Appropriate to the title of the workshop, Engineering a Learning Healthcare System: A Look at the Future, the final session was devoted to exploring critical policy areas that must be engaged in order to advance engineering approaches to transformational changes in health care, including those that might trigger “disruptive innovations.” Five panelists provided context and policy recommendations, drawing from widely varying experiences in academic medical centers, community hospitals, integrated care delivery organizations, ambulatory clinics, and skilled nursing facilities: Paul F. Conlon, senior vice president for Clinical Quality and Patient Safety at Trinity Health; Denis A. Cortese, president and chief executive officer of the Mayo Clinic; Mary Jane Koren, assistant vice president of The Commonwealth Fund; Louise L. Liang, senior vice president of quality and clinical systems support for Kaiser Foundation Health Plan and Kaiser Foundation Hospitals; and Douglas W. Lowery-North, vice chair of clinical operations at Emory Healthcare Department of Emergency Medicine.
Each panelist offered brief reflections on his or her vision for changes in practice, policy, and culture. Recurring themes included the need for delivery of best practices, both clinical and administrative; process standardization and improvement at care interfaces; and leveraging human capital.
PROCESS STANDARDIZATION AND IMPROVEMENT
A key element of the panel's discussion was the notion of interfaces between engineers and providers, and among multiple processes. Cortese discussed the importance of medical school admissions selection criteria to ensure that medical education and training include fundamental engineering concepts. He provided the Mayo medical school as an example of a medical school where the size will not increase until the training program adds incrementally to historical practices. American health care is not lacking for resources, Conlon said. Those resources are probably abundant, but they suffer from poor distribution and use. He spoke of the importance of the intersection of engineering with health care in helping to build an understanding of the systems we use for creating the product that at this point is so rife with inefficiency and waste.
Lowery-North also highlighted the gap in uptake of healthcare engineers. He attributed it to language differences between medicine and health care. Fortifying the interface between health care and engineering will provide additional perspectives on the opportunities in health care for accelerated improvement. Cortese said that healthcare educators need to ensure there is a basic understanding of systems engineering in their programs and that their students, the future healthcare practitioners, need to understand how to handle data, turn it into information, and turn that information into knowledge, as well as effective communication tools. Cortese also indicated that engineering schools can play an important role in integrating health information training into engineering curriculums and master's and postgraduate programs through relationships developed with academic medical centers. The Regenstrief Institute at Purdue has one such program; other examples can be found at Georgia Tech, the University of Wisconsin, and North Carolina State.
The roles of the federal government and the private sector could be to create multidisciplinary centers to address issues of quality, value, and waste. Such centers could link the work of researchers, practitioners, educators, and engineers, and could include both basic and applied research, according to Cortese. The centers could demonstrate and disseminate tools, technologies, and knowledge, and they could perhaps identify a federal agency to take a lead role. Perhaps the government and private sources could ensure stable and adequate funding. Such an approach could help overcome barriers to the application of systems engineering, information technologies (ITs), and communication technologies, and it could play an important role in educating students. Public education would also have a role. If health care is to be improved through engineering, the government has to work to improve public education.
Reflecting on the intersection of engineering and healthcare delivery and on the kind of policies needed to help increase value, Conlon proposed using technology to hardwire some best clinical practices. For example, if a patient anywhere in a system is identified as being at risk for ulcers or falling, the act of entering that information in the electronic health record (EHR) could trigger a set of evidence-based nursing orders designed to mitigate against those risks. Individual nurses would have the opportunity to modify those orders as appropriate. Using technology to facilitate coordination of care is vitally important, yet Conlon expressed concern over the debate with IT vendors about whether they are implementing the right information in the right systems for the changes needed.
Liang addressed the measurements, measuring systems, and metrics used by large health plans and purchasers to identify process measures, all of them driven by claims data. Such data are available, Liang noted, but the measures only feed current activities instead of encouraging better outcomes or processes. Data collection is time intensive, Liang noted, and she urged an examination of the benefits provided vs. the burden created in achieving those benefits. Cortese added another layer by challenging the Joint Commission to completely change the way it does business and instead become a conduit for sharing information—a reporting center that could encourage the learning process throughout health care.
