5

Fostering Systems Change to Drive Continuous Learning in Health Care

INTRODUCTION

The vision of the Institute of Medicine’s (IOM’s) Roundtable on Evidence-Based Medicine (now the Roundtable on Value & Science-Driven Health Care) is “the development of a learning healthcare system that is designed to generate and apply the best evidence for the collaborative health care choices of each patient and provider; to drive the process of discovery as a natural outgrowth of patient care; and to ensure innovation, quality, safety, and value in health care” (Charter pp. xi–xii). How to realize the vision of continuous learning was the focus of the fourth session of the workshop.

The publication The Learning Healthcare System: Workshop Summary (IOM, 2007), based on an earlier Roundtable workshop, identified several common characteristics of a system with continuous learning, including a culture that emphasizes transparency and learning through continuous feedback loops, care as a seamless team process, best practices that are embedded in system design, information systems that reliably deliver evidence and capture results, and results that are bundled to improve the level of practice and the state of the science. With those characteristics in mind, the contributors in this chapter looked closely at how specific aspects of feedback and performance can be improved in the healthcare organizational culture, in the development of accessible knowledge, in the management of information and technology, and in the organization of information systems.

Steven J. Spear, senior lecturer at Massachusetts Institute of Technology and a Senior Fellow at the Institute for Healthcare Improvement, observed



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



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

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

OCR for page 237
5 Fostering Systems Change to Drive Continuous Learning in Health Care INTRODUCTION The vision of the Institute of Medicine’s (IOM’s) Roundtable on Evidence-Based Medicine (now the Roundtable on Value & Science-Driven Health Care) is “the development of a learning healthcare system that is designed to generate and apply the best evidence for the collaborative health care choices of each patient and provider; to drive the process of discovery as a natural outgrowth of patient care; and to ensure innovation, quality, safety, and value in health care” (Charter pp. xi–xii). How to realize the vision of continuous learning was the focus of the fourth session of the workshop. The publication The Learning Healthcare System: Workshop Summary (IOM, 2007), based on an earlier Roundtable workshop, identified several common characteristics of a system with continuous learning, including a culture that emphasizes transparency and learning through continuous feed- back loops, care as a seamless team process, best practices that are embed- ded in system design, information systems that reliably deliver evidence and capture results, and results that are bundled to improve the level of practice and the state of the science. With those characteristics in mind, the con- tributors in this chapter looked closely at how specific aspects of feedback and performance can be improved in the healthcare organizational culture, in the development of accessible knowledge, in the management of informa- tion and technology, and in the organization of information systems. Steven J. Spear, senior lecturer at Massachusetts Institute of Technology and a Senior Fellow at the Institute for Healthcare Improvement, observed 2

OCR for page 237
28 ENGINEERING A LEARNING HEALTHCARE SYSTEM that in many sectors of the service and manufacturing economies a few high-performance organizations seem to be the leaders, and their competi- tors essentially compete for second place. These pioneers deliver value with less effort and cost, even though they have similar—or identical—tools, customers, suppliers, labor, and regulations. Spear said that these organiza- tional leaders continue to push the envelope through differences in systems management and that the lessons from their success might offer perspectives on value, efficiency, quality, and other areas that are important to produc- ing a learning, team-oriented, patient-centric culture within health care. Based on his close observations of Toyota, Southwest Airlines, Alcoa, and other industry leaders, Spear reported that, in contrast to organizations that address systems anomalies with workarounds, industry leaders care- fully analyze adverse events and use them as sentinels for investigation into causes. Spear hypothesized that by adopting similar techniques, healthcare systems may be able to deliver better care to more people at less cost and with less effort—on the order of twice as good for twice as many people at half the cost. Examining the value of knowledge management, access, and use, Donald E. Detmer, president and chief executive officer (CEO) of the American Medical Informatics Association (AMIA) and professor of medi- cal education at the University of Virginia, argued that improved man- agement of information applied to clinical decision support (CDS) will require structured policies and complementary agendas for informatics education and research. Detmer discussed the CDS Roadmap for National Action developed by the AMIA, which is based on the principles of (1) best knowledge available when needed, (2) high adoption and effective use, and (3) continuous improvement of CDS methods and knowledge. Detmer also discussed the Morningside Initiative, which seeks to share information broadly for CDS. Detmer highlighted the AMIA–Association of Academic Health Centers’ (AAHC’s) current collaboration to develop enhanced in- formatics curriculums for health professional and continuing education students. He also discussed current developments in CDS policy and infra- structure and identified areas for further investigation and efforts. Looking to the future, he emphasized the importance of determining the appropriate mechanism for integrating personal health records with electronic health records (EHRs). Stephen J. Swensen, director of quality for the Mayo Clinic and profes- sor of radiology at the Mayo Clinic College of Medicine, said the healthcare industry must address specific elements of technology management in order to drive systems change. He described work in technology management at the Mayo Clinic to develop networks that embody optimal reliability, per- mit nimble and effective diffusion of best practices, have built-in safety nets, and support optimal organizational learning and communication. Swensen

OCR for page 237
29 FOSTERING SYSTEMS CHANGE emphasized that technology management should leverage human capital and should embody a decision-making process whereby decisions are made with an organizational perspective by cross-functional, physician-led teams. Swensen’s discussion encompassed five facets of technology management— policy, appropriateness, reliability, diffusion, and social capital. In ensuring appropriate care, for example, Swensen observed that health systems face the complex task of first making sure that the right policies are in place to encourage medical centers, physicians, and other providers to use tech- nology and deliver the appropriate care in the best setting, and then they must ensure that patients are connected in the most efficient way with the technology assets most appropriate to their needs. Discussing the link between the organization and management of in- formation systems and the quality and safety of patient care, David C. Classen, a physician at Computer Sciences Corporation, described current approaches to the evaluation of clinical information systems. He detailed a new simulation tool that has been developed and used by healthcare organizations to evaluate the effectiveness of clinical information systems implementations in improving the safety of care for patients. Classen dem- onstrated how such tools have been used by organizations to learn about the capabilities of their implemented clinical information systems and to assess system shortfalls, and he showed how organizations have used these tools to improve clinical information systems. CHASING THE RABBIT: WHAT HEALTHCARE ORGANIZATIONS CAN LEARN FROM THE WORLD’S GREATEST ORGANIZATIONS Steen J. Spear, D.B.A., M.S., M.S., , Massachusetts Institute of Technology, Institute for Healthcare Improement In manufacturing, heavy industry, high tech, services, aviation, the military, and elsewhere, a small number of organizations always race to the front of the pack in their sector, leaving everyone else competing for runner- up. Although these organizations use similar science and technology to meet the needs of a similar customer base, are dependent on the same group of suppliers, hire from the same labor pools, and are subject to the same regulations as their competitors, they deliver far more value with much less effort and at lower cost. They gain and sustain leadership by managing the complex systems of work on which they depend in markedly different ways. Healthcare organizations can learn—and have learned—from these exemplars, with outstanding results in efficacy, efficiency, safety, and qual- ity of care. The proposition considered here is that it is possible to deliver much better care then we currently do, to many more people than we currently

