Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
4 Information and Communications Technology Infrastructure: A "Paperless" Health Care System SUMMARY DESCRIPTION The wise use of information and communications technology (ICT) has the potential to improve the quality and safety of health care while at the same time enhancing access and reducing waste, unnecessary delays, and administrative costs. Some degree of improvement results from use of the recent rapid and dramatic advances in moving information among a variety of clinicians and experts. Even more important over the long run, however, is the capacity of ICT to make it possible to work differently and better. The provision of health care consists of very complex processes, including initiating efforts to diagnose or treat a patient's problem; receiving and acting on the results of laboratory, radiology, anti other diagnostic tests; communicating with patients about results and progress; and monitoring patient progress and ensuring appropriate follow-up (Middieton et al., forthcoming). At every step of the way, communication among providers and between clinicians and patients is critical, and infor- mation and knowledge must be made available to users when they need it. . Demonstration projects in this category would be 5 years in duration and are intended to result in the establishment of a state-of-the-art health care ICT infrastructure in a state, community, or a multi- state region that is accessible to all providers and all consumers. Most patient information should be computer-based, virtually eliminating processes based on moving, finding, or filing paper. There would still be some situations in which paper is used (for example, an attending physician may find it useful to carry a concise summary of patients' clinical information when making rounds or to make handwritten notes that are subsequently scanned into the computer-based record). However, the volume of paper would be significantly reduced. More important, ambulatory practices, hospitals, [A
ICT Infrastructure ancillary providers, and others would be inter- connected, and the availability of patient and other information to support decision making at the time of care delivery would be dramatically improved. All demonstration projects in this category would include the formation of some form of public-private partnership. In general, the demonstrations would encompass three phases: 1. A planning and preparation phase, which includes the establishment of a public- private partnership. 2. 3. Establishment of a secure platform for com- munication and sharing of clinical and other data between patients and providers and among providers. This infrastructure- building phase is intended to allow rapid movement of computer-based information to multiple sites on a need-to-know and right-to-know basis. Rapid incorporation of the information made accessible through the platform to support (1) the steady migration of adminis- trati~e and business processes to the plat- form (e.g., appointment scheduling, insur- ance eligibility checking, billing and payment), (2) development and application of knowledge management and decision support tools, and (3) development of new e-health delivery modes. Critical to accom- plishing this phase is the computerization of all or nearly all types of patient data. Phase ~ should be accomplished within 6 months and phase 2 between months 7 and 24. During phase 2, the community would begin to reap tangible benefits from the demonstration project. Phase 3 would proceed in parallel with phase 2 and continue indefinitely, producing stepwise benefits with enhancements to the applications performed using the platform. Demonstration projects in this category would be greatly facilitated by an immediate emphasis on accelerated development of 58 1~ national data standards in certain key areas. All projects should be required to conform to national standards where they exist and to feed back experience on the use and utility of these new standards, thus facilitating the development of even more robust standards. Furthermore, there should be an expectation that the projects will lend their expertise to and share their tech- nology with other geographic areas. The initial demonstration projects should serve as the first generation or starter nodes of a national health information infrastructure. The committee recommends that a total of ~ to 10 demonstration projects be funded in this cate- gory, with the expectation that a second genera- tion will be funded in 2005. One-time-only federal financial support would be required to carry out phases ~ and 2. In general, private-sector health care providers should commit to making the ongoing financial investments necessary to carry out the develop- mental activities in phase 3. Payment structures would need to be realigned so that providers will benefit financially from the ongoing invest- ments required to sustain the [CT infrastructure. Some combined federal and state assistance to safety net providers would be needed to enable their full participation in the ICT infrastructure. BACKGROUND The potential for information technology to improve the quality and safety of health care is enormous (Institute of Medicine, 1991, 1997, 2001~. A strong ICT infrastructure can help patients be more informed and better prepared to engage in decision making and to carry out their treatment plan. From a clinician's perspec- tive, [CT includes many powerful tools that are useful in applying scientific evidence to clinical decision making and in providing care that is safe. Specifically, the ICT infrastructure can contribute to improvements in the following areas: . Communication Web-based interactions between patients and clinicians and among
