Building a Better Delivery System: A New Engineering/Health Care Partnership (2005)

The preceding chapter describes an array of systems-engineering tools and associated techniques for analyzing, designing, controlling, and improving health care delivery processes and systems. This chapter is focused on the application of information and communications technologies to the delivery of safe, effective, timely, patient-centered, efficient, and equitable health care, a review of progress toward the establishment of a National Health Information Infrastructure (NHII), and a description of the tasks that lie ahead. The committee highlights the complementary nature of information/ communications technologies and systems engineering.

Information and information exchange are crucial to the delivery of care on all levels of the health care delivery system—the patient, the care team, the health care organization, and the encompassing political-economic environment. To diagnose and treat individual patients effectively, individual care providers and care teams must have access to at least three major types of clinical information—the patient’s health record, the rapidly changing medical-evidence base, and provider orders guiding the process of patient care. In addition, they need information on patient preferences and values and important administrative information, such as the status and availability of supporting resources (personnel, hospital beds, etc.).

To integrate these critical information streams, they will also need training/education, decision-support, information-management, and communications tools. For individual patients to participate as informed, “controlling” partners in the design and administration of their own care, they must also have access to much the same kind of information and education, decision-support, and communications tools—in a “patient-accessible/usable” form.

At the organizational level, hospitals and clinics need clinical, financial, and administrative data/information to measure, assess, control, and improve the quality and productivity of their operations. At the environmental level, federal/state funding and regulatory agencies and research institutions need information on the health status of populations and the quality and productivity/performance of care providers and organizations to execute regulatory oversight, protect and advance the public health (surveillance/monitoring), evaluate new forms of care, accelerate research, and disseminate new medical knowledge/evidence.

As discussed in Chapter 3, information and information exchange are also critical to the tactical and strategic applications of systems-engineering tools at all four levels of the system, especially for strategic applications of enterprise-management tools and risk analysis and management tools at the organizational and environmental levels.

Although information gathering, processing, communication, and management are essential to health care delivery, the health care sector as a whole has historically trailed far behind most other industries in investments in information/ communications technologies (DOC, 1999). Moreover, most health care-related information/communications technologies investments to date have been concentrated on the administrative side of the business, rather than on clinical care. As a result of this prolonged underinvestment, little overall progress has been made toward meeting the information needs of patients, providers, hospitals, clinics, and the broad regulatory, financial, and research environment in which they operate. A number of localized efforts have been made to develop and implement electronic patient records and other clinical applications of information/communications technologies since the 1960s, but little progress has been made in closing the gap.

Many factors have contributed to the information/ communications technology deficit: (1) the atomistic structure of the industry (the prevalence of relatively undercapitalized small businesses/provider groups); (2) payment/reimbursement regimes and the lack of transparency in the market for health care services, both of which have discouraged private-sector investment in information/communications systems; (3) historical weaknesses in the managerial culture for health care; (4) cultural and organizational barriers related to the hierarchical nature and rigid division of labor in health professions; and (5) the relative technical/functional immaturity (until very recently) of available commercial clinical information/communications systems.

The idea of transforming paper medical records into electronic medical records (EMRs) was first considered in the mid-1960s, when early prototype systems were developed. A number of large integrated health care provider organizations were early adopters of EMR systems, including Massachusetts General Hospital (COSTAR) in the 1960s, Indiana University Medical School (Regenstrief Medical Record System) in the early 1970s, and others (Kass-Bartelmes et al., 2002; Lindberg, 1979). However, there was little diffusion of these systems in the next two decades. In 1991 and 1997, IOM issued reports documenting the magnitude and implications of the large information-technology gap in U.S. health care and called for the adoption of EMRs as a first, critical step in moving health care delivery toward information/communications-technology-supported improvements in quality performance achieved in other industries (IOM, 1991, 1997).

Building on these studies, a series of reports by IOM, the National Committee on Vital and Health Statistics (NCVHS), and other organizations in the past five years have documented the profound negative impact of the information/ communications technology deficit on patient safety, the number of medical errors, and the quality and cost of care; every one of these reports calls for the development of a comprehensive health care information infrastructure (e.g., NHII) to help close the gap (IOM, 2000, 2001, 2003, 2004; NCHVS, 2001; NRC, 2000).

In Information for Health: A Strategy for Building the National Health Information Infrastructure, NCVHS described the NHII as both infrastructure and a defined set of components linked explicitly to health care delivery processes (NCVHS, 2001). IOM (2004) summarized the NCVHS definition as follows:

This stream of reports from IOM, NCVHS, and others catalyzed a number of actions in the private and public sectors intended to lay the groundwork for and build momentum toward realization of the NHII (IOM, 2004; PITAC, 2004; Thompson and Brailer, 2004; Yasnoff et al., 2004). Inspired by the 1999 IOM report, To Err Is Human, the Leapfrog Group for Patient Safety, a coalition of large companies established expressly for the purpose of using their market power as major purchasers of health care to encourage care providers to improve the safety, quality, and efficiency of health care. The Leapfrog Group called on all health care provider organizations serving Leapfrog members’ employees to use information/communications systems (EMRs and computerized physician order entry [CPOE] systems in particular) (see paper by Milstein in this volume).

In April 2004, progress toward an NHII was given new impetus when President Bush called for national implementation of EMRs and announced the creation of the Office of the National Coordinator for Health Information Technology (ONCHIT) in the U.S. Department of Health and Human Services (DHHS); Dr. David Brailer was appointed the first national coordinator. In July, DHHS released a report outlining a 10-year plan to build an NHII, including the creation of electronic health records (EHRs), for all Americans (Thompson and Brailer, 2004). In November 2004, ONCHIT issued a Request for Information (RFI) for a National Health Information Network (NHIN), soliciting proposals for ways to advance interoperability and standards. As of early 2005, ONCHIT had received more than 500 responses from a wide variety of organizations and collaboratives.

One of the respondents to the RFI, the Interoperability Consortium, an alliance of eight information-technology systems vendors (Accenture, Cisco, CSC, Hewlett-Packard, IBM, Intel, Microsoft, and Oracle), describes the current challenges to interoperability:

In its preliminary blueprint for NHIN, the Interoperability Consortium (2005) stresses that the NHIN must be part of an

agenda for the comprehensive transformation of health care delivery:

To meet these requirements, the NHII/NHIN must be a secure, reliable, and adaptable national infrastructure capable of connecting and supporting highly distributed, varied, independently managed, multi-tiered, intra-institutional, clinical information/communications technology systems and applications. This infrastructure would vastly expand the information gathering, exchange, processing, and application capabilities of stakeholders at all four levels of the health care system.

The NHII would provide a platform for the application of a wide range of proven and emerging information/ communications technologies that could have a dramatic impact on health care processes and outcomes. The following discussion explores the promise of an NHII for each level of the health care delivery system.

At the patient level, progress toward an NHII would greatly empower individual patients to assume a much more active, controlling role in decision making and in implementing their own health care (i.e., applications that could help bring about a shift from hospital/clinic-based, clinician-directed care to home-based, clinician-guided self-care). The foundations for this shift have been laid by the emergence of the Internet and the World Wide Web, which have provided patients with unprecedented access to information (albeit of mixed quality) and made possible more continuous, asynchronous communication between patients and care providers.

Progress in systems interoperability and data standards is likely to advance the development of patient remote access to self-care educational tools, individual patient health records, and health care provider and insurer services (scheduling, billing, etc.) (see papers by Gustafson and Halamka in this volume). In time, the NHII would also provide a platform for the implementation of new information/communications systems, such as wireless integrated microsystems (WIMS, sensors combined with microelectronics and wireless interfaces), which would enable the remote capture and continuous communication of a patient’s physiological data to care professionals, thereby increasing the likelihood of the timely diagnosis and treatment of illnesses.

An improvement in patients’ ability to assume greater control and responsibility for care decisions enabled by information/communications technologies would also advance many of IOM’s six aims for patient-centered, quality health care. Information/communications technology systems would give patients access to timely, effective, and convenient care; improve patient compliance with guidance/treatment protocols, including preventive measures; and enable continuous, or at least much more frequent, monitoring of patient conditions by care professionals/care teams. Greater compliance with clinicians’ guidance—preventive or palliative—and more timely intervention in case of illness would not only benefit the health of the patient but would also reduce the costs of caring for the patient over time.

At the care team level, progress toward an NHII would accelerate the development, diffusion, and use of a broad spectrum of information/communications technologies to help care providers capture, tap into, and integrate critical information streams for patient-centered care—the patient’s health record, information on the patient’s preferences and values, the evolving medical-evidence base, the status of clinical orders, administrative information, and a range of process/system performance data—essentially all of the data and information necessary to diagnose and prescribe treatment, as well as to analyze, control, and optimize the performance of the delivery system and subsystems.

Over the past decade, several core clinical applications have been developed to support the clinical information needs of frontline care teams. These include, EHR systems linking various information resources related to clinical care; CPOE systems, through which physicians enter orders for tests, drugs, and other procedures; decision-support tools that draw on clinical-data repositories, and databases that collect and store patient care information from diverse data sources.

