changes in provider actions, patient responsibilities, or organizational approaches. It would be difficult to use this multifaceted, comprehensive care approach except at the level of a specific condition. Organize and Coordinate Care Around Patient Needs The primary purpose of identifying priority conditions is to facilitate the organization of care around the patientâs perspective and needs rather than, as in the current system, around types of professionals and organizations. For ex- ample, the current system may require patients to travel to multiple locations to receive care (usually Monday through Friday, between 8:00 a.m. and 5:00 p.m.) instead of using modern technologies to facilitate access even for patients with mobility problems or those living in rural areas. Most hospitals are organized around physician specialties (such as thoracic surgery or internal medicine), not around common clinical needs of patients, which may cross departmental bound- aries. (An example is diabetes care, which may require general medicine, endo- crinology, ophthalmology, and vascular services.) Organizing care around prior- ity conditions emphasizes meeting the needs of patients with those conditions, regardless of who provides their care or where. Attention must be paid to how care is coordinated across settings and provider types. A surgical procedure may be performed perfectly, but if there is inadequate postoperative care, follow-up care, home care, or other supports, the patient may encounter complications that compromise the quality of the episode of care. It is also important to recognize that patients may have to manage multiple conditions simultaneously, because they either have more than one chronic condition or have one chronic condition and an unrelated acute event. Indeed, there is evidence that patients actively receiving care for one chronic condition may not receive treatment for other, unrelated conditions (Redelmeier et al., 1998). Thus, one of the challenges of designing care around specific conditions is to avoid defining patients solely by their disease or condition. There are several mechanisms for coordinating care across priority condi- tions. First, coordination could be performed by a health professional acting as a
TAKING THE FIRST STEPS 99 liaison across patientsâ multiple needs, ensuring the exchange of information and any necessary follow-up2 (Bodenheimer et al., 1999). This individual could be a physician, nurse, case manager, or other type of professional working in the care delivery system. Second, some consumers might choose to actively coordinate their own care. This is a growing possibility as consumers and patients have access to more health information and are able to make use of the evidence base and practice guidelines for their specific conditions. Third, as information tech- nology becomes more sophisticated, computer algorithms can be used to coordi- nate many activities, for example, sending reminders of needed follow-up or identifying missing information, such as test results. Finally, coordination could occur through a combination of all these methods, involving a health profes- sional, the patient, and technological support. Various approaches in practice today offer insight that can be applied in organizing care around specific conditions or types of needs, including disease management programs and centers of excellence. Each is briefly reviewed below. Disease Management Programs Multiple definitions of disease management programs have been put forth (Blumenthal and Buntin, 1998; Ellrodt et al., 1997; Homer, 1997). In general, they describe a systematic and comprehensive approach to improving the man- agement of a condition. This approach involves improving coordination of care and controlling costs through the integration of components across the entire delivery system and the application of appropriate tools (e.g., guidelines, proto- cols, information systems) specifically designed for the population in question. Disease management programs share some of the features envisioned for organization around priority conditions as described above, but also differ in important ways. The two are similar in that disease management represents a systematic approach to designing care, uses multidisciplinary teams to deliver care, and potentially includes services across the entire continuum of care. How- ever, disease management programs differ in that they are frequently perceived primarily as a method for controlling costs (Bodenheimer, 2000; Homer, 1997; Ketner, 1999). They are often applied only to the most severely ill patients (generally categorized by historical costs) instead of to the entire population with a specific condition (Hunter, 2000; Ketner, 1999). This means not all patients with a condition are able to benefit from such programs. Although the programs are developed to improve care for patients, patients may not have much say in whether they receive care through such programs. Finally, there are no clear 2It should be noted that this discussion of coordination of care is not meant to imply support for or opposition to a gate-keeping function used by many groups to ensure appropriate access. Coordina- tion can be provided with or without a gate-keeping role.