ICT Infrastructure members of the care team, including e-mail, home monitoring, teleconsulting, and other applications. . . . Access to patient information Computer- based health and clinical information for each patient that is complete, organized, and available in real time to the patient and the patient's providers, while at the same time being confidential and secure. Knowledge management Easy access to reliable information from the science base in forms that are useful to both clinicians and patients (and accommodate both English- and non-English-speaking patients and different levels of health literacy). Decision support Computer-aided deci- sion support tools for both patients and clinicians, such as reminder systems targeted at clinicians or patients, medication order entry systems, and chronic disease management systems. Properly structured ICT also has great potential to reduce some administrative costs and burden. Administrative costs are estimated to account for 20 to 30 percent of the nation's health care expenditures (Evans and Roos, 1999; WooThandier, 1997; Woolhandler and Himmelstein, 2002~. The development of a secure ICT platform to support clinical, admin- istrative, and financial transactions, as well as the use of computer-based clinical records, should over time reduce some administrative costs and dramatically improve the effective- ness, safety, and timeliness of the health care system. As the ICT infrastructure expands beyond the boundaries of a single or several organiza- tions to span a community, a state, and eventu- ally the nation, there is the potential for it to support many additional clinical, research, professional education, and public health appli- cations (National Research Council, 2000~. A community-wide infrastructure can facilitate access to far more complete clinical information by all of the patient's providers, not just those within a given institution (e.g., hospital or health system). This interconnectedness offered by a Web-based platform, accompanied by protocols for data exchange, has great potential to improve quality and efficiency, especially in communities where health care delivery is highly decentralized. Clinical research applica- tions include the development of large-scale biomedical databases that can support popula- tion-based, longitudinal studies. Professional education applications include improved access to the knowledge base, virtual conferences, and other distance-learning applications. Of particular importance in light of height- ened concerns about chemical and biological terrorism are the public health applications of an ICT infrastructure. These applications include improved tracking and surveillance for bioter- rorism and other public health threats, and rapid dissemination of important health and medical information to providers and possibly citizens (Tang, 2002~. With the passage of the Health Insurance Portability and Accountability Act of ~ 996 (HIPAA) (Public Law 104-191) and the Depart- ment of Health and Human Services' recent promulgation of regulations to protect data r ~ -A privacy and security, one major impediment to the establishment of an ICT infrastructure has been removed. The time is right for widespread adoption of computer-based clinical records. A health information infrastructure as envi- sioned by the National Committee on Vital and Health Statistics (2001) involves three dimen- s~ons: Personal health cttimension~reation of a personal health record by an individual (and his or her family or caregivers), which includes both information from the clinical record that is relevant to patients and non- cTinical information' such as self-care tracl`- ers and directories of health care and public health service providers. Health care provider dimension~reation of a computer-based clinical record includ- .~
f TOT Infrastructure ing such information as clinical orders, prescriptions, and clinician notes; ready access to practice guidelines and the science base; and the application of decision support tools. Population health dimension- Collection of information on the health of the population and the salient influences, making it possi- ble to identify and track health threats, assess population health, sponsor targeted health education campaigns, and conduct research. Much of the necessary data can come from deidentified personal and patient records that are accessed on a right- and need-to-know basis. Although the potential applications and benefits likely increase logarithmically as the boundaries of the [CT infrastructure expand, so, too, does the complexity of implementation. Establishing such an infrastructure involves the development of data