Another basic challenge to system improvement is the adoption of health IT. For example, nursing homes have not been at the technological forefront in terms of IT, Koren said. Although the homes are starting to use IT, it is a disruptive innovation, and too many of them see it as something that you buy, you plug in, and then you teach somebody to press the button. Few vendors engage the possibilities of IT as a change management tool. Therefore, Koren said, we need to think about teaching nursing homes how to do things like process mapping and workflow design and to use those tools to optimal advantage. And because doctors often do not like to go to nursing homes, more effective ways are needed to use telehealth and telemonitoring tools to ensure that patients get the best medical care even at those times when a physician is not present.
Delivery of Value
Koren also highlighted some opportunities for skilled nursing care facilities to increase the value derived from services. She urged the audience to consider the best ways to design facilities, employing engineering systems insights that support caring for a frail, usually older population. Cortese followed up on this line of thinking by advocating that the federal government simply pay for value, a policy proposal that has gained significant traction as offering an alternative to reimbursement. Liang discussed the idea of generating value through the creation and use of medical knowledge, and she noted the significant workshop conversation about the barrier of financial incentives. Right now the healthcare system pays for activity and, as expected, activity is the result. However, value, quality, service, and better outcomes should be the focus of reimbursement.
For full attainment of the value potential, Liang said that better use of informatics is needed, and this means that the federal government needs to do more to address in a straightforward fashion the privacy concerns of institutions and patients that arise from the use of information for clinical knowledge generation. In light of the recent challenges experienced by other institutions, she said, legitimate privacy concerns need to be clarified in order to allow and support full leverage of the significant information becoming available to the healthcare community.
LEVERAGING PEOPLE FOR HEALTHCARE IMPROVEMENT
Culture and the Learning Process
Culture is generally the most important barrier to change, Cortese observed, and this is especially true with health care. Lowery-North cautioned against losing the component of human systems engineering when evaluating engineering approaches to culture change. Organizational composition, diffusion of innovation, and change management strategies are areas in which health care continues to lag behind other sectors, Lowery-North said, and this is probably why it has been so difficult to effect change. The pharmaceutical sales industry can offer insight into how to change physician behavior, panelists said. Examining the experiences of that group of people may offer some lessons about how to change the behaviors of physicians, who in practice may have little, if any, incentive to change. Cortese noted that one facet of necessary cultural change is found in the current emphasis on research and even teaching over patient care. In academic centers, most often the motivator is research; many academic medical centers exist because of research. If someone has full funding, they get a tenured position; if they lose funding, they are often required to dedicate more time to teaching. In the face of such concerns, it is unfortunately the case that medical care is totally secondary. Rather than separate the two, Cortese said, what is important is to draw research and patient care closer together so that every patient experience becomes a learning opportunity.
Conlon identified the omnipresent measurement culture, monitoring the effectiveness of the healthcare system and identifying opportunities for improvement, as a barrier to change, explaining that the growing burden of data capture is potentially beginning to exceed the value of information. That is a problem because nurses and physicians spend too much time marking off checklists in order to be able to prove, for example, that ACE inhibitors or beta-blockers were used properly. This is a great opportunity to automate that data capture—and correspondingly shift the culture—so that practitioners can devote more of their time directly to patient care.
Conlon noted that EHR implementation is about more than simply documenting health information in EHRs; it can serve as a catalyst to really transform how we deliver care. The records provide the opportunity to actually look at processes of care and to redesign them. Liang also offered a word of caution by citing the work of Ronald Heifetz, who wrote Leadership Without Easy Answers: “One of the most common leadership mistakes is expecting technical solution to solve adaptive problems” (Heifetz, 1994).
Liang also argued that cultural challenges are a major issue everywhere, even in the Kaiser Permanente system in which physicians and nurses are more aligned than perhaps anywhere else. The fact remains that the fundamental guild or craftsman culture of healthcare professionals is still a significant problem, Liang said. For Kaiser, the biggest factor enabling that culture change has been the availability of transparent, specific data that are comparable across the organization and which allow different locations of care to see what is possible in other parts of the organization and what is possible in terms of national benchmarking—as well as where their individual performances stand. That information needs to be made available, but it also needs to be combined with good evidence about the right pathway and with contextual knowledge about how the clinic next door does so much better. Such context is not necessarily found in the data; the data say where to go look, where to have the conversations about exactly what someone is doing that could offer lessons to others.
