cclogoSMALL.gif (1930 bytes)

competetextgrey.gif (3110 bytes)



Table of Contents

Highway to Health: Transforming U.S. Health Care in the Information Age


Executive Summary

Chapter 1. Setting the Stage

Chapter 2. Remote Care

Chapter 3. Personal Health Information and Management

Chapter 4. Integration of Health Information Systems

Chapter 5. Health Care Research and Education

















cclogoSMALL.gif (1930 bytes)

competetextgrey.gif (3110 bytes)



U.S. R&D Policy for Competitiveness



Now, more than ever, research and development (R&D) drives the process of innovation that underpins our nation's economic well-being and national security. Together with the federal government, the members of the Council on Competitiveness -- private sector companies, academic institutions and labor unions -- are key participants in the U.S. R&D enterprise.

At no time in the last 50 years have there been so many forces and changes affecting the R&D enterprise as those that exist today. The most salient ones include:

  • the unprecedented mobility of capital and technology and use of human resources that have helped othernations develop their own R&D capabilities;
  • the rapid ascent of computer, information and communication technology to a level from which entirely new industries have evolved;
  • the emergence of civilian and commercial interests as the primary drivers of leading edge technology, rather than the defense sector;
  • the accelerated pace, increasing complexity, and new disciplines in science and technology that have transformed the innovation process itself; and
  • the mission to balance the federal budget.

These forces have overtaken the post-1945 system of innovation that developed and sustained U.S. technological preeminence. A new paradigm of R&D partnerships is emerging, based on the collaboration rather than the separation of key participants in the R&D enterprise. This new pattern is one in which the roles of the private and public sectors overlap in their missions to provide a viable science and technology infrastructure. It is based on concurrent technology processes, instead of the time-consuming, step-by-step procedures that in the past have transformed fundamental science into products for the market.

This report by the Council on Competitiveness provides guidelines for industry, government and academia that will enable the United States to make the paradigm shift that is necessary in the post-Cold War era. The Appendices of the report are in-depth assessments of R&D in six industry sectors that are central to U.S. competitiveness: Aircraft, Automotive, Chemical, Electronics, Information Technologies, and Pharmaceuticals.

While the lessons of these individual sector studies helped shape this report, the major findings emerged from the deliberations of a distinguished advisory committee of experts from academia, government and industry. We owe special thanks to the members of this group for their invaluable insights, although the Council assumes responsibility for the conclusions of the report.

The title itself, Endless Frontier, Limited Resources: U.S. R&D Policy for Competitiveness, makes a key point. The promises and expectations of R&D are increasing, but the resources needed to sustain the R&D effort are decreasing. In this environment, priorities not only need to be set, but knowledge and resources must be shared through wide-ranging collaborations involving companies, universities, and government agencies and laboratories.

This report contributes to the current assessment of the future of U.S. R&D but stands apart from other reports, not just by putting special emphasis on the need for partnerships but by framing guidelines for policy that are equally demanding for the industrial, government and academic sectors. Endless Frontier, Limited Resources also calls for a reasoned end to the unproductive ideological debate over the federal government's proper role in R&D.

The Council on Competitiveness and its members have a unique role to play in forging national consensus over the future of the R&D enterprise. The institutions from which our members are drawn operate at the cutting edge of innovation. They have first-hand experience with the new realities of the laboratory, and they provide the foundation in science, technology and education upon which U.S. competitive strength rests. In addition, the Council serves as a nonpartisan reservoir of expertise from which both the executive branch and Congress have drawn.

The Council feels an obligation to act on our recommendations and conclusions by:

  • convening a national discussion on the future of the R&D enterprise that will focus attention on the need to set priorities, foster partnerships and shape policies to position the United States for technological leadership in the 21st century;
  • stimulating our members to address the guidelines for industry and academia; and
  • working with Congress and the administration in a nonpartisan way to arrive at R&D strategies and programs that further the United States' competitiveness and standard of living.


Paul Allaire

Council Chairman

Chairman and Chief Executive Officer

Xerox Corporation

Jack Sheinkman

Council Vice Chairman

Amalgamated Bank of New York

Thomas E. Everhart

Council Vice Chairman


California Institute of Technology






cclogoSMALL.gif (1930 bytes)

competetextgrey.gif (3110 bytes)


Chapter 1 - Setting the Stage

Highway to Health: Transforming U.S. Health Care in the Information Age



With the increase in health care costs, the graying of the American population, and the shift toward managed care, few people are unaffected by the changes sweeping the health care industry. Attempts to reduce costs yet deliver quality care to as many citizens as possible have prompted those involved in providing health care products and services to look to technologies inherent in the national information infrastructure (NII) for solutions. A brief description of the market drivers for incorporating NII-related technologies into health care delivery is given followed by a conceptual framework for envisioning the interactions among the elements involved in creating an NII-based health care system.


Cost of Health Care

Health care costs are the fastest growing segment of government expenditures (federal, state, and local). The United States was expected to spend more than $1 trillion on health care in 1995, or a full 14 percent of the U.S. gross domestic product (GDP). This percentage is expected to grow to 16 percent of the GDP by the year 2000 and to reach 18 percent by 2005. This increase is being fueled by both a rising standard of living, which allows more to be expended for health care, and a faster rate of price increases for health care than for other segments of the economy (CBO 1995). (See Figure 1-1.)

As a nation, the U.S. leads the world in health care spending yet does not have the best outcomes overall. U.S. employers spend, on average, 8.2 percent of their payroll on health insurance while their counterparts in Germany and Japan spend approximately 6.4 percent and 4.2 percent respectively. Not only are governments and employers spending more on health care, so are patients. (See Figure 1-2.)


Shift to Managed Care

In an attempt to control escalating costs, the nation's private and public third-party payers have turned to shared-risk strategies, including managed care plans. Managed care, the most rapidly growing of these approaches, can be characterized as a health care delivery system that "integrates the financing and delivery of appropriate medical care by means of the following: contracts with selected physicians and hospitals that furnish a comprehensive set of health care services to enrolled members, usually for a predetermined monthly premium; utilization and quality controls that contracting providers agree to accept; financial incentives for patients to use the providers and facilities associated with the plan; and the assumption of some financial risk by doctors, thus fundamentally altering their role from serving as agent for the patient's welfare to balancing the patient's needs against the need for cost control." (Iglehart 1992). In order to balance care and cost, managed care companies are pursuing wellness and prevention programs, as well as disease management programs, to improve health outcomes. Managed care and other risk-sharing approaches to care delivery are evolving into capitation, where practitioners are paid to deliver primary care on a per member per month basis, and where they share hospital and specialty care costs with the managed care organizations. The feasibility of capitation depends on the use of data to predict lifetime health care costs based on outcomes.

The growth in managed care has been explosive. In 1976, only 6 million people were enrolled in a health maintenance organization (HMO). By 1992, that figure had risen to an estimated 41 million. Data from 1993 indicate an increase to 45 million enrollees with projections for 1995 topping 50 million (HIAA 1994). (See Figure 1-3.) The number of managed care plans has also increased, from 174 in 1976 to 551 in 1993 (Merrill Lynch 1995).

Despite the proliferation of managed care plans, data on their effectiveness in controlling costs have been mixed (Iglehart 1992). Larger employers seem to be benefitting more than smaller companies from the cost savings. Companies with 500 or more employees reported an average 1.9 percent decrease in benefit costs in 1993, while smaller employers saw their costs rise 6.5 percent (Currents 1995).


Aging Population

The changing age distribution in the U.S. is illustrated in Figure 1-4. The over-65 population is increasing, with the over-85 group increasing the most rapidly. Life expectancy rates have increased as well due to changes in public health and medical technology. As shown in Figure 1-5, they are expected to continue to increase. And as the population ages, its need for health care increases. This is compounded by the fact that women make up greater portions of the more advanced age groups. At present, women constitute 51 percent of the U.S. population, yet they account for 60 percent of the population over age 65 and more than 70 percent of the population above age 85 (HHS 1995b). This is significant because "women face health problems that accompany old age--such as osteoporosis, depression, Alzheimer's disease--in greater numbers than do men. Moreover, throughout their lives, women tend to suffer far more illness and chronic, debilitating conditions than do men. In fact, women's activities are limited by poor health approximately 25 percent more days each year than are men's activities, and women are bedridden 35 percent more days than men because of infectious diseases, respiratory problems, digestive disease, injuries, and other chronic conditions." (Blumenthal 1995)

Not only are Americans living longer, but they are requiring more care that is not covered by insurance plans. The Health Care Financing Administration (HCFA) reports that nursing home care costs were $66.3 billion in 1992, up 11 percent over 1991. Medicare accounted for $2.7 billion while Medicaid's portion was $34.4 billion. Private, long-term care insurance contributed about $700 million. That means that the remainder, over $26 billion, was paid for by patients and their families.



The NII can be thought of as four interdependent parts: a set of widely accessible and interoperable communications networks; digital libraries, information databases, and services; easy-to-use information appliances and computer systems; and trained people who can build, maintain, and operate these resources. (For a comprehensive overview of the NII, see Council on Competitiveness 1993, 1994.) The U.S. NII is robust and rapidly evolving. Telephone, cable, and wireless delivery systems are linking together and new applications are being introduced in many markets. The use of the NII in the delivery of health care has the potential to transform the current health care system into a prototype telehealth model. Although the promise has yet to be realized, the convergence of the NII and the health care system is creating new market opportunities to deliver better, more cost-effective health care to every American.

For the purposes of this report, telehealth is defined as the provision of remotely located health information or services. The convergence of information and services within the telehealth model is intergal to the four market sectors covered in this report. Although remote care, personal health information management, systems integration and research and education are not the only examples of where markets are being influenced by the NII, they are illustrative and constitute significant market potential. Multimedia databases and information resources for patients, practitioners, health care delivery organizations, medical researchers and payers are some of the areas discussed in the context of information based products. Services reviewed include consulting, monitoring and providing decision-support software where the individual, as patient or consumer, is not physically located in the same place as either the resource or the practitioner. Figure 1-6 is a conceptual picture of a telehealth system made possible as the NII and the health care markets converge. It illustrates the components necessary to ensure that health-related information and services are available anywhere, anytime.

Beginning at the center of the concentric ring diagram and working outward, the first ring represents the technologies involved in creating a telehealth system Basically, any telehealth system such as telemedicine--the provision of remote care--consists of two components: (1) the equipment necessary to create the interface between the patient or individual requesting information or service and the practitioner or information resource, and (2) the infrastructure that allows the interchange to take place. The equipment, including computer-based, is comprised of a variety of hardware and software for audio, video, and data devices ranging from telephones to full-motion, interactive video devices. The NII provides the information and telecommunications infrastructure. Information focuses on such items as patient record keeping, scheduling, creation of health information databases, and tracking administrative/business transactions. Telecommunications is the means of transferring that information by wireless (e.g., microwave, satellite, and cellular), cable, or wireline-telephone systems.

Ring 2 consists of the stakeholders in telehealth, who all interact in a variety of complex ways to determine the composition, availability, and cost of telehealth systems. Stakeholders are the major participants in the health care system and consist of: patients, including individuals who seek information for wellness; employers who are involved in determining health care benefits for their personnel; practitioners, such as doctors and nurses; health care delivery organizations such as hospitals, clinics, outpatient facilities, and managed care organizations; payers who are the private and public insurance plans; medical researchers; and vendors who provide both infrastructure and interface hardware and software including integrated, turnkey systems.

Ring 3 represents the legal, regulatory, financial, and policy components that affect telehealth system design, development, and deployment. The legal environment deals with such issues as malpractice, liability, licensing, privacy, and confidentiality. The regulatory environment is driven by such agencies as the Food and Drug Administration (FDA), HCFA, and the Federal Communications Commission (FCC), whose regulations govern the manufacture/sale/distribution of health care and communications products and services. The financial environment drives the availability of funds to pay for infrastructure and services. The policy environment is created by federal and state administrative and legislative bodies. But standards-setting bodies such as the American Medical Association (AMA), the American College of Radiologists (ACR), and the Joint Commission on the Accreditation of Health Organizations (JCAHO), and the American Telemedicine Association (ATA) for the medical profession; and the National Electrical Manufacturers Association (NEMA), the Electronic Industries Association (EIA), and the Health Industry Manufacturers Association (HIMA) for equipment manufacturers also create important policies and greatly influence the government's policymaking process.

Ring 4 depicts the telehealth markets. If the necessary interface and infrastructure equipment and services are available, if principal stakeholders perceive benefits, and if suitable legal, regulatory, and policy environments exist, so should a viable marketplace. Moving inward from Ring 4 back to the center, each ring represents the barriers that must be overcome to move from market potential to a viable telehealth market.


































cclogoSMALL.gif (1930 bytes)

competetextgrey.gif (3110 bytes)


Chapter 2 - Remote Care

Highway to Health: Transforming U.S. Health Care in the Information Age




The provision of health care to medically disenfranchised patients, those who for reasons of distance or circumstance do not have ready access to medical care, presents both a challenge to the medical community and an opportunity for the use of the NII. Telemedicine--the combined use of telecommunications and information technologies to link medical practitioners with their patients for consultation--is an emerging application of the NII that is addressing this challenge. Through telemedicine, remotely located patients, such as those residing in communities with limited medical resources, inmates in isolated correctional facilities, soldiers in the field, or patients in their homes, can access health care services that are currently unavailable to them. Telemedicine also offers the promise of better health care by providing more timely access to specialists and vital medical information, benefiting all health care delivery system stakeholders and society as a whole.

This chapter examines scenarios combining NII-related technologies and the delivery of remote health care in three settings:

  • Traditional rural/urban markets.
  • Markets that are immune to certain regulations and economic considerations, such as the government, correctional facility, and international markets.
  • The home.

It also explores the differences in applications of telemedicine within different health care delivery environments, including managed care and fee-for-service. For illustrative purposes, aspects of the development and diffusion of teleradiology are compared with other telemedicine sub-specialty applications. The market today and the market potential are analyzed, including the cultural, regulatory, legal, psychological, and economic implementation barriers that stand in the way of increased utilization. Finally, ways are suggested for overcoming existing barriers and realizing a robust commercial market.