standards, laws, regula- tions, business practices, and technologies (National Committee on Vital and Health Statis- tics, 2000~. Payment policies must be modified to recognize new methods of care delivery (e.g., e-visits, remote consultations, remote monitor- ing of patients). In many of these areas, federal leadership will be needed to move forward, because numerous issues of clinical data owner- ship, security, and privacy have yet to be addressed. The up-front capital investments required to establish the ICT infrastructure will be quite siz- able. The United States has successfully confronted similar challenges, however. Follow- ing World War II, the federal government supported the development of the Interstate highway system, and years later, the Defense Advanced Research Projects Agency fitnded the work that led to the modern Internet (National Research Council, 2000; Weingroff, ~ 996~. Today, the challenge is not just one of ensuring access to public health data for homeland secu- rity, but also one of ensuring national health and economic productivity through a healthy work- force. The committee is confident that over time clinicians 1 the return on investment in terms of dollars and health will rival the success of these other major national infrastructure investments, and that Americans of all socioeconomic levels will benefit (U.S. Department of Commerce, 2000~. GOALS Demonstration projects in this category are intended to result in a practice environment that relies on computer-based patient records, and allows clinicians and patients immediate access to evidence and other information needed to support decision making. The goal is for the following capabilities to be available, for the most part, in real time: 1. Ready access to information by clinicians and patients on a right- and need-to-know basis Eligibility, appointment, and account status information Computer-based patient records; for example, a summary of current prob- lems, medications, and allergies, together with all results, notes, and summaries Disease management guidelines 2. Support tools for patients Educational materials Tools to monitor individual health status and needs Tools to track progress in meeting clini- cal goals and compliance with treatment plans Tools to model preferences and the impact of alternative treatment deci- sions and outcomes Tools to access patient records and to contribute information to these records 3. E-maiT and audiovisual communication between patients and clinicians and among
f ICT Infrastructure Nonepisodic care (e.g., short questions and answers) Home monitoring Remote consultation 4. Data capture and decision support tools for .. . . clinicians Reminder systems Computer-based order entry and pre- scription writing, with dosage and inter- action checking Note capture with immediate coding for billing Individual registries for chronic disease management 5. Management Appointment scheduling and admis- sions Workload scheduling and staffing Automated charge capture Contract modeling Enrollment in insurance programs . - Claim subrn~ssion, with immediate veri- fication of completeness and eligibility for payment 6. Performance measurement data for ongoing assessment of quality and safety improve- ments 7. Accountability Compliance with performance reporting requirements in the areas of safety and quality Demonstration of improvements in safety and quality Reporting of relevant data to public health authorities DEMONSTRATION ATTRIBUTES Demonstration projects in this category 6 WOUi] likely include the three phases enumer- ated earlier. The work in phases ~ and 2 should establish an initial track record of stakeholders working together on a substantial statewide undertaking that has enormous potential to benefit patients and the community but also requires extensive collaboration and shared decision making. Although critical, these phases would accomplish only a subset of the goals listed above. Phase 3 would entail extensive change within every health care organization. Emphasis is on the computerization of many types of patient and other data that currently are paper-based in most health care settings. This phase would also involve doing work differ- ently. The increasing computerization of clinical data and the availability of knowledge manage- ment and other decision support tools would open up new opportunities to redesign care processes in ways that would improve safety, effectiveness, and efficiency and be more responsive to patient preferences and needs. Phase 3 would continue indefinitely as advances in medicine, science, and technology offer new possibilities, but by year 5, each demonstration site should be within reach of the goal of a near "paperless" health care system. Each of the three phases is discussed in detail below. Phase I: Planning and Preparation During this phase, a lead organization, such as a state agency, an academic health center, a consortium of providers, or a state hospitaV health system association, would be responsible for establishing a broad-based public-private partnership. It is critical that the commitment of all major stakeholders in the community be secured at this stage. The demonstration projects would directly impact patients, physi- cians, nurses, hospitals, nursing homes, home health agencies, free-standing clinical laborato- ries, imaging centers, and others. Careful atten- tion should be paid to ensuring that each of these stakeholders derives both near- and iong- term benefits Tom the ICT infrastructure. Strong leadership from professional and hospi- tal associations, academic health centers, the state government, and consumer groups would be needed. IT