100 CROSSING THE QUALITY CHASM definitions of what is included or excluded from such programs, making it diffi- cult to compare their effectiveness in treating similar populations. One of the main concerns associated with disease management programs is the potential for fragmenting care, especially if the patientâs primary care physi- cian is not involved in the program. Again, this concern highlights the impor- tance of coordination across conditions and the need to design such coordination systematically into care processes. Primary care models offer one approach to coordination, in which the primary care practitioner is both provider and coordi- nator of care. Alternatively, programs can be designed specifically to include or link an individualâs primary care practitioners in care planning and assessment. The committee does not recommend one approach over another, but emphasizes the importance of designing coordination into care to avoid fragmentation. Centers of Excellence There is a growing body of evidence on the relationship between volume of service and outcomes. The IOM conducted a workshop in May 2000 to explore the volumeâoutcome relationship (Hewitt, 2000). A systematic review of the literature conducted for the workshop encompassed 88 studies concerning eight conditions and procedures.3 This review led to the conclusion that for a wide variety of procedures, higher volume (by either the hospital or physician) is associated with better health outcomes. Statistically significant associations be- tween higher volume and better outcomes were found in 79 percent of studies of hospital volume and 77 percent of studies of physician volume. None of the studies showed a negative effect of volume. The Health Care Financing Administration has pursued Centers of Excel- lence as a way of operationalizing the volumeâoutcome relationship. Centers of Excellence are hospitals and physician groups that meet high quality standards, for which they are paid a single bundled fee for all services related to specific, complex procedures (Health Care Financing Administration, 1999). Evaluation of the experience with cardiac surgery has indicated that the approach can offer cost savings without compromising quality, measured as mortality rates (Crom- well et al., 1997). Although developed as a payment policy, the approach bears some similarity to organizing around priority conditions in that it gives health care organizations and professionals the flexibility to organize care appropriately for a specific population group. The approach is dissimilar in that it is focused primarily on selected complex procedures with a strong emphasis on costs, rather than being solely a quality-driven strategy (Cromwell et al., 1997; Health Care Financing Administration, 1999). 3Coronary artery bypass graft surgery, pediatric cardiac surgery, carotid endarterectomy, abdomi- nal aortic aneurysm repair, cancer surgery, percutaneous transluminal coronary angioplasty, acute myocardial infarction, and acquired immunodeficiency syndrome (AIDS).
TAKING THE FIRST STEPS 101 Provide a Common Base for the Development of Information Technology Priority conditions can provide a framework for the development of an infor- mation infrastructure that is aligned around the clinical conditions frequently faced by patients. The absence of well-designed care processes is currently an impediment to the development and application of effective information technol- ogy systems. To be most useful, information technology must be designed to support the work of the care team. Consciously and skillfully designed care processes for priority conditions are an important step in establishing a founda- tion from which to design supportive information technology applications. Common information technology systems are needed to effectively measure outcomes and processes of care and to provide benchmarks for continuous im- provement. Currently, each provider group may implement its own information system, but incompatibilities inhibit communication among the many people caring for an individual patient. Priority conditions can provide a focus for the development of standards and terminology for use in managing and using infor- mation technology to improve care for patients. Best practices can help define standard information needs and guide the development of information technolo- gies that can be used to implement best practices (e.g., decision support systems). A significantly enhanced information infrastructure is critical to achieving the aims set forth in Chapter 2 and the other potential applications of the priority conditions. Synthesizing the evidence base, linking it to clinical practice, and making it accessible to a variety of potential users will require good information systems and, most likely, greater use of the Internet. Greater use of the Internet and telemedicine should in turn facilitate access to clinical expertise and support care for patients in their own communities, especially in rural areas. Better flows of information are also necessary to improve the ways care is organized and coordinated, especially across settings and over time. Using payment methods to reward quality will require stronger information systems to track costs and link them to processes of care. Finally, measurement and evaluation cannot be ad- vanced without better technology for data collection and management (see Chap- ter 7 for further discussion on using information technology). Reduce Suboptimization in Payment A major barrier to quality improvement is the lack of reward that character- izes the most common payment methods used today (see Chapter 8). The current payment system often reinforces fragmentation by paying separately according to the setting of care and provider type, and by not giving providers the flexibility needed to customize care for individual patients. Furthermore, common payment methods can inhibit quality improvement to the extent that organizations that improve certain aspects of quality (e.g., by reducing readmission rates or office visits) can experience a reduction in their revenues, which serves as a disincen- tive for continuous improvement.