OCR for page 237
20 ENGINEERING A LEARNING HEALTHCARE SYSTEM do, and at much less cost and with less effort than is currently the case. The envisioned improvements are not incremental; instead, I am speaking of a product that is twice as good for twice as many people at half the cost. The proposition is not based on hypothesis or conjecture, but is supported by good clinical evidence. This paper begins by examining what needs to be done and in particular, the lessons healthcare organizations can learn from other complex, high-performing organizations in other industries so as to achieve the goal of better care for more people at less cost. Twenty years ago I was an employee of Congress at one of the congres- sional agencies. At the time, the competitiveness of U.S. manufacturing with that of Japan was a major concern, focused on the idea that the Japanese were gaining an advantage from what was essentially unfair competition. People sensed that financing arrangements, competition, and domestic and international markets were being manipulated. There were accusations of dumping of various types of goods, steel not the least of these. The notion was that the appropriate response to declining competitiveness on the part of American companies was for Congress, regulators, and the executive branch to act similarly to how they perceived the Japanese to be acting. A few years later, there was a fundamental shift in what people saw as the causes of competitive differences. Replacing the focus on large macro- national elements was recognition that the differences between the countries were rooted in the differences between companies—that what was being done in companies such as Sony, Toshiba, and Hitachi was fundamentally different from what was happening in their U.S. counterparts and that what was taking place at Toyota and Honda was fundamentally different from what was going on at General Motors (GM), Chrysler, and Ford. This realization was good news because it meant that the solution to the problem did not depend on consensus among the Majority Leader of the Senate, the President, and the Speaker of the House on the source of the problem or the solution. This good news, however, meant that U.S. compa- nies bore a great responsibility, and that managers of individual companies and of business units within those companies had enormous influence on the outcome of their organizations’ efforts. To link this discussion to health care, let us start with a statement of the problem: too few people have access, the costs are too high, and so on. Much of the discussion among politicians focuses on whether more resources should be committed to the system. But if we pursue the parallels with the manufacturing sector, that may not be the answer. I am not going to argue against spending altogether. Certain changes are needed in terms of how information is reported and how coverage is provided for those who are least able to care for themselves. There are also separate issues of transfer of wealth and caring for the least well in our society. Continuing the focus on the delivery of care, the experience from

OCR for page 237
21 FOSTERING SYSTEMS CHANGE manufacturing suggests that presidents of hospitals, presidents of systems, managers of hospitals, and deliverers of care—from small practices to very large organizations such as the Office of Veterans Affairs—can have an enormous influence on outcomes. That is good news because it means that improving the quality of health care will not depend on a confluence of interest and perspective among three people, but can arise from the efforts of thousands or tens of thousands of people who work together to move the system in the right direction. Returning to the perception of the competition between the United States and Japan, originally the perception was that the key competition was between Tokyo and the Diet (Japan’s legislative body) and Washing- ton and Congress. In a sense, the idea was that somehow the Japanese had unleveled the playing field, that they were not playing by the same rules or were playing by the same rules but cheating. But this perception changed, and people started to recognize that the playing field was in fact quite level. Consider that to compete today, industries must compete in every region around the world. When they do so, they compete head to head with all of their competitors, so they cannot lock up markets, regions, or customers. For example, many towns have the equivalent of Boston’s “Auto Mile,” where one can walk into a Buick dealer, and if that dealer does not have what one wants, one can visit the Chrysler dealer next door and, if necessary, move on to the Ford dealer, the Toyota dealer, and so forth—all literally within walking distance. Given this phenomenon, major auto com- panies cannot lock up customers. How, then, can they gain a competitive advantage? If a monopolistic relationship with one’s customers is impossible, a company might try to lock up its suppliers. That cannot be done with automobiles, however, and, generally speaking, all automobile manufac- turers are subject to the same regulations and innate market preferences. Customers are paying the same price for a gallon of gasoline whether they put that gas into a Ford or a Toyota or a Chrysler. The playing field is ex- traordinarily level. And when the playing field is level, this parity of rules can be expected to lead to a parity of outcomes. When everything is the same in terms of customers, suppliers, labor pools, and so forth, people can be expected to gain and lose leadership, gain and lose profits, in a very fluid, dynamic situation. In the automobile industry, Ford, Chrysler, GM, Volkswagen, and their competitors do indeed fluctuate between very hot and very cold years. They are engaged in intense competition, but they are all competing for second place. In first place is Toyota, which has experienced extraordinary profit- ability and growth in market share and revenue. By other measures, such as market capitalization vs. profitability, there is an enormous disproportion