Occasionally it takes people time to accept the data. There are times when some physicians and departments have to go through a dialogue of “The data [are] wrong, my patients are sicker, you just don't understand,” but eventually they come to accept the system. Kaiser has seen a huge decrease in its variation and a large improvement overall, Liang said, based fundamentally on the availability of the data to identify issues and help people grapple with the fact that, at the moment, everything is not possible.
Communicating With and Engaging Patients
Communication was a central point of discussion in the final session of the workshop. Three primary communication themes arose: (1) interoperability of systems in order to facilitate communication, (2) communication between care team members, and (3) enlisting the patient in support of knowledge development.
Noting the lack of portability of patient records among care settings, patients, and providers, Conlon discussed the predominance of incompatible software that effectively precludes information sharing. There is a great policy opportunity for consistency in the interoperability of these systems and in the exchange of the information associated with them. Conlon also urged the adoption of policies that enable information to follow patients and to exist in a form that can be easily shared and transferred, regardless of location. Liang added that the federal government should set interoperability standards, particularly in areas that hold outstanding promise, such as home monitoring and other similar medical devices, as well as standards that will help give all patients the right to take their medical records with them. Right now these records can be provided in print, on a compact disc, or on a memory stick, but that is still a far cry from what it should be. Lowery-North emphasized that developing interoperability standards will be critical. A favorable factor is that physician practitioners are gradually migrating to larger group settings, which are more likely to adopt information systems that have interoperability standards because they are more likely to have been created with a systems engineering approach in mind.
Another area with policy implications is the team nature of health care; however, we are not teaching people to work in teams, according to Koren. Educational policy is needed to make sure that we have people who are skilled and working in an interdisciplinary or multidisciplinary manner. Moving doctors into such teams is an innate problem because, for example, doctors typically don't like to go to nursing homes. These teams are largely led by nurses, with paraprofessionals working at the bottom. An issue, therefore, is how we can integrate those paraprofessionals into that team. We need engineering to help us think about how to bring workers into that team and effectively listen to the knowledge being generated at the front lines of care. How do we best use that knowledge to make the system better and more responsive to what people want?
Finally, it was noted that if patients are to become more engaged in the research process, several rules will have to change. Privacy remains a barrier to knowledge generation from patient data, yet the concerns are largely perceptual in nature. Research is an opportunity to shift the culture in health care through getting people to understand that research in the name of patient care improvement is legitimate, publishable, hypothesis-testing research.
RECURRING THEMES FOR ROUNDTABLE ATTENTION
The presentations and discussions within the workshop zeroed in on a number of specific ideas and themes concerning the best ways to use engineering to improve healthcare delivery. In addition, they provided a variety of insights into engineering approaches to dealing with systems complexity and identified critical areas needing attention in health care. The recurring themes of discussion throughout the 2 days of the workshop are summarized below. While perhaps intuitively obvious—hence the reason for their recurrence—they were nonetheless noted as worthy of attention and engagement by the Roundtable members.
- The system's processes must be centered on the right target—the patient. Patient-centered care was defined in the 2001 Institute of Medicine (IOM) report Crossing the Quality Chasm as providing care that is respectful of and responsive to individual patient preferences, needs, and values and ensuring that patient values guide all clinical decisions (IOM, 2001). However, health care is by nature highly complex, involving multiple participants and parallel activities that sometimes take on a character of their own, independent of patient needs or desires. Throughout several sessions, workshop participants emphasized the need to ensure that processes support patients—and that patients are not forced into processes. Patient needs and perspectives must be at the center of all process design, technology application, and clinician engagement.
- System excellence is created by the reliable delivery of established best practice. Identifying and embedding practices that work best, and developing the system processes to ensure their delivery every time, help to define excellence in system performance and to focus the system on delivering the best possible care for patients. In health care, establishing practices from the best available evidence and building them as routines into practice patterns, as well as developing systems to document results and update best practices as the evidence evolves, will integrate some of the best elements from the engineering disciplines into healthcare issues. Participants often cited the need for better integration of the development and communication of best practices in healthcare systems, as well as the need for process systems to track care details and outcomes, with feedback for practice refinement and better patient outcomes.