102 CROSSING THE QUALITY CHASM Priority conditions offer a framework for linking payment with patient needs and for designing incentives to reward quality. Alternative payment methods (e.g., fee for service or capitation) could be adapted to facilitate the delivery of care around priority conditions, consistent with the evidence base. Priority con- ditions could also provide a framework for purchasers to use in assessing the value of their purchases. See Chapter 8 for a detailed discussion on the relation- ship between payment and quality improvement. Simplify Quality Measurement, Evaluation of Performance, and Feedback Priority conditions improve the feasibility of quality measurement by offer- ing a framework for the development of standards to guide the necessary data collection. At present, quality measurement for external accountability tends to focus on institutions or discrete services; there is little comparative information available for patients seeking specific care or physicians referring care. For example, a patient can obtain information on mammogram rates, but will find little information on methods of treatment or outcomes for breast care programs. Priority conditions can offer a framework for the development of core measures that address both processes and outcomes of care. Part of the difficulty involved in obtaining such information is due to meth- odological barriers in measurement. The services of an individual physician are usually too small a unit for measurement of many aspects of clinical care pro- cesses and outcomes (Hofer et al., 1999). Even the typically sized medical group may be too small to provide reliable information on outcomes. Health plans may aggregate information, but clinicians are often affiliated with multiple plans. The delineation of priority conditions, the organization of services around these con- ditions, and the development of core sets of measures may help overcome some of these barriers to measurement. Public- and private-sector oversight organizations are already organizing some of their activities around particular conditions. For example, the Founda- tion for Accountability has developed population- or condition-specific quality measurement guides related to adult asthma, alcohol misuse, breast cancer, dia- betes, health status under age 65, and major depressive disorders (Foundation for Accountability, 1999a) and continues to work on quality measurement and con- sumer reporting approaches in the areas of child and adolescent health, coronary artery disease, end of life, and HIV/AIDS (Foundation for Accountability, 1999b). The Foundationâs model organizes comparative information about quality perfor- mance into five categories based on how consumers think about their care: the basics, staying healthy, getting better, living with illness, and changing needs. The Joint Commission on Accreditation of Healthcare Organizations (2000) has identified five specific areas for the development of indicators to assess hospital care: acute myocardial infarction, heart failure, pneumonia, surgical procedures and complications, and pregnancy and related conditions. Accredita-
TAKING THE FIRST STEPS 103 tion by the National Committee for Quality Assurance includes measures related to how well a health plan cares for people when they have a chronic illness in such areas as cardiovascular disease, cancer, asthma, pneumonia and influenza, and diabetes (National Committee for Quality Assurance, 1999). Peer Review Organizations focus their national quality improvement efforts on six clinical priority areas: acute myocardial infarction, breast cancer, diabetes, heart failure, pneumonia, and stroke (Health Care Financing Administration, 2000). Finally, the National Quality Forum is developing a comprehensive quality measurement and reporting strategy that will address priorities for quality measurement that are consistent with the national aims for quality improvement in health care set forth in this report (National Quality Forum for Health Care Quality Measurement and Reporting, 2000). CRITERIA FOR IDENTIFYING PRIORITY CONDITIONS Various criteria can be used to identify the priority conditions. Two IOM committees have suggested criteria for setting priorities among conditions: one committee focused on how to set priorities for guideline development, the other on how to set priorities for technology assessment. The common criteria from both processes included prevalence, burden of illness, cost, variability in prac- tice, and the potential to improve outcomes or reduce costs (Institute of Medicine, 1992, 1995). As noted earlier, this committee suggests starting with the priority conditions identified in the MEPS. Some are long-term life-threatening conditions, such as cancer, diabetes, emphysema, high cholesterol, HIV/AIDS, hypertension, is- chemic heart disease, and stroke. Others, such as arthritis, asthma, gall bladder disease, stomach ulcers, and back problems of any kind, are categorized as chronic manageable conditions. The list also includes Alzheimerâs disease, depression, and anxiety disorders. MEPS obtains a larger sample size for seven of the conditionsâhypertension, ischemic heart disease, asthma, diabetes, stroke, em- physema, and arthritisâto make population estimates. Although other sources are also available, the advantage of starting with the MEPS listing is its represen- tative population sample, as opposed to claims data that rely on services having been used. PROVIDING THE RESOURCES NEEDED TO INITIATE CHANGE The health care system in the United States needs significant redesign. Given the magnitude of the change required, the innovation fund recommended earlier is needed to seed projects that can help apply the concepts described in this report. A Health Care Quality Innovation Fund should finance the demonstration