Although the utility and functionality of these first-generation core clinical applications have been severely limited by the absence of comprehensive clinical information systems throughout much of the health care delivery system, progress toward the NHII would lead to the development and implementation of next-generation clinical applications that are more fully integrated and capable of translating clinicians’ orders into dynamic, automated execution routines, as well as tracking and notifying clinicians of the status of their patients automatically. These applications could lead to changes in the role of the care team and individual care professionals, enabling them to spend less time executing and verifying the execution of

orders and more time focusing on healing relationships with individual patients. Implementation of these technologies would also facilitate continuous learning in the care delivery system.

At the level of the organization, steps toward an NHII would greatly facilitate the capture, integration, and analysis of clinical, administrative, and financial data for measuring and improving the quality, patient-centeredness, and efficiency of health care. As noted in Chapter 3, integration is essential to the application of data-intensive systems tools for systems design, analysis, and control. Beginning in the 1980s, a select group of health care provider organizations and networks began the integration process by adding clinical-department systems to their billing and administrative systems. It is worth noting that most of these forerunner, integrated systems were used by organizations with corporate-type structures and management (e.g., the Mayo Clinic, Kaiser-Permanente, Veterans Health Administration, and others with salaried physicians and wholly owned hospitals and ancillary functions). Only in the last decade have leading hospitals and integrated institutions begun to leverage their information systems by adapting and deploying systems-engineering tools and techniques to analyze, control, and optimize aspects of their operations.

As NHII (and interoperability and data-interchange standards in particular) advances, more and more health care organizations would be able integrate their clinical, administrative, and financial information systems internally, as well as link their systems with those of insurers, vendors, regulatory bodies, and other elements of the extended health care delivery enterprise. This capacity, in turn, would enable provider organizations to make more extensive global or strategic use of data/information-intensive systems-engineering tools, such as enterprise management, financial engineering for risk management, and knowledge discovery in databases.

The NHII would lead to significant improvements on the environmental level of the health care delivery system. With advances in interoperability standards and other tools and technologies, the NHII would enable connectivity both within and across levels of the delivery system. This, in turn, would facilitate the aggregation and more timely exchange of useful data between and among providers at the organizational level and elements/stakeholder organizations at the environmental level (i.e., public and private payer organizations [insurers, employers], regulatory bodies, and the research community).

A functioning NHII could provide a rich pool of data to support regulation and oversight of the health care delivery system, population health surveillance, and the continuing development of the clinical knowledge/research database. For example, the NHII could accelerate the flow of health care quality data from providers to the Center for Medicare and Medicaid Services and private insurers, data on evidence-based-medicine trials to the Agency for Healthcare Research and Quality, data on infectious diseases and bio-hazards to the Centers for Disease Control, and data on post-introduction adverse drug events to the Food and Drug Administration (FDA).

NHII could also accelerate the interfacing of the expanding genomic and phenotypic (clinical) knowledge databases. The application of high-level systems-engineering tools (risk analysis) to these massive linked data sets could support significant advances in “predictive medicine”—mathematical and statistical techniques to identify and treat high-risk patients and to personalize treatment strategies.

Although much of the information/communications technology necessary for the realization of NHII exists today, and will certainly improve in the decade ahead, there will be many challenges to putting it in place. Very serious privacy concerns must be addressed, as well as training issues at all levels of the health care system. There are also serious challenges associated with making information/communications systems reliable enough to ensure that records are not lost. Ensuring reliability will require a very large-scale distributed computing system.

Paper-based systems are still the norm at most hospitals, which are all but “drowning” in paperwork. Clearly, it will take a national effort to develop an infrastructure capable of connecting, integrating, and supporting diverse information systems and applications at facilities nationwide. Although individual functions might still vary from facility to facility, the operating framework used for storing records and the protocols by which information is passed between locations and systems must be standardized.

Indeed, interoperability among diverse information/ communications systems and messaging standards will be critical to the realization of an information/communications technology-enabled health care system, a programmable system with the capacity for mass customization to meet the needs of individual patients. At every level of the health care system, the focus should be on the patient, and the goal should be to ensure effective interactions between the patient and doctor or health care delivery team. Developing such a system in the coming decade is not an option. It is an absolute necessity for achieving the IOM vision of a patient-centered, high quality health care system.

The remainder of this chapter is divided into two sections. The first focuses on the current status of major components of the emerging NHII, identifies technical challenges and opportunities, identifies economic and cultural/organizational barriers to implementation, and provides recommendations for building on current momentum. The second focuses on emerging technologies based on wireless communications and microelectronic systems that have the potential to

radically change the structure of the health care delivery system and advance the patient-centeredness and quality performance of the system. Although the widespread implementation of emerging technologies represents a longer term agenda than upgrading and/or diffusing existing clinical information/communications technology applications, the NHII has a 10-year time horizon for realization that can accommodate the incorporation of new technologies. Above all, the implementation of NHII must be part of a comprehensive transformation of health care delivery.

The components of a national health information infrastructure can be divided into three interrelated categories: (1) health care data standards and technical infrastructure; (2) core clinical applications, including EHRs, CPOE systems, digital sources of medical knowledge, and decision-support tools; and (3) information/communications systems.

If health care data are standardized, they become understandable to all users. The IOM report (2004), Patient Safety, considered three key groups of standards:

• Data interchange formats are standard formats for electronically encoding data elements (including sequencing and error handling). Interchange standards can also include document architectures for structuring data elements as they are exchanged and information models that define relationships among data elements in a message.

• Terminologies are the medical terms and concepts used to describe, classify, and code the data elements and data-expression languages and syntax that describe relationships among the terms/concepts.

• Knowledge representation refers to standard methods of electronically representing medical literature, clinical guidelines, and other information required for computerized decision support.

For each group of standards, IOM identified critical challenges and described ongoing efforts led and/or funded by the federal government to address them. In the area of data-interchange formats, in which engineering has played an important role, a number of mature standards, recently endorsed by the secretary of DHHS, address some of the required domains:

• administrative data (the X12 standard of the Accrediting Standards Committee, Subcommittee on Insurance, Working Group 12)

• clinical data (Health Level 7)

• medical images (digital imaging and communications in medicine [DICOM])

• prescription data (National Council for Prescription Drug Programs [NCPDP] Script)

• medical device data (Institute for Electrical and Electronics Engineers [IEEE] Standard 1073)

In its data standards “action plan,” IOM called for the rapid development of the next version (version 3.0) of the Health Level 7 clinical-data standards “to support increased interoperability of systems and comparability of clinical data, as well as patient safety,” and underscored the need for “implementation guides and conformance testing/ certification procedures…to insure consistent application of the standards in commercial systems” (IOM, 2004).

In the area of medical terminologies, IOM called for the identification of a “core group of well-integrated, non-redundant clinical terminologies…needed to serve as the backbone of clinical information and patient safety systems.” With respect to knowledge representation, IOM identified a need for standards “for the representation of clinical guidelines and the implementation of automated triggers” (IOM, 2004).

To accelerate the development and adoption of health care data standards, IOM recommended a significant increase in the technical and material support provided by the federal government to ongoing public-private partnerships in this area (IOM, 2004). IOM also put forward a six-point federal government “work plan.”¹ As noted above, the establishment of ONCHIT and the subsequent RFI were focused on interoperability and standards for an NHIN, demonstrating the urgency of the clinical information/communications

technology challenge at the national level and the need for renewed efforts to engage the private sector in developing solutions.

To ensure that the emerging NHII can support next-generation clinical information systems and applications, it is critical that research on advanced interface standards and protocols continue apace and that standards-related issues concerning the protection of data integrity, controlled access to data, data security, and the integration of large-scale wireless communications be addressed early on. There is also a pressing need for low-cost tools for standardizing new and legacy digital data without disrupting the clinical work flow (PITAC, 2004). Other industries that had to accommodate conflicting standards (e.g., computer networks and computer graphic design) used translators to allow the best standard to emerge. Stable funding for research in all of these areas will be essential.

These challenges are neither new nor unique to health care. Indeed, engineers, computer scientists, and researchers and practitioners in other disciplines have been working on them for more than a decade to meet the needs of financial services, telecommunications, and national defense. Much of this work has been supported by federal research and mission agencies (NITRD, 2004). Cross-sector research and learning in the area of information/communications technology standards among federal agencies, health care insurers, and health care providers represents a potentially vast source of knowledge and advancement. To realize this potential, the President’s Information Technology Advisory Council has called for increased coordination of federally supported research and development related to standards, computer infrastructure, privacy issues, security issues, and other topics relevant to health care through the Networking and Information Technology R&D (NITRD) Program, an 11-agency program that includes NSF, National Institutes of Health, Agency for Health Care Research and Quality, National Institute of Standards and Technology, Defense Advanced Research Projects Agency, U.S. Department of Energy, and others (NITRD, 2004).

Clinical information systems provide a mechanism for sharing data collected from various sources (e.g., EHRs in care settings that may include personal health record systems maintained by patients or their representatives). Data become available to clinical information systems via direct entry at the point of care, off-line entry through abstraction from other media, such as handwritten notes, and data collected by other systems, such as laboratory systems or monitoring devices. The data can take many forms—including free text, coded data, speech, document imaging, clinical imaging (e.g., x-rays), and video. In the following section, four core components of clinical information systems are described: (1) EHRs; (2) CPOE systems; (3) digital sources of medical evidence; and (4) decision-support tools. These descriptions are followed by a discussion of human/ information systems interface design and software dependability issues.