OCR for page 237
22 ENGINEERING A LEARNING HEALTHCARE SYSTEM between Toyota and U.S. companies. Not only has Toyota grown, but if one looks at the ratios, the market expects it to continue to grow at a sustained rate over many years. One might think Toyota is an anomaly and decide to look at another playing field—say, commercial aviation. To make an apples-to-apples com- parison, players in this field fly the same planes out of the same airports, hire from the same labor pool, pay the same price for jet fuel, and are subject to the same regulations. By and large, the playing field is level with parity of contest and parity of outcomes. Accordingly, in this competitive environment, Delta, Northwest, Continental, United, and American regu- larly have good and bad years, and they regularly gain and lose share. It turns out, though, that this is a competition for second place because there is Southwest, with some 35 years of profitable growth, year after year. Even when things go bad at Southwest, they go bad on a much smaller scale than at other companies. For example, when Southwest and American failed similar Federal Aviation Administration inspections, Southwest paid a fine and kept flying, while American essentially shut down for a week. In terms of parity of playing field and parity of outcome, one can see such anomalous patterns in industry after industry—automobile manufac- turing, the aluminum and steel industries, commercial aviation, govern- ment services, and on and on. One begins to realize that there are not just anomalies, but a population of anomalous outcomes. When one examines what the leaders—Toyota, Southwest, Alcoa, the Navy’s nuclear reactor program—have in common, one finds that they have solved a problem that plagues every industry—complexity. For any product or service, the number of elements necessary to make it function is far greater than it was 5, 10, or 20 years ago. The number of interdependencies and interconnections among those elements is far greater than ever before. The basic problem with a complex system is that, at some point, once there are enough elements and connections and interdepen- dences, it is nearly impossible to understand the structure of the system and to understand or predict its behavior perfectly. This is where the divide begins between the companies or organizations that are in first place and those that are stuck competing for second place. Two fundamental differences in behavior have direct application to health care, which, of course, is a complex system of work to deliver care to patients. The first is that those who are competing tend to organize them- selves functionally around specialty silos, whereas those who are highly successful tend to place tremendous emphasis on building functional tech- nical skill because they need it to compete, but this skill is in service of the process by which they deliver value to customers or patients or users. The difference is between a functional view and a functional view plus service of process and system.

OCR for page 237
2 FOSTERING SYSTEMS CHANGE The second difference is that those who are less successful put a lot of effort into the design of systems. I am not going to try to diminish that effort. But once a complex system is running and operating, its behavior is going to be somewhat unpredictable because of the problem, noted earlier, of the inevitable limits on understanding the system and the way its parts interrelate. Those who focus on designing systems tolerate things going contrary to expectations, as is inevitable with a complex system, and they dismiss this as the inevitable noise of what they do. In contrast, when highly successful companies design and start operat- ing systems, they do not dismiss such chatter as noise. They do not live in a world of signal and noise, where the signal is what they wanted to get and the noise is something contrary to that. Instead, they live in a world of signal and signal. The expected signals are things that happen that confirm what they believed about the system’s structure and behavior; the chatter or noise points them to the things they did not understand. A key difference between the highly successful organizations and the others is that the others tolerate, encourage, and depend on an environment where fighting fires, working around problems, coping, and otherwise mak- ing do is how work is accomplished. The problem with that approach for the people who work in those organizations is it means that every day they know they are going to go to work and fail to some degree. Another basic problem with complex systems is that sometimes these little failures come together in idiosyncratic fashion. Not only are there the normal daily annoyances of doing work in a flawed system, but sometimes these things combine catastrophically. In contrast, those who are very good at dealing with complexity will design a system, but when they operate it, they place tremendous emphasis on identifying things that go contrary to expectations. Those signals tell them where they have to invest in building more knowledge. When they see that something has gone wrong, they are quick to deal with the problem because they know that the time to address problems is when they are still hot. Think, for example, of doctors rushing to a patient who is crashing or detectives getting to a crime scene while the evidence is still fresh. It is in dealing with problems while they are still hot that new knowledge can be generated about how the system behaves. When that knowledge is gained locally by an individual, great effort can be made to ensure that this knowledge is shared with everyone else involved. As an example, the nuclear navy has modeled very well the behavior of constant dynamic discovery, of creating a high-velocity organization. The navy thinks about it in terms of an operator sitting down to run a nuclear reactor on board a submarine. The person may be just 22 years old, per- haps just graduated from Annapolis with a year’s training in the Nuclear Reactor Program. This person is not running the reactor as if he or she has had just a year’s experience, but as if he or she has had the 5,700 reactor-

OCR for page 237
2 ENGINEERING A LEARNING HEALTHCARE SYSTEM years of experience the navy has accumulated over the past 50 years of running reactors on board warships. I started looking at the anomalies in manufacturing first at Toyota and then at Alcoa. I also got involved in the Pittsburgh Regional Health Initiative. Our first effort in Pittsburgh was to look at medication admin- istration. This is a process problem because it involves doctors making diagnoses and writing prescriptions, prescriptions going to the pharmacy, orders being filled and delivered, nurses providing medication, and so on. It turned out there was a problem with how orders were being transmitted and delivered. People in the pharmacy and in nursing came up with a solution to the problem that would save a tremendous amount of nursing time and reduce and nearly eliminate any chance of error that would result in giving the wrong medication to a patient—a particularly serious problem in a trans- plant case. They tried this solution in a low-cost way, then tested it again through a variety of pilots and realized it was a great idea. They wanted to institutionalize it and make their learning valuable to the organization, so they tried to find the person who owned the bridge between nursing and pharmacy. They knew this was not someone in nursing or in pharmacy or in their particular domain, so they started looking elsewhere in the orga- nization. It was not the charge nurse or the person who played a similar role in the pharmacy, and it turned out it was not even the president of the hospital who owned the bridge between the two because the hospital was part of a larger system. Eventually they found that the first person who had formal authority over the bridge between this pharmacy and this nursing unit in a much larger system was the CEO of the hospital. Everything else was managed through functional silos and disciplines—orthopedics, ob- stetrics, and so on; nursing separate from medicine, medicine from surgery. Consider how difficult it is to institutionalize all the micro-changes neces- sary so an organization has on a daily basis a homeostatic self-correcting, self-improving dynamic. In that case it was impossible. To return to my original proposition, it is possible to deliver much better care to many more people with much less cost and effort than is cur- rently the case. We need not wait for the President, the Majority Leader, and the Speaker of the House to come to some kind of agreement. What we do need is for people who are responsible for systems and organizations to understand that although managing functions is necessary, it is not suf- ficient. They need to manage processes—not just pharmacy and nursing, but also medication administration, and not in a static fashion whereby one designs a process and hopes it will run well, but in a dynamic fashion so that chatter is not treated as inevitable noise, but as an indication of where one needs to improve.