For the purposes of this report, the concept of "remote" is defined as a patient and health care practitioner separated by distance, regardless of the amount of that distance. It includes, for example, interactions among health care facilities co-located in urban environments (e.g., inner city clinic and academic medical center); tertiary medical centers consulting with rural hospitals; academic medical centers providing medical expertise overseas; contract medical service providers consulting with practitioners located in correctional institutions; Department of Veterans' Affairs (VA) or Department of Defense (DOD) facilities; health care providers monitoring and consulting with patients located in their homes; and regional centers providing a particular type of expertise such as radiology to other providers in that region.

Three practitioner scenarios are common in telemedicine interactions: (1) a physician at one location seeking consultative help from a specialist at another location; (2) a non-physician practitioner consulting with a remotely located medical practitioner; and (3) a patient consulting with a medical practitioner such as a primary care provider, a specialist referred by a primary care practitioner, or a monitoring nurse. The technologies for linking the participants in these scenarios can range from the most simple and available of telecommunications connections based solely on audio links (such as telephone voice mail) to two-way, interactive, full-motion, high-definition video requiring wide bandwidth communication capability. The interactions can involve a range of transmission options, from store-forward technology, where the consulting physician calls up previously recorded information for review and forwards recommendations to be processed at a convenient time by the practitioner initiating the request, to a scheduled real-time, interactive, full-motion video accompanied by a variety of medical instrumentation linking all participants simultaneously. The necessary telecommunications infrastructure, depending on the type of telemedicine scenario, can include telephone lines (standard, T-1, DS-3, ISDN), cable systems, wireless systems, or any combination of these. Although the telephone is used routinely to conduct consultations, consultations via audio links alone will not be considered for purposes of this report.

The types of telemedicine consultations conducted in remote care scenarios are limited only by the range of medical problems to be addressed, the availability of the required practitioners, and the appropriate equipment, including telecommunications infrastructure. Consultations have already been documented in all sub-specialties. In addition, several projects under way are demonstrating telemedicine's potential for providing home care services to two distinct populations: (1) individuals who are receiving treatment or require monitoring on a time-limited basis for a particular episodic illness, and (2) patients suffering chronic illnesses who require frequent re-hospitalizations for emergency care and/or frequent nurse visits. Currently, teleradiology is the most common form of telemedicine. Lessons learned from the extensive use of teleradiology by both government and commercial health care delivery organizations offer valuable insights into the use and market potential of telemedicine in general, and are highlighted throughout this chapter.



Historically, what we are seeing today marks the second wave of telemedicine initiatives. From the late 1950s through the 1970s, several telemedicine applications were launched. But even though they were viewed by some as successful from both clinical and patient perspectives, the projects were terminated because of the elimination of government funding, the high cost of complex, technically immature systems, and/or their lack of widespread acceptance by and integration into the medical community. Other stifling factors were the lack of a robust communications infrastructure in some locales and high operational costs. More than 20 years later, the resurgence of interest in alternative delivery systems, such as telemedicine, and an explosion of applications has been sparked by a variety of tensions in the health care industry, including: competition among providers prompted by the need to become profitable, the need to reduce the cost of care, the desire to increase access to care for those who have none, and/or the need to improve the quality of care. Advances in technology that enable the transmission and remote display of images and information over digital communication pathways are also contributing to renewed interest.

The assessment of the market for telemedicine was based on literature searches, interviews, surveys of products in development or already on the market, anecdotal reports, and the experiences of Council Advisory Committee members in the health care marketplace. Because the use of telemedicine is in the early adoption/implementation phase, it lacks robust program evaluation (as do other medical systems more mature in terms of time in use). Therefore anecdotal and personal reports were particularly valuable sources of information. Several extensive reviews on the state of telemedicine today (Allen 1995, Grigsby et al. 1994, Scott and Neuberger 1994) were consulted for background information. Also, a number of the existing U.S. telemedicine sites listed in the Medical College of Georgia database Telis and an on-line telemedicine database maintained under government contract (TRC 1995) were contacted.

Although this site survey data seems reasonably complete, there is no commercially available market analysis that details the number of public or private dollars currently being spent to establish or maintain general telemedicine applications, either nationally or internationally. Estimates of telemedicine and related technology funding by government organizations suggest that several hundreds of millions of dollars in grants and contracts might have been available in fiscal 1995 through various federal agencies and individual states. At least one estimate, from the Koop Institute, puts today's U.S. market at $20 billion (Banks 1995). However, a breakdown of this funding for services, equipment, and infrastructure is unavailable.

The one telemedicine application that has been tracked extensively is teleradiology, with an estimated 7,000 X-ray film digitizer units in place. Annual market estimates exist for both medical imaging systems and picture archiving and communication systems (PACS), both of which depend on teleradiology (TMG 1995, Frost & Sullivan 1994). The 1995 U.S. commercial market for all medical imaging products has been sized at $473 million (Frost & Sullivan 1994). The market for teleradiology has matured more rapidly because of several characteristics of this speciality: practice patterns of physicians consulting with distant radiologists have been established; radiologists are familiar with digital technology and accept its use for a number of diagnostic procedures (CT and MRI) that are amenable to rapid transmission and archiving; radiologists are relatively high-cost specialists and hospitals need to maintain their productivity; favorable medical-effectiveness evaluation data exist for the use of digital techniques for some applications; and reimbursement is available.

Business strategies behind the establishment of teleradiology sites are propelled by a desire to increase patient volume, lower cost, and obtain a greater market share. For example:

  • Telequest, a for-profit consortium of academic radiology departments, intends to connect regional diagnostic imaging centers across the country with its members to provide cost-effective radiology reading services. The first imaging center in Columbus, Ohio, was connected to radiology departments at the University of Pennsylvania, the University of California at San Francisco, and Brigham and Woman's Hospital in Boston in the last quarter of 1995.
  • Teleradiology Associates in Durham, N.C., uses a central facility to provide radiology services to 20 rural sites.
  • A radiology practice in Los Angeles provides teleradiology services to four hospitals during the night shift at a lower cost than in-house, on-call staff.
  • Oklahoma is in the process of installing more than 40 film digitizers throughout rural areas of the state. The digitizers send images to radiology reading sites in the state, which then send diagnoses back.
  • To reduce the cost of radiology service for its many employees and their dependents, Harris Corporation of Melbourne, Fla., has designed its own teleradiology system linking it to the University of California at Los Angeles (UCLA). UCLA receives the images and performs all requisite radiological services under contract.
  • In rural Yakima Valley, Wash., radiology services are now provided among small hospitals through digital links.

These enterprises are demonstrating how telemedicine can contribute significantly to the effective delivery of quality services. In addition to the numerous teleradiology practices, hundreds of telemedicine projects involving many hundreds of individual sites are operating in the U.S. today. These projects are providing a variety of specialty consultations regionally, nationally, and internationally by using a number of telecommunications strategies for transmitting data. The sites encompass five broad categories: (1) domestic commercial initiatives including both fee-for-service and managed care, (2) government initiatives, (3) international commercial initiatives, (4) correctional facility initiatives, and (5) home care initiatives.


Domestic Commercial Initiatives

Health care delivery organizations in fee-for-service and managed care environments in the U.S. are employing telemedicine systems for a variety of reasons. The adopting sites range from system wide approaches, such as Allina Health Systems in Minneapolis, to application-specific sites, such as the ultrasound program at the Methodist Hospital of Indiana. Both are profiled in this chapter. Allina recently installed a telemedicine network linking 22 separate campuses. (See Case Study 2-1.) Statewide systems in Georgia, Louisiana, Oklahoma, West Texas, and Iowa were initiated to deliver health care to their populations that were far removed from a health care facility, or in some cases, had no access at all. The Mayo Clinic initially conceived of its telemedicine system as a means to address specialty staffing issues opened new clinics in other regions in the United States. Additionally, as more hospitals merge and consolidate to manage costs, duplication of services could be eliminated through the use of telemedicine links between the facilities.


Case Study 2-1: Allina Health Systems




Allina Health Systems, a large managed care provider in Minnesota and Wisconsin, formed a telemedicine network with the Rural Health Alliance (R.A.), a group of rural communities in central Minnesota. The network is designed to support consultations, teleradiology, medical education, administration, and community education and development; although specific use is expected to vary from site to site. The network became active in May 1995, with three rural and five urban sites. In September six more rural sites were added, and in January 1996, an additional six rural and two urban sites completed the initial network. Allina anticipates adding three to five additional sites per year.



The network was originally developed as a tool to reduce costs in preparation for a capitated environment and to expand market share into the rural communities not part of Allina's existing plan. Currently, the Blue Cross and Blue Shield insurance plans are uncontested in rural areas. Allina hopes the telemedicine network will generate referrals and potential insurance sales there.



Initial funding was provided through a combination of grants and investment from Allina. R.A. received a $642,000 grant to purchase the equipment located at each rural site. This funding provided the necessary justification for Allina's investment for the link to their tertiary care campuses and other ancillary costs. Allina pays for the T- I telephone lines-Creating the hub center and links to other Allina sites; the rural alliance members pay for the spokes to the hub and divide the costs equally among themselves.



Allina's insurance division agreed to reimburse specialists for consultations conducted over the network. It has established a fund to pay for any MediCare/Medicaid patients because HCFA has not approved reimbursement for telemedicine consultations. Allina believes that HCFA may eventually change this reimbursement policy. However, if a capitated market exists, HCFA reimbursement will not matter.



Because the Allina system is new, cost-benefit data is sl im. Allina reports, however, that the system is generating new referrals and demonstrable administrative savings. Participating physicians accept the technology. In addition, early trials report that patients are very accepting.



Allina views its principal remaining barriers as lack of reimbursement while in the fee-for-service arena and lack of wider physician acceptance.





Government Health Care Initiatives

The DOD, the VA, and the Office of Indian Health Services have all used teleradiology to provide services to their members, both nationally and internationally. The VA has been transferring computer-based patient radiology records for some time using a PACS modeled after the successful DOD installations at Madigan and Brooke Army Medical Centers. The Office of Indian Health Services, in conjunction with the National Aeronautics and Space Administration (NASA), has used telemedicine to deliver care to Native Americans in rural locations. DOD has invested heavily in research and development on the use of telemedicine to provide earlier intervention for combat casualties and thus reduce morbidity and mortality rates. DOD has also used telemedicine for care of geographically dispersed beneficiaries deployed far from a military clinic or hospital.


Military telemedicine systems benefit from having an extensive communications network already established, including the latest in technology. In addition, these systems have not been constrained by licensing and credentialing regulations, HCFA regulations for reimbursement, or profit motives. DOD has pioneered the development of digital PACS, including teleradiology, and the use of telemedicine to care for soldiers in distant war zones. In December 1994 the DOD Telemedicine Testbed was established by the Army, Air Force and Navy to synchronize their efforts to rapidly exploit emerging technologies. The Medical Advanced Technology Management Office (MATMO), located at Ft. Detrick, MD, oversees the combined forces initiatives which include aggressive use of the Internet for medical information, development of ruggedized health data storage for medical use in the field (Project Meditag) and use of multi-media E-mail to coordinate care among central and remote sites. Walter Reed Army Medical Center, Tripler Army Medical Center, the National Naval Medical Center at Bethesda, and the Naval Medical Center in San Diego have been engaged in telemedicine projects to Europe, Somalia, the Pacific Islands, Japan, and ships at sea, as well as within the U.S. The Air Force has also established the Medical Defense Performance Review (MDPR) to evaluate the delivery of medical care within the military. MDPR is rapidly developing technological initiatives to enable a wide range of applications, including video conferencing, administrative networks, and telemedicine.


International Commercial Initiatives

In addition to DOD, some health care delivery organizations within the private sector are initiating international telemedicine consultations to export their medical services. WellCare Holdings, N.V. provides consultations primarily in the Middle East. (See Case Study 2-2.) Stanford University Hospital has a link to Singapore for continuing medical education. The Johns Hopkins Oncology Center is making preparations to link with Gleneagles Hospital in Singapore to provide clinical consultations, medical education, and research opportunities for health care providers at Gleneagles and affiliated hospitals in Southeast Asia. The Mayo Clinic has a link to Jordan and is establishing a link with a hospital in Greece.


Case Study 2-2: WellCare Holdings, NV



WellCare Holdings, NV (WellCare) was founded in 1992 by Pillar, a Paris-based international business development group, and Medical Science Partners, a venture capital group in Wellesley, Mass., that specializes in the development of new health care ventures. In August 1995, WellCare purchased MD/Dx from Massachusetts General Hospital (MGH). MID/Dx had developed, in conjunction with MGH, a robust international teleradiology business linking MD/Dx-MG trained radiologists with hospitals primarily in the Middle East.



NM/Dx and MG saw an opportunity for the export of medical expertise, partly in response to the shrinking domestic market due to capitation, cutbacks in government funding, and pressures to enhance revenues. An international outreach activity could also sustain funds for reaching, education, and research, as well as provide care delivery. Governments and patients in this market were seen as the driving forces for using telemedicine, not physicians.



MD/Dx-WellCare was partially funded by the venture capital firm Medical Science Partners, and by funds from WellCare and Allstate Venture Capital.



Reimbursement is not an issue. WellCare operates as a fee-for-service provider and customers pay for the serv ices rendered.



The majority of international patients who would have had to travel for medical consultations can now stay in their local environments, resulting in huge savings for both the foreign government and the patient. If additional treatment is needed, patients are referred to Massachusetts General. WellCare is expanding its relationships to include leading medical centers in other parts of the world with special expertise in radiology, pathology, dermatology, and other medical sub-specialties.



Barriers to international telemedicine are market oriented and reflect the difficulties any company encounters in establishing new operations, particularly in a foreign country cultural differences, differences in business practices, and die challenges of creating a new market. WellCare does not face the interstate licensing and reimbursement issues that inhibit U.S. telemedicine projects.





International markets are attractive for a number of reasons. They can generate fee-for-service referrals; they may provide requisite commercial offsets for foreign military sales; they are not hampered by the lack of clear-cut insurance reimbursement policies; and they are not constrained by the interstate licensing requirements that are frustrating many domestic commercial telemedicine projects.