The electronic capture of patient-specific clinical information is critical to many health care information/ communications technology applications. Attention has been focused in the creation of EHRs since the 1960s, and in 1991, IOM set forth a vision and issued a call for nationwide implementation of computer-based patient records that would be paperless and instantly available throughout the health care system in forms readily understandable to physicians and other providers at point of care and specialists, perhaps in a different location (IOM, 1991). However, the rate of progress toward realizing this vision has been glacial.

Only a fraction of hospitals have implemented comprehensive EHR systems, although many have made progress in certain areas, such as computerized reporting of laboratory results (Brailer, 2003). Rates of adoption of EHR systems are higher in ambulatory care settings—probably about 5 to 10 percent of physician’s offices—but these systems vary greatly in content and functionality (IOM, 2004). Although some cases of failed EHR systems have been documented, many more examples show cost savings and quality improvements yielded by EHR systems (Clayton in this volume; Littlejohns et al., 2003; Pestotnik et al., 1996; Wang et al. 2003).

EHRs have been instituted in health care settings in the public and private sectors, and a few communities and systems have implemented secure systems for the exchange of data among providers, suppliers, patients, and other authorized users. Among these are the Veterans Health Administration (see Box 4-1), Mayo Clinic (see Box 4-2), New England Healthcare Electronic Data Interchange Network, Indiana Network for Patient Care, Santa Barbara County Care Data Exchange, Patient Safety Institute’s National Benefit Trust Network, and the Markle Foundation Healthcare Collaborative Network (CareScience, 2003; Kolodner and Douglas, 1997; Markle Foundation, 2003; New England Healthcare EDI Network, 2002; Overhage, 2003; Patient Safety Institute, 2002; Zachariah in this volume).

All of these are exceptions to the rule, however. In most hospitals, orders for medications, laboratory tests, and other services are still written on paper, and many hospitals do not even have the capability of delivering laboratory results and other test results in automated form. The same situation prevails in most small practice settings, where little if any progress has been made toward creating electronic records (IOM, 2004).

A patient’s EHR must also include long-term data and information about the patient’s daily life. This information will be useful not only in the planning and delivery of

progressive care, but will also provide evidence for assessing different clinical interventions. Systems-engineering tools and techniques are available for modeling and determining the information needs of a “system” that can deliver progressive care and evaluate that system’s performance.

Patient-centered health care delivery in the broadest sense must also focus on what the patient really wants from the entire health care community—the best physical and mental function in daily living possible within the constraints of the patient’s physical condition. The key word here is “system,” that is, coordinated care, including care in the clinic, the hospital, home, rehabilitation facility, skilled nursing facility, long-term care facility, hospice, and perhaps social and societal programs. NHII is a first step toward obtaining data and information necessary for coordinating care in the clinic and hospital.

The management of large databases, which are essential to comprehensive core clinical applications for information/ communications systems, remains a critical determinant. Although databases are effectively managed in select locations, efforts must continue to develop secure, dispersed, multiagent databases that can serve both providers and patients effectively and efficiently.

Using CPOE systems for entering orders for tests, drugs, and other procedures has led to reductions in transcription errors, which have led to demonstrable improvements in patient safety. When CPOE systems are integrated with other core clinical applications, their impact on patient safety is even greater. One component of a CPOE system is computerized decision support. CPOE systems that include data on patient diagnoses, current medications, and history of drug interactions or allergies can significantly reduce prescribing errors (Bates et al., 1998, 1999; Leapfrog Group, 2000).

CPOE systems also improve the quality of care by increasing clinician compliance with standard guidelines of care, thereby reducing variations in care. For example, a 1998 study by Shojania et al. found that CPOE, combined with the use of a vancomycin guideline, reduced the use of this over-prescribed antibiotic by 32 percent. A study of CPOE use at one large academic medical center (Brigham and Women’s Hospital) by Teich et al. (2000) estimated that the overall annual cost savings from reductions in drug costs, laboratory tests, and diagnostic studies and the prevention of adverse drug events were roughly $5 to 10 million annually.

Despite many documented benefits of CPOE systems—improvements in the quality of patient care, decreases in medication errors, and decreases in overall costs—they have not been widely implemented. In the only study that has rigorously examined the adoption of CPOE by hospitals in the United States, less than 2 percent of hospitals were found to have CPOE systems completely or partially available and to require that physicians use them (Ash et al., 1998). Nevertheless, a few success stories have been well documented, notably the Brigham and Women’s Hospital in Boston, Massachusetts, and the Regenstrief Medical Record Systems.

Studies indicate that there are multiple barriers to the effective use of CPOE systems, including the lack of education and training of physicians, problems with user-interface designs, concerns about accuracy and reliability, high upfront fixed and ongoing maintenance costs, a lack of leadership commitment, difficulties in coordinating the introduction of new applications among varied care delivery settings and functions, and poor integration of CPOE systems with existing work processes and other information/ communications systems, both clinical (e.g., digital sources of evidence, decision-support tools) and administrative (Boodman, 2005; Durieux, 2005; Garg et al., 2005; Sarata, 2002; Tang in this volume; Wears and Berg, 2005).

To address these problems, a template for patients, based on the current database, could be customized by the physician using evidence-based standards as the “orders” for each patient. One of the most frequent causes of errors and failures to carry out planned treatments has been a lack of integration of orders and results. Branching logic based on results can be used to verify that each step in the treatment is accomplished. The system described would not only reduce errors, such as missed handoffs and unnecessary waiting times, it would also interact with enterprise systems for supply-chain management and capacity planning.

Another key component of the health information infrastructure, digital sources of evidence—including bibliographic references, evidence-based clinical guidelines, and comparative databases—is essential for evidence-based practice. Currently, most digital sources of evidence are stand-alone systems that are not integrated into clinical information systems. The challenge for practitioners is to use these sources of evidence in combination with their experience and expertise to make clinical decisions (Bakken, 2001). However, as the medical-evidence base continues to expand exponentially and more and more clinicians accept the validity of best-demonstrated practices for diagnosis and treatment, there is mounting interest in integrating rapidly expanding digital sources of evidence (including genomic and phenotypic [clinical] data) into decision-support tools that can be fully integrated into care processes.

At the same time, fueled by the rapidly expanding medical-evidence base, there is a growing awareness among care professionals of the need for customization of best demonstrated practice rules for almost all patients. In the past five years, a new field has emerged, “predictive medicine” (i.e., the use of mathematical and statistical strategies to mine phenotypic [clinical] databases to identify and treat high-risk patients and to individualize treatment strategies).

Another emerging area is translational medicine, the use of the results of the genome project to predict and customize treatment.

The standardization of health care data, the development of digital sources of medical evidence and knowledge, and the creation of EHRs will all facilitate the use of decision-support tools, which are key components of clinical information systems. Decision-support tools that are fully integrated into the care process will enable both care providers and patients to access medical knowledge relevant to the patient’s care. They may, for example, identify negative interactions between a drug the patient is already taking and an additional drug that might be prescribed.

A necessary platform for decision-support tools is the clinical-data repository, a database that collects and stores patient care information from diverse sources. Clinical-event monitors, which work with clinical-data repositories in support of real-time delivery of care, are usually triggered by clinical events (e.g., a patient visit, a medication order, a new laboratory result), either when data representing the event enter a repository or when a provider uses a clinical information system. The event monitor combines clinical rules, the triggering event, and information present in the repository to generate alerts, reminders, and other messages important to the delivery of care.

For more than 20 years, departmental systems (e.g., laboratory, x-ray) have had internal computerized systems that control operations and report results. But there is no health care process-management system in which all information concerning a patient’s history is gathered in one place in standardized text where the appropriateness and strategy of orders for patient care can be checked. Equally important, a health care process-management system would ensure that the result of each step in treatment was entered into the record and communicated to all relevant parties. The collection of data, the consideration of the decision support offered, followed by the ordering and carrying out of the diagnostic and or treatment plan is an iterative process. As results are entered, the next steps in the care process are instituted.

Because the value of computerized clinical systems depends on how well they support care decisions in the service of patients, the development and implementation of information/communications systems that provide support and increase connectivity among health care providers will require “human-factors” research. This area of research, which combines expertise in cognitive and software engineering, behavioral science and cooperative work, and computer and cognitive sciences, focuses on the development of techniques and concepts that facilitate interactions between people and computers (Winograd and Woods, 1997; Woods, 2000).

Usefulness and usability in software-intensive systems cannot be achieved by patching “user-friendly” interfaces onto user-hostile system architectures. Health care computer systems have been administrator-centered or billing-centered systems rather than provider-centered or patient-centered systems. However, software and telecommunications capabilities are being expanded, although slowly, to achieve continuity of care without losing sight of economic and other pressures (Box 4-3).

A recent study concludes that there is an urgent need for more “research and development in innovative and efficient human-machine interfaces that are optimized for use in the health care sector” (PITAC, 2004). Areas for research include hardware interfaces, as well as sociological and psychological aspects of the use of computerized systems by physicians and other health care workers. The human-computer/information systems interface should be a high priority for health care. The study identifies targets for research: “improved medical-domain voice-recognition data conversion systems; improved automated entry of instrument data; and automated methods for converting both new and legacy electronic data to normalized form” (PITAC, 2004).