- Complexity compels reasoned allowance for tailored adjustments. Established routines may need circumstance-specific adjustments related to differences in the appropriateness of established health-care regimens for various individuals, variations in caregiver skill, the evolving nature of the science base—or all three. Mass customization and other engineering practices can help assure a consistency that can accelerate the recognition of the need for tailoring and delivering the most appropriate care—with the best prospects for improved outcomes—for the patient. Participants pointed to the need for the development of a system of care flexible enough to incorporate these considerations and to leverage the lessons learned from their employment in a process of continuous learning.
- Learning is a non-linear process. The focus on an established hierarchy of scientific evidence as a basis for evaluation and decision making cannot fully accommodate the fact that much of the sound learning in complex systems occurs in local and individual settings. Participants cited the need to bridge the gap between dependence on formal trials, such as randomized controlled trials, and the experience of local improvement in order to speed learning and avoid impractical costs.
- Emphasize interdependence and tend to the process interfaces. A system is most vulnerable at links between critical processes. In health care, attention to the nature of relationships and hand-offs between elements of the patient care and administrative processes is therefore vital and a crucial component of focusing the process on the patient experience and improving outcomes.
- Teamwork and cross-checks trump command and control. Especially in systems designed to guarantee safety, system performance that is effective and efficient requires careful coordination and teamwork as well as a culture that encourages parity among all those with established responsibilities. During the workshop, several examples were cited of other industries that have used systems design and social engineering to better integrate and strengthen their systems processes with great improvements in efficiency and safety.
- Performance, transparency, and feedback serve as the engine for improvement. Continuous learning and improvement in patient care requires transparency in processes and outcomes as well as the ability to capture feedback and make adjustments.
- Expect errors in the performance of individuals, perfection in the performance of systems. Human error is inevitable in any system, and should be assumed. On the other hand, safeguards and designed redundancies can deliver perfection in system performance. Mapping processes, embedding prompts, cross-checks, and information loops can assure best outcomes and allow human capacity to focus on what can not be programmed—compassion and individual patient needs. Several workshop presentations shared success stories and lessons learned from other industries, such as the automotive and airline industries, that have effectively incorporated this strategy.
- Align rewards on the key elements of continuous improvement. Incentives, standards, and measurement requirements can serve as powerful change agents. Therefore, it is vital that they be carefully considered and directed to the targets most important to improving the patient and provider experiences. Participants noted that it is vital that incentives be carefully considered and directed to the targets most important to improving the efficiency, effectiveness, and safety of the system—and ultimately patient outcomes—as well as taking into consideration the patient and provider experiences.
- Education and research can facilitate understanding and partnerships between engineering and the health professions. The relevance of systems engineering principles to health care and the impressive transformation brought to other industries, speaks to the merits of developing common vocabularies, concepts, and ongoing joint education and research activities that help generate stronger questions and solutions. Workshop participants pointed to the dearth of training opportunities bridging these two professions and spoke of the need to encourage greater collaborative work between them.
- Foster a leadership culture, language, and style that reinforce teamwork and results. Positive leadership cultures foster and celebrate consensus goals, teamwork, multidisciplinary efforts, transparency, and continuous monitoring and improvement. In citing examples of successful learning systems, participants highlighted the need for a supportive and integrated leadership.
AREAS FOR INNOVATION AND COLLABORATIVE ACTION
Presentations and discussions during the workshop offered insight into the opportunities for Roundtable members to consider possible follow-up actions for ongoing multi-stakeholder involvement to advance the integration of engineering sciences into healthcare systems improvement.
Discussions during the breakout sessions provided the opportunity for workshop attendees, in both the health and engineering fields, to engage with each other and identify novel opportunities for innovative work that might yield breakthroughs that capture more value in health care. Participants felt that the opportunities were great for various engineering approaches to streamline processes and improve efficiency, but they struggled with the ambiguity of the definition of value in health care. The result was that they largely referred back to themes covered in the workshop presentations and summarized elsewhere in this summary. This suggests that there is still much work to do in laying a foundation at the intersection of engineering and health care if drilling down with greater specificity is to add substantially to value.