104 CROSSING THE QUALITY CHASM and evaluation of programs designed to implement the types of changes recom- mended in this report. Although a specific agenda should be established, the areas of interest for funding should address one or more of the issues covered in this report: techniques for implementing the rules for redesigning care set forth in Chapter 3, applying evidence to health care delivery, using information tech- nology, aligning payment policies for quality, and preparing the workforce. Emphasis should be placed on funding projects that will integrate the result- ing innovations into processes of care. The goal is not simply to fund âgood ideas,â but rather to fund the implementation of good ideas in real-life settings, focusing on innovations that have a good likelihood of broad applicability to other sites. Evaluations will need to be carefully structured to be able to assess the programmatic features that contributed to a projectâs successful implementa- tion, including how technical, cultural, and economic factors were addressed. Barriers encountered should also be identified, as well as how they were over- come or whether they presented too great an obstacle. Funding may be provided to individual local health care organizations, pri- vate partnerships (e.g., those between purchasers and delivery systems), or pub- licâprivate partnerships (e.g., those among delivery systems, local public health agencies, and consumer groups). Possible projects may relate to the direct deliv- ery of health care for a specific population or to the development of an infrastruc- ture that facilitates needed change (e.g., approaches for sharing data). A portion of the Health Care Quality Innovation Fund should be set aside to provide resources to answer critical research questions. Implementation and evaluation of innovative projects are important, but some areas may require additional understanding to guide demonstrations and their implementation. For example, the change to using relative value units as a payment approach for physicians was legislated in 1989 for implementation in 1992, following about 10 years of research, testing, and evaluation (Hsaio and Stason, 1979). Possible areas requiring additional organizational research include understanding how financial and other types of incentives relate to organizational setting, how phy- sician and nonphysician members of the care team can optimally interact and complement each other, what components and interactions of systems of care are most important for improving quality, and how to organize care for people with chronic conditions. The committee views public support as important for catalyzing the needed changes for several reasons. First, a commitment of funds over several years can ensure a sustained and stable funding source. Projects funded by health care organizations through operating revenues represent a valuable contribution, but the stability and level of funding can be unpredictable, and perhaps unsustain- able, in a rapidly changing marketplace. Second, public support can provide partial funding for the up-front costs that health care organizations face in undertaking the changes recommended in this report. Organizations should be prepared to support the continuing costs of any
TAKING THE FIRST STEPS 105 initiatives, but public funding for some portion of up-front costs can be a valuable resource for an organization that is interested, willing, and ready to redesign the delivery of health care to improve quality. Thus, the Health Care Quality Innova- tion Fund represents a publicâprivate approach to change, with the public sector providing seed money and the private sector using operating revenue to fund some of the up-front costs and any ongoing costs. Third, rather than trying to identify large programs aimed at reforming the entire system, smaller applied projects of varying size and focus should be per- mitted to flourish. Public funding for a mix of projects would permit midcourse corrections to be made as greater understanding is gained on what types of projects work or fail. Use of public seed money can also require an objective evaluation of demonstration projects and public access to the tools and tech- niques used. Rather than remaining in the private domain, the information be- comes a public good for use by all to learn how to improve health care quality. Research and demonstrations for organizational redesign in health care occur today in