In systems in which software is an element in the critical path, a variety of serious problems have plagued organizations, including lack of dependability and/or usability, the high cost of system failure, high maintenance requirements, and difficulties in updating systems. Because software-intensive systems perform valuable functions, the consequences of failure are generally serious. For example, developers may assemble modules, each apparently dependable, but, when they are integrated, problems and weaknesses emerge. Usability failures are also an issue. For example, if there are too many steps in a program for a user to follow or if a program is too complicated, various “work arounds” will be developed, and patient safety or some other critical parameter may be compromised.

In some cases, the initial software-intensive system may be dependable, but changes in use over time may lead to changes in the software that lead, in turn, to unnoticed side effects that can introduce weaknesses in the system. Another type of failure can occur when cost overruns in the development process prevent the project from ever reaching the commercialization stage. In some instances, noncritical software that interacts directly or indirectly with critical functions introduces failures and weaknesses (NRC, 2004; Rae et al., 2003). As these and other forms of software-system failure show, investments in clinical information systems must be complemented by investments in research on software dependability.

The implementation of core clinical applications of information/communications systems has progressed very slowly because of costs, possible disruptions to current operations, problems with overlapping legacy systems, and problems with the use and integration of various systems. Opportunities for improvement (and research) include: better human-computer system interfaces; software to improve the interoperability of systems from various vendors; systems and accompanying business models for spreading costs among multiple users; and software dependability in the context of health care delivery.

The delivery of quality care, especially in a highly fragmented delivery system, requires that both clinicians and patients have access to complete patient information and decision-support tools and that communications among clinicians and between clinicians and patients are effective. The Internet and the World Wide Web have provided patients with unprecedented access to health information and made possible more continuous, asynchronous communication between patients and their care providers. These technologies for asynchronous communication have enabled the development of self-care educational tools/modules, such as University of Wisconsin’s Comprehensive Health Enhancement Support System (CHESS), which promotes informed health monitoring and decision making by giving patients access to disease-specific information (see paper by Gustafson in this volume). The case example of CareGroup Healthcare System (see Halamka in this volume) illustrates many of the challenges and opportunities associated with building fixed-line information/communication networks that increase connectivity and information exchange between patients and clinicians, as well as among dispersed elements of the care team.

Meeting the current and emerging communications needs of health care will require a combination of wireless and fixed-line networks. Because of financial constraints, creating different systems for different settings will not be feasible, however. Vendors of hardware and software

components of the system will need system transparency, which can only be achieved once standards have been adopted. The challenge will be to generate a robust, but flexible system that can be duplicated in many different circumstances without requiring major modifications; the system must be based on technology that can be rapidly diffused and at low cost. Five technical factors are important in planning for the implementation of communication networks: (1) bandwidth requirements and availability; (2) latency in transmission throughout the network; (3) continuous availability of the network; (4) confidentiality and security of data; and (5) ubiquity of access to the network (NRC, 2000).

Enabling patients to communicate effectively with health care providers without face-to-face meetings will require many improvements in electronic communications. The Internet and World Wide Web provide a framework for communication links, and a few large provider organizations have demonstrated the potential of these technologies. But making them accessible to large populations in a health care community will require experimentation and research (Perlin et al., 2004). Other issues that must be addressed include ensuring the confidentiality and security of transmissions and health care data. In the long term, sensors that register a patient’s vital signs and transmit data via wireless links could greatly improve the “connectivity” between patients and health care providers.

There is considerable evidence linking the use of advanced clinical information/communications systems to improvements in the quality, safety, and patient-centeredness of care (Breslow in this volume; Casalino et al., 2003; Clayton in this volume; Demakis et al., 2000; Lansky, 2002; Littlejohns et al., 2003; Miller and Bovbjerg, 2002; Wang et al., 2003; Weingarten et al., 2002). One recent estimate of potential savings to the nation’s health care system from widespread implementation of clinical information/ communications systems concluded that a fully interoperable network of EHRs would yield $77.8 billion a year in net benefits, or roughly 5 percent of the nation’s total annual health care spending (Walker et al., 2005). In spite of the demonstrated benefits to society as a whole, however, many barriers stand in the way of widespread implementation of clinical information/communications systems in the United States.

In the preceding discussion of major components of the NHII, a number of technical impediments to implementation of these systems were identified (e.g., lack of interoperability standards, human-factors barriers; patient and caregivers’ concerns about usability, reliability, and security; patients’ concerns about the privacy of integrated health information, etc.). In addition, there are economic, cultural/organizational, and educational barriers that must by overcome. (Educational barriers are discussed in Chapter 5.)

Significant up-front and continuing costs of implementing clinical information/communications systems (e.g., EMRs, CPOE systems, decision-support tools) are particularly burdensome for individual care providers or small provider organizations, that is, the vast majority of care providers. These costs include not only the cost of hardware, software, and technical support, but also the costs of intensive training of patients and care providers in the use of these technologies, as well as costs associated with the adaptation of work processes, the roles of professionals and support staff, and the infrastructure necessary for information/ communications systems to be effective components in the delivery of health care.

At the present time, several factors severely undercut the returns health care providers might expect to capture on their investments. First, the lack of technical interoperability standards for information/communications technologies and components and the lack of standard vocabularies have impeded information connectivity within the highly fragmented health care delivery system. This lack of connectivity, in turn, has severely limited improvements in efficiency and quality.

Presently, the scarcity of comparative quality and cost-performance data and the corresponding lack of quality/cost transparency in the market for health care services prevent patients from making informed choices among care providers on the basis of quality or value (quality/cost) (see Safran, this volume; Rosenthal et al., 2004). At the same time, the current “market” for health care services provides little reward to care providers who improve the quality of their processes and outcomes through investments in systems engineering tools, information/communications technologies, or other innovations (Hellinger, 1998; Leape, 2004; Leatherman et al., 2003; Miller and Luft, 1994, 2002; Robinson, 2001).

Another major barrier is the prevailing reimbursement arrangement for health care services, which does not reimburse care providers differentially on the basis of quality of care. Accordingly, providers have little incentive to invest in information/communications systems or process-management tools in support of quality improvement, unless they directly generate revenue or demonstrate immediate improvements in operating efficiency. (Contrast this with incentives for provider organizations to invest in new diagnostic equipment, such as MRI machines, which begin to generate revenue as soon as they are up and running). Moreover, most private and public insurance reimbursement models actively discourage delivery-related applications of information/ communications technology by care providers, for example, by refusing to reimburse patient care/consultations delivered via e-mail (Leape, 2004; Leatherman, et al., 2003; Robinson, 2001).

The mandate of the Medicare Modernization Act and efforts by the Leapfrog Group and other buyers, insurers,

and accreditation agencies to remove reimbursement- and regulation-related barriers to the use of information/ communications systems in health care represent positive developments (CMS, 2004; Milstein in this volume). Nevertheless, the barriers persist. New financing and networking models will be necessary to encourage small businesses, which employ the vast majority of physicians, to take advantage of information/communications technologies without compromising the care of patients who are not computer literate. A number of public-sector and private-sector entities are already working on a cost-effective way to accomplish this (PITAC, 2004; SNL, 1996; Thompson and Brailer, 2004; Yasnoff et al., 2004).

The committee believes that as conceptual and material progress is made in measuring quality and productivity in health care, significant returns on investment at all levels of the health care system will be demonstrated (NRC, 2002; Triplett, 1999, 2001). But developing and validating system options for measuring the impact of information/ communications technologies will require much more support from federal agencies (e.g., National Institutes of Health, National Science Foundation, Agency for Healthcare Research and Quality, Veterans Health Administration, and others). In the meantime, although the anticipated quality and productivity returns to the overall system from widespread application of systems engineering, information/ communications technology, and related innovations may be great, in the current context, most individual provider organizations are not convinced that they can capture a large enough fraction of the total “social returns” on their private investments to warrant making these investments in the first place.

Clearly, many questions remain to be answered about the potential benefits of advanced information/communications systems in the health care industry, and answering these questions will mean overcoming many barriers. The introduction of systems analysis, systems redesign, and new information/communications systems are likely to cause significant disruptions to organizations and the structure of work processes at all four levels of the health care system. In addition, many clinicians have a very limited understanding of the potential uses, impacts, and benefits of advanced information systems for the production and delivery of care. Thus, the benefits of change are not immediately visible, but the costs are. Not surprisingly, then, there has been significant resistance to innovation and changes in work processes and the division of labor among health care professionals.

The cultural and organizational factors that have contributed to a rigid division of labor in many areas of health care often impede the introduction and exploitation of tools, technologies, and other innovations that could improve quality and productivity in health care (see Bohmer, this volume; Christenson et al., 2000). Ultimately, the benefits offered by many of these tools and technologies can only be realized if management has the authority and/or capacity to persuade care providers to change their work practices and organization. Not surprisingly, the health care provider organizations most advanced in the use of systems tools and information/ communications technologies have corporate management structures—all of their health care professionals are employees and are part of a clearly defined managerial hierarchy.