That said, workshop participants identified several areas for collaborative work that merit follow-up. With particular emphasis on the need for ongoing means of communication and collaboration that will bring better perspective and nurtured understanding from the two fields, areas mentioned for possible Roundtable follow-up include the following:
- Clarify terms: The ability of healthcare professionals to draw upon relevant and helpful engineering principles for system improvement could be facilitated by a better mutual understanding of the terminology. A collaborative effort by the IOM and the National Academy of Engineering could create a targeted glossary and develop potentially bridging terminology for use as appropriate.
- Identify best practices: Three areas of systems orientation are particularly important to improving the efficiency and effectiveness of health care: (1) focusing the system elements more directly on the key outcome—the patient experience, (2) ensuring transparency in the performance of the system and its players and components, and (3) establishing a culture that emphasizes teamwork, consistency, and excellence. Progress could be accelerated by identifying and disseminating examples of best practices from health care and from engineering on each of these dimensions.
- Explore health professions education change: In the face of a rapidly changing environment in health care—expanding diagnostic and treatment options, much greater knowledge available, movement beyond the point at which any one individual can personally hold all the information necessary, and IT that opens new capabilities—changes to the education of health professionals can advance caregiver skills in knowledge navigation, teamwork, patient-provider partnership, and process awareness.
- Advance the science of payment for value: With cost increases in health care consistently outstripping gains in performance by most measures, progress toward counteracting this trend could be achieved with a stronger focus on ways to enhance both health and economic returns from healthcare investments. This could include work in the areas of understanding, measuring, and providing incentives for value in health care.
- Explore fostering the development of a science of waste assessment and engagement: Similarly, and directly related, an exploration of the elements of inefficiency in health care, how to define and measure waste, and how to mobilize responses to eliminating waste could contribute to increasing value within healthcare systems.
- Support the development of a robust health IT system: The development of a health IT system, designed with systems-related continuous improvement principles in mind, must lie at the core of an efficient, effective learning system. Beginning with challenges to EHR adoption, much work remains in order to achieve such a system that allows for continuous learning; permits data sharing, including the construction of databases; employs consistent standards; and addresses privacy and security concerns. Health IT is a natural place for collaborative work between engineers and caregivers, beginning with better resolution of barriers to the achievement of such a system through the employment of both expert lenses.
As healthcare and engineering professionals consider these areas for collaboration and innovation, it is important to emphasize that the focus of all the engineering applications to health care discussed in the workshop was, ultimately, improving patient outcomes. The reforms that were discussed are all focused on to bringing the right care to the right person at the right time at the right price. The essential questions are straightforward: Can it work? Will it work—for this patient, in this setting? Do the benefits outweigh any harms? Do the benefits justify the costs? Do the reforms offer important advantages over existing alternatives?
If full advantage is to be taken of this potential, much work remains to bridge the gaps between the professions of health care and engineering. As the problems within healthcare systems become increasingly better defined, the opportunity increases for true collaborative approaches that go beyond joint acknowledgment and parallel approaches. This workshop, while limited by the chosen areas of emphasis and the specific backgrounds of the participants, identified a number of important prospects for advancing the discussion and sharing of ideas as a more frequent and routine activity.
Better coordination, collaboration, public-private partnerships, and priority setting are central challenges for the U.S. healthcare system. The discussions summarized in this report highlight engineering's potential contribution to progress toward the Roundtable membership's concept of a learning health system with a stated goal: that by the year 2020, 90 percent of clinical decisions will be supported by accurate, timely, and up-to-date clinical information and will reflect the best available evidence.
- Heifetz RA. Leadership without easy answers. Cambridge, MA: Belknap Press of Harvard University Press; 1994.
- IOM (Institute of Medicine) Crossing the quality chasm: A new health systems for the 21st century. Washington, DC: National Academy Press; 2001.
National Academies Press (US), Washington (DC)
Institute of Medicine (US) and National Academy of Engineering (US) Roundtable on Value & Science-Driven Health Care. Engineering a Learning Healthcare System: A Look at the Future: Workshop Summary. Washington (DC): National Academies Press (US); 2011. 6, Next Steps: Aligning Policies with Leadership Opportunities.