both the private and public sectors, although the level of effort appears to be modest given the size of the task ahead. In the private sector, one of the main sources of funds for organizational design research is foundations. The Robert Wood Johnson Foundation has sponsored the Changes in Health Care Financing and Organization Program since 1988. It has provided over $50 million to stimulate research into new strategies in the financing and organization of health care and the impact of changes in the delivery system on quality, access, and costs. The program funds research, policy analysis, demonstrations, and evalua- tions to provide timely information to policy makers, purchasers, providers, and researchers. Some health care organizations also devote a portion of their revenues to research and development projects. For example, Kaiser-Permanente has con- ducted work on risk adjustment methods for payment policy; Group Health Co- operative of Puget Sound has conducted extensive work on improving care for populations with chronic illness; and Intermountain Health Care in Salt Lake City has developed data systems for evaluating and improving process of care. Organizational design research is also being conducted at universities across the country. On the public side, the primary source of funds for organizational design research is AHRQ. The Center for Organization and Delivery Studies was cre- ated in 1996 to provide leadership for research on health care markets, delivery systems and organizations (Agency for Healthcare Research and Quality, 2000a). The Integrated Delivery System Research Network is a new model of field-based research that partners health services researchers with large health care systems to develop and disseminate evidence on data and measurement systems and orga- nizational best practices (Agency for Healthcare Research and Quality, 2000b). Approximately $4 million has been allocated over a 3-year period. Other AHRQ projects also contribute to innovation in health care delivery, such as work in
106 CROSSING THE QUALITY CHASM medical informatics and patient safety. Other investments in organizational rede- sign in the public sector include those of the VHA, whose Quality Enhancement Research Initiative effort was described earlier in this chapter, and the Health Care Financing Administrationâs Office of Research and Development, funded at approximately $55 million in fiscal year 2001 (a decline of 11 percent from the prior year) (U.S. Department of Health and Human Services, 2000a). A portion of those funds supports projects that are consistent with redesigning health care delivery, including work related to competitive pricing, coordinated care for the chronically ill, and Centers of Excellence. A great many resources are devoted to technological innovation in health care in the areas of pharmaceuticals, medical devices, and biotechnology. Invest- ment in research and development was estimated at almost $36 billion in 1995, or about 3.5 percent of total health care spending (Neumann and Sandberg, 1998). Over $30 billion has been invested each year since 1993. Just over half of the investments were made in the private sector; the remainder of the spending was in the public sector, primarily the National Institutes of Health. Although such investments have produced great advances in technological innovation, however, they have produced little innovation in the organization and delivery of care. The irony is that the current health care system cannot ensure that the new technologies are delivered effectively, efficiently, and safely to the people that can most appropriately benefit from them. This dilemma will likely worsen in the future with expected advances in genome research, tissue reengineering, pharmacogenetics, and other areas. As noted earlier, the committee has not recommended a specific amount for a Health Care Quality Innovation Fund, but believes that an amount on the order of $1 billion over 3 to 5 years is needed. This amount represents one-quarter of 1 percent of the almost $400 billion the federal government currently spends on health care. By comparison, the top ten funded diseases at the National Institutes of Health were funded at $3.6 billion just for the year 1996 (Gross et al., 1999), while approximately $1 billion was devoted to the human genome project over the last 4 years (U.S. Department of Health and Human Services, 2000b). There is no central source for determining the extent of public and private investments in innovation in health care organization and delivery; however, the public- and private-sector initiatives identified above may total $100 million annually. This is far short of the amount needed for the magnitude of changes required and far less than the $36 billion expended annually for technology research. The following are examples of the types of projects that might be supported through the recommended fund.4 4These examples draw on some of the approaches to redesign being pursued by clinical leaders who were interviewed as part of an Institute of Medicine study aimed at identifying exemplary practices (Donaldson and Mohr, 2000).