Finding 4-1. A fully implemented National Health Information Infrastructure would support distributed, independently managed, multi-tiered, intra-institutional information/communications systems and would dramatically improve the collection, exchange, and processing of information on all levels of the health care system.

Finding 4-2. A critical step toward realizing the National Health Information Infrastructure will be the development and widespread adoption of network standards for health care data and software. Research must focus on standards-related issues concerning the integrity of data, controlled access to data, data security, and the integration of large-scale wireless communications. There is also a pressing need for low-cost tools for standardizing new and legacy digital data without disrupting clinical work flows.

Finding 4-3. Interoperability standards for diverse information/ communications systems and messaging standards will be critical to the realization of an information/communications technology-enabled health care system that has the capacity for mass customization to meet the needs of individual patients.

Finding 4-4. Progress in systems interoperability and data standards is likely to improve remote access to self-care educational tools, patient health records, and health care provider and insurer services (scheduling, billing, etc.).

Finding 4-5. Cross-sector learning and research on information and communications standards among federal agencies, health care insurers, and health care providers represents a potentially vast source of knowledge and advancement.

Finding 4-6. The Internet and World Wide Web provide a framework for communication links, but making them accessible to large populations in a health care community to promote communication between patients and health care providers will require experimentation and research, particularly to ensure the confidentiality and security of transmissions of health care data.

Finding 4-7. Opportunities for improvement of core clinical applications of information/communications technologies

include: better human-computer interfaces; software to improve the interoperability of systems from various vendors; clinical information systems and accompanying business models for spreading costs among multiple users; the development and management of large, multi-agent databases; and software dependability in the context of health care delivery.

Finding 4-8. A National Health Information Infrastructure could provide a platform for the implementation of new information/communications technologies, such as wireless integrated microsystems, which would enable the remote capture and communication of patients’ physiological data to care professionals, thereby increasing the likelihood of timely diagnoses and treatments of illnesses. In the long term, sensors that register a patient’s vital signs and transmit data via wireless links could greatly improve the “connectivity” between patients and health care providers.

Finding 4-9. Much of the information/communications technology necessary for the development of the NHII, on all four levels of the health care delivery system, exists today. However, many barriers will have to be overcome before it can be implemented.

Finding 4-10. Although considerable evidence shows that advanced clinical information/communications systems lead to improvements in the quality, safety, and patient-centeredness of care, the health care sector as a whole trails far behind most other industries in investments in these systems. Many factors have contributed to this deficit: the atomistic structure of the industry; current payment/ reimbursement regimes; the lack of transparency in the market for health care services; weaknesses in the managerial culture; the hierarchical structure and rigid division of labor; and (until very recently) the immaturity of available commercial clinical information/communications systems.

Recommendation 4-1. The committee endorses the recommendations made by the Institute of Medicine Committee on Data Standards for Patient Safety, which called for continued development of health care data standards and a significant increase in the technical and material support provided by the federal government for public-private partnerships in this area.

Recommendation 4-2. The committee endorses the recommendations of the President’s Information Technology Advisory Council that call for: (1) application of lessons learned from advances in other fields (e.g., computer infrastructure, privacy issues, and security issues); and (2) increased coordination of federally supported research and development in these areas through the Networking and Information Technology Research and Development Program.

Recommendation 4-3. Research and development in the following areas should be supported:

• human-information/communications technology system interfaces

• voice-recognition systems

• software that improves interoperability and connectivity among systems from different vendors

• systems that spread costs among multiple users

• software dependability in systems critical to health care delivery

• secure, dispersed, multi-agent databases that meet the needs of both providers and patients

• measurement of the impact of information/ communications technologies on the quality and productivity of health care

Recommendation 4-4. The committee applauds the U.S. Department of Health and Human Services 10-year plan for the creation of the National Health Information Infrastructure and the high priority given to the creation of standards for the complex network necessary for communications among highly dispersed providers and patients. To ensure that the emerging National Health Information Infrastructure can support current and next-generation clinical information/ communications systems and the application of systems tools, research should focus immediately on advanced interface standards and protocols and standards-related issues concerning access, security, and the integration of large-scale wireless communications. Special attention should be given to issues related to large-scale integration. Funding for research in all of these areas will be critical to moving forward.

Recommendation 4-5. The committee recommends that public- and private-sector initiatives to reduce or offset current regulatory, accreditation, and reimbursement-related barriers to more extensive use of information/communications technologies in health care be expanded. These initiatives include efforts to reimburse providers for care episodes or other bundling techniques (e.g., severity-adjusted capitation; disease-related groups, etc.), public and private support of community-based health information network demonstration projects, the Leapfrog Group’s purchaser-mediated rewards to providers that use information/communications technologies, and others.

The emerging technologies in wireless communications and microelectronic systems described in this section have the potential to advance the patient-centeredness and quality performance of the health care delivery system and to change the structure of care delivery in the process. Microelectronics promises to be a powerful tool for meeting quality and

productivity challenges in health care delivery, provided that resources can be marshaled in a rational way. The microelectronics revolution began in the 1950s with the advent of integrated circuits and has since revolutionized data processing, communications, and control. The number of transistors that can be integrated on a silicon chip the size of a finger-nail has increased from about 2,000 on the first micro-processor (1971) to about 200,000,000 today; the speed of these chips has increased more than a thousand-fold. At the same time, the number of bits of memory on a chip has increased by a factor of more than a million, and costs have decreased just as precipitously. Low-cost disk storage is now approaching a density of more than 40 gigabytes per square inch. In short, the processing and storage of data, the creation of information and knowledge based on those data, and the efficacy of decisions have improved exponentially.

In the coming decades, as the number of nurses and physicians decreases, monitoring and diagnostics will have to improve dramatically. Efforts to develop sensors using integrated circuit technology has resulted in microelectro-mechanical systems, which can be combined with microelectronics and wireless interfaces to create wireless integrated microsystems (WIMS) for use in health care delivery. In the near future, WIMS will be merged with sensors with embedded microcomputers and minute wireless transceivers (a cubic centimeter in size or smaller) that operate at power levels of less than 1 milliwatt, consistent with long-term operation fueled by batteries maintained by energy scavenged from the environment (Wise, 1996, 2002).

These new devices could potentially provide continuous monitoring of critical functions, thereby turning every hospital room into an intensive care facility. WIMS devices small enough to be worn comfortably and unobtrusively could communicate with a bedside receiver that communicates, in turn, with monitoring stations and a larger health care facility. The system just described would go a long way toward meeting the objective of the Leapfrog Group of having an ICU physician present in every hospital at all times (Leapfrog Group, 2000).

WIMS systems are still scarce, and their performance is limited, but they are emerging. Blood oximeters, heart rate monitors, and temperature sensors could all be components of WIMS; swallowable capsules for viewing the digestive tract are already in use (Fireman, 2004; Pelletier, 2004; Pennazio et al., 2004). Wearable devices that monitor blood pressure (hypertension), breathing patterns (sleep apnea), and other variables will certainly be available in the near future (see Budinger in this volume). The major challenges to their use are interfaces with the body itself.

Swallowable capsules for all kinds of internal viewing and measurements could significantly improve diagnoses of a variety of conditions and thus could improve the quality of health care. DNA analysis chips will bring advances in genetics into the hospital, and even the local doctor’s office (Burns et al., 1998; Mastrangelo et al., 1998), and should lead to improvements in both diagnostics and preventive health care. However, the impact of these developments on costs will be indirect. In addition, privacy issues must be addressed before they can be widely used.

WIMS for health care are expected to be technically feasible in the coming decade, but to reduce costs, they must be part of a complete system. Bedside receivers and wearable monitors might be technical triumphs, but they could also lead to economic disaster for the company that produces them unless they fit into a larger system.

A similar situation has existed for at least 20 years in the process-control industry. Although prototypes of sophisticated sensors have been produced, they are still not widely used because controllers that can exploit their features have not yet been developed. In the transportation industry, the entire control system of the automobile engine had to be redesigned to take advantage of microprocessors and electronic sensing. Comparable redesigning of the health care system will be necessary at every level to take advantage of WIMS.

The application of WIMS technologies in the hospital promises to significantly improve the quality and patient-centeredness of inpatient and ambulatory care. The potential impact of WIMS on home care and the quality of life for senior citizens and chronically ill patients is even greater (Whitten et al., 2003). Moving WIMS technology into the home is being seriously considered by makers of home communications equipment. With properly integrated home-based WIMS systems, patients could be monitored on a continuous basis and care professionals alerted automatically when events merit attention. Continuous or at least more frequent home monitoring of the health status of elderly and chronic care patients could notify clinicians of the need for timely therapeutic interventions that could avoid hospitalizations and shorten hospital stays, thus reducing the costs associated with the care of the patient over time (see Budinger in this volume). Moreover, home-based WIMS could facilitate safe home environments and the activities of daily living that are so important for the health of the elderly and chronically ill.

The main technical problems in the development of WIMS are largely related to reliable interfaces between sensors and the body and ensuring that sensors are capable of differentiating between instrumentation artifacts and physiological events. If these problems can be solved and such systems can “piggyback” on existing communication networks, they could be implemented within the coming decade.