U.S. health care organizations are not alone in initiating international telemedicine projects. Foreign health care delivery organizations are also investigating telemedicine activities, often in partnership with U.S. telecommunications providers. Hughes Electronics Corporation and the Mexican government are sponsoring a pilot telemedicine program that connects the General Hospital "Dr. Belizario Domingues," a full-service, regional medical facility in Chiapas, Mexico, with the Centro Medico Nacional "20 de Noviembre" Hospital in Mexico City. In addition, Mexico's National Academy of Medicine, the Mexican Academy of Surgery, and the Medical Faculty of the Universidad Nacional Autonoma de Mexico have been invited to participate in medical education and research applications of the pilot telemedicine system.


Correctional Facility Initiatives

The U.S. correctional facility population consists of approximately 1.5 million inmates and continues to grow. They are housed in more than 300 locations, many of them in non-urban areas. In general, the average age of the inmate population is increasing, resulting in the need for more frequent health care services. Major medical concerns include accidents, orthopedics, problems associated with drug use (hepatitis), and infectious diseases (tuberculosis and AIDS). All of these conditions can require multiple consultations with specialists over extended periods of time. This population's right to health care is guaranteed under the Constitution, and recent litigation established strict guidelines for inmate access to primary and specialty care. Medical treatment outside the facility, however, can involve risks and is seen as an opportunity for temporary release. It is also expensive to provide because inmates must be accompanied by guards and transported in specially secured vans. Screening, and treatment of medical conditions through consultations with remotely located physicians can greatly reduce safety concerns, as well as associated transportation costs.

Telemedicine is currently being used successfully in several states to deliver health care more efficiently to incarcerated individuals. A partnership between the University of Texas Medical Branch in Galveston and Texas Tech University is providing telemedicine links to two-thirds of the Texas correctional system (the second largest in the U.S. with 127,654 inmates as of January 1996). At least one national contract provider of correctional facility health care services has also used telemedicine to deliver quality medical services at a lower cost, thus gaining a competitive edge in this cost-driven services marketplace. The East Carolina University telemedicine system had its origins in the delivery of health care to inmates and now has expanded its services to other, non-correctional sites.


Home Health Care Initiatives

The need for home health care is being driven by several factors, including: demographic trends; the shift in health care to more cost-effective approaches such as managed care and other risk-sharing systems; and the desire of patients, health care delivery organizations, practitioners, payers, and employers to dramatically curtail costs while still providing quality care. Not surprisingly, a survey of 59 home care chains showed that revenues increased 47 percent from 1992, to $5.1 billion in 1993, with the average chain growing from $83 million to $123 million a year. This is occurring in conjunction with major acquisitions as health care delivery organizations move into new lines of the home care business (Scott, L. 1994).

Integrated Health Services (IHS) of Owings Mills, MD, is one example of a health care delivery organization aggressively expanding into the home care market. IHS has built a $2 billion dollar business in post acute care, with over 280 facilities in 30 states. It is rapidly expanding into the home health care arena through Symphony, its recently launched home health care services division. Based on the purchase of seven home health care companies, Symphony now offers home health care and home medical equipment services to managed care organizations in 17 states. The emerging importance of home care and the move toward seamless delivery within managed care were primary motivators for IHS to enter this arena.

Home care practitioners can make, on average, only four visits a day due to the time spent in transit. Telemedicine offers an effective means of increasing the efficiency of their services while maintaining high-quality personal care for patients, often at a reduced cost. Several testbeds are exploring the feasibility of using telemedicine to provide care to patients in their homes. The Medical College of Georgia, in conjunction with the U.S. Army's Center for Total Access Program, the Georgia Institute of Technology, and Jones Intercable, is developing the "electronic house call." The project will link 25 homes of patients with chronic illnesses to practitioners via the local cable television infrastructure, using a personal telemedicine system with two-way interactive video, audio, and medical diagnostic instrumentation.

The Home-based Electronic Link to Professionals (HELP) Innovations Project in Kansas employed a regional cable television company to link four patients with chronic diseases to practitioners for daily monitoring. The system is now being fielded at the Hays (Kansas) Medical Center's Home Health Agency for additional market testing, with near-term plans for commercialization. Another project, Home Health-Telecare, has been funded by the California Research and Education Network (CalREN) to establish interactive data and video links between individuals with chronic obstructive pulmonary disease and a remotely located nurse or physician. Sutter Health is leading this project in conjunction with Fujitsu. Figure 2-1 provides additional information on other emerging home telemedicine products.


Figure 2-1: Emerging Telemedicine Products for Home Health Care


Several companies are introducing commercial telemedicine systems to address the home market.

American Telecare, Inc., in Minneapolis was the first to develop and test a home telemedicine system that provides around-the-clock nursing, allows for more complex care to be delivered outside of hospitals and nursing/extended care facilities, prevents and shortens stays at such facilities, and significantly reduces health care costs. The system relies on standard telephone lines for the necessary infrastructure to connect homes with the monitoring center, and equipment is leased on a daily fee basis (Mahmud and Lenz 1995).

Another commercial telemedicine, the Home Assisted Nursing Care (HANC) Network, is scheduled to enter the market in July 1996 pending FDA approval. Developed by HealthTech Services Corporation in Northbrook, IL, HANC is designed to facility 24-hour patient monitoring by health care professionals located at a central monitoring station within a home health care agency, hospital, or physician's office. The HANC network uses a combination of data, voice, and video transmissions that allow the practitioner to view the patient, monitor the patient's vital signs (e.g., blood pressure, temperature and pulse), administer up to 10 medications over a two-week period, and present self-care training screens.

HealthNet of Lubbock, TX has developed a briefcase-sized portable system, TeleDoc Junior, for ambulances, remote clinics, or in-home health care requirements. Several other companies are also testing prototype products.

One vendor estimates the immediate market for these types of personal telemedicine systems to be several hundred thousand units. Still, actual sales data will not be available for some time. Because demand has not yet consolidated, these systems are being marketed primarily to managed care organizations to gain critical mass. Therefore, they are generally not available to the 40 percent to 60 percent of the population not covered by such organizations.





Because telemedicine is in the early implementation stage, there are no firm data by which to estimate market potential. Early predictions by some organizations suggest the market opportunity may be as large as $100 billion (Banks 1995), but there is little to substantiate this claim. Nevertheless, two market drivers have emerged that will force serious consideration of commercial telemedicine to provide health care: (1) the need for providers to lower their costs while maintaining quality service, and (2) the need for health care delivery organizations to expand their market share in a competitive environment to increase profitability (this includes increasing access for patients in under served areas). The need for an updated telecommunications infrastructure in some locations will drive the pace of implementation. Financing issues will also influence the application of telemedicine. These activities affect each of the five market segments highlighted in the previous section.


The Need for Providers to Lower Costs

Even without an official national health care reform movement, the nation's health care industry is undergoing tremendous restructuring. Concern for accurate and timely use of health care resources, the movement toward capitation and other risk-sharing mechanisms to contain costs in the commercial health care market, and the trend toward discharging patients from hospitals earlier, even when they may still need monitoring, are causing many to examine the potential benefits of telemedicine.


According to a recent study by the Sachs Group of Evanston, Ill., the total number of hospital inpatient days was expected to decrease by 34 percent from 1994 to 1995. Total discharges were expected to decline by 26 percent, with the average length of stay also declining 11 percent, from 6.1 to 5.5 days. (Sachs Group 1995). In this environment, the market opportunity for the use of telemedicine to monitor complex medical conditions while patients are at home is easy to appreciate.

Several home health care agencies have conducted independent evaluations of the effectiveness of HealthTech Services Corporation's HANC telemedicine system, profiled in Figure 2-1. An evaluation of hospital readmissions found that 29 of 43 patients (67 percent) were suitable HANC candidates and that use of HANC could have saved approximately $5,000 per patient/admission. The average length of stay for these 43 patients was 11 days, at an average cost of $800 per day. For this group, total potential savings could have reached $255,000. Thirty percent of intravenous therapy/AIDS patients reviewed (9 of 30) could have been assisted by HANC with a resulting savings of $473 per episode. An additional retrospective chart review study reported similar potential savings for intravenous therapy, wound care, diabetes, stroke, and orthopedic cases. Clinical trials have begun at six sites to further quantify the potential of this telemedicine tool as a cost-saving device for a variety of home care patient populations.

Some view telemedicine in risk-sharing environments as a way to deliver quality health care at the lowest available cost point (Siwicki 1995, Goodall and Murphy 1995). The movement toward capitation, where health costs are paid by a predetermined fee and a single insurer/provider cares for a given person across a particular period of time or for management of a particular disease, may also accelerate the use of telemedicine for clinical as well as preventive applications. Health care delivery organizations in a fee-for-service environment already are competing with those in managed care settings. Now managed care organizations are also competing with one another, forcing them to consider new delivery mechanisms, such as telemedicine, that can lower their fixed costs by using practitioners more efficiently. The health care industry will continue to evolve to include different competitive approaches, such as the entry of medical practice groups that seek to reduce the role of third-party payers and that may also further the development of commercial telemedicine systems for remote care applications.

Preliminary data from the Hughes-sponsored testbed in Mexico, described earlier in this chapter in the section on international commercial initiatives, also illustrate telemedicine's potential to reduce costs. Figures 2-2 and 2-3 indicate some of the cost-savings that are being realized. Based on this initial feedback, Hughes believes that it will be possible to amortize the telemedicine system installation costs for the linked hospitals in the first quarter of use.

Figure 2-2: Cost Savings Associated with Patient Transfers


Month (1995)

Reduction in Number of Patients Transferred

Savings Associated with Reduced Transportation Costs






















Figure 2-3: Continuing Medical Education



Subjects Covered

Hours of Instruction

Estimated Savings


Course 1

Chemotherapy, Overall Treatment Applicability, Patient Care, Drugs

46 hrs.



Course 2

Pediatric Nephrology, Overall Treatment Applicability

20 hrs.







The increasing health care costs associated with the aging of the population are also creating market opportunities for telemedicine. In 1993, the nation spent $80 billion on long-term care for approximately 7.3 million people. Nursing home care accounted for approximately three-quarters of this spending, with home care making up the bulk of the remainder. Spending in this area is expected to continue its rapid pace as an estimated 10 million to 14 million people will require some form of long-term care by 2020, climbing to 14 to 24 million by 2060. With nursing home care averaging $35,000 per person per year, home care is becoming an attractive alternative, particularly as telemedicine permits the delivery of more sophisticated and less expensive care to the home (CRS 1995).

It should also be noted that a significant number of people who reside in nursing homes are there more for health "security" reasons than for health care "needs." For example, market data suggest that of the two million residents in extended care facilities, perhaps as many as 10 percent could be cared for at home at significantly reduced costs if the appropriate telemedicine tools were available to enable remote monitoring. Additionally, many of the home health visits conducted today are based on the need to observe or monitor a patient's status, a function that could be accomplished through interactive video systems coupled with the appropriate instrumentation and a simple-to-use interface. Figure 2-4 illustrates the potential cost-savings that could be realized by delivering care to the home through telemedicine.


Figure 2-4: Average Cost of Care





Hospital Inpatient*



Nursing Home**



Home Care Visit***



Telemedicine to the Home****





*Average expense of community hospital per inpatient day in 1992 (HIAA 1994)

**Average expense of nursing home facility per day in 1993 (CRS 1995a)

***Skilled nursing visit (CRS 1995a)

****Estimate for NANC (HealthTech Services Corporation)


Securing or Expanding Market Share

In addition to containing costs, health care delivery organizations must also secure or expand market share to maintain the revenue stream necessary to provide health care services and to finance advanced medical research and education. Given the emphasis on health care delivery as a revenue generator, and the fact that telemedicine often lessens the need to refer patients away from their local medical institutions, rural facilities are embracing this new technology as a means of maintaining their market share. By using telemedicine consultations to provide access to specialist care they can retain patients who otherwise would be referred to larger, regional medical centers. The ability to maintain and profitably operate these small hospitals in today's expensive health care environment is a genuine concern; on average, 10 to 12 close each month. Yet not only do these facilities provide much-needed care at lower costs for their patients; they are also often economic anchors in their small communities, providing jobs and attracting and maintaining other businesses that want to ensure access to health care services for their employees. In fact, some telemedicine providers, like HealthNet, see this mission as their most vital function. A rural hospital in Louisiana has experienced positive results through telemedicine, increasing its bed census and adding an additional physician. (See Case Study 2-3.)


Case Study 2-3: Louisiana State Health Care Authority




The Louisiana Health Care Authority (LHCA) operates 10 hospitals within the state. It began the "TELEMEDicine Project" in the fall of 1994 and modeled it after the Georgia system.



The Louisiana State University (LSU) Medical School, Department of Health and Hospitals, and other health care providers were interested in creating a statewide telemedicine network to enhance the quality of health care available to the rural and under served residents of the state. Louisiana has the highest infant mortality rate in the country (18 percent), as well as higher-than-normal accident victim mortality. Since 1979, 39 small hospitals have closed in the state.



The emphasis on reaching rural populations resulted in funding from the state's Department of Rural Development. The project is still running on state funding because an anticipated federal grants rescinded. Louisiana State University (LSU) provides system management, expertise, and the participation of every clinical departmental the university medical center. Recommendations include a plan to move away from government funding, as well as exploring the lease rather than the purchase of equipment. If one additional rural hospital bed was filled each day for the year, the system would be self-sustaining.



Reimbursement is currently handled through a special reimbursement fund established by LHCA to pay urban physicians the current Medicaid rate. The funds available will allow operation to continue for another year, after which time LHCA will have to seriously address the reimbursement rate.



A recent intense evaluation of telemedicine consultations involving a rural hospital, a regional LHCA facility, and the tertiary facility associated with LSU Medical School has been completed. Patient savings are potentially significant the rural hospital charges, on average, $600 per day, compared to$1,200 per day in the urban hospital. Three additional rural facilities are currently being added to the TELEMEDicine system.

The effectiveness of the system is being judged by physician usage rates rather than fixed equipment costs. Costs per clinical encounter appear high. However, the "aura of excellence"perceived by patients and the increased reputation of those connected with the system are considered significant payback. Education/counseling appear to be significant benefits bit are difficult to quantify. The system allows physicians to participate in weekly/monthly conferences presented by various LSU clinical departments.