OCR for page 237
2 FOSTERING SYSTEMS CHANGE KNOWLEDGE MANAGEMENT FOR CLINICAL CARE Donald E. Detmer, M.D., M.A., Uniersity of Virginia, Charlottesille Knowledge management for CDS requires a policy framework as well as an education and research agenda. In its CDS Roadmap for National Ac- tion, the American Medical Informatics Association (AMIA) recommended a structure with the following three pillars: (1) best knowledge available when needed, (2) high adoption and effective use, and (3) continuous improvement of CDS methods and knowledge. The roadmap led to the Morningside Initiative, which aims to develop a Knowledge Management Repository for CDS through a public–private partnership. The goal is to create a shared repository of executable knowledge for CDS that will be broadly available. The hope is that these and related initiatives funded recently by the Agency for Healthcare Research and Quality (AHRQ) will eventually result in a sustainable infrastructure. Educational initiatives and relevant informatics research are needed. The AMIA, in collaboration with the AAHC Affiliate Roundtable, will collaborate to create a two-stage, integrated, multimodular informatics cur- riculum for all students studying to become health professionals. The initial course will be appropriate for students entering professional education, and the second is to be pursued just before students begin professional practice. These initiatives, combined with AMIA’s 10 × 10 program for those in practice, will help address basic professional educational needs, especially in applied clinical informatics. Finally, there are major informatics research issues that need attention. One critical research and development area, for instance, concerns patients’ use of their own EHRs for chronic illness man- agement in collaboration with their clinicians via secure Web portals. The AMIA is clearly interested in trying to foster change for pur- poses of improving both health and healthcare delivery. In particular, we are challenged to integrate the carbon dimensions with the silicone dimensions—that is, to bring informatics to bear on the problem. Today, we lack the right policy infrastructure to accomplish this integration. This paper looks briefly at some relevant IOM work and a project that the AMIA carried out for the Office of the National Coordinator on Clini- cal Decision Support, and then offers some ideas about what a national roadmap for knowledge management should look like. The 1991 IOM study Computer-Based Patient Record: An Essential Technology for Health Care (reissued in 1997) (IOM, 1997) identified EHRs as an essential technology for health care. It is fascinating that this remains the case 17 years later, yet one would not think so based on the us- age of EHRs in the United States today. While the 1991 report emphasized

OCR for page 237
2 ENGINEERING A LEARNING HEALTHCARE SYSTEM the importance of EHRs for quality improvement and for clinical decision support, these dimensions of EHRs were not really obvious at the time in terms of widely held national policy perspectives. Whereas the 1991 report addressed essential skills needed in the work- force, the IOM’s 2001 Crossing the Quality Chasm report firmly tied policy to how EHRs and EHR systems and communications technologies should be used to move from a costly, inefficient, and highly variable system to a system that is equitable, safe, patient centered, efficient, effective, and timely (IOM, 2001). Another crucial IOM effort that has not received as much attention as it deserves is the Health Professions Education Summit. That meeting and the ensuing report, entitled Health Professions Education: A Bridge to Quality 200, addressed many relevant dimensions of health care, includ- ing the need for aligned reimbursement incentives and regulatory require- ments, robust information infrastructure, widespread use of evidence-based medicine, and a workforce skilled in evidence-based medicine, information technology (IT), and process improvement (IOM, 2003). From the perspective of a policy background, a national roadmap for knowledge management with decision support is clearly lacking. In fact, although many developed economies around the world have put in place a good basic information infrastructure, decision support remains immature. Even Denmark has a long way to go. The AMIA developed the CDS Roadmap for National Action between 2005 and 2007 with the support of many groups and individuals (AMIA, 2006). Some findings have just recently been approved by the American Health Information Community (AHIC) as a guide for U.S. policy in this domain. Essentially, the roadmap was intended to create a blueprint for coordinated nationwide action to ensure that usable and effective CDS will be widely used by clinicians and patients. The challenge was seen as devel- oping decision support that is equally usable by patients and their clinicians to improve health care. Three pillars were envisioned as the foundation of the model: A system must continually develop the best knowledge available and make that knowledge available at the point and time it is needed. Knowledge must be both current, “right” to the best standards of the day (ideally both generally and locally), and accessible. There needs to be high adoption and effective use—performance is key to this. Methods must be improved continuously in addition to the knowledge base. With these three pillars in mind, a coherent structure will enable prog- ress. The following objectives are crucial: develop practical, standard for- mats for representing CDS knowledge and interventions; establish standard approaches for collecting, organizing, and distributing CDS; address policy,

OCR for page 237
2 FOSTERING SYSTEMS CHANGE legal, and financial barriers, and create additional support and enablers; compile and disseminate best practices for usability and implementation; develop methods for collecting, learning from, and sharing national experi- ence with CDS; and use EHR data systematically to advance knowledge. The roadmap recommends a series of activities to improve the de- velopment, implementation, and use of CDS. It identifies work products. Objectives include organizing and facilitating the creation of a conven- ing coordinating body and establishing a CDS technical assistance center. (Whether this should be one center or a cluster is not clear, but the goal is to establish consensus groups to answer key questions that arise on road- map development.) We wish to assemble a best-practice synthesis, conduct training and education, and develop prototypes. Many pilot demonstrations are recommended, including supporting and facilitating related nationwide initiatives, developing practical standard formats to share knowledge and interventions, and collecting and disseminating best practices for usability and implementation.1 Looking at current policy and national structure, one can see that the roadmap has had an impact. AHRQ is supporting some activities through its national resource center and Centers for Education and Research on Therapeutics grants, including knowledge management CDS grants. There have been presentations to the National Committee on Vital and Health Statistics, and the Morningside Initiative begun by the Telemedicine and Advanced Technology Research Center (TATRC) is now gaining some in- stitutional linkage to the AMIA. The Department of Health and Human Services (HHS) AHIC meeting on April 22, 2008, approved recommendations of the ad hoc CDS work- group; AHIC considers the AMIA’s CDS Roadmap for National Action to be a foundational document. At the meeting, three priorities were identi- fied: (1) drive measurable progress toward priority performance goals for healthcare quality improvement, (2) explore options to establish or lever- age a public–private entity to facilitate collaboration across CDS develop- ment and deployment, and (3) accelerate CDS development and adoption through federal programs and collaborations. All activities relate to seeking measurable progress through quality improvement. Another recommenda- tion was that by October 31, 2008, HHS and relevant partners should have explored options for establishing or leveraging a public–private entity (e.g., AHIC 2.0) to convene public and private organizations and stakeholders for the purpose of promoting effective CDS development and adoption and addressing gaps in CDS capabilities through planning, facilitation, and coordination of activities across diverse constituencies. 1 More CDS Roadmap Information is available at www.amia.org/inside/initiatives/cds/ (ac- cessed September 20, 2010).