WIMS may also have therapeutic uses. The development of wireless implantable microsystems has been the subject of research for 40 years or more, but, to date, few devices have been developed besides pacemakers. Pacemakers have become increasingly sophisticated electronically, but their interfaces with the body are primarily via electrodes. Nevertheless, they have set the stage for the emergence of new devices in the coming decade. For example, cardiovascular catheters have been used for diagnosing cardiac conditions for many years, and pressure sensors small enough to be mounted directly on catheters have existed for some time (Chau and Wise, 1988; Ji et al., 1992). In fact, catheter-based electronics for improving diagnostic capabilities are long overdue. Another example is stents, which are widely used for treating coronary occlusions and now have chemical coatings to prevent re-stenosis. In the near future, stents may also be used as platforms for instrumentation, such as wireless sensors for monitoring blood pressure or blood flow that could be activated by a radio frequency wand positioned over the chest. Significant challenges remain involving range, accuracy, and size, but such systems may be feasible soon (Collins, 1967; DeHennis and Wise, 2002; Stangel et al., 2001).

Wireless sensors could also be used in intracranial, intraocular (glaucoma), and intra-arterial applications. Miniature biocompatible packages that can exist for many decades in the body are also being developed for long-term use in chronic conditions (Ziaie et al., 1996).

WIMS could also have a dramatic impact on the treatment of conditions involving the central nervous system. More than 90,000 cochlear implants are in use worldwide today, enabling many profoundly deaf and severely hearing-impaired individuals to function normally in a hearing world (House and Berliner, 1991; Spelman, 1999). Even though their performance is still limited and there is some opposition to them in the deaf community, these devices may render most kinds of deafness treatable disorders in the next two decades. In the United States alone, more than 2 million people are profoundly deaf, and 20 million are severely hearing impaired.

There is considerable interest in treating other neurological disorders using WIMS. Visual prostheses have recently received considerable attention but are still at a very early stage of development (Lui, 2002). The same is true of prostheses for severe epilepsy and paralysis. For example, an implanted electrode array might detect the onset of an epileptic seizure and provide local electrical stimulation or drug delivery to prevent the spread of the seizure. Functional neuromuscular stimulation (FNS) is being used to help quadriplegics stand and even walk, and the use of dense electrode arrays to capture control signals directly from the motor cortex has recently enabled primates to control robotic arms (Chapin et al., 1999; Serruya et al., 2002; Taylor et al., 2002) and humans to control cursors in operating a computer interface (Donoghue, 2004). Combining FNS with cortical control could lead to at least limited closed-loop activation of paralyzed limbs (Wise et al., 2004). And the use of deep brain stimulation in the subthalamic nucleus to eliminate the manifestations of Parkinson’s disease has yielded impressive results and is now approved for human use (Limousin et al., 1998). Although all of these devices are still at a relatively early stage of development (Table 4-1), some are gaining acceptance now, and many could be in wide use in the next 20 years, which could substantially impact the quality of health care and the costs of rehabilitation.

Microsystems implemented as wearable and implantable devices connected to clinical information systems through wireless communications could provide diagnostic data and deliver therapeutic agents for the treatment of a variety of chronic conditions. In fact, WIMS could potentially restructure care delivery in the hospital. There is no question that microdevices can and will significantly improve the daily lives of many people.

The barriers to the realization of this vision are significant, however. For patients to take on greater control and responsibility for their own care, they will have to be educated or able to educate themselves. In addition, patients must continue to have access to trusted sources of advice and counsel.

Changes in the division of labor between patients and care teams implicit in the self-care model will also have a profound impact on the roles, work processes, and division of labor among members of the patient’s care team. Current work rules, licensing requirements, staffing requirements, and regulations designed to ensure the safety, reliability, and quality of care in a hospital/clinic/provider-centered delivery system will also present impediments to a shift to the self-care model. Resistance to change, especially if roles, authority, and jobs are threatened, may arise among care professionals and organizations that deliver services both within and outside of hospital setting (e.g., testing labs, etc.). Current reimbursement systems may also present barriers if care providers are not reimbursed for e-visits, patient modules, remote care services, and so on.

The implications of the self-care model for the health care industry are profoundly disruptive. The move toward self-care could be considered threatening to businesses (e.g., testing laboratories, etc.) and individual care providers whose services will be less in demand. The current complex mix of professional licensing, regulation, liability law, and other constructs established to ensure the health care safety and reliability also pose barriers. The current hierarchical culture and rigid division of labor in the health care profession could make the reallocation of responsibilities and changes in the roles of care team members extremely contentious.

Finding 4-11. Wireless integrated microsystems could have an enormous beneficial impact on the quality and cost of health care, especially home health care. Microsystems implemented as wearable and implantable devices connected to clinical information systems through wireless communications could provide diagnostic data and deliver therapeutic agents for the treatment of a variety of chronic conditions, thereby improving the quality of life for senior citizens and chronically ill patients.

Finding 4-12. The use of wireless integrated microsystems technologies in hospitals and clinics promises significant improvements in the quality and patient-centeredness of inpatient and ambulatory care. Microdevices that could provide continuous monitoring of critical functions could turn every hospital room into an intensive care facility.

Finding 4-13. Wireless integrated microsystems for health care are expected to be technically feasible in the coming decade, but to reduce costs, they must be part of a complete system.

Finding 4-14. Significant cultural and organizational barriers will have to be overcome for the full benefit of WIMS to be realized.

Recommendation 4-6. Public- and private-sector support for research on the development of very small, low-power, biocompatible devices will be essential for improving health care delivery.

Recommendation 4-7. Engineering research should be focused on defining an architecture capable of incorporating data from microsystems into the wider health care network and developing interface standards and protocols to implement this larger network. Microsystems research should be focused on the following areas:

• integration, packaging, and miniaturization (to sizes consistent with implantation in the body)

• tissue interfaces and biocompatibility for long-term implantation

• interfaces and approaches to noninvasive (wearable) devices for measuring a broad range of physiological parameters

• low-power, embedded computing systems and wireless interfaces consistent with in vivo use

• systems that can transform data reliably and accurately into information and information into knowledge as a basis for treatment decisions

Timely, accurate information is critical to the efficient operation of large dispersed systems. Although the health care system has been slow to recognize this, efforts are now under way to rectify the situation. But it is imperative that research, development, demonstration, and training be expanded and accelerated.

Putting together a system that can make use of information microtechnology, nanotechnology, and biotechnology and ensure that applications are widely available and affordable will require coordination at the national level among device manufacturers, clinicians, and hospital systems. A successful health care system would use information/ communications technologies in ways that would be largely invisible to patients but would improve care, reduce costs, and provide patient-centered care. However, unless the approach is coordinated, the impact of new technologies could improve health care for a few but increase costs for everyone else and move the overall system even farther away from providing patient-centered care.

Ash, J.S., R.N. Gorman, and W.R. Hersh. 1998. Physician Order Entry in U.S. Hospitals. Pp. 235–239 in Proceedings/AMIA Annual Symposium. Bethesda, Md.: American Medical Informatics Association.

Bakken, S. 2001. An informatics infrastructure is essential for evidence-based practice. Journal of the American Medical Informatics Association 8(3): 199–201.

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 computerized physician order entry and a team intervention on prevention of serious medication errors. Journal of the American Medical Association 280(15): 1311–1316.

Bates, D.W., J.M. Teich, J. Lee, D. Seger, G.J. Kuperman, N. Ma’Luf, D. Boyle, and L. Leape. 1999. The impact of computerized physician order entry on medication error prevention. Journal of the American Medical Informatics Association 6(4): 313–321.

Brailer, D.J. 2003. Use and Adoption of Computer-Based Patient Records in the United States: A Review and Update. PowerPoint presentation to the IOM Committee on Data Standards for Patient Safety, Irvine, California, January 23, 2003.

Boodman, S.G. 2005. Not Quite Fail-Safe; Computerizing Isn’t a Panacea for Dangerous Drug Errors, Study Shows. Washington Post, March 22, 2005, p. F.01.

Burns, M.A., B.N. Johnson, S.N. Brahmasandra, K. Handique, J.R. Webster, M. Krishnan, T.S. Sammarco, P.M. Man, D. Jones, D. Heldsinger, C.H. Mastrangelo, and D.T. Burke. 1998. An integrated nanoliter DNA analysis device. Science 282(5388): 484–487.

CareScience. 2003. Santa Barbara County Care Data Exchange. Available online at: http://www.carescience.com/healthcare_providers/cde/care_data_exchange_santabarbara_cde.shtml.

Carroll, J.M., W.A. Kellogg, and M.B. Rosson. 1992. Getting around the task-artifact cycle: how to make claims and design by scenario. ACM Transactions on Information Systems 10(2): 181–212.

Casalino, L., R.R. Gillies, S.M. Shortell, J.A. Schmittdiel, T. Bodenheimer, J.C. Robinson, T. Rundall, N. Oswald, H. Schauffler, and M.C. Wang. 2003. External incentives, information technology, and organized processes to improve health care quality for patients with chronic diseases. Journal of the American Medical Association 289(4): 434–441.

Center for Health Transformation. 2005a. Transforming Examples: Mayo Clinic, Jacksonville, FL. Available online at: http://www.healthtransformation.net/Transforming_Examples/transforming_examples_resource_center/129.cfm.