~ ICT Infrastructure This first phase would also involve the development of a detailed operational plan for carrying out the 5-year demonstration project. The operational plan should include proposed contractual agreements with all participants and vendors; a business and financial plan; and poli- cies and procedures for data access and manage- ment, including security and authentication. Phase 2: Data Exchange PIafform The primary advantage of establishing a data exchange platform is that it facilitates access to computer-based patient inflation in real time regardless of the form in which that information is currently being collected, coded, and classified by the provider or ancillary service organization. Establishment of a plat- form for sharing data poses a reasonably low risk of failure because it does not require providers (e.g., hospitals, health systems, physi- cian groups) or other participating organizations (e.g., laboratories, imaging centers) to make major changes in their existing systems. Effect- ing the collaboration required for such data exchanges can be challenging, however, and hence leadership from the federal government could be quite useful in this regard. Perhaps the best-known example of a data exchange platform for patient information is operated by CareScience in Santa Barbara There are other examples of data sharing County, California (carescience' 2001~. Care networks. With support from the National Data Exchange was initiated in 1991 as a low- Library of Medicine ~M), the Regenstrief cost public utility mode] that would allow Institute has developed a community-wide elec- participat~ng organizations to share data through Ironic medical record system called the Indiana an Tnternet-based system About 25 participat- Network for Patient Care (lNPC) (Regenstrief sing organizations accounting for the majority of Institute, 2002~. The lNPC is being used by ~ ~ care provided in Santa Barbara County include geographically separated hospitals in the Indian- hosp~tals and health systems, physician prac- apolis metropolitan area to improve care for tices laboratories and ancillary providers clin- ~ ~ patients presenting to their emergency depart- ics and outpatient facilities public health agen- ~ ~ ~ ~ meets. In W~nona, Minnesota, Cerner Corpora- cies, health plans and payers, pharmacies, an] lion and Hiawatha Broadband Communications employers. The key characteristics of the initiated. a partnership with Winona Health to system are as follows create W~nona Health Online, a network con- necting the local health system, clinics, physi- cians, and patients (Chin, 2000~. In addition, the Patient Safety Institute (PSI), a nonprofit organization that has developed a secure com- . Users (i.e., clinicians, hospitals, laborato- ries) need only a Web browser and an Inter- net connection to participate. Patient data remain at the* original sources (e.g., hospital legacy system, imaging center data system), but authorized users are allowed to view the data. Participants determine who can and cannot access patient data according to protocols. For instance, when a participant requests patient information, the data exchange veri- fies the requester's digital credentials. Patients do not have unique identifiers; rather, the data exchange maintains a file with patient demographic data and corre- lates these data with those maintained by the provider organization to produce a vali- dated patient search. The locations of the patient records are then stored with the patient's demographic data as "pointers" or "locators." The data exchange maintains a secure portal through which data may be exchanged "peer-to-peer." The data exchange maintains audit logs of all requests (i.e., who, when, what data requested). t 62
ICT Infrastructure munications network for real-time access to clinical information at the point of care, has selected two hospital sites in Washington State to demonstrate this network (Patient Safety In- stitute, 2002~. One of the major benefits of a data exchange is that it makes more complete clinical data available at the point of service. The availability of this information should result in fewer medical errors, less waste in terms of redundant services, and improved timeliness of services. The magnitude of benefits derived from a data exchange depends on the extent to which patient data are computer-based. The types of transactions and applications that can be carried out over the network expand dramatically with the availability of more computer-based clinical and other patient data. The committee expects that the mere existence of a data exchange within a community would generate momentum to add other types of data. Phase 3: Comprehensive ICT Infrastructure All of the demonstration projects are intended to result in a health care