Although patients were enthusiastic about the system, physicians were initially skeptical. Concerns about being able to interact with the patient face-to-face were largely alleviated once a telemedical consult was attempted. Rural/regional practitioners were initially concerned that their practices would be negatively affected by telemedicine. Guidelines and principles were implemented that alleviated their fears. These policies protect the practice of the rural and regional physicians while making available referrals to LSU specialists.





Large regional/academic medical centers also benefit from telemedicine links to rural facilities. These health care delivery organizations can increase revenues through additional referrals by extending services to under served populations via telemedicine, thus expanding their patient volume and achieving a more efficient use of resources, such as physician time. Practitioners participating in the telemedicine consults also receive the benefit of enhanced revenue.

An example of the use of telemedicine to potentially secure and expand market share comes from the state of Georgia. Preliminary findings from the first rural hospital linked to the Medical College of Georgia indicate that 85 percent of the patients examined by telemedicine, who otherwise would have been transferred to a secondary or tertiary care facility, were retained locally following the introduction of telemedicine. The projected increased bed census translates into more revenue for the rural hospital ($1,100 per day per bed), reduced hospital costs for patients ($1,100 per day versus $1,850 per day at a larger medical center), and referral of appropriate patients to the Medical College. A study is ongoing to assess whether this has increased revenue for the Medical College.

Less easily quantified, but just as important, are the additional benefits that patients experience, such as elimination of travel time and associated travel costs, more rapid care, closeness to family, and reduction in lost wages. Additionally, businesses experience less productivity loss when patients and their families do not need to take time from work to travel. The primary care practitioner also benefits. The consultative process can become a valuable educational opportunity for the local practitioner and, over time, should translate into reduced need to obtain referral expertise. The increased access to colleagues through the consultative process also reduces the professional isolation many of these practitioners experience.

Preliminary data from the Hughes-supported pilot in Mexico also illustrate how medical facilities can secure and maintain their local market share. Figure 2-5 shows the reduction in the number of patients transferred from the regional medical facility in Chiapas to Mexico City after the telemedicine linkup was installed. In addition to helping Chiapas treat its patients locally, the reduction in transfers to Mexico City also helped reduce overcrowding at the "20 de Noviembre" Hospital.

Several health care delivery organizations are seeking to expand market share through international telemedicine linkages. As mentioned previously, the Mayo Clinic, Massachusetts General Hospital, Stanford, Johns Hopkins, and for-profit corporations such as WellCare all are developing relationships with overseas medical facilities. These efforts should be encouraged since they promise to be "win-win" opportunities for both the U.S. health care providers and the international recipients of their services. The telemedicine linkups permit the U.S. health care delivery organizations to provide medical services which these international populations lack. In turn, the revenue these services generate can be an important source of additional funding to help support valuable medical research and education at U.S. academic medical centers, as well as contribute to bottom line profitability. Thus, exporting U.S. medical services will help contribute to U.S. economic growth and competitiveness.



For telemedicine to achieve its full market potential, it must be as easy to use and pay for as a telephone or automated teller machine (ATM). In most statewide, regional, and international systems, telecommunications providers are strategic partners and are involved throughout all stages of development and operation. They provide the infrastructure that allows the consultations to occur, and they also account for and underwrite a significant portion of the day-to-day operational costs for telephone/cable/wireless transmission. Some, such as NYNEX, GTE, Sprint, AT&T, Southwestern Bell, Ameritech, and US West, even offer turnkey telemedicine systems through agreements with vendors. Project Total Access, a DOD initiative, seeks to lead the way in creating a medical infrastructure that will allow the military and their dependents instant access to their records and personal physicians whenever and wherever they might require them (Blakeslee 1995). Hughes entered the telemedicine market as a means to sell satellite communications offered by its Spaceway program, and is now using its technical expertise to capture international telemedicine opportunities.

Unfortunately, current telecommunications costs and the need for an updated telecommunications infrastructure in some rural areas (some communities are still dependent on rotary-style telephones) are inhibiting the delivery of telemedicine services. The recently passed federal telecommunications legislation should result in decreased infrastructure costs as competition among cable companies, telephone companies, and wireless communication providers increases. With increased competition, there should be a larger array of services to select from at competitive prices. Strategic partnerships between the health care and infrastructure providers should speed the development of advanced telemedicine systems.

Unfortunately, while those who currently do not have access or enjoy only limited access to quality care may stand to benefit the most from telemedicine, they also may be the least able to pay for these services. Without some form of payment-support mechanism, infrastructure or health care providers may not consider telemedicine alone to be capable of delivering a sufficient return to justify their investment. However, if multiple applications are available to use the infrastructure, such as those related to education or entertainment, the infrastructure costs can be shared among them and the overall investment return could be increased.


Reimbursement and Funding

The most significant factors in determining market potential may be whether services provided via telemedicine are reimbursable and whether funds are available to initiate telemedicine projects. Reimbursement varies depending on the insurer, and it is often unclear which types of telemedicine interactions, as well as which health care providers participating in these events, are covered. Although HCFA routinely reimburses teleradiology and telepathology consults, it does not reimburse other telemedicine consultations for Medicare patients. At least one estimate projects that Medicare reimbursement for telemedicine consultations could increase that budget by $30 billion to $40 billion over the first three to five years of use (Grigsby 1995). Private insurers are, for the most part, following HCFA's lead. However, some self-insured health care systems are providing coverage, and in cases where managed care organizations are self-insured, reimbursement is being paid to physicians and facilities. Also, some physicians participating in telemedicine projects are providing consultations without reimbursement because it decreases their need to travel to conduct consultations, they gain experience, and they can gather data on telemedicine's effectiveness and potential for increasing referrals.

Herein lies the public policy dilemma. On the one hand, telemedicine can significantly enhance the availability of medical care to people who currently do not have adequate access and can decrease the individual cost of care. On the other, it could also increase the number of patients filing claims for reimbursement and thus increase private insurance and Medicare payouts. This would be true for any delivery systems that increases access to care. And although patients would receive the benefits of earlier and better care, and save on costs associated with travel to distant medical facilities, the government may have to adjust its budget or its rates for Medicare to finance this social objective.

Given these scenarios, telemedicine may be available more quickly in the less traditional market segments such as the military, correctional, veterans, Indian Health, and international arenas, where cost containment and access are the drivers and telemedicine technology is seen as an investment that leads to lower cost in the long-term. Telemedicine applications may also develop more quickly in HMOs operating on a capitated basis within a single state. These organizations have strong incentives to treat patients effectively at their lower-cost clinics rather than at their tertiary care hospitals. But the lack of adequate reimbursement mechanisms may also cause telemedicine to be viewed as "gold-plated" health care and restrict its application to those populations that can pay for the service "out-of-pocket."


In addition to reimbursement issues, the ability of organizations to obtain sufficient funding for equipment and operations will also influence the growth of the commercial telemedicine market. Although the technology is migrating to the desktop and equipment costs are dropping, start-up costs for some telemedicine systems are not insignificant (ranging from $50,000 to $100,000 for individual site equipment to $8 million for a statewide system). Thus some view the ability to obtain local, state, or federal grant money as a necessary first step.

A survey of non-teleradiology sites conducted during the summer of 1995 revealed that the majority were subsidized in their initial phases by some form of grant or contract. Only a few of the sites had used internal funding. However, these sites had based their operations on becoming financially self-sustaining. (See Case Study 2-4.) In contrast, few of the sites dependent on external funding had detailed plans for supporting operational costs after their funding ends. These trends seem to be borne out by the site information contained in the reviews and databases mentioned previously. This is causing concern within the medical community because the previous wave of interest in telemedicine in the 1970s was sustained only so long as government funding was available. With reliable evaluation data two to three years away, some of these early adopter sites may not have sufficient volume and payment strategies in place to become self-sufficient within the time frame allotted in their grants. Thus, the best opportunity for assessing market potential may come from those sites which initiated projects to address a well-defined need, that considered business realities prior to implementation, and that are not reliant on external funding sources.


Case Study 2-4: Methodist Hospital of Indiana




Methodist Hospital, Indianapolis, IN, initiated a telemedicine program in October 1994 focusing on an ID40 clinic-to-practitioner application, ultrasound. Five radiology sites are now linked to the main hospital campus with several more sites coming on-line for differing applications.



This telemedicine Program grew out of earlier projects in digital acquisition and transmission of CT and URI scans. Prior to using telemedicine, the ultrasound procedure was completed and evaluated by a technician at a remote facility. The tape was then sent to Methodist Hospital by a courier. Turnaround for readings averaged three days except for abnormal cases, which were sent to emergency.

While the principle drivers are clinical, business considerations are given high priority also. From the beginning, Methodist emphasized proven technology and a programmatic approach to lowering costs. The program uses fewer and less expensive technicians and sends the images to a central location staffed with appropriate sub-specialists.



Telemedicine at Methodist is internally funded and must pay for itself. Each potential application is evaluated on its own merit and payback must be sufficient so that it is financially sustainable from inception. No overall plan has been developed for migrating telemedicine to other areas but Methodist is adding applications on a case-by-case basis, and expanding on a standard scalable infrastructure for the projects that are self-sustaining.



Reimbursement is an issue and may limit the program as Methodist moves into more consultative uses.



Radiologists observe the ultrasound in real time, making immediate diagnoses. Thus far, Methodist has achieved savings of $4,000 to $5,000 per month by avoiding emergency room visits, return visits, and additional ultrasounds. Less tangible, "soft" savings are one-stop care for patients and faster answers to their concerns. Staff find that early or immediate access to specialists reduces diagnostic cycle time, delivers more timely and effective care, and subsequently reduces the cost of healthcare. With increasing competition Methodist views telemedicine not only as a valuable tool to deliver high-quality care, but also as a means to bolster bottom-line competitiveness.



Barriers include reimbursement and acceptance issues. Clinical and administrative staff uncertainty has inhibited the universal acceptance of telemedicine and has frustrated its inclusion in standard care regimes. Affiliated institutions are also uncertain about the ramifications of telemedicine and have therefore been slow to embrace it.





Discussions with various telemedicine project directors, including those with overseas operations, uncovered little activity by the venture capital community. This was confirmed through discussions with a half dozen leading venture capital firms. Some see investment opportunities in leading-edge or unique telemedicine products (medically related hardware and software). But only Medical Science Partners in Wellesley, Mass. has made a significant investment in the business of delivering health care services through telemedicine links, and it has focused initially on the international market.

Venture capital firms have been disinclined to consider domestic investment opportunities for several reasons: the uncertainties surrounding reimbursement; the lack of evaluation data supporting the profitability of telemedicine; issues surrounding ownership, licensing, and patents; and the current focus on medical service delivery rather than product development. Private funding to support telemedicine projects seems to be coming from partnership agreements between vendors, telecommunications companies, and health care delivery organizations. Funding is less problematic for teleradiology because the market is more mature and reimbursement is available. Teleradiology systems become "money makers" for investors, and hospital boards are beginning to treat them like other capital purchases.


Summarizing Market Potential

Unfortunately, firm market projections do not exist for remote care. However, the potential opportunity can be illustrated by the populations that can benefit from telemedicine. (See Figure 2-6.) Because telemedicine has the capability to benefit every person, the patient populations in the different market segments discussed in this chapter are reflected. Clearly, a telemedicine market exists. There are early adopters, vendors, a perceived need, and some degree of supporting infrastructure, all of which suggest a growing market, despite the absence of any widely accepted commercial market forecast. Anecdotal evidence suggests that company-proprietary market analyses are driving the continued investment. However, without prompt resolution of key barriers to broad domestic market entry and growth, the U.S. market potential for telemedicine may never become fully realized. Health care and infrastructure providers will be hesitant to commit significant resources to this area, and may be reluctant or unable to divert internationally invested resources back to the U.S.



Many of the issues that led to the termination of the telemedicine projects of the 1970s remain unresolved today. Outlined below are cultural, legal, regulatory, and implementation barriers that stand in the way of using the NII to create a robust commercial market for the delivery of remote care. None of the barriers discussed is common to all telemedicine applications or market segments. Some may not be hindrances at all, depending on the particular context in which implementation is attempted.


Cultural/Human Barriers

These barriers revolve around practitioner, health care delivery organization, and patient reactions to the use of telemedicine for consultations.

  • Lack of widespread physician/practitioner familiarity and acceptance of telemedicine system technology, including computer and video-based systems. Some physicians believe that because remote links remove the opportunity for touch, consultation via telemedicine is rendered less effective. Others believe it diminishes the traditional physician/patient relationship. Some think they would be more comfortable if they received more training in its use. And others who are convinced of telemedicine's benefits are frustrated by cumbersome technology. Practitioners usually must leave their offices and conduct their telemedicine consultations from another location, which may or may not be in the same building. Until telemedicine can be conducted "from the desktop," some practitioners may find that it is too inconvenient to use regularly.
  • Lack of physician acceptance. Physicians may have less power in determining how they practice under evolving models of health care and may believe that telemedicine will have a negative impact on their practices. Some physicians exposed to telemedicine systems believe that their practices will fall victim to the "swoop and scoop" strategy whereby their patients will be linked to other practitioners who will lure them away. Others do not feel comfortable consulting with physicians whom they do not know. Some physicians also fear that less skilled, lower-cost practitioners will be used to facilitate telemedicine consultations. Some believe telemedicine will reduce their referral sources. As patients and primary care providers gain greater access to any specialist without regard to distance, traditional referral patterns based on proximity may be altered. In effect, telemedicine creates greater competition for the same patient population.
  • Uncertain patient reaction to/acceptance of telemedicine, including perceptions of quality and efficacy. Patient reaction may be significantly determined by the health care market segment that serves them or the characteristics of the practitioners involved in the consultation. Some of the issues involved in a telemedicine consultation that have not been studied include communication styles, appearance, and room characteristics (Cukor and Baer 1994). Rather than having the opportunity to choose it, patients may be told by their health care provider to use telemedicine or have it as their sole consultation option. One study (Allen and Hays 1995) reports satisfaction with telemedicine when weather precluded actual visits with an oncology specialist. In some instances patients report feeling special when selected for telemedicine. However, in other instances involving pediatric consultations, patients have opted to wait for a later appointment that required travel to the physician's location rather than have immediate access to a specialist via a telemedicine linkup.