Center for Health Transformation. 2005b. Transforming Examples: Veterans Administration Healthcare System—My HealtheVet. Available online at: http://www.healthtransformation.net/Transforming_Examples/transforming_examples_resource_center/142.cfm.

Chapin, J.K., K.A. Moxon, R.S. Markowitz, and M.A. Nicolelis. 1999. Real-time control of a robot arm using simultaneously recorded neurons in the motor cortex. Nature Neuroscience 2(7): 664–670.

Chau, H.-L., and K.D. Wise. 1988. An ultraminiature solid-state pressure sensor for a cardiovascular catheter. IEEE Transactions on Electron Devices 35(12): 2355–2362.

Christenson, C.M., R. Bohmer, and J. Kenagy. 2000. Will disruptive innovations cure health care? Harvard Business Review (September-October): 103–111.

CMS (Centers for Medicare and Medicaid Services). 2004. Medicare Modernization Act. Available online at: http://www.cms.hhs.gov/medicarereform/.

Collins, C.C. 1967. Miniature passive pressure transducer for implanting in the eye. IEEE Transactions on Bio-medical Engineering 14(2): 74–83.

Cook, R.I., D.D. Woods, M.B. Howie, J.C. Harrow, and D.M. Gaba. 1992. Unintentional delivery of vasoactive drugs with an electromechanical infusion device. Journal of Cardiothoracic and Vascular Anesthesia 6(2): 238–244.

Cook, R.I., D.D. Woods, and C. Miller. 1998. A Tale of Two Stories: Contrasting Views of Patient Safety. Report from a Workshop on Assembling the Scientific Basis for Progress on Patient Safety. Chicago, Ill.: National Patient Safety Foundation.

DeHennis, A., and K.D. Wise. 2002. A Double-Sided Single-Chip Wireless Pressure Sensor. Pp. 252–255 in Digest of the IEEE MEMS Conference, Las Vegas, Nevada. New York: IEEE.

Demakis, J.G., C. Beauchamp, W.L. Cull, R. Denwood, S.A. Eisen, R. Lofgren, K. Nichol, J. Woolliscroft, and W.G. Henderson. 2000. Improving residents’ compliance with standards of ambulatory care: results from the VA cooperative study on computerized reminders. JAMA 284(11): 1411–1416.

DOC (U.S. Department of Commerce). 1999. The Emerging Digital Economy II: Appendices. Washington, D.C.: DOC.

Donoghue, J.P. 2004. Presentation at Neural Interfaces Workshop, Bethesda, Maryland, November 15–17, 2004.

Durieux, P. 2005. Electronic medical alerts—so simple, so complex. New England Journal of Medicine 352(10): 1034-1036.

Fireman, Z. 2004. The light from the beginning to the end of the tunnel. Gastroenterology 126(3): 914–919.

Flach, J.M., and C.O. Dominguez. 1995. Use-centered design. Ergonomics in Design 3(3): 19–24.

Garg, A.X., N.K.J. Adhikari, H. McDonald; P. Rosas-Arellano, P.J. Devereaux, J. Beyene, J. Sam, R.B. Haynes. 2005. Effects of computerized clinical decision support systems on practitioner performance and patient outcomes: a systematic review. Journal of the American Medical Association 293: 1223–1238.

Guerlain, S., P.J. Smith, J.H. Obradovich, S. Rudmann, P. Strohm, J.W. Smith, J. Svirbely, and L. Sachs. 1999. Interactive critiquing as a form of decision support: an empirical evaluation. Human Factors 41(1): 72–89.

Hellinger, F.J. 1998. The effect of managed care on quality. Archives of Internal Medicine 158: 833–841.

House, W.F., and K.I. Berliner. 1991. Cochlear Implants: From Idea to Clinical Practice. Pp. 9–33 in Cochlear Implants: A Practical Guide, edited by H. Cooper. San Diego, Calif.: Singular Publishing Group, Inc.

Interoperability Consortium. 2005. Development and Adoption of a National Health Information Network: Response to ONCHIT’s RFI. January 18, 2005.

IOM (Institute of Medicine). 1991. The Computer-Based Patient Record: An Essential Technology for Health Care. Washington, D.C.: National Academy Press.

IOM. 1997. The Computer-Based Patient Record: An Essential Technology for Health Care. Revised edition. Washington, D.C.: National Academy Press.

IOM. 2000. To Err Is Human: Building a Safer Health System, edited by L.T. Kohn, J.M. Corrigan, and M.S. Donaldson. Washington, D.C.: National Academy Press.

IOM. 2001. Crossing the Quality Chasm: A New Health System for the 21^st Century. Washington, D.C.: National Academy Press.

IOM. 2003. Key Capabilities of an Electronic Health Record System. Washington, D.C.: National Academies Press.

IOM. 2004. Patient Safety: Achieving a New Standard of Care. Washington, D.C.: National Academies Press.

Ji, J., S.T. Cho, Y. Zhang, K. Najafi, and K.D. Wise. 1992. An ultraminiature CMOS pressure sensor for a multiplexed cardiovascular catheter. IEEE Transactions on Electron Devices 39(10): 2260–2267.

Kass-Bartelmes, B.L., E. Ortiz, and M.K. Rutherford. 2002. Using informatics for better and safer health care. Research in Action Issue 6. AHRQ Pub. No. 02-0031. Rockville, Md.: Agency for Healthcare Research and Quality.

Kolodner, R.M., and J.V. Douglas, eds. 1997. Computerized Large Integrated Health Networks: The VA Success. New York: Springer-Verlag.

Lansky, D. 2002. Improving quality through public disclosure of performance information. Health Affairs (Millwood) 21: 52-62.

Leape, L.L. 2004. Human factors meets health care: the ultimate challenge. Ergonomics in Design (summer): 6–12.

Leapfrog Group. 2000. Leapfrog Patient Safety Standards: The Potential Benefit of Universal Adoption. Available online at: http://www.leapfroggroup.org.

Leatherman, S., D. Berwick, D. Iles, L.S. Lwein, F. Davidoff, T. Nolan, and M. Bisognano. 2003. The business case for quality: case studies and an analysis. Health Affairs 22(2): 17–30.

Limousin, P., P. Krack, P. Pollack, A. Benazzouz, C. Ardouin, D. Hoffmann, and A.L. Benabid. 1998. Electrical stimulation of the subthalamic nucleus in advanced Parkinson’s Disease. New England Journal of Medicine 339(16): 1105–1111.

Lin, L., R. Isla, K. Doniz, H. Harkness, K.J. Vicente, and D.J. Doyle. 1998. Applying human factors to the design of medical equipment: patient-controlled analgesia. Journal of Clinical Monitoring and Computing 14(4): 253–263.

Lindberg, D.A.B. 1979. The Development and Diffusion of a Medical Technology: Medical Information Systems. Pp. 201–239 in Medical Technology and the Health Care System: A Study of the Diffusion of an Equipment-Embodied Technology. Washington, D.C.: National Academy of Sciences.

Littlejohns, P., J.C. Wyatt, and L. Garvican. 2003. Evaluating computerised health information systems: hard lessons still to be learnt. British Medical Journal 326(7394): 860–863.

Lui, W. 2002. Retinal Implant: Bridging Engineering and Medicine. Pp. 492–495 in Digest of the IEEE International Electron Devices Meeting, San Francisco, California, December 2002. New York: IEEE.

Markle Foundation. 2003. Connecting for Health Unites over 100 Organizations to Bring American Healthcare System into Information Age. Press release. Available online at: http://www.markle.org/resources/press_center/press_releases/2003/press_release_06052003.php.

Mastrangelo, C.H., M.A. Burns, and D.T. Burke. 1998. Microfabricated devices for genetic diagnostics. Proceedings of the IEEE 86(8): 1769–1787.

Miller, R.H., and R.R. Bovbjerg. 2002. Efforts to improve patient safety in large, capitated medical groups: description and conceptual model. Journal of Health Politics, Policy and Law 27(3): 401–440.

Miller, R.H., and H.S. Luft. 1994. Managed care plan performance since 1980: a literature analysis. Journal of the American Medical Association 271(19): 1512–1519.

Miller, R.H., and H.S. Luft. 2002. HMO plan performance update: an analysis of the literature, 1997–2001. Health Affairs 21(4): 63–86.

NCHVS (National Committee on Vital and Health Statistics). 2001. Information for Health: A Strategy for Building the National Health Information Infrastructure. Washington, D.C.: U.S. Department of Health and Human Services. Available online at: http://ncvhs.hhs.gov/nhiilayo.pdf.

New England Healthcare EDI Network. 2002. About NEHEN. Available online at: http://www.nehen.net/.

Nielsen, J. 1993. Usability Engineering. Boston, Mass.: Boston Academic Press.

NITRD (National Coordination Office for Information Technology Research and Development). 2004. Revolutionizing Health Care Through Information Technology. Arlington, Va.: NITRD. Available online at: http://www.itrd.gov/pitac/reports/20040721_hit_report.pdf.

Norman, D.A. 1990. The “problem” of automation: inappropriate feedback and interaction, not “over-automation.” Philosophical Transactions of the Royal Society of London B327: 585–593.