system that relies on computer-based patient records. The committee recognizes that not all sites will achieve this goal by year 5, but believes that most can make considerable progress in com- puterizing key aspects of the patient record, such as laboratory, imaging, and prescription drugs. . . . ~ ~ . · . The immediate objective of these demon- strations is to provide electronic access to those portions of a patient's clinical record that are computerized, and to encourage steady progress towards more complete computerization of patient information. Of course, access would be limited to authorized users (e.g., clinicians, ancillary service organizations, patients and their families). The ICT infrastructure should facilitate Internet-based communication between patients and clinicians, between patients and patient groups, and among cTini- cians. Over time, the infrastructure should be capable of securely supporting a wealth of applications, including (~) insurance enrollment and eligibility verification, (2) claims process- ing and payment, (3) clinical knowledge man- agement for clinicians and patients (e.g., guide- lines, chronic disease management sites), (4) decision support tools (e.g.' reminders, pro- vider order entry systems, (5) telemedicine (e.g., specialist consults, in-home monitoring), (6) disease surveillance, and (7) a public health rapid alert system. The demonstration sites selected should be ones for which there is a reasonable expectation that this more extensive ICT infrastructure can be established within 5 years. This may mean selecting sites in geographic areas in which some investments in building an ICT infrastruc- ture have already been made, the necessary expertise exists, and there is strong interest in working toward the accomplishment of commu- nity-wide goals. Sites that already have some building blocks in place will be able to move more expeditiously and will be more likely to achieve the 5-year goal. The potential for resis- tance to change from one or more important stakeholders should not be underestimated. Applicants should also be asked to detail the benefits that each major stakeholder is expected to derive from the demonstration project during the first few years. In many geographic areas, there are pockets of innovation where strong ICT infrastructures are already available. A number of academic health centers, health systems, and large multis- pecialty groups have developed strong TCT capabilities (Doolan and Bates, 2002; McDon- ald et al., 1999; Overhage et al., 2000; Teich et al., 1999), and it should be possible for some communities and regions to build upon these successful efforts. In the public sector, the Veterans Health Administration (VHA) has one of the largest and most sophisticated health information systems in the nation, serving about 5 million veterans annually in 22 designated regions. The VHA system includes an architec- ture that supports information exchange across multiple clinical disciplines and lines of busi- ness, a computer-based patient record system for clinical documentation and information
f ICT ]:nfrastructure retrieval, and an e-health communications system to provide veterans with online access to their medical records and other health informa- tion (Christopherson, 2002~. Careful consideration should be given to the best means of creating a public-private partner- ship in a geographic area to leverage existing resources and to ensure that no providers or consumers are excluded. One possibility might be for the state government, VHA, and private- sector health care organizations and vendors to work in partnership to establish the ICT infra- structure. Additional support might be provided to VHA so it can offer safety net providers (e.g., public hospitals, community health centers) the opportunity to participate in its ICT system and receive technical assistance for that purpose. Another possibility for providing assistance to safety net providers would be to allow a state Medicaid program to work collaboratively with such providers (who generally receive a sizable proportion of their revenues from Medicaid and the State Children' s Health Insurance Program) to establish the necessary ICT infrastructure, and for the federal government to provide a 90 percent federal matching rate for these expenditures in the same way it pays for state- leve! Medicaid Management Information System development expenditures under Medi- caid (Congressional Research Service, 1993~. MAKING PROGRESS TOWARD A NATIONAL ICT INFRASTRUCTURE There is a good deal of truth to the saying that "all health care is local." For the most part, health care is provided by local institutions and clinicians. Communities take great pride in their health systems, and efforts to develop TCT infra- structure would best build on this foundation. As important as grassroots support may be, however, a strong federal role in setting stan- dards for ICT infrastructure is critical for several reasons. First, patient care-seeking behavior does not respect strict geographic boundaries, and this will likely be the case increasingly in the fixture. An estimated 25 per- cent of the U.S. population resides in a metro- politan area that crosses state boundaries (Salinsky, 2002), and many likely receive services in more than one state. The Internet has opened up many opportunities for telemedicine, a method of health care delivery that often crosses state and sometimes national bounda- ries. A patient' s computer-based record should be accessible to all health care providers regard- less of geographic location. Second, the capability to monitor and respond to infectious disease outbreaks and bioterrorist attacks requires a national if not global ICT infrastructure. The ICT infrastruc- tures developed by the various demonstration sites should be able to interface with each other, and over time these state- or community-based systems should become part of a nationwide infrastructure. Third, the return on investment in knowI- edge management and other clinical decision support tools would be much greater if the soft- ware can be readily used in multiple geographic settings. The promulgation of national standards in some areas is critical to achieving this objec- tive. Fourth, many health care administrative functions, such as insurance and federal regula- tory requirements, are national in scope. National data standards in certain areas are im- portant to lessen the burden associated with provider compliance with these requirements. The committee wishes to emphasize the importance of viewing the state- or community- based ICT demonstrations as part of a broader strategy intended to result in the establishment of a national ICT infrastructure. Specifically, demonstration projects in this category are intended to accomplish two objectives: . Each individual demonstration project should result in the establishment of a com- munity- or state-wide ICT infrastructure that begins to yield benefits to the commu- nity in the near term. Collectively, the demonstration projects in this category should within 3 years consti-
I:CT Infrastructure tute primary or starter nodes of what will be expanded to constitute a national ICT infrastructure . One impediment to establishing an ICT infrastructure either locally or nationally is the difficulty of getting all participating health care providers and vendors to agree on uniform data standards (i.e., the methods, protocols, and terminology adhered to by everyone for the purpose of allowing disparate information systems to interoperate successfully). The term "uniform data standards" is not intended to im- ply that all clinical and other information in the patient record must be coded and classified in exactly the same way, although in some instances strict adherence to what is known as "controlled vocabulary" may be desirable. In many cases, the development of"reference stan- dards or terminology" may be adequate, with all participants agreeing to code data at a sufficient level of detail to allow mapping of local data to the national reference standard (Stead et al., 2000). forward by implementing the standards recom- mended by the National Committee on Vital and Health Statistics in reports sent to the Secre- tary in the past 3 years. In particular, the report entitled Uniform Data Stanclards for Patient Medical Record Information contains a number of standards that deserve prompt use in the Medicare and Medicaid programs and within all portions of the federal government that utilize health data standards (National Committee on Vital and Health Statistics, 2000~. A partial list of these standards is Health Level 7 (HL7), Version 2; Digital Imaging and Communica- tions in Medicine (DICOM); National Council for Prescription Drug Programs (NCPDP); and for early adoption, Institute of Electrical and Electronics Engineers, Series 1073 (IEEE 1073) Medical Device Communications and HL7, Version 3. It will be critical for the Secretary of Health and Human Services, working with the Presi- dent and Congress, to pursue an aggressive agenda for the adoption of national health data standards. Some degree of standardization is important to improve the interoperability of ICT systems; the comparability and usefulness of the data; and the quality, accountability' and integ- rity of the data. Numerous expert panels have recommended a stronger federal role in the promulgation of standards to be used nation- wide (National Committee on Vital and Health Statistics, 2001; National Research Council, 2000), but to date, progress in this area has been slow. The proposed demonstration projects are unlikely to thrive in the absence of progress in establishing data standards. The American public will reap a far greater return on invest- ments in these demonstration projects if they are pursued in tandem with national and global efforts to establish data standards. The Secretary of Health and Human Services can immediately move this agenda Of particular importance is the work of HL7, one of the largest pnvate-sector standards- setting organizations, focusing on Version 3 standards for data interchange. With some addi- tional financial resources, the work of this group could be accelerated, thereby completing some important aspects of standards setting within about 18 months (Hammond, 2002~. The com- pletion of this work would greatly facilitate the development of the ICT demonstration projects proposed in this report and