Legal and Regulatory Barriers

Some of the issues discussed below are seen as specific to particular market segments or to intrastate systems. Others represent major stumbling blocks to interstate systems and must be resolved if telemedicine is to enjoy widespread implementation.

1. The need for interstate licenses. Interstate licensing requires that a physician maintain a license in every state where regular or frequent consultations occur, with several states offering various exceptions to this rule. The time, cost, and administrative burden of obtaining a different license in every state could have a dampening effect on the spread of telemedicine. The Federation of State Medical Boards has recently proposed a Model Act to regulate the practice of telemedicine across state lines. The Model Act defines "the practice of medicine across state lines" and would allow limited licensure for this purpose if a physician holds a full and unrestricted license to practice medicine in any state. If a state chooses to adopt this act, physicians with a valid "telemedicine" license in another state would be granted reciprocity.

Such a telemedicine license would be simpler and less costly to obtain than a full and unrestricted medical license. However, for a limited license approach to be implemented, each state medical board would have to independently legislate it. Each state also would retain its own jurisdiction on disciplinary and patient confidentiality matters. So while some see this proposal as a means to cut through current licensing red tape, others fear that it would do little to encourage interstate telemedicine consults because each state would still retain its own practice rules and regulations. Only time will tell which states, if any, will adopt the federation's Model Act and in what form. Figure 2-7 illustrates the complexity of current interstate licensing laws. Case Study 2-5 amplifies the current legal situation.


Case Study 2-5: Interstate Licensing


Physicians and other health care providers interested in interstate telemedicine need to examine state physician licensing laws carefully before establishing multi-state practices. In most instances, state licensing laws were drafted decades ago and never contemplated the use of advanced communications systems to provide long-distance medical care.

Each of the 50 states, the District of Columbia, and the United States territories and their respective boards of medical licensure have rules that govern the ability of health care practitioners, including physicians, to practice medicine. These laws and regulations have developed in piecemeal fashion, creating a patchwork of laws that can vary significantly.

When a practitioner uses telemedicine across state lines (or even internationally) to consult on medical decisions, numerous medical licensure questions are raised. First and foremost, anyone who is practicing using interstate telemedicine should examine the state medical licensure laws in each of those states. It is possible that the practitioner is unintentionally, but illegally, practicing medicine without a license. State legislatures and the state Boards of Medicine are only now beginning to consider legislative or regulatory changes specifically aimed at telemedicine.

Figure 2-7 depicts the various questions and concerns regarding telemedicine across state lines. The questions presented in the chart must be answered in each state in which the health care provider intends to provide health care services by telemedicine.

In some states, health care providers who practice interstate telemedicine may find relief in state consultation provisions. Approximately a dozen states have consultation provisions in their medical licensure laws. These consultation provisions were not enacted with telemedicine consults specifically in mind, but they may apply. Arizona Revised Statutes 32-1421 is a good example of a consultation provision that creates an exemption from licensing requirements. A.R.S. 32-1421 (B) expressly provides that the state licensure requirements do not apply " any doctor of medicine residing in another state, federal jurisdiction or country who is authorized to practice medicine in that jurisdiction, if he engages in actual single or infrequent consultation with a doctor of medicine licensed in this state and if the consultation regards a specific patient or patients."

An interesting variation on a consultation provision is found in Idaho. Idaho Code 54-1804 provides that a doctor licensed in another state or jurisdiction is allowed " consult if called in consultation by doctor licensed in Idaho or for medical education purposes so long as he (or she) does not open an office or appoint a place to meet patients or receive calls in (Idaho)."

Kansas was one of the first states to take action aimed specifically at telemedicine. The Kansas State Board of Healing Arts. at the behest of the Kansas Medical Society, issued a regulation in 1994 requiring a physician who treats, prescribes, practices, or diagnoses a condition, illness, ailment, etc. of an individual who is located in Kansas to obtain a Kansas medical license. (K.A.R. 100-26-1 (1994)). The regulation was widely erroneously reported as a now Kansas law passed by the state legislature. In fact, the new licensure requirement is a result of an interpretation by the state board of healing arts of the existing state licensure statute.

Although the regulation does not explicitly mention "telemedicine," it is widely referred to as the "telemedicine regulation" in Kansas, in part because it effectively prevents a physician legally practicing medicine in a state other than Kansas from using telemedicine to treat or diagnoses patient located in Kansas if the physician is not licensed in Kansas. Thus, any physician who establishes a regular telemedicine link with that state must obtain a Kansas license. However, neither the state attorney general or a court has interpreted the telemedicine regulation or considered its validity.



Prepared for the Council on Competitiveness by the law firm of Arent Fox Plotkin & Kahn.

Licensing is not a barrier in the military, VA, public health, and Indian Health markets, which require only a single license. As a result, telemedicine may grow more rapidly in these markets than in the traditional fee-for-service or managed care markets. Although the correctional market has been developing intrastate, licensing requirements could deter organizations from instituting cost-effective interstate programs.

2. Lack of HCFA reimbursement for providers in a fee-for-service environment. Until the completion of its own evaluation study, HCFA is not reimbursing for telemedicine consults (other than for teleradiology and telepathology), nor is it reimbursing for the personal telemedicine systems that are beginning to be used in the home. Because sufficient data does not exist about the outcomes and cost-benefits of telemedicine, HCFA has considered funding a three-year telemedicine evaluation trial in four states (North Carolina, Iowa, Georgia, and West Virginia) to gather this information. Until the results become available and HCFA makes a ruling, health care providers in a fee-for-service environment cannot receive reimbursement for telemedicine consultations with Medicare patients.

With the exception of radiology and pathology, HCFA denies reimbursement because its practice guidelines require a face-to-face consultation. HCFA's Evaluation and Management Services Guidelines define a reimbursable consultation as, "...that time the physician spends face-to-face with the patient." And because HCFA policy influences private insurance policy, most private insurers will not reimburse. Therefore providers in a fee-for-service environment are more likely to view telemedicine as an additional cost, particularly for the technical equipment required, rather than as a means for increasing efficiency and service while lowering cost, and they may be slower to encourage adoption.

Reimbursement is not an issue in the DOD, VA, and Indian Health Services markets where physicians receive their salaries from government organizations, nor is it stalling implementation in the correctional market; state governments view telemedicine as a tool to deliver appropriate care to inmate populations within a fixed operating budget. It also may not be a barrier in other managed care organizations. And it has not inhibited the growth of international telemedicine consultations. International patients are not constrained by U.S. insurance policies. In addition, many foreign governments that provide national health care coverage for their populations recognize that it is often cheaper to support telemedicine applications than to pay for their patients travel to distant countries to receive necessary care. They view telemedicine much as some managed care organizations do--as an investment that leads to overall lower costs of care.

3. Potentially cumbersome credentialing requirements. Physician credentialing, although not a major barrier for telemedicine today, could become a more prominent issue if consults include non-physician practitioners. The Joint Commission on Accreditation of Healthcare Organizations (JCAHO) recently reaffirmed its two-year-old decision that a physician practicing to a remote site does not have to be credentialed at that site so long as the primary care physician who has referred the patient is credentialed at the site and writes the orders. In this scenario, the consultation is viewed as a recommendation only, and all medical orders are written by the on-site physician. However, if a physician is consulting with a non-physician practitioner who is not supervised, the consulting physician would be considered responsible and thus would have to be credentialed at the distant site.

4. Uncertainty regarding legal liability for the use of telemedicine products. An issue of growing concern is whether traditional product liability insurance is adequate to cover manufacturers of products and technologies used in the practice of telemedicine. Telecommunications companies and equipment manufacturers may be particularly vulnerable if inappropriate care or misdiagnosis could be linked, for example, to such things as software problems, data lost through compression, and degradation or failure of communications links. To date, most manufacturers consider their liability exposure for the use of their products in telemedicine scenarios similar to the risk they must assume in other health care environments. For example, one teleradiology equipment manufacturer includes a limited liability clause in its contract, treats its teleradiology systems no differently than other medical devices, and complies with FDA reporting requirements. As with any medical device, the potential for litigation is present. Thus telecommunications providers for telemedicine systems are including an indemnity clause in their contracts. This may be reasonably effective for current prototype tests, but its adequacy has yet to be tested in court. Until this issue is resolved satisfactorily, it could inhibit the widespread use of telemedicine.

5. Uncertainty regarding malpractice exposure and the risk of malpractice litigation. The issue of malpractice is perhaps the greatest unknown legal barrier for some telemedicine market segments. A key question is whether a distant physician who performs a telemedicine consultation will be held subject to the jurisdiction of the courts in the patient's judicial district. If so, physicians may be faced with the cost of additional malpractice insurance to cover their added liability risks in the distant state.

Regular telemedicine contacts with a distant state create an additional problem if a physician or institution is regarded as "doing business" in that state. Persons who regularly conduct telemedicine consults (i.e., "do business") in a state may be subject to suit in that state whether or not that state has any connection to the lawsuit. The potential exists for litigious patients to engage in "venue shopping," whereby they search for the most favorable malpractice-award state when suing a physician or institution practicing via telemedicine at multiple sites, regardless of where the medical consult occurred. Some physicians or institutions may therefore choose not to engage in telemedicine consults in certain states if they risk exposing their entire practice to a suit in that state. This is not as great an issue in international and some government markets.

While some providers have elected not to engage in telemedicine consultation with various states due to the uncertainties regarding malpractice litigation, insurance companies do not appear to be changing their policies because of telemedicine. Conversations with some of the physicians and organizations conducting teleradiology indicate that their malpractice insurance premiums have not changed. For example, the largest medical malpractice insurance company in the country, St. Paul Fire and Marine Insurance Company, does not distinguish between medical services provided in person, over the telephone or through video linkups and has not changed its medical malpractice coverage policies with the advent of telemedicine applications. Coverage for radiologists who receive images electronically is based on the rate for the state in which the image is read, not the state(s) from which the image is transmitted. If state boards defined the practice of telemedicine as occurring at the location where the consulting physician is located, all the issues regarding licenses, credentialing, and malpractice would be obviated. 6. Lack of FDA guidelines regarding use of hardware and software in telemedicine systems. There is considerable uncertainty in the remote care market sector because vendors and providers do not know what the FDA will regulate, much less the extent of the regulations or the time it will take to gain regulatory approval. Currently the FDA requires approval for complete medical imaging systems. All teleradiology devices used for primary diagnosis are regulated, and the agency is currently examining teleradiology software that interfaces with hospital information systems. However, it has issued no guidelines on what constitutes a telemedicine system. It is unclear whether the entire network and/or supporting software would be subject to FDA regulations because they support telemedicine.

7. Uncertainty regarding rights to patient information/privacy and confidentiality issues. Patients, as well as insurers and employers, are becoming more concerned about these issues because of the highly personal nature of health information and the uncertainty surrounding who would have access to it and for what purpose. Privacy and confidentiality may be enhanced through the use of NII-related technology, compared with traditional paper-based medical record systems. To date, only 28 states allow individuals to access their medical records and individual state approaches to privacy vary greatly. A more detailed discussion of privacy and confidentiality is contained in Chapter 4, "Integration of Health Information Systems."


Other Barriers

The issues raised here are based on the experiences of those who are already in the early adopter stage.

1. Lack of near-term, methodologically sound evaluation data on cost-effectiveness and medical usefulness. Several groups are advocating the creation of a common evaluation platform so that reliable data on telemedicine is available. Sites may differ significantly, making such data comparisons less useful than was hoped for. The scarcity of evaluation data also makes it difficult for vendors to justify the use of limited internal resources to promote their involvement in telemedicine. Many barriers to implementation will fall if hypotheses on effectiveness and usefulness are proven.

2. High cost of telecommunications infrastructure. Installing the necessary infrastructure remains a high cost of entry in some situations, particularly in rural communities. The cost for transmitting over the network may also be a barrier, depending on rate of use and transmission mode; teleradiology consults are reimbursed, but infrastructure costs remain a major obstacle to delivering these services. If telecommunications deregulation promotes competition and, subsequently, lowers costs and increases accessibility, a more conducive environment for telemedicine may exist. Lower telecommunications costs may also spur information infrastructure development for education, entertainment, and economic development. If this occurs, telemedicine applications might advance more rapidly because they could "ride" the infrastructure developed for these other applications.

Market Impact

The barriers enumerated above, if not overcome or circumvented, will have widespread effect on the role of telemedicine in the U.S. commercial marketplace. Telemedicine will remain a largely regional or intrastate activity, determined by health care providers and insurers, and will not reach its potential of becoming a fully integrated part of the NII for the delivery of health care in the near-term. Adoption will continue to focus on early users who have secured outside funding for demonstration testbeds rather than on fully integrated operational systems. The private sector may be slow to invest, viewing telemedicine more as an add-on to entertainment or administrative priorities than as a leading application. And because the largest U.S. market, the domestic commercial market, is also the market with the most barriers to entry, specialized markets (correctional, Indian Health, VA, and DOD) or those who can pay out-of-pocket may be the principle recipients of telemedicine's benefits. (See Figure 2-8.)

The international market will continue to attract investment that could be directed to the U.S. market. This market is eager for the U.S. technologies that make telemedicine possible, both for linkups within foreign countries and for linkups to the U.S. Many countries whose national health care programs help pay for their citizens to receive specialty care in the U.S. are recognizing the potential cost-savings available through the use of telemedicine. The best expertise in U.S. health care thus may be exported overseas to an environment unfettered by the complex legal and regulatory issues that are slowing adoption in the U.S. market. An opportunity to significantly affect the access to, quality of, and cost of health care in the U.S. may be delayed or missed entirely.



The following recommendations, if adopted, should promote the use of telemedicine to expand access to quality care and reduce the cost of that care.

1. HCFA should base its reimbursement policies on data collected from telemedicine projects where reimbursement is provided so that the cost-benefit data collected will be realistic. HCFA is not yet convinced of telemedicine's clinical and cost effectiveness and does not provide reimbursement for Medicare patients who receive care through telemedicine (except for teleradiology, telepathology, and other diagnostic procedures that do not require face-to-face contact). And because HCFA will not provide reimbursement, most private insurers will not either.

To assist HCFA in its decision-making process, many health care delivery organizations are conducting demonstration projects to test the clinical and cost-effectiveness of telemedicine to deliver care. Most of these efforts, though, were not framed in a competitive market where telemedicine services would be supported by the operating budgets of health care delivery organizations and health care practitioners would be paid for their work. Instead, they are operating in an artificial environment, supported by government grants that often do not permit reimbursement for physician consultations.