OCR for page 237
20 ENGINEERING A LEARNING HEALTHCARE SYSTEM TABLE 5-2 Leapfrog Computerized Provider Order Entry Evaluation Tool Procedure Steps in the Evaluation Procedure Activities • Obtain the password used to participate in Register for the computerized provider order entry (CPOE) evaluation the Leapfrog survey • Sign on to the Web application • Enter hospital information • Sign up for adult or pediatric evaluation • Assemble teams for patient set-up and order entry • Ensure that the test system mirrors the production system, or make plans to use the production system • When ready to begin set-up for the sample Download test patient information (e.g., age, weight, allergies, lab values) test or full evaluation, sign on to the Web application • Print the list of test patients • Set up test patients • Ensure that patients are “active” (may require nursing unit and bed before orders can be written and signed) • When ready to begin the sample test Download test orders or full evaluation, sign on to the Web application • Print test orders, instructions, and answer sheets • Ensure that the physician performing the evaluation has system authorizations required for order entry in CPOE (may be a test user) BREAKOUT SESSION: CAPTURING MORE VALUE IN HEALTH CARE During a breakout session, participants broke into small groups to dis- cuss how to capture more value in health care. They were asked to discuss three issues: (1) how much more value (health returned for dollars invested) could be obtained through the application of systems engineering principles in health care, (2) which one area had the potential for the greatest value to be returned from applying these principles, and (3) which actions could do the most to facilitate the needed changes. The main points of their discus- sions were reported back to the entire group. In response to the question of how much value could be obtained from

OCR for page 237
21 FOSTERING SYSTEMS CHANGE TABLE 5-2 Continued Steps in the Evaluation Procedure Activities • Enter test orders for specified test patients Enter orders into the CPOE application • Sign every test order (or pair of orders) • Record the system responses on the answer sheet • Discontinue each test order (or pair of orders) • Sign on to the Web application Enter and submit results • Submit information from the answer sheet as instructed • Use automatic scoring of success in Scoring providing decision support to avert common, harmful medication errors for each order category and the evaluation overall • Print or view the feedback report Reporting immediately available (scores for each order category) • Aggregate the score available for posting along with hospital survey results SOURCE: Reprinted with permission from Patient Safety & Quality Healthcare. Metzger et al., 2008. the application of systems engineering principles in health care, respondents began by pointing out that the definition of value was problematic. They discussed the fact that value is hard to measure because it is composed of different components that are measured in different ways, including safety, quality and cost. Some groups concluded that value can be construed as a measure with many definitions, and the particular definition used will de- pend on the stakeholder’s point of view. One group identified the problem of not having a common definition of value among stakeholders as one of the barriers to a patient-centered healthcare system and pointed to the need to align the value space as an interesting point for potential follow-up and additional research. The work of the Commonwealth Commission on High

OCR for page 237
22 ENGINEERING A LEARNING HEALTHCARE SYSTEM TABLE 5-3 Medication Order Categories in the Leapfrog Computerized Provider Order Entry Evaluation Order Category Description Examples Therapeutic duplication Medication with Codeine and Tylenol #3 therapeutic overlap with another new or active order; may be same drug, within drug class, or involve components of combination products Single and cumulative dose Medication with a Ten-fold excess dose of limits specified dose that exceeds Methotrexate recommended dose ranges or that will result in a cumulative dose that exceeds recommended ranges Allergies and cross-allergies Medication for which Penicillin prescribed for patient allergy has patient with documented been documented or penicillin allergy allergy to other drug in same category has been documented Contraindicated route of Order specifying a route of Tylenol to be administered administration administration (e.g., oral, intravenously intramuscular, intravenous) not appropriate for the identified medication Drug–drug and drug–food Medication that Digoxin and quinidine interactions results in a known, dangerous interaction when administered in combination with a different medication in a new or existing order for the patient or results in an interaction in combination with a food or food group

OCR for page 237
2 FOSTERING SYSTEMS CHANGE TABLE 5-3 Continued Order Category Description Examples Contraindication/dose Medication either Nonspecific beta blocker in limits based on patient contraindicated based patient with asthma diagnosis on patient diagnosis or diagnosis affects appropriate dosing Contraindication dose Medication either Adult dose of antibiotic in a limits based on patient age contraindicated for this newborn and weight patient based on age and weight or for which age and weight must be considered in appropriate dosing Contraindication/dose Medication either Normal adult dose regimen of limits based on laboratory contraindicated for this renally eliminated medication studies patient based on laboratory in patient with elevated studies or for which creatinine relevant laboratory results must be considered in appropriate dosing Contraindication/dose Medication contraindicated Medication prescribed known limits based on radiology for this patient based on to interact with iodine to studies interaction with contrast be used as contrast medium medium in recent or in ordered head computed ordered radiology study tomography exam Corollary Intervention that requires Prompt to order drug an associated or secondary levels when ordering order to meet the standard aminoglycoside of care Cost of care Test that duplicates a Repeat test for digoxin level service within a time frame within 2 hours in which there are typically minimal benefits from repeating the test SOURCE: Reprinted with permission from Patient Safety & Quality Healthcare. Metzger et. al., 2008.

OCR for page 237
2 ENGINEERING A LEARNING HEALTHCARE SYSTEM BOX 5-1 Use of Simulation Tool to Evaluate Computerized Physician Order Entry: Case Study 1 The Setting • Academic medical center • Commercially available computerized provider order entry (CPOE) in use for many years Lessons Learned • Verified poor results in some areas: drug–lab, drug–disease, dose limits • Surprising results in drug–drug and drug–allergy interaction checking • Pointed out new areas to pursue: wrong route, corollary orders, duplicate test Actions Taken • Initiated pharmacy review of preconfigured allergy and drug–drug alerts • Planned to reduce redundant drug–drug alerting by building from the ground up • Reviewed important food allergies and how to handle • Began pharmacy/physician review of circumstances in which corollary orders are important • Began work with third-party drug knowledge vendor on content needed for dosing-related messages • Plan to incorporate new functions into next big rebuild of CPOE Performance Healthcare Systems3 was cited as an important reference in de- fining of policy areas that could affect significant cost savings in the system as a way of approaching increased value. Other groups took a pragmatic approach to the question of how much more value could be obtained and based their estimation on the figures presented during the workshop, which had suggested the existence of up to 50 percent waste in the current system. Based on this, they concluded that it was reasonable to assume that a dou- bling of value was attainable through the application of systems engineering principles. They went on to identify some of the key changes that would be needed to bring about this increased value. These included a realignment of payment incentives away from volume of services, the institution of a comprehensive EHR and health IT system for greater efficiency and as a 3 For more information, see http://www.commonwealthfund.org/Content/Program-Areas/ Commission-on-a-High-Performance-Health-System.aspx.