perhaps even reduce the costs of such demonstrations. There are some important standards-setting efforts in the federal government as well. The work of the Consolidated Health Informatics Initiative (CHTI) and NLM is particularly note- worthy and can be leveraged. CHIl was created under the auspices of the White House Office of Management and Budget in 2001 to facilitate the development of standards that would ensure compatible ICT systems across the government health programs (Office of Management and Budget, 2002~. NLM has made extensive efforts through the Unified Medical Language System project to consolidate the various controlled vocabularies used in medicine and to provide tools needed to make the relationships among terms clear to human and machine alike (Humphreys et al., l 998; Lindberg et al., l993~. _
f ICT Infrastructure These groups are well positioned to play impor- tant roles in the further development of clinical data standards. Recent private-sector efforts also present an excellent opportunity for the federal govern- ment to collaborate in and obtain broad-based support for national standards-setting initiatives. Specifically, the Markle Foundation's Connect- ing for Health Initiative is focused on building consensus on and accelerating the rate of adop- tion of clinical data standards within health care. The group has already obtained general agree- ment among a broad stakeholder group on several clinical data standards that are ready for adoption, and more work is under way. An early product of this effort is an inventory of stan- dards domains necessary to support computer- based records; these include reference informa- tion models, data types, terminology, clinical documents, clinical templates, data interchange, implementation manuals, rules, toot sets, termi- nology services, security, unique identifiers for patients and clinicians, and guidelines (Marble Foundation, 2002~. The IOM Committee on Pa- tient Safety Data Standards, scheduled to release a report in fall 2003, is currently work- ing to identify the types of standards needed for monitoring safety and quality, and is providing guidance to the federal government on actions it might take to facilitate the development of such standards. Although a more intensive effort at the national level to establish data standards is de- sirable, the committee does not intend to imply that the development of ICT infrastructures at the state or community level should be post- poned until national standards are promulgated. The setting of national standards will be an ongoing process for many years if not decades to come. All information systems will need to accommodate frequent updates in standards. Also, as discussed above, some national data standards provide room for local sites to map their coding systems to a national reference standard. The goal of having these initial demonstra- tion projects serve as the hubs of a national ICT infrastructure has implications for the selection 66 11 of demonstration sites. Consideration should be given to ensuring balanced geographic represen- tation when selecting the demonstration sites. All major regions of the country (northeast, south, central, northwest, and southwest) and both urban and rural areas should be represented since the issues involved in implementing ICT capabilities differ greatly depending on the size and [CT sophistication of clinical groups. All participants in the demonstration program should be willing to sponsor site visits so others can learn from their experiences. The Agency for Healthcare Research and Quality should be charged with establishing a central coordinating unit for the demonstrations in this category. Such a unit might be estab- lished in an academic or other applied research setting. The coordinating unit should be tasked with the following: Monitoring and evaluating the progress of the demonstrations Documenting successful and unsuccessful change management strategies pursued by the various demonstration sites Establishing ongoing communication among demonstration sites to share learning experiences and to discuss how best to over- come challenges Quantifying the benefits and costs associ- ated with the demonstrations, and identify- ing and communicating best practices in care redesign This last point deserves special attention. Reap- ing a return on investment in the ICT infrastruc- ture will require changes in care processes. The costs associated with making these changes should be captured and included in the cost- benefit analysis. Another complicating factor is that benefits from investing in ICT accrue to many stakeholders, and not necessarily those who made the investment. Substantiating who benefits and who pays should be part of the evaluation.