As HCFA considers whether to reimburse for telemedicine consultations, it needs realistic cost-benefit data that can be measured against the real world where medical services are routinely reimbursed. The federal government, through the Office of Rural Health Policy, and state governments, should therefore fund demonstration projects that are coordinated with HCFA. HCFA should ensure reimbursement for their duration. These projects should include telemedicine linkages into the home. HCFA should also draw on outcomes data from projects that were designed to be self-sustaining and where reimbursement is provided, such as in a managed care environment. It should not overlook cost-benefit data available from specialized market activities such as telemedicine in correctional facilities and in the international arena. Here cost containment and/or profitability, coupled with increased access to quality care, are the drivers. HCFA should make every effort to complete its analysis and release its policy as rapidly as possible.

2. Speciality medical societies should develop evaluation platforms for telemedicine consultations for patient care, such as the initiative undertaken by the American College of Radiology to create teleradiology standards along with practice guidelines and standards. Payers and providers need improved information to determine whether telemedicine cost-effectively improves access to care and enhances the quality of care delivered, as well as whether patients are satisfied with the care they receive. The medical specialty societies would advance the effort to collect this data if they take the initiative and develop, where feasible, common evaluation platforms in order to maximize sample sizes and results.

3. Clear limits should be set on when software and telecommunications infrastructure are subject to FDA regulations. A number of organizations are advocating FDA reform for device manufacturers. FDA policies are needed pertaining to telemedicine products and services, including clear policy on what constitutes a new device in the integration of a telemedicine system. The FDA should work in partnership with medical specialty societies, manufacturers, and cognizant organizations as it develops these policies to ensure that they meet appropriate standards for patient care and are not so restrictive that they inhibit widespread use of these technologies.

4. States should pass legislation that creates a uniform special license across all 50 states to permit physicians to practice telemedicine across state lines if they are already licensed to practice medicine in any state. As discussed in this chapter, current requirements for physicians to obtain licenses in every state where a telemedicine consultation occurs are costly and time-consuming and are retarding the development of the commercial telemedicine market. This market includes most of the U.S. population, who will be denied the benefits of increased access to care and lower cost until this issue is resolved. A special telemedicine license would be simple to administer and the process could be monitored by a cognizant agency. The license should be uniform in administrative requirements, fee structure, and scope of practice.

5. Initiatives to clarify legal, liability, and malpractice issues involved in the practice of telemedicine should be adopted. Malpractice insurance coverage must cover all states where telemedicine activities occur and must be available at a reasonable cost. Legal action may be required to prevent venue shopping and to cap jury awards. Hospitals may need to provide indemnification at a reasonable cost for physicians engaged in telemedicine consulting.

6. Medical schools and continuing medical education programs should provide telemedicine training in their curricula. Many physicians have had little or no exposure to telemedicine or to the use of related technologies. The most promising way to overcome the lack of widespread physician acceptance of telemedicine and the lack of familiarity with related technologies is through training. Training in these areas should be required in medical school and made available through continuing medical education programs.

7. Manufacturers should accelerate technology development so that telemedicine consultations can be conducted from desktop equipment. In order for telemedicine to be incorporated into the daily practice of medicine, telemedicine platforms must be conveniently located on the practitioner's desktop. The supporting technology must be as easy to use and access as the telephone. Currently, health care practitioners are usually required to leave their offices and go to specially equipped rooms to conduct telemedicine consultations. This is often a sufficient deterrent to the wider use of the technology. Patients must also be able to easily access and use the technology so that medical care can be delivered at the point-of-need. This is usually the home, not the office of the physician. This suggests that similar desktop or television-based technology will be required for the home.





















cclogoSMALL.gif (1930 bytes)

competetextgrey.gif (3110 bytes)

Chapter 3 - Personal Health Information and Management

Highway to Health: Transforming U.S. Health Care in the Information Age



The public is becoming more demanding as well as more sophisticated about the information it wants and actively seeks, particularly as new technologies usher in more ways to obtain more information. This is evident in the health arena as individuals are being asked to take more responsibility for their own well-being and to make more informed choices regarding their medical care. Individuals are revealing broad information interests as well as very specific information needs. Increasingly, they are demanding more control over, choice of, and "connectedness" to their health care practitioner and provider organization, as well as the health care delivery process. Practitioners and health care delivery organizations in turn are creating more patient-centered models of care to ensure that the most timely, cost-effective treatment is delivered.

Unfortunately, most people lack the information necessary to assume these new responsibilities. In 1993 the Advanced Research Projects Agency (ARPA) sponsored a conference to address the critical issues surrounding the NII and its application to health care. The personal care subgroup at that conference reported that the individual makes 95 percent of the first-line health decisions with support from family and friends but without the aid of, or access to, health information and health decision-making tools (DOD 1993). In a survey published by the Medical Library Association (Highlighting 1994), almost 70 percent of those sampled reported problems in gaining access to appropriate health information. When queried, 60 percent said that they would be willing to pay for an integrated resource to provide such health and wellness information. A preliminary study commissioned by HHS for its 1995 conference on Networked Health Information for the Public included a survey to determine consumer health interests among adults (Deering and Harris 1995). Among the 1,005 adults surveyed, 35 percent indicated they had had a health-related problem in the previous 12 months. Fully 86 percent of those people had sought information about their condition, suggesting that once a health event occurs, patients have a high demand for easy access to information concerning that condition. Sixty-eight percent of the survey respondents stated that they have "some" or "many" unanswered questions about their personal health. To meet these ever growing information needs, a market for health-related information resources is rapidly emerging.

This chapter focuses on that market and the potential use of NII-related tools. It describes health information consumers and the significance of two kinds of information that the emerging market is delivering: "wellness" or "prevention" information to assist people in staying healthy longer and "disease management" information to assist them in better managing their own illnesses or those of loved ones. It then surveys the market today and the drivers for consumer-oriented health systems. The market potential for health information is then projected, followed by an exploration of the barriers that may inhibit the anticipated growth. Recommendations are offered to facilitate more rapid commercial development.



The Health Information Consumer

A recent report from the May 1995 conference Partnerships for Networked Health Information for the Public held in Rancho Mirage, Calif., offered a number of observations about today's consumers of health information. As people grow older, their consumption of health information increases along with their use of health care. As a result, older people are generally better informed about health-related matters than younger people. But younger people seem more likely to use new sources for information, such as the Internet. Women tend to be slightly better informed than men, often because of their greater use of health care services. Also, women are often seeking information on behalf of other people, such as their children. Finally, the disabled person is more likely to seek health information yet seems more dissatisfied with its availability.

The desire for more and better health information cuts across demographic lines. Access, though, does not. This might be a result of language barriers as well as disparities in income and education. People with lower levels of income and education are half as likely to read health-related materials, a third as likely to seek health information when they have a health problem, but twice as likely to have problems getting that information.


The Health Information Market

Significant commercial opportunities are emerging to assist individuals in their health decision-making process by providing more targeted information through a variety of media formats. The "ICE" industries of information, cable, and entertainment, for example, see health care for the individual as a lucrative market that can be served by their expanding information delivery infrastructure. Managed care organizations (MCOs) are beginning to recognize the importance of investing in the long-term health of their plan participants and are disseminating wellness and disease management information. In the past, MCOs have focused more on the broad distribution of wellness information when there has been a short-term payback (less than a year), when it has been mandated and paid for by their payer clients, and/or when it has been seen as a tool to encourage enrollment. Today, as these organizations seek to further reduce costs, increase profitability, deliver quality service, and retain members, they are focusing on longer-term strategies to keep their members healthier, thereby reducing future health expenditures and improving the quality of life of their membership.

Preventable illnesses are estimated to account for 70 percent of all medical treatments (Faltermeyer 1994). These include illnesses related to alcohol, tobacco, drugs, and sexually transmitted diseases. Unfortunately, many of these problems seem to be increasing. For example, a recent report by the National Parents' Resource Institute for Drug Education (PRIDE) indicated that alcohol, tobacco, and drug use by high school students is on the rise (Teenagers 1995). Encouraging healthy lifestyles is therefore critical to reducing the nation's health problems and controlling the associated costs over the long-term. More than 75 percent of the doctors polled in a nationwide survey by the Times Mirror Center for The People and The Press indicated that encouraging preventive care was as high a priority for them as eliminating paperwork and limiting malpractice suits (Times Mirror 1993). The NII offers one additional approach to complement or expand this effort.

Studies also suggest that a small portion (5 percent to 20 percent) of members in any health plan, including HMOs, account for a disproportionate share (25 percent to 50 percent) of health care expenditures. For example, 5 percent of Medicare enrollees account for 50 percent of the program's expenditures, and 25 percent account for 91 percent (HHS 1995a). These tend to be patients with major and/or chronic diseases such as diabetes, cancer, and asthma. For this reason, HMOs in particular have focused primarily on delivering more efficient health care to these disease-specific groups in order to reduce cost. According to a recent study of 62 managed care organizations by the Zitter Group, a San Francisco research firm, more than half were considering developing or implementing disease management systems (Terry 1995).

The key to the success of these systems is their ability to identify changes early on that indicate potential problems so that effective treatment can be promptly administered. The cost-savings associated with this approach can be significant. The Humana San Antonio Diabetes program outlined in Case Study 3-1 is an example of one HMO's approach.

Case Study 3-1: Humana San Antonio Diabetes Program



Working with the American Diabetes Association Humana developed an innovative program known as the Health Works' Key to Managing Diabetes. The program is designed to promote community awareness of the effect of lifestyle on health and to empower diabetic members and their families to become actively involved in managing their disease. Once diagnosed with the disease, the member is asked to call Health Works' toll-free telephone number to schedule a two-day diabetes education seminar. After the seminar, the member meets with a Certified Disease Educator (CDE), who helps the member plan an individualized disease management program. A physician, a dietitian, and a registered pharmacist meet with the member and their CDE every three months, or more if necessary, to review their progress. Documented progress is forwarded to the member's physician.

In terms of positive impact, 16 members are now off medication. A random sampling of 377 program participants showed no hospital admissions for retinopathy, nephropathy, or neuropathy during the first six months of 1995.





In 1991, the Harvard Community Health Plan initiated a pilot program to identify pediatric asthma patients who were at high risk for emergency room treatment or hospitalization. The program cut hospitalizations for this high-risk group by 79 percent and emergency room visits by 86 percent by teaching patients and their families how to manage the disease with the aid of home monitoring equipment. As a result of this success, plan-wide asthma programs were developed for both children and adults in 1992. Preliminary data suggest that these programs are significantly reducing hospital admissions.

Managed care organizations could be early adopters, potential investors, and market drivers if providing health information to individual members in their homes or work environments helps maintain healthier patients. However, health care delivery organizations will not drive the use of the NII for disseminating wellness information unless they increase these activities substantially.



The rapid development of new technology is making health information for wellness and disease management much easier to disseminate and access. In fact, health information represents one of the largest single information markets in the United States. The 1994 Standard Periodical Directory, the most comprehensive catalog of periodicals available, lists twice as many health-related titles as appear in the next largest subject area, business and industry (Oxbridge 1994b).

Two recent national surveys released at a September 1995 meeting of the Radio and TV News Directors Foundation (RTNDF) reflect the public's desire for health-related information. Frank N. Magid Associates, Inc. (1995), a media consulting firm in St. Louis, conducted a survey to explore the mood of the electorate in anticipation of the 1996 presidential elections. Respondents voted health care second to education/schools as the issue of most personal importance to them (and also one with insufficient local news coverage). Princeton Survey Research Associates (1995) conducted a survey on local television news and health care. Respondents indicated by large margins that they were "very interested" in seeing more news about medical breakthroughs (66 percent) and diseases and epidemics (60 percent). More than a third of the respondents wanted more news about the relationship between doctors and patients (37 percent) and changes in hospitals (34 percent). Also in September, the Towers Perrin (1995) management consulting firm released a survey on what Americans know, and need to know, about health care. More than a third of the respondents (36 percent) indicated that they need "a lot more" medical information such as information on certain illnesses and conditions.

Unfortunately, a comprehensive, reliable market survey for consumer health information is not yet available. This is because health-related information targeted for the patient/consumer market is only just beginning to be widely disseminated. An overview of the health-related newsletters, magazines, and television programs, though, does provide a snapshot of the market today for current health information.



According to the Newsletter Foundation, a non-profit association representing newsletter publishers and specialized information services, the market today for newsletters most likely exceeds $2 billion. Most newsletters target select audiences and address specific topics. They are produced and distributed by a wide array of entities, including trade and professional organizations, medical centers, consumer advocacy groups, for-profit companies, schools, and private foundations.

One way to gauge today's market appetite for up-to-date health information is to examine the number and growth of health-related newsletters. In 1989, the Oxbridge Directory of Newsletters listed 553 health-related newsletters. In 1994, the number had climbed to 785, a 42 percent increase (Oxbridge 1989, 1994). Current circulation for all health newsletters in the United States and Canada is estimated to be between three and four million. The top five health-related newsletters today are profiled in Figure 3-1. Note that the largest newsletter, based on paid circulation, has more than tripled its circulation over the last five years.

Figure 3-1: Top Five Health-Related Newsletters




Date Founded


Circulation 1989

Circulation 1994


The Nutrition Action Health Letter


Latest research and advice on nutrition, diet, and related health issues.




University of California at Berkeley Wellness Letter


Focus on nutrition, fitness, and stress management




Johns Hopkins Health After 50


Health newsletter for people over 50.




Dr. Julian Whitaker's Health and Healing


Newsletter by a leading expert on alternative health.




Mayo Clinic Health Letter


Issues Relating to Good Health










Source: Oxbridge Communications and Phillips Publishing, Inc.