TAKING THE FIRST STEPS 107 Example 1âUsing Information Technology to Improve the Timeliness of Ser- vices in a Hospital Emergency Department An emergency department sought to redesign the process of care by reduc- ing the time required to provide complete care to a patientâthe cycle time. Reducing cycle times improves both patient satisfaction and productivity. The staff identified key processes of care, such as x-ray cycle time, time between arrival at the emergency department and seeing a physician, and time needed to get a patient admitted to a bed. They introduced the concept of parallel process- ing and designed algorithms to permit the simultaneous performance of multi- ple tasks. Additionally, they developed their own tracking system (since noth- ing acceptable for the purpose was available from vendors) to track where patients are in the process of care, as well as the status of the system, in 15- minute increments (almost real time). A touch screen informs staff instantly of any problems in specific care processes so they can intervene quickly. These efforts have reduced total cycle time for less urgent patients from 92 to 47 minutes, time between arrival and seeing a physician from 32 to 18 minutes, time between decision to admit and getting the patient to the floor from 210 to 60 minutes, and x-ray cycle time from 92 to 32 minutes. If supported by an innovation fund, this project would share its algorithms and tracking system, along with pitfalls encountered during the redesign, so that the improved pro- cess would be disseminated to other emergency departments. Example 2âA Partnership to Improve Chronic Care A hospital, two small primary care practices, and an endocrinologist decid- ed to collaborate on the development of a state-of-the-art diabetic care program. They began by reviewing the practice guidelines and agreeing on the key ele- ments of preventive, acute, and chronic care. In an attempt to identify best practices, they visited two of the leading diabetic care programs. They then reached agreement on the key elements of a care process, and on the quality measures to be collected and used in assessing the care process and patient outcomes. They also worked to establish an interactive patient education pro- gram, which could be used online or viewed on video in the office. A diabetic care team was formed, consisting of two primary care physicians, a nurse prac- titioner, and an endocrinologist. Relying heavily on e-mail, they were able to establish procedures for ongoing communication and management of diabetic patients. Example 3âReorganizing Staff for Patient-Centered Primary Care A primary care center with multiple offices served 270,000 patients with 110 full-time equivalents (FTEs). The center had a 55-day waiting period for appointments, resulting in packed schedules for staff, a chaotic office environ- ment, an overused urgent care clinic, and unhappy patients. The center decided to develop an open-access system so that patients could get an appointment either the same day they called or the next day, with their own physician when- ever possible. Under the old system, patients calling for appointments would be sorted according to need: wellness care, acute illness, or chronic care. This approach was ineffective because patients might have two or three needs simul-
108 CROSSING THE QUALITY CHASM taneously; thus, they would make multiple appointments to meet different needs or go to the urgent care clinic, which further stressed the system. In the redesigned system, the staff was organized into 15 teams, each with 7 to 9 physicians and nurse practitioners; the urgent care clinic was closed, and that staff was reassigned to the regular offices. Patients were no longer catego- rized according to the nurseâs or receptionistâs assessment of their need, but were seen based on their own perception of need. Whereas the number of visits was expected to increase because of âpatientsâ insatiable demands,â the total number of patient visits declined by 7 percent, and the no-show rate went from 20 percent to being âtoo small to show up in the statistics.â Patients could meet all their needs in one visit with their regular doctor (rather than in one visit to the urgent care clinic and another visit to their regular doctor). The rate of patients able to see their own physician increased from 47 to 75 percent. The provision of preventive services also increased. Additionally, fewer patient charts were lost because they were pulled the day the patients came in, and with fewer lost charts, clinicians were more likely to have the information they need- ed when seeing a patient. Overall costs decreased because of fewer visits and the closure of the urgent care clinic. Additionally, the center thought it would be necessary to hire additional staff, but when the operational system was im- proved, this was not the case, so cost increases were avoided. If such a project were funded by an innovation fund, the primary care center would share its tools for appointment scheduling and staffing design. The above examples illustrate the range and depth of redesign efforts that should occur. Because such efforts can be disruptive to current operations and take extended periods of time to accomplish, health care professionals and orga- nizations need extra support and incentive to undertake them. An innovation fund should support the implementation of projects that could not otherwise be conducted during the routine course of business because they would be too dis- ruptive for the patients and staff. As suggested in Example 1 above, use of a new information technology can be quite disruptive to the provision of services and staff functioning. In some cases, redesign may be so fundamental that temporary closure of a service may be required. For example, in another case that was part of an IOM study of exemplary practices (Donaldson and Mohr, 2000), one medi- cal group had to close its offices for a short period of time to make the changes they deemed necessary. Few health care professionals and organizations can undertake such drastic steps to substantially reorganize their care processes with- out special (and temporary) assistance. Additional examples of reengineering and redesign projects are provided in Chapter 5. REFERENCES Agency for Healthcare Research and Quality. 2000a. âOverview: Center for Organization and De- livery Studies.â Online. Available at http://www.ahrq.gov/about/cods/codsover.htm [accessed Jan. 29, 2001].