f ICT Infrastructure The coordinating unit should have a strong communication capacity. It should be responsi- ble for reaching out to various stakeholders, such as professional societies, health care asso- ciations, academic health centers, and others, to assist in disseminating knowledge gained Tom the demonstrations and in building a strong base of support for the establishment of a national ICT infrastructure. and D. D. Detmer. Washington, D.C.: National Academy Press. . 2001. Crossing the Quality Chasm: A New Health System for the 21st Century. Washington DC: National Academy Press. Lindberg, D. A., B. L. Humphreys, and A. T. McCray. 1993. The Unified Medical Language System.MethodsInfMed 32~4~:281-91. Markle Foundation. 2002. Connecting for Health Data Standards Working Group: Standards As- REFERENCES sessment Matrix (Draft). CareScience. 2001. 'Care Data Exchange." Online. Available at http://www.carescience.com/ healthcare~roviders/cde/care_data_exchange. shtml "accessed Sept. 5, 20024. Chin, T. 2000. "The Winona Project: Developing an Electronic Link." Online. Available at hips// www.ama-assn.org/sci-pubs/amnews/pick_02/ bisaO311.h~n [accessed Oct. 28, 20023. Christopherson, G. (VHA). 6 June 2002. Personal communication to Barbara Smith (Institute of Medicine). Congressional Research Service. 1993. Medicaid Source Book: Background Data and Analysis. Committee Print 103-A. Washington, D.C.: U.S. Government Printing Office. Doolan, D. F., and D. W. Bates. 2002. Computerized Physician Order Entry Systems in Hospitals: Mandates and Incentives. Health Aff (Millwood) 21~4~:180-8. Evans, R., and N. P. Roos. 1999. What Is Right About the Canadian Health Care System? Mil- ba7~k Q 77 (3):393-99. Hammond, W. E. 2002. Memorandum on HL-7 Acceleration Project. Humphreys, B. L., D. A. Lindberg, H. M. School- man, and G. O. Barnett. 1998. The Unified Medical Language System: an Informatics Research Collaboration. JAm Med Inform Assoc 5 (1~:1-11. Institute of Medicine. 1991. The Computer-Based Patient Record: Art Essential Technology for Health Care. eds. R. S. Dick and E. B. Steen. Washington, D.C.: National Academy Press. . 1997. The Computer-Based Patient Record: An Essential Technology for Health Care (Revised Edition9. eds. R. S. Dick, E. B. Steen, McDonald, C. J., J. M. Overhage, W. M. Tierney, P. R. Dexter, D. K. Martin, J. G. Suico, A. Zafar, G. Schadow, L. Blevins, T. Glazener, J. Meeks- Johnson, L. Lemmon, J. Warvel, B. Porterfield, J. Warvel, P. Cassidy, D. Lindbergh, A. Belsito, M. Tucker, B. Williams, and C. Wodniak. 1999. The Regenstrief Medical Record System: a Quarter Century Experience. Int J Med Inf 54 (3~:225-53. Middleton, B., T. K. Gandhi, and D. W. Bates. forth- coming. "The Role of Information Technology in Ambulatory Care Patient Safety." Online. Available at http://www.partners.org/cird/allref. asp Accessed in draft hard copy form]. National Committee on Vital and Health Statistics. 2000. "Uniform Data Standards for Patient Medical Record Information." Online. Available at http ://ncvhs.hhs . gov/hipaaO00706.pdf "accessed Aug. 30, 20023. . 2001. Information for Health. A Strategy for Building the National Health Information Infra- structure. Washington DC: U.S. Department of Health and Human Services. National Research Council. 2000. Networking Health: Prescriptions for the Internet. Wash- ington, D.C.: National Academy Press. Office of Management and Budget. 2002. "E- Government Strategy: Implementing the Presi- dent's Management Agenda for E-Government." Online. Available at http://www.whitehouse. gov/omb/inforeg/egovstrategy.pdf "accessed Sept. 30, 20023. Overhage, J., C. J. McDonald, and J. G. Suico. 2000. The Regenstrief Medical Record System 2000:Expanding the Breadth and Depth of a Community Wide EMR. Pro c A MIA Symp :1173. _ Lit
i ICT Infrastructure Patient Safety Institute. 2002. "PSI Selects Two Washington State Communities to Demonstrate Innovative Patient Safety Communications Net- work." Online. Available at ht~p://www.ptsafety. Orglnews/nr020502.asp [accessed Oct. 28, 2002]. Regenstrief Institute. 2002. "Electronic Medical Re- cords Development." Online. Available at hthp:// informatics.regenstrief.org/what [accessed Oct. 28, 2002]. Salinsky, E. 2002. Will the Nation Be Ready for the Next Bioterrorism Attack? Mending Gaps in the Public Health Infrastructure. NHPF Issue Brief: No. 776. Washington, D.C.: National Health Policy Forum. Stead, W. W., R. A. Miller, M. A. Musen, and W. R. Hersh. 2000. Integration and Beyond: Linking Information From Disparate Sources and into Workflow. J Am Med Inform Assoc 7 (2~: 135- 45. Tang, P. C. 2002. AMID Advocates National Health Information System in Fight Against National Health Threats. J Am Med Inform Assoc 9 (2~: 123-4. Teich, J. M., J. P. Glaser, R. F. Beckley, M. Aranow, D. W. Bates, G. J. Kuperman, M. E. Ward, and C. D. Spurr. 1999. The Brigham Integrated Computing System (BICS): Advanced Clinical Systems in an Academic Hospital Environment. IntJMedInf 54~3~:197-208. U.S. Department of Commerce. 2000. "Falling Through the Net: Toward Digital Inclusion. A Report on American's Access to Technology Tools.'t Online. Available at ht~://www.ntia. doc.gov/ntiahome/digitaldivide/ "accessed Sept. 30, 2002]. Weingroff, R. F. 1996. "Federal-Aid Highway Act of 1956: Creating the Interstate System." Online. Available at http:llwww. fhwa.dot.govlllllll infrastructure/rw96e.htm [accessed Nov. 5, 20024. Woolhandler, S. 1997. Costs of Care and Admini- stration at for-Profit and Other Hospitals in the United States. NEngl JMed 337 (24~:1783. Woolhandler, S., and D. U. Himmelstein. 2002. Na- tional Health Insurance - Liberal Benefits, Con- servative Spending. Arch Intern Med 162 (9~:973-75.