The Mayo Clinic was the first academic medical center to craft a newsletter that successfully targeted the growing consumer desire for reliable and understandable health information. Driven by a public service mentality and an institutional marketing orientation, Mayo Medical Ventures, a for-profit subsidiary of the Mayo Foundation that provides consumer-oriented health information, launched the Mayo Clinic Health Letter in 1983. A direct mail campaign was initiated with the hope of obtaining 10,000 to 15,000 subscribers. The response was overwhelming. Fifty thousand requests were received and current circulation stands at 380,000, reflecting the fact that people were willing to pay for what they perceived to be reliable health care information. In the process, Mayo's reputation was enhanced and the revenues generated for the non-profit Mayo Foundation continue to support advanced medical research. Interestingly, Mayo estimates that only 10 percent of the subscribers have a relationship with its institution beyond the newsletter.

Mayo attributes its success in providing consumer information to its ability to put health information into a context that consumers easily understand, enabling them to ask the right questions about care and assisting them in sorting good from bad information. Mayo ensures the technical accuracy and readability of its information products whether they are distributed in print or electronically. Its success has not been lost on other leading academic medical centers and universities. Johns Hopkins launched its Health After 50 newsletter in 1989. Others who have initiated their own health-related newsletters include Stanford University Medical Center, Harvard University, the University of California at Berkeley, the University of Texas, and Tufts University. Many others provide more informal pamphlets on specific health-related topics as part of their patient education programs.



Health-related magazines, like newsletters, are an additional source of timely and relevant health information for the public. In 1994, Prevention (Rodale Press) was the top health-related magazine. Its paid subscriber circulation of 3,427,803 made it the 14th most popular magazine in the country, beating out such popular titles as Sports Illustrated, Newsweek, and People. It was also the fastest growing magazine among the top 20. Other health magazines are enjoying increased popularity as well. Men's Health, also published by Rodale Press, grew 37 percent between 1993 and 1994, and was the fastest growing title of the 50 largest magazines in the nation based on paid circulation.

According to Ad Age (1990-1992) magazine and Standard Rate and Data Service (SRDS), which publishes media statistics for advertisers, health magazines have consistently generated exceptionally high revenue growth over the past five years. For example, in 1990 health-related magazines had combined total revenues of $116 million, representing a 12 percent increase over the previous year. This made health the fastest growing magazine category, ahead of news, fashion, and sports. Revenues grew another 24 percent by 1992, underscoring the market demand for health information.



The popularity and current market penetration of video cassette recorders (VCRs)--90 million households in the U.S. have at least one--suggests a market opportunity for distributing prepackaged health information on videocassettes. But although some educational videos have enjoyed particularly strong sales, such as those based on popular educational television shows like the PBS "Magic School Bus" series for children, educational videos on consumer health topics have not been entirely successful. Attempts have been made to market them through pharmacies, but consumer response has been weak. Several companies have terminated sales because of poor volume. However, a recent joint venture will attempt again to exploit consumer VCRs for delivering health information via videocassette. Time-Life Medical, a joint venture between Time Warner and the Dr. C. Everett Koop, intends to produce and distribute a series of half-hour videos on medical topics targeted to the consumer market. The videos will cover such conditions as coronary artery disease, cataracts, diabetes, and migraines. A panel of medical experts will oversee content quality and will advise on each of the selected topics.


Cable/Broadcast Television

Entertainment channels have begun broadcasting health programming with toll-free number call-ins that involve either a talk show format or a toll-free number referral for product or program information. Anchor sponsors use strong cross-promotional efforts to maintain these programs. Delivery is via broadcast or cable television, with both national and local content and scope. They are supported by a variety of health information services, including on-line and referral/counseling services. Thus the programs are actually providing an interactive format through which consumers can request and obtain specific health information. Interestingly, the previously mentioned survey by Princeton Survey Research Associates indicated that respondents would prefer to get information on medical breakthroughs, diseases and epidemics, health care costs, the relationship between doctors and patients, changes in hospitals, and how HMOs work from local television news rather than from national news, radio news, newspapers, or magazines.

Despite the fact that most Americans receive their scientific and medical information from television, this market has also been characterized by failure. At least three medical cable networks--Lifetime Medical Television, Medical News Network, and American Medical Television--have ceased operation due to poor financial performance. It should be noted, though, that these ventures used a mass market broadcast medium to deliver information directed primarily to a narrow market segment of health care providers.



Some studies suggest that the increasing popularity and availability of consumer health information could have a tangible impact on the health of consumers. HHS data indicate, for example, that wellness information and prevention programs for the elderly have been successful in promoting smoking cessation, good nutrition, and physical activity. Acknowledging these trends, managed care organizations will increasingly be involved in screening, counseling, and immunization programs designed to maintain health and promote prevention so that individuals can live independently. Although their current focus is on delivering such health-related information services to their plan members, in the future these organizations may target the broader public market, becoming "electronic safe deposits" of critical health information.

While the abundance of printed health-related information suggests a public appetite for health care information, the market potential may be better understood by examining the growth of interactive technologies and their potential impact on the delivery of health information. In 1995, 37 percent of all households had personal computers (PCs), and penetration was expected to reach 40 percent by the beginning of 1996. Although only 10 percent of the PCs in homes today are equipped with CD ROMs, half of all PCs sold now include them. Sixteen percent of home computers have modems (EIA 1995a,b). In addition, increasing numbers of consumers are believed to be reaching public venues of information through the use of PCs at their work sites, libraries, senior citizen centers, and other community-based resources.

As access to this technology increases, particularly in the home, so does the desire to use it to acquire health-related information. In a recent Harris Poll exploring the educational and communication aspects of the emerging NII, 79 percent of the respondents indicated that they were "somewhat to very interested" in the opportunity to obtain health information via their computer or their television; in contrast, only 32 percent were somewhat to very interested in shopping on-line. (See Figure 3-2.)


Figure 3-2: Interest in Interactive Services



Source: Louis Harris 1994

An internal survey conducted by a leading HMO anticipates that interactive technology could be used effectively to provide general information about health-related conditions and procedures, access to journals and reference books, patient support for specific care decisions, preventive medicine guidelines, health education video games, informed consent, videos on demand, personalized health maintenance programs, and health risk appraisal. Other opportunities include E-mail services between practitioners, patients, health delivery organizations, expert and peer advice bulletin board services, and on-line support groups. The growth of health-related CD-ROMs, information kiosks, and, especially, on-line information indicate that some of these services are beginning to emerge. The desire for health information is prompting the market to respond with information products that take advantage of the new interactive delivery mechanisms.



Only a few years old, the consumer market for CD-ROMs was approximately $648 million in 1994. This represents about 22.8 million units, up considerably from the 5 million that were sold in 1993 for an estimated $202 million. Of the CD-ROMs sold in 1994, about 3 percent were health related, covering such diverse topics as anatomy, drugs, mental health, and women's health.

The consumer health market for CD-ROMs was virtually born in 1992, when IVI Publishers released the Mayo Clinic Family Health CD-ROM. The story of Mayo's decision to publish is instructive. Based on the success of its newsletter, Mayo decided to publish three consumer health information texts. Subsequently, a CD-ROM version of one of these, Mayo Clinic Family Health, has sold over one million copies.

Although there is a great potential for this consumer market, the technology is still relatively expensive and may not become available to a large segment of the population for some time. However, many publicly accessible institutions and retail firms are providing health information using CD-ROM databases. Public libraries and hospital/clinic settings are increasingly making such systems available to the public.


Information Kiosks

Information kiosks that use touch screen and CD-ROM technologies to answer consumer questions are one example of such publicly available information outlets. HealthPartners, a managed care organization in Minneapolis, is installing kiosks in its affiliated clinics to provide information on its more than 4,500 physicians. The system allows people to search for information about a particular physician or team of physicians, including how well they follow guidelines for medical procedures compared with others in the HealthPartners system and other physicians nationwide. Patients feedback is also provided (Jossi 1996).

In March 1995, Kroger placed kiosks in all 64 of its Dallas grocery stores to provide customers with health care information and over-the-counter drug counseling. Two independent pharmacies and Wal-Mart stores are also testing kiosks provided by Info-Touch. A similar product is also built by Health Touch. Consumers can receive information about 30 health care categories at an Info-Touch kiosk through a combination of text, video, sound, and still photographs. The kiosks are leased for around $2,000 a month, which can be offset through advertising private label products, featured specials, or coupon sales. Kroger expects to be on the leading edge of the application of this type of technology, as it was 14 years ago when it introduced blood pressure monitors in its supermarkets. Kroger believes the kiosks provide both a marketing advantage and a consumer service (DeNitto 1995).


On-line Information

The most promising opportunity for delivering health information involves the Internet, a worldwide network that provides computer access to nearly unlimited amounts of information. The Internet is the fastest growing communications medium in history. In 1994, it doubled in size, as it has done yearly since 1988, and now encompasses nearly five million "host computers" on over 28,000 networks. It is now estimated to reach between 15 million and 30 million people, encompassing at least 30,000 databases, with one million new users logging on each month. These new users, along with daily advances in technology, are generating a spectacular increase in information traffic of 15 percent per month. (CRS 1994a, ITU 1996).

Usage rates are also increasing. Ninety percent of on-line users sign on at least three or four times a week and 54 percent sign on once a day, suggesting that this is becoming a regular activity and reflecting a growing consumer acceptance of on-line services as a mainstream medium (ISA 1995).

Making health information available in a networked environment offers the potential for timely delivery of the most current data. The World Wide Web (the "Web"), a subset of computers on the Internet whose information is collectively linked, permits users to select data in one document on a Web computer, receive an index of documents with related information, and automatically connect to any one of these other sources, even if they reside on other computers on the Web. Many Internet users are retrieving health information in this manner, joining discussion groups on specific diseases, and perusing bulletin boards that focus on a number of medical topics and provide personal E-mail referrals to resource groups related to specific diseases. Internet services are also being combined with more traditional forms of communication to provide user-friendly access to relevant health information. Both Medical Alliance consultants of Alexandria, Va., and Global Success Corporation of Naples, Fla., intend to launch newsletters in 1996 to update consumers on the latest on-line health developments.

CW Henderson Publishers of Atlanta, Ga., a large distributor of health information, has set up a home page "NewsFile" on the Web that provides a selection of material from its 12 health-related newsletters. The newest version of the Mayo Clinic Family Health CD-ROM will facilitate an on-line connection to a Mayo Web site. The Web site will include such features as back issues of newsletters and on-line seminars with doctors. Mayo is betting that they will be tapping into a new market and will not cannibalize sales of their existing health information products.


Commercial on-line service companies are also working actively to satisfy the market's desire for health information. The three largest in the country, and examples of health information services they offer, are profiled in Figure 3-3. In addition, Microsoft recently previewed an on-line multimedia health information service that is currently being distributed on the Microsoft Network (MSN). The initial release focuses on providing consumers with information on pregnancy and pediatrics. The resource includes over 400 illustrated articles and an interactive Find By Symptom feature that uses knowledge bases built and evaluated by leading experts in pediatrics. The intelligent index allows users to access information by inputting sets of symptoms.

Figure 3-3: Health Information Available Online








Started: 1979

Subscribers: four million

(Health and Fitness among the top 5 categories of interest among subscribers.)

Health and Fitness

--Consumer Reports' Complete Drug Reference.

--CompuServe-created medical reference library.

--Various health-related forums, such as Diabetes Forum, which has 18,000 members and holds conferences and maintains a bulletin board.

--Services to search various medical databases including the National Library of Medicine's Medline and the National Cancer Institute's CancerLit.


America Online

Started: 1987

Subscribers: Four million

Health Channel

--A number of health-related databases, including Medline and the Consumer Reports' Complete Drug reference.

--Live chat rooms hosting numerous prescheduled support groups each month on tooics ranging from depression to caregiving to cancer.

--Message boards covering more than 1,000 health-related topics.

--Direct links to selected health-related Web sites.



Started: 1990

Subscribers: two million

--Health and medical support bulletin boards.

--Primers and other related articles under Health Topics section.





Source: Compuserve, America Online, and Prodigy


A brief look at the health-related products offered by America Online (AOL) illustrates the growing popularity of on-line health information. In 1993 AOL retained Health ResponseAbility, a multimedia contractor that provides health information, to supplement its health forums and databases. Prior to 1993, AOL's "health channel" consisted of simple health message boards and chat rooms. With the inauguration of Health ResponseAbility's Better Health and Fitness Forum on AOL that year, health information services were expanded and hits to the forum, defined as the number of times the site is accessed, increased 100-fold.

During this time, on-line surveys available to all of AOL's subscribers indicated that Medline was the single most requested service that AOL did not provide. In response, AOL added Medline in September 1995 and hits to Health ResponseAbility's forum doubled from a half million to one million a month. Partly as a result of offering Medline, the number of hits to the health forum is now estimated to have increased 200-fold since its inception in 1993. This growth is roughly 24 times the rate of growth of AOL's membership as a whole.

While the NII is providing access to an enormous amount of data through the Web, users face the daunting challenge of sorting through this tremendous volume of data to obtain meaningful information. For example, a recent search on the Web of the word "health," where a "strong match" was requested, uncovered more than 67,000 documents. Many companies, recognizing the demand for detailed, accessible health information, are rushing to provide consumers with a reliable, user-friendly means of obtaining the data they want. Some, like the Lancet Online Corporation of Cambridge, Mass., have set up Web sites that provide pointers to significant health information on the Web. Others, like Aspen Publishers of Frederick, Md., and PhysiciansOnline of Tarrytown, N.Y., have initiated commercial on-line networks for health care professionals that provide E-mail capabilities, databases, and access to health care regulatory information from the Federal Register. At least one Internet service provider, Far Net, is teaching consumers how to retrieve health information from the Internet. Ohio State University is successfully running a support group for hemophiliacs using the Internet.


Clearly, on-line services can expand access to health data. However, industry experts have expressed concern over the large and growing number of uncredentialed information sources and the cost and consequences to consumers when they act on unsubstantiated or incorrect information (Roan 1995). To address this concern, some of the health industry's most respected institutions are investing in this new medium. On August 1, 1995, the American Medical Association (AMA) launched its own Web site to "serve as a civilized outpost of medical information" in the Internet's "information jungle." Its stated goal is to "communicate with physicians, health science communities, and the general public to improve the practice and delivery of health care." It provides tables of contents and linked abstracts from the Journal of the American Medical Association and the nine Archives journals, as well as other AMA sources.