OCR for page 237
2 FOSTERING SYSTEMS CHANGE BOX 5-2 Use of Simulation Tool to Evaluate Computerized Physician Order Entry: Case Study 2 The Setting • A 750-bed academic medical center where computerized provider order entry (CPOE) is in use house-wide • Very proud of work and accomplishments in safety and quality Lessons Learned • Only order category covered was drug–allergy checking • Some categories (patient-specific dose checking based on renal status, weight) being done in pharmacy application, but not delivered to physicians Actions Taken • Evaluated order categories in simulation tool against local experience (phar- macist interventions) to assign priorities for advancing clinical decision support (CDS) in CPOE • Launched aggressive effort to advance CDS source of data for continuous learning and improvement, and, finally, better systems integration. Breakout groups were also asked to identify the area in which the great- est value could be returned. Participants pointed to several areas within the healthcare system that were discussed during the workshop and also to some themes that appeared in several presentations. The major area identi- fied was the use of health IT systems in the form of EHRs and a coordinated system for the transfer of knowledge and communication of best practices, as well as a resource for research and improvement. Participants pointed to these information systems as potential conduits for better systemic coordi- nation and informed decision making as a way to increase value. The area of health provider education was also cited as one that could yield increased value. Participants pointed to the various workshop presen- tations that touched on the need for change in the culture of the healthcare system and suggested that modifying the way that caregivers are trained would be one way to initiate these changes. They identified several potential modifications to training, including greater interdisciplinary exposure and more emphasis on the team-based nature of modern health care. Increasing the use of a collaborative approach among caregivers and between disciplines was identified as another area that should be targeted

OCR for page 237
2 ENGINEERING A LEARNING HEALTHCARE SYSTEM for increasing value. Increasing efficiency and efficacy through better inte- gration of systems was also discussed, along with the adoption of practices that translate to use and evaluation as a part of execution. Groups further identified the area of payment as one with great poten- tial to extract increased value. They suggested that incentives be realigned in order to promote best practices instead of favoring greater volume, which the current fee-for-service compensation system rewards. In response to the question of what actions could do the most to facili- tate the changes needed to capture more value in health care, participants returned to some of the areas and themes mentioned previously and de- scribed strategies that could be taken to carry out these actions. Participants noted that the particular approach to reform is itself an important consid- eration. They suggested that reform start with easy, manageable issues and then progress to broader, more difficult reforms. This two-tiered approach would allow for a demonstration of the potential for improvement within the system, and it would give those orchestrating the reform the oppor- tunity to get greater buy-in from stakeholders. One group described the necessity to be prepared to undergo constant evolution and to not have a predetermined end state. Several groups mentioned the need to encourage a more collaborative approach to the care process and to involve multidisciplinary groups. Par- ticipants mentioned the need to overcome barriers created by the current culture in order to allow for more integrated care; reforming the models of education for healthcare providers would be one way to approach this problem. The need for greater collaboration between process engineers and medical professionals was also mentioned as an area for action in achieving higher value from health care. Groups discussed what steps might be taken to encourage greater interdisciplinary research, including changing the way engineers and health professionals are educated and developing funding mechanisms. Specific suggestions included the creation of a master’s of engineering in engineering and healthcare systems and the establishment of combined interdisciplinary institutes for research and practice. Changes in the availability, implementation, and application of EHRs and health IT were discussed as ways to better communicate best practices, to allow for better analysis of process and outcomes data that could be fed back and used to improve the system, and to create better continuity of care. One group described the health IT system as the glue that ties everything together and makes it act like a system. In order to achieve con- nectedness, however, interfaces between technology and users need to be redesigned to allow for ease of use and seamless integration into the care process. Steps in creating a successful health IT system will include using simulation to validate the systems before implementation and inculcating the expectation that systems will improve with use and learning over time.

OCR for page 237
2 FOSTERING SYSTEMS CHANGE Use of data from health IT systems to model and optimize care processes would be a natural application of systems engineering to health care. One of the groups discussed the value of examining existing processes to get a better understanding of what needs to be done, what may be done better, and what may not need to be done at all and then using this evaluation as a basis to reengineer systems. Several groups shared ideas for specific projects or approaches that could take the field further down the path to greater value. Exploiting the EHR system as a resource for research through data mining was one sug- gestion. Another was to combine healthcare economics models with process engineering models in order to get a better grasp on measuring value and outlining strategies for further action. One group recommended subjecting healthcare processes in which engineering is particularly experienced, such as resource allocation and queuing prioritization, to more rigorous study through the lens of operations research. Additionally, there was widespread support for an effort to clarify nomenclature between the two fields in order to simplify future collaboration. Development of best practices that incor- porate systems engineering principles was discussed, as well as the creation of a web portal for the dissemination of these best practices; this portal could be supervised by a joint IOM/National Academy of Engineering committee or subcontracted to a university. Participants suggested that the financial engineering community should be engaged to design more effective incentives for wellness was suggested. Finally, several groups reiterated the need to better define value in the context of a learning healthcare system and from the perspective of all of the stakeholders involved. This would allow the creation of processes that measure value and make it possible to include value in decision-making processes. REFERENCES Adams, M., D. Bates, and G. Coffman. 2008. Saing lies, saing money. The impera- tie for computerized physician order entry in Massachusetts hospitals. Cambridge: Massachusetts Technology Collaborative and New England Healthcare Institute. AMIA (American Medical Informatics Association). 2006. A roadmap for national action on clinical decisions support. https://www.amia.org/files/cdsroadmap.pdf (accessed January 28, 2010). Ash, J. S., P. N. Gorman, V. Seshadri, and W. R. Hersh. 2004. Computerized physician order entry in U.S. Hospitals: Results of a 2002 survey. Journal of the American Medical In- formatics Association 11(2):95–99. Bates, D. W. 1998. Drugs and adverse drug reactions: How worried should we be? Journal of the American Medical Association 279(15):1216–1217.