TAKING THE FIRST STEPS 109 âââ. 2000b. âIntegrated Delivery System Research Network (ISDRN): Field Partnerships to Conduct and Use Research, Fact Sheet. AHRQ Publication No. 01-P004.â Online. Available at http://www.ahrq.gov/research/idsrn.htm [accessed Jan. 29, 2001]. Berk, Marc L. and Alan C. Monheit. The Concentration of Health Expenditures: An Update. Health Affairs 11(4):145â9, 1992. Blumenthal, David and Melinda Beeuwkes Buntin. Carve Outs: Definition, Experience, and Choice Among Candidate Conditions. American Journal of Managed Care 4(Suppl):SP45â57, 1998. Bodenheimer, Thomas. Disease Management in the American Market. BMJ 320:563â6, 2000. Bodenheimer, Thomas, Bernard Lo, and Lawrence Casalino. Primary Care Physicians Should Be Coordinators, Not Gatekeepers. JAMA 281(21):2045â9, 1999. Brook, Robert H., Elizabeth A. McGlynn, and Paul D. Cleary. Part 2: Measuring Quality of Care. N EngI J Med 335(13):966â70, 1996. Centers for Disease Control and Prevention. 1999. âChronic Diseases and Their Risk Factors: The Nationâs Leading Causes of Death.â Online. Available at http://www.cdc.gov/nccdphp/statbook/ statbook.htm [accessed Dec. 7, 2000]. Cleary, Paul D. and Susan Edgman-Levitan. Health Care Quality: Incorporating Consumer Perspec- tives. JAMA 278(19):1608â12, 1997. Cromwell, Jerry, Debra A. Dayhoff, and Armen H. Thoumaian. Cost Savings and Physician Re- sponses to Global Bundled Payments for Medicare Heart Bypass Surgery. Health Care Financ- ing Review 19(1):41â57, 1997. Demakis, John G. and Lynn McQueen, July 11, 2000. Office of Health Services Research and Development Services, Veterans Administration. Personal communication: telephone conver- sation. Demakis, John G., Lynn McQueen, Kenneth W. Kizer, and John R. Feussner. Quality Enhancement Research Initiative (QUERI): A collaboration between research and clinical practice. Medical Care 38(6, Suppl 1):I17â25, 2000. Donaldson, Molla S. and Julie J. Mohr. Exploring Innovation and Quality Improvement in Health Care Micro-Systems: A Cross-Case Analysis. Washington, D.C.: Institute of Medicine, Na- tional Academy Press, 2000. Ellrodt, Gray, Deborah J. Cook, Jean Lee, et al. Evidence-Based Disease Management. JAMA 278(20):1687â92, 1997. Fennell, Mary L. and Ann B. Flood. Key Challenges in Studying Organizational Issues in Delivery of Healthcare to Older Americans. Health Services Research 33(June;Part II):424â33, 1998. Fishman, Paul, Michael Von Korff, Paula Lozano, and Julia Hecht. Chronic Care Costs In Managed Care. The first single-plan comparison of chronic care costs shows why cost-effective clinical approaches are critical to managed care success. Health Affairs 16(3):239â47, 1997. Foundation for Accountability. Accountability Action 3(2), 1999a. âââ. 1999b. âFoundation for Accountability (FACCT) homepage.â Online. Available at http:// www.facct.org [accessed Dec. 7, 2000]. Friedman, Maria A. Issues in Measuring and Improving Health Care Quality. Health Care Financing Review 16(4):1â13, 1995. Gross, Cary P., Gerard F. Anderson, and Neil R. Powe. The Relation between Funding by the National Institutes of Health and the Burden of Disease. N EngI J Med 340(24):1881â7, 1999. Health Care Financing Administration. 1999. âFact Sheet: Medicareâs Prudent Purchasing Initia- tive.â Online. Available at http://www.hcfa.gov/facts/fsprud.htm [accessed Dec. 7, 2000]. âââ. 2000. âPeer Review Organizations (PROs). Medicareâs Health Care Quality Improvement Program v5â2.â Online. Available at http://www.hcfa.gov/quality/5b1.htm [accessed Dec. 7, 2000]. Hewitt, Maria for the Committee on the Quality of Health Care in America and the National Cancer Policy Board. Interpreting the VolumeâOutcome Relationship in the Context of Health Care Quality. Washington, D.C.: Institute of Medicine, National Academy Press, 2000. Online. Available at http://books.nap.edu/catalog/10005.html [accessed Jan. 29, 2001].