NRC (National Research Council). 2000. Networking Health: Prescriptions for the Internet. Washington, D.C.: National Academy Press.

NRC. 2002. At What Price?: Conceptualizing and Measuring Cost-of-Living and Price Indexes, edited by C.L. Schultze and C. Mackie. Washington, D.C.: National Academy Press.

NRC. 2004. Summary of a Workshop on Software Certification and Dependability. Washington, D.C.: National Academies Press.

Obradovich, J.H., and D.D. Woods. 1996. Users as designers: how people cope with poor HCI design in computer-based medical devices. Human Factors 38(4): 574–592.

Overhage, J.M. 2003. Improving Patient Safety in Chronic Diseases Using Electronic Medical Records. PowerPoint presentation to the IOM Committee on Data Standards for Patient Safety, Washington, D.C., April 1, 2003.

Patient Safety Institute. 2002. Presentation to the 5th National HIPAA Summit—Beyond HIPAA: Clinical Data Standards and the Creation of an Interconnected Electronic Health Information Infrastructure. Available online at: http://www.ehcca.com/presentations/HIPAA5/walker.pdf.

Patterson, E.S., M.L. Rogers, and M.L. Render. 2004. Fifteen best practice recommendations to improve the effectiveness of bar code medication administration. Joint Commission Journal on Quality and Safety 30(7): 355–365.

Pelletier, F. 2004. Wireless tech allies with low-power gear. EE Times, August 16, 2004. Available online at: http://www.eetimes.com/article/showArticle.jhtml?articleId=26806906.

Pennazio, M., R. Santucci, E. Rondonotti, C. Abbiati, G. Beccari, F.P. Rossini, and R. De Franchis. 2004. Outcome of patients with obscure gastrointestinal bleeding after capsule endoscopy: report of 100 consecutive cases. Gastroenterology 126(3): 643–653.

Perlin, J.B., R.M. Kolodner, and R.H. Roswell. 2004. The Veterans Health Administration: quality, value, accountability, and information as transforming strategies for patient-centered care. American Journal of Managed Care 10(11 Pt 2): 828–836.

Pestotnik, S.L., D.C. Classen, R.S. Evans, and J.P. Burke. 1996. Implementing antibiotic practice guidelines through computer-assisted decision support: clinical and financial outcomes. Annals of Internal Medicine 124(10): 884–890.

PITAC (President’s Information Technology Advisory Committee). 2004. Revolutionizing Health Care through Information Technology. Arlington, Va.: National Coordination Office for Information Technology Research and Development.

Rae, A., P. Ramanan, D. Jackson, and J. Flanz. 2003. Critical Feature Analysis of a Radiotherapy Machine. Pp. 231–234 in Computer Safety, Reliability, and Security: Proceedings of the 22nd International Conference (SAFECOMP 2003), Edinburgh, U.K., September 23–26, 2003. New York: Springer-Verlag.

Robinson, J.C. 2001. Theory and practice in the design of physician payment incentives. Milbank Quarterly 79(2): 149–177.

Rogers, W.A., A.L. Mykityshyn, R.H. Campbell, and A.D. Fisk. 2001. Analysis of a “simple” medical device. Ergonomics in Design 9(1): 6–14.

Rosenthal, M.B., R. Fernandopulle, H.R. Song, and B. Landon. 2004. Paying for quality: providers’ incentives for quality improvement. Health Affairs 23(2): 127–141.

Sarata, A. 2002. Adverse Drug Events, Medication Errors, and CPOE: A Case Study. Prepared for the Committee on Engineering and the Health Care System. Washington, D.C.: National Academy of Engineering.

Sarter, N.B., D.D. Woods, and C.E. Billings. 1997. Automation Surprises. Pp. 1926–1943 in Handbook of Human Factors/Ergonomics, 2nd ed., edited by G. Salvendy. New York: John Wiley and Sons.

Serruya, M.D., N.G. Hatsopoulos, L. Paninski, M.R. Fellows, and J.P. Donoghue. 2002. Instant neural control of a movement signal. Nature 416(6877): 141–142.

Shojania, K.G., D. Yokoe, R. Platt, J. Fiskio, N. Ma’luf, and D.W. Bates. 1998. Reducing vancomycin use utilizing a computer guideline: results of a randomized controlled trial. Journal of the American Medical Informatics Association 5(6): 554–562.

SNL (Sandia National Laboratories). 1996. Strategies for the Future: The Role of Technology in Reducing Health Care Costs. Albuquerque, N.M.: Sandia National Laboratories.

Spelman, F.A. 1999. The past, present, and future of cochlear prostheses. IEEE Engineering in Medicine and Biology Magazine 18(3): 27–33.

Stangel, K., S. Kolnsberg, D. Hammerschmidt, B.J. Hosticka, H.K. Trieu, and W. Mokwa. 2001. A programmable intraocular CMOS pressure

sensor system implant. IEEE Journal of Solid-State Circuits 36(7): 1094–1100.

Taylor, D.M., S.I.H. Tillery, and A.B. Schwartz. 2002. Direct cortical control of 3-D neuroprosthetic devices. Science 296(5574): 1829–1832.

Teich, J.M., P.R. Merchia, J.L. Schmiz, G.J. Kuperman, C.D. Spurr, and D.W. Bates. 2000. Effects of computerized physician order entry on prescribing practices. Archives of Internal Medicine 160(18): 2741–2747.

Thompson, T.G., and D.J. Brailer. 2004. The Decade of Health Information Technology: Delivering Consumer-centric and Information-Rich Health Care: Framework for Strategic Action. Washington, D.C.: U.S. Department of Health and Human Services.

Triplett, J.E., ed. 1999. Measuring the Prices of Medical Treatments. Washington, D.C.: Brookings Institution Press.

Triplett, J.E., ed. 2001. Measuring Health Output: The Draft Eurostat Handbook on Price and Volume Measures in National Accounts. Presentation at the Eurostat-CBS Seminar, Voorburg, Netherlands, March 2001.

VHA (Veterans Health Administration). 2005a. Health Community IT Sharing. Available online at: http://www1.va.gov/vhaitsharing/page.cfm?pg=7.

VHA. 2005b. What is VistA? Available online at: http://www.vistasoftware.org/vista/.

Walker, J., E. Pan, D. Johnston, J. Adler-Milstein, D.W. Bates, and B. Middleton. 2005. The value of health care information exchange and interoperability. Health Affairs Web Exclusive (Jan. 19): W5-10–W5-18.

Wang, S.J., B. Middleton, L.A. Prosser, C.G. Bardon, C.D. Spurr, P.J. Carchidi, A.F. Kittler, R.C. Goldszer, D.G. Fairchild, A.J. Sussman, G.J. Kuperman, and D.W. Bates. 2003. A cost-benefit analysis of electronic medical records in primary care. American Journal of Medicine 114(5): 397–403.

Wears, R.L., and M. Berg. 2005. Computer technology and clinical work still waiting for Godot. Journal of the American Medical Association 293(10): 1261–1263.

Weingarten, S.R., J.M. Henning, E. Badamgarav, K. Knight, V. Hasselblad, A. Gano Jr, and J.J. Ofman. 2002. Interventions used in disease management programmes for patients with chronic illness—which ones work? Meta-analysis of published reports. British Medical Journal 325: 925–942.

Whitten, P., G. Doolittle, M. Mackert, and T. Rush. 2003. Telehospice: end-of-life care over the lines. Nursing Management. 34(11): 36–39.

Winograd, T., and D.D. Woods. 1997. Challenges for Human-Centered Design. In Human-Centered Systems: Information, Interactivity, and Intelligence, edited by J. Flanagan, T. Huang, P. Jones, and S. Kasif. Washington, D.C.: National Science Foundation.

Wise, K.D. 1996. Microelectromechanical Systems: Interfacing Electronics to a Non-Electronic World. Pp. 11–18 in International Electron Devices Meeting Technical Digest, 1996. New York: IEEE.

Wise, K.D. 2002. Wireless Implantable MicroSystems: Coming Break-throughs in Health Care. Pp. 106–109 in Symposium on VLSI Circuits Digest of Technical Papers, 2002. New York: IEEE.

Wise, K.D., D.J. Anderson, J.F. Hetke, D.R. Kipke, and K. Najafi. 2004. Wireless Implantable Microsystems: Electronic Interfaces to the Nervous System. Proceedings of the IEEE 92(1): 76–97.

Woods, D.D. 2000. Behind Human Error: Human Factors Research to Improve Patient Safety. Washington, D.C.: American Psychological Association.

Yasnoff, W.A., B.L. Humphreys, J.M. Overhage, D.E. Detmer, P.F. Brennan, R.W. Morris, B. Middleton, D.W. Bates, and J.P. Fanning. 2004. A consensus action agenda for achieving the National Health Information Infrastructure. Journal of the American Medical Informatics Association 11(4): 332–338.

Ziaie, B., J.A.V. Arx, M.R. Dokmeci, and K. Najafi. 1996. A hermetic glass-silicon micropackage with high-density on-chip feedthroughs for sensors and actuators. IEEE Journal of Microelectromechanical Systems 5(3): 166–179.

This page intentionally left blank.