OCR for page 237
28 ENGINEERING A LEARNING HEALTHCARE SYSTEM Bates, D. W., D. J. Cullen, N. Laird, L. A. Petersen, S. D. Small, D. Servi, G. Laffel, B. J. Sweitzer, B. F. Shea, R. Hallisey, M. V. V. R. Nemeskal, P. Hojnowski-Diaz, R. Stephen Petrycki, M. Cotugno, H. Patterson, B. F. Shea, M. Hickey, S. Kleefield, J. Cooper, D. J. Cullen, E. Kinneally, R. Nemeskal, B. J. Sweitzer, S. D. Small, H. J. Demonaco, M. D. Clapp, R. Hallisey, T. Gallivan, J. Ives, K. Porter, B. T. Thompson, G. Laffel, J. R. Hackman, and A. Edmondson. 1995. Incidence of adverse drug events and potential adverse drug events. Implications for prevention. ADE Prevention Study Group. Journal of the American Medical Association 274(1):29–34. Bates, D. W., L. L. Leape, D. J. Cullen, N. Laird, L. A. Petersen, J. M. Teich, E. Burdick, M. Hickey, S. Kleefield, B. Shea, M. Vander Vliet, and D. L. Seger. 1998. Effect of computer- ized physician order entry and a team intervention on prevention of serious medication errors. Journal of the American Medical Association 280(15):1311–1316. Beam, C. A., E. F. Conant, E. A. Sickles, and S. P. Weinstein. 2003. Evaluation of proscrip- proscrip- tive health care policy implementation in screening mammography. Radiology 229(2): 534–540. CBO (Congressional Budget Office). 2008. Technological change and the growth of health care spending (Pub. No. 2). Washington, DC: CBO. Classen, D. C., J. P. Burke, S. L. Pestotnik, R. S. Evans, and L. E. Stevens. 1991. Surveillance for quality assessment: IV. Surveillance using a hospital information system. Infection Control and Hospital Epidemiology 12(4):239–244. Classen, D. C., R. S. Evans, S. L. Pestotnik, S. D. Horn, R. L. Menlove, and J. P. Burke. 1992a. The timing of prophylactic administration of antibiotics and the risk of surgical-wound infection. New England Journal of Medicine 326(5):281–286. Classen, D. C., S. L. Pestotnik, R. S. Evans, and J. P. Burke. 1992b. Description of a computer- ized adverse drug event monitor using a hospital information system. Hospital Pharmacy 27(9):774, 776-779, 783. Classen, D. C., S. L. Pestotnik, R. S. Evans, J. F. Lloyd, and J. P. Burke. 1997. Adverse drug events in hospitalized patients. Excess length of stay, extra costs, and attributable mortal- ity. Journal of the American Medical Association 277(4):301–306. Crosson, J. C., P. A. Ohman-Strickland, K. A. Hahn, B. DiCicco-Bloom, E. Shaw, A. J. Orzano, and B. F. Crabtree. 2007. Electronic medical records and diabetes quality of care: Results from a sample of family medicine practices. Annals of Family Medicine 5(3):209–215. Dawson, D., and K. Reid. 1997. Fatigue, alcohol and performance impairment. Nature 388(6639):235. de Brantes, F. 2003. Bridges to excellence: A program to start closing the quality chasm in healthcare. Journal for Healthcare Quality 25(2):2, 11. Evans, R. S., S. L. Pestotnik, D. C. Classen, S. D. Horn, S. B. Bass, and J. P. Burke. 1994. Preventing adverse drug events in hospitalized patients. Annals of Pharmacotherapy 28(4):523–527. Evans, R. S., S. L. Pestotnik, D. C. Classen, T. P. Clemmer, L. K. Weaver, J. F. Orme, Jr., J. F. Lloyd, and J. P. Burke. 1998. A computer-assisted management program for antibiotics and other antiinfective agents. New England Journal of Medicine 338(4):232–238. Han, Y. Y., J. A. Carcillo, S. T. Venkataraman, R. S. Clark, R. S. Watson, T. C. Nguyen, H. Bayir, and R. A. Orr. 2005. Unexpected increased mortality after implementation of a commer- cially sold computerized physician order entry system. Pediatrics 116(6):1506–1512. Hillman, B. J., C. A. Joseph, M. R. Mabry, J. H. Sunshine, S. D. Kennedy, and M. Noether. 1990. Frequency and costs of diagnostic imaging in office practice—a comparison of self-referring and radiologist-referring physicians. New England Journal of Medicine 323(23):1604–1608.