Other health care stakeholders also have begun to develop and explore the opportunities these networks present. A few forward-looking insurers, including Aetna of Arizona, U.S. Healthcare, Kaiser Permanente, Humana, and FHP, were among the first stakeholders to set up home pages. At the moment, these Web pages are used primarily as marketing tools, but many of them soon will offer detailed wellness information as well as information on their health care practitioners. The Towers Perrin survey (Towers Perrin 1995) found that 66 percent of respondents indicated that information on health care delivery organizations would be "very valuable" in choosing a health plan. As for the types of information that would be desirable, 58 percent of those surveyed would like access to consumer satisfaction ratings of different physicians; 52 percent would like report cards comparing health plans; and 58 percent of the respondents said that an opportunity to compare the percentage of plan members who receive important preventive medical services would influence their choice of a health plan "a lot."

Humana is also using an on-line service to provide information to the 38,000 physicians in its network. Physicians can use it to access practice guidelines, Humana medical news, Medline, and other information. It recently made PhysiciansOnline available to any of its participating physicians.


Summarizing Market Potential

The demand for health-related information is burgeoning as individuals are being asked to take more responsibility for their own health and as health care delivery organizations recognize the value of providing timely information to help people remain well longer and/or better manage their illnesses. The NII offers tremendous potential for providing vast amounts of health-related information to satisfy these needs, and the opportunities to address the mass consumer market are growing, particularly as the "ICE" industries converge. However, as commercial firms direct resources to available lucrative markets, they may deny these new information services to the already underserved population, particularly if they rely exclusively on electronic delivery methods. This is discussed in more detail below.



The barriers listed below are restricting individuals' easy access to health-related information for maintaining their health and for better managing their illnesses.

1. Uncertainty regarding use of information. Once information is obtained, it is unclear how individuals will use it to improve their health. For example, the dangers of smoking have been known to the public for years. But, achieving lasting, widespread behavioral change has been difficult.

2. Changes in the patient/practitioner relationship. Informed health care consumers will alter the traditional patient/practitioner relationship as they participate more actively in the decision-making process for their health care. Emerging trends in "patient's rights" are having the added impact of generating demands for performance information on medical professionals. Resistance to change by both parties is bound to occur.

3. Uncertainty regarding accuracy and/or authenticity of health-related information. The authenticity and accuracy of information received over the Internet or any other mediummay become an issue as information sources other than the medical community enter the wellness market. Whether information should be certified in some way and who would do that is unclear.

4. Lack of easy access to understandable health information. Unfortunately, most people do not have ready access to the kinds of information they need to assist them in assuming more responsibility for their own health. Information providers may be exacerbating the problem by aggressively pursuing new electronic delivery methods that are not yet widely available to the general public. In addition, some of the information available was developed to meet the needs of the medical professional, not the information needs of a diverse population. This may be one reason why consumers' first choice for information has been an intermediary, such as a medical practitioner, who can translate technical medical jargon into more understandable terms. Given the high cost of educating individuals through their health care practitioners, there is a growing need for easier access to more "user friendly," accurate health-related information.

5. Uncertainty regarding liability for disseminating health-related information. It is not clear who will be liable if inappropriate information is dispensed that results in medical harm. Would it be the author or possibly the information provider? Publishers of printed information are not usually liable for fraud, willful or intentional misrepresentation, or negligence. The question of an author's liability is separate and there is a lack of case law in this area. Some believe that First Amendment protection would make a suit difficult. However, it is not clear whether health plans, for example, would be held liable for wellness or preventive information distributed electronically to patients if medical harm resulted.

6. Cost of electronic publishing. Though inexpensive compared to print, electronic publishing can be a costly add-on to existing means used by academic medical centers and non-profit organizations to disseminate information. Hence, most detailed information is currently posted electronically by those who have an economic incentive to do so, such as managed care organizations, drug companies, and commercial on-line services.



The following recommendations, if adopted, should help the public gain access to the information it needs to support the health care decision-making process.

1. Major medical associations and professional societies as well as health care delivery organizations should establish guidelines for themselves and their members for screening and authenticating health-related information before it is publicly disseminated. Given the tremendous proliferation of health information, it is increasingly difficult for the public to assess its quality or determine its source. The process for screening this information prior to public dissemination is informal and incomplete. Confusion is bound to arise as information increasingly is delivered in formats traditionally associated with the entertainment industry. Peer-reviewed journals are available but are generally written for the medical professional, not the general public. Some academic medical centers and managed care organizations are carefully selecting the information they distribute in print and electronically to ensure its veracity and to ensure that it is understandable to the general public.

Because it would be difficult to develop a national "seal of authenticity" for publicly disseminated health information (and nearly impossible to administer), major medical associations, professional societies, and health care delivery organizations should take the lead and implement procedures to carefully screen and verify all health-related information they distribute, particularly electronically. They should encourage their members to adopt a similar screening process. This would increase the sources of accurate information for the public.

The public should also recognize that the quantity of health and wellness information available today does not necessarily reflect equivalent quality. Just as individuals must discriminate between authoritative and hearsay information on other topics, the same diligence must be exercised when considering health-related information.


2. Vendors, health care delivery organizations, and payers must consider the cultural, economic, and generational diversity of the population when developing health and wellness information products and services. Individuals must have ready access to the information they need to better manage their own health. But the market for health and wellness information is not a single market. It is a variety of specialized, unique populations. Given this diversity, information providers (including health care delivery organizations) must take great care in selecting both the means and the format of communication. Issues to consider include the technical (il)literacy of the targeted population as well as its access to certain information delivery tools such as cable television or PCs. For example, while PCs show great promise in their ability to deliver sophisticated, interactive health and wellness information, more homes have other media, such as print and television. Additionally, the populations most likely to use PCs to access health information may not be those most "at risk" for certain diseases and illnesses and who need the information most.





























cclogoSMALL.gif (1930 bytes)

competetextgrey.gif (3110 bytes)

Chapter 4 - Integration of Health Information Systems

Highway to Health: Transforming U.S. Healthcare in the Information Age




The health care industry is undergoing a massive restructuring in response to concerns about rising costs and access to services. Payers, health care delivery organizations, and practitioners (in different combinations) are linking together through mergers, acquisitions, and alliances, transforming themselves from standalone operations and businesses into health care "systems" supporting a continuum of care. This metamorphosis requires integration across many dimensions. Geographies must be bridged to facilitate new communications, new governance and financial models must be created, and clinical practices must be coordinated.

Integration must also occur at another level. Once stakeholders are linked into real or virtual health care delivery systems, they must strategically align their information technology with their new business objectives, often adding additional functionality. Not only must information technologies be integrated within each individual organization (intra organizational); they must also work cohesively and seamlessly across the multiple organizations that have linked together into new health care delivery systems (enterprise wide) and, ultimately, across multiple delivery systems (interenterprise). (See Figure 4-1.) The extent to which they can achieve this will pace their ability to deliver higher-quality, more cost-effective, patient-centered care and, hence, their long-term competitiveness. The task, though, is daunting. In a recent national survey of health care professionals, a majority indicated that a national health care information system will either take longer than 10 years to develop or will never develop. (See Figure 4-2.) Just two years ago, 52 percent had responded that such a system would happen within five years.

Figure 4-1: Information Systems Integration Evolution


Categories of Integration



Intra organizational

Within a single health care delivery organization such as a health plan, health, or clinic.


Enterprise wide

Across mutual health care organizations that are linked together and acting as a single entity, such as a regional or national managed care organization with delivery capability.



Across all major health care delivery enterprises: government-funded, fee-for-service, managed care, and other risk-sharing enterprising.





Source: Council on Competitiveness

Figure 4-2: When Will We Have a National Health Care Information Infrastructure?


[Figure 4-2]

Source: HIMSS/HP 1995

The potentially lucrative health care market for information integration has attracted both telecommunications and systems integration companies to enter the business. As both infrastructure and networking experts, the Regional Bell Operating Companies (RBOCs) have aggressively pursued this opportunity. Pacific Bell has created a 100-member health care marketing group targeting improvements in telecommunications and administrative services. Ameritech, also with more than 100 members in its Health Connections Group, is involved in implementing regional and enterprise wide health networks. NYNEX develops infrastructure, software applications, and telemedicine systems. The remaining RBOCs (Bell Atlantic, BellSouth, US West, and Southwestern Bell) all have significant health care systems activities. All of the RBOCs have been operating in their own regions as restricted by law. However, recently passed telecommunications legislation permits them to expand their services to other domestic geographies when certain conditions are met.

Information-solution suppliers and systems integrators such as 3M Health Information Systems, IBM Healthcare Solutions, Science Applications International Corporation (SAIC), and the EDS Healthcare Division have identified information systems integration for health care as a major business thrust. Industry teams are being created with hardware, software, network, and imaging system partners that can provide fully integrated, open systems. Many consumer-oriented software vendors, such as Microsoft, are introducing products specifically designed for the health care information systems market.

This chapter focuses on the integration of information technologies across multiple organizations, in support of the integration of health care stakeholders into new delivery systems. It explores the evolving health care information systems integration market from the perspective of the various stakeholders; characterizes today's piecemeal information systems purchasing environment; emphasizes the key roles of privacy, standards, and quality issues; and outlines future markets, including barriers to widespread commercialization. It points out the NII-related elements of infrastructure and information that are necessary to build truly integrated, open systems across all health care components and participants (interenterprise systems). Such systems will necessarily focus on the myriad of financial, administrative, clinical, and quality functions of health care delivery while meeting the specific needs of each of the stakeholders: patients, practitioners, health care delivery organizations, employers, and payers.



Understanding the Need

Historically, the health care industry has not invested as high a share of revenue in information technology as other information-intensive industries such as banking and finance. This, despite the fact that there is a tremendous need for integrated information technologies and information management tools. For example, many of today's paper-driven processes add significantly to health care costs and inefficiencies. Between 20 percent and 30 percent of our national health care expenditures are associated with informational paperwork for the hundreds of millions of transactions that take place every month (Woolhandler 1993). A recent grant solicitation from the Commerce Department's Advanced Technology Program (ATP) indicated that in one site alone, physicians spent 35 percent of their time, and nurses 50 percent of their time, doing paperwork (DOC 1994b). Some estimates suggest that as many as 13 percent of the one billion to two billion claims filed annually are returned for error correction (NJIT and Edison 1994). The Workgroup for Electronic Data Interchange (WEDI 1993) has estimated that half of all existing paper-based medical records are either missing or contain incomplete data.


Other indicators also point to the need for information systems integration. Often, important information is unavailable or inaccessible because it is spread across multiple information systems and/or organizations with differing systems. This can result in poor care and increased illness and mortality. Concerns about rising costs, accountability, and quality have resulted in more reporting and documentation requirements. This has led to an increase in both administrative costs and errors. The Institute of Medicine has reported that 11 percent of laboratory tests must be reordered because of lost results and that 30 percent of treatment orders are not documented at all. The Advanced Technology grant program has estimated that diagnosis results are not recorded 40 percent of the time.

The advent of managed care has also stimulated the need for systems integration. In fact, a survey of health care information executives by the Health Information and Management Systems Society (HIMSS) and Hewlett-Packard (HP) indicates that the number one reason for chief information officers (CIOs) of health organizations to adopt integrated information systems is the movement toward managed care (HIMSS/HP 1995). The need to integrate existing information systems to provide usable administrative, clinical, and quality/performance data will increase as cost-containment and the move toward capitation take greater hold. Other significant drivers for systems integration, according to the same survey, are the need to collect comparative outcomes data, develop clinical practice guidelines, and create a computer-based patient record.


Understanding the Challenge

Traditional health care information systems investments typically extended only to the procurement of the hardware, software, and administrative components within a specific department of a hospital, such as radiology or finance; or among departments (intra organizational). Decision-making was decentralized, and individual departments had only to address their own requirements. As a result, "islands of automation" developed, as separate departments often worked independently with vendors who created proprietary solutions to service the unique needs of these different users. Past integration efforts created intra organizational information clearinghouses with administrative transactions (typically for billing or cost-accounting) at their core.

Future efforts must embrace both a wider set of health care constituents and a broader range of information. They will involve the combination of information processing and telecommunications, computer-based patient records, and point-of-care order entry, ensuring access to the full spectrum of health information. Reflecting the strategic importance of these activities, a survey of health care institutions indicated that 70 percent have instituted CIO positions in their governing structures, up from 50 percent in 1992/1993 (Healthcare 1995).


Information is seen as the common thread connecting all health care industry stakeholders as they seek to create integrated delivery systems. However, the image of "building an information superhighway" may be masking the true nature of the systems integration challenges facing the health care industry. Such imagery implies that the major effort needed to achieve integration is creating the physical communications and computing infrastructure. More important will be the language agreements such as rules governing formatting, semantics, and the coding definitions that enable electronic communication of health-related information. Given a common language, stakeholders can then begin sharing what is now transmitted on paper, such as clinical data shared by labs, hospitals, and doctors. Once agreements among stakeholders have been reached, health care data warehouses can emerge to facilitate collaborative efforts in outcomes research, such as determining regional and national health care norms. (See Chapter 5, "Health Care Research and Education.")

Other industries, such as banking, airline transportation, and real estate, have effectively accomplished industry wide information integration to improve services while reducing costs. They decided what information would be provided to whom and devised the standards for presentation and sharing of data. For example, a single financial statement containing ATM transactions and assessed fees can be obtained from one bank even though those transactions may have been conducted through other banks or institutions. These changes were brought on by consumer demand to have access to money and accounts from any automated teller machine (ATM) worldwide. Consumer demand for access to real estate information is prompting that industry to display roughly 600,000 nationwide listings on Internet home pages maintained by local real estate agencies.

The health care industry, however, remains fragmented and has less direct and explicit consumer (patient) demand for integration. Information integration occurs mainly within each stakeholder's organization. Stakeholders are struggling to adopt enterprise wide integration. It is unclear whether they can achieve interenterprise integration. For example, requests for payment for medical services come from individual entities such as the physician, hospital, and pharmacy, and are sent to both the patient and, usually, a third-party payer for claims processing. The future market challenge is to determine whether the majority of stakeholders will opt for more open systems that can reap more benefits by enabling common processes and the sharing of information industry wide, across all stakeholders, including competitors.

Today, access to health care data is seen as a competitive advantage. In an open system environment, competition would be based on how information is used rather than on access to it. Increasingly, collaborative stakeholder initiatives look to an