110 CROSSING THE QUALITY CHASM Hofer, Timothy P., Rodney A. Hayward, Sheldon Greenfield, et al. The Unreliability of Individual Physician âReport Cardsâ for Assessing the Costs and Quality of Care of a Chronic Disease. JAMA 281(22):2098â105, 1999. Hoffman, Catherine, Dorothy P. Rice, and Hai-Yen Sung. Persons with Chronic Conditions. Their Prevalence and Costs. JAMA 276(18): 1473â9, 1996. Homer, Charles J. Asthma Disease Management. N EngI J Med 337(20):1461â3, 1997. Hsaio, W. C. and W. B. Stason. Toward Developing a Relative Value Scale for Medical and Surgical Services. Health Care Financing Review 1(23):23â38, 1979. Hunter, David J. Disease Management: Has it a Future? BMJ 320:530, 2000. Institute of Medicine Setting Priorities for Health Technologies Assessment: A Model Process. Molla S. Donaldson and Harold C. Sox, Jr., Eds. Washington, D.C.: National Academy Press, 1992. âââ. Setting Priorities for Clinical Practice Guidelines. Marilyn J. Field, ed. Washington, D.C.: National Academy Press, 1995. Joint Commission on Accreditation of Healthcare Organizations. 2000. âORYX: The Next Evolu- tion in Accreditation. Questions and Answers about the Joint Commissionâs Planned Integra- tion of Performance Measures into the Accreditation Process.â Online. Available at http:// www.jcaho.org/perfmeas/oryx%5Fqa.html [accessed Dec. 7, 2000]. Ketner, Lisa. Population Management Takes Disease Management to the Next Level. Healthcare Financial Management 53(8):36â9, 1999. Lasker, Roz D. The Diabetes Control and Complications Trial, Implications for Policy and Practice. N EngI J Med 329(14):1035â6, 1993. Medical Expenditure Panel Survey. 2000. âMEPS HC-006R: 1996 Medical Conditions.â Online. Available at http://www.meps.ahrq.gov/catlist.htm [accessed Dec. 7, 2000]. National Committee for Quality Assurance. 1999. âNCQA Releases Final HEDIS 2000 Measures; Focus is on Heart Disease, Asthma, Womenâs Health.â Online. Available at http://www. ncqa.org/Pages/Communications/news/H2KFINREL.html [accessed Dec. 7, 2000]. National Quality Forum for Health Care Quality Measurement and Reporting. 2000. âNational Qual- ity Forum Mission.â Online. Available at http://www.qualityforum.org/mission/home.htm [ac- cessed Dec. 7, 2000]. Neumann, Peter J. and Eileen A. Sandberg. Trends in Health Care R&D and Technology Innovation. Health Affairs 17(6):111â9, 1998. Ray, G. Thomas, Tracy Lieu, Bruce Fireman, et al. The Cost of Health Conditions in a Health Maintenance Organization. Medical Care Research and Review 57(1):92â109, 2000. Redelmeier, Donald A., Siew H. Tan, and Gillian L. Booth. The Treatment of Unrelated Disorders in Patients with Chronic Medical Diseases. N EngI J Med 338(21):1516â20, 1998. The Robert Wood Johnson Foundation. Chronic Care in America: A 21st Century Challenge. Princeton, NJ: The Robert Wood Johnson Foundation, 1996. Online. Available at http:// www.rwjf.org/library/chrcare/ [accessed Sept. 19, 2000]. U.S. Department of Health and Human Services. 2000a. âFiscal Year 2001 Budget Request, Health Care Financing Administration.â Online. Available at http://www.hcfa.gov/testimony/2000/ 000208.htm [accessed Jan. 29, 2001]. U.S. Department of Health and Human Services, National Institutes of Health National Human Genome Research Institute. 2000b. âFY 2001 Budget.â Online. Available at http://www. nhgri.nih.gov/NEWS/fy2001.pdf [accessed Dec. 7, 2000]. Wagner, Edward H. The Role of Patient Care Teams in Chronic Disease Management. BMJ 320:569â 72, 2000. Wagner, Edward H., Brian T. Austin, and Michael Von Korff. Improving Outcomes in Chronic Illness. Managed Care Quarterly 42(2):12â25, 1996a. âââ. Organizing Care for Patients with Chronic Illness. Milbank Quarterly 74(4):511â42, 1996b.