OCR for page 237
29 FOSTERING SYSTEMS CHANGE Hillman, B. J., G. T. Olson, P. E. Griffith, J. H. Sunshine, C. A. Joseph, S. D. Kennedy, W. R. Nelson, and L. B. Bernhardt. 1992. Physicians’ utilization and charges for outpatient diagnostic imaging in a medicare population. Journal of the American Medical Associa- tion 268(15):2050–2054. Hillman, B. J., G. T. Olson, R. W. Colbert, and L. B. Bernhardt. 1995. Responses to a payment policy denying professional charges for diagnostic imaging by nonradiologist physicians. Journal of the American Medical Association 274(11):885–887. IOM (Institute of Medicine) 1997. Computer-based patient record: An essential technology for health care (revised edition). 1997. Washington, DC: National Academy Press. _____. 2001. Crossing the quality chasm: A new health systems for the 21st century. Wash- ington, DC: National Academy Press. _____. 2003. Health professions education: A bridge to quality. Washington, DC: The Na- tional Academies Press. _____. 2007. The learning healthcare system: Workshop summary. Washington, DC: The National Academies Press Kaushal, R., D. W. Bates, C. Landrigan, K. J. McKenna, M. D. Clapp, F. Federico, and D. A. Goldmann. 2001. Medication errors and adverse drug events in pediatric inpatients. Journal of the American Medical Association 285(16):2114–2120. Kilbridge, P., D. C. Classen, and E. Welebob. 2001. Oeriew of the Leapfrog Group test standard for computerized physician order entry. Report by first consulting group to the Leapfrog Group. http://www.leapfroggroup.org/media/file/Leapfrog-CPOE_Evaluation2. pdf (accessed June 21, 2010). Kilbridge, P. M., D. C. Classen, D. W. Bates, and C. R. Denham. 2006a. The national quality forum safe practice standard for computerized physician order entry: Updating a critical patient safety practice. Journal of Patient Safety 2:183–188. Kilbridge, P. M., E. M. Welebob, and D. C. Classen. 2006b. Development of the leapfrog methodology for evaluating hospital implemented inpatient computerized physician order entry systems. Quality and Safety in Health Care 15(2):81–84. Koppel, R., J. P. Metlay, A. Cohen, B. Abaluck, A. R. Localio, S. E. Kimmel, and B. L. Strom. 2005. Role of computerized physician order entry systems in facilitating medication er- rors. Journal of the American Medical Association 293(10):1197–1203. Leach, D. C., and I. Philibert. 2006. High-quality learning for high-quality health care: Getting it right. Journal of the American Medical Association 296(9):1132–1134. Leape, L. L., D. W. Bates, D. J. Cullen, J. Cooper, H. J. Demonaco, T. Gallivan, R. Hallisey, J. Ives, N. Laird, G. Laffel, R. Nemeskal, L. A. Petersen, K. Porter, D. Servi, B. F. Shea, S. D. Small, B. J. Sweitzer, B. T. Thompson, M. Vander Vliet, ADE Prevention Study Group, D. Bates, P. Hojnowski-Diaz, S. Petrycki, M. Cotugno, H. Patterson, M. Hickey, S. Kleefield, E. Kinneally, M. D. Clapp, J. R. Hackman, and A. Edmondson. 1995. Systems analysis of adverse drug events. ADE Prevention Study Group. Journal of the American Medical Association 274(1):35–43. The Leapfrog Group. 2008. Factsheet: Computer physician order entry. http://www.leapfrog- group.org/media/file/FactSheet_CPOE.pdf (accessed June 21, 2010). Masys, D. 2007. AMIA informatics 200 year in reiew. http://dbmichair.mc.vanderbilt. edu/amia2007/ (accessed January 28, 2010). Metzger, J., and F. Turisco. 2001. Computerized physician order entry: A look at the endor marketplace and getting started. The Leapfrog Group. http://www.informatics-review. com/thoughts/cpoe-leap.html (accessed June 21, 2010). Metzger, J., E. Welebob, M. Del Beccaro, and C. Spurr. 2007. Taking the measure of inpatient EHRs. Hospitals inch closer to certified products. Journal of the American Health Infor- mation Management Association 78(6):24–30; quiz, 33–24.

OCR for page 237
20 ENGINEERING A LEARNING HEALTHCARE SYSTEM Metzger, J. B., E. Welebob, F. Turisco, and D. C. Classen. 2008. The Leapfrog Group’s CPOE 2008. standard and evaluation tool. Patient Safety & Quality Healthcare 5(4):22–25. www. psqh.com/julaug08/cpoe.html (accessed January 27, 2009). Mizrahi, T. 1984. Managing medical mistakes: Ideology, insularity and accountability among internists-in-training. Social Science & Medicine 19(2):135–146. Nebeker, J. R., J. M. Hoffman, C. R. Weir, C. L. Bennett, and J. F. Hurdle. 2005. High rates of adverse drug events in a highly computerized hospital. Archies of Internal Medicine 165(10):1111–1116. Simon, S. R., R. Kaushal, P. D. Cleary, C. A. Jenter, L. A. Volk, E. J. Orav, E. Burdick, E. G. Poon, and D. W. Bates. 2007. Physicians and electronic health records: A statewide sur- vey. Archies of Internal Medicine 167(5):507–512. Smits, M., D. W. Dippel, G. G. de Haan, H. M. Dekker, P. E. Vos, D. R. Kool, P. J. Nederkoorn, P. A. Hofman, A. Twijnstra, H. L. Tanghe, and M. G. Hunink. 2005. External validation of the Canadian CT head rule and the New Orleans criteria for CT scanning in patients with minor head injury. Journal of the American Medical Associa- tion 294(12):1519–1525. Swensen, S. J., and D. A. Cortese. 2008. Transparency and the “end result idea.” Chest 133(1):233–235. Thrall, J. H. 2004. The emerging role of pay-for-performance contracting for health care services. Radiology 233(3):637–640. Ting, H. H., H. M. Krumholz, E. H. Bradley, D. C. Cone, J. P. Curtis, B. J. Drew, J. M. Field, W. J. French, W. B. Gibler, D. C. Goff, A. K. Jacobs, B. K. Nallamothu, R. E. O’Connor, and J. D. Schuur. 2008. Implementation and integration of prehospital ECGs into systems of care for acute coronary syndrome: A scientific statement from the American Heart Association Interdisciplinary Council on Quality of Care and Outcomes Research, Emer- gency Cardiovascular Care Committee, Council on Cardiovascular Nursing, and Council on Clinical Cardiology. Circulation 118(10):1066–1079. Tosczak, M. 2004. UnitedHealth trying to limit imaging use. The Business Journal December 24. http://triad.bizjournals.com/triad/stories/2004/12/27/story2.html (accessed May 31, 2010). Walsh, K. E., C. P. Landrigan, W. G. Adams, R. J. Vinci, J. B. Chessare, M. R. Cooper, P. M. Hebert, E. G. Schainker, T. J. McLaughlin, and H. Bauchner. 2008. Effect of computer order entry on prevention of serious medication errors in hospitalized children. Pediatrics 121(3):e421–e427. Wennberg, J. E., E. S. Fisher, J. S. Skinner, and K. K. Bronner. 2007a. Extending the P4P agenda, part 2: How Medicare can reduce waste and improve the care of the chronically ill. Health Affairs (Millwood) 26(6):1575–1585. Wennberg, J. E., A. M. O’Connor, E. D. Collins, and J. N. Weinstein. 2007b. Extending the P4P agenda, part 1: How Medicare can improve patient decision making and reduce unnecessary care. Health Affairs (Millwood) 26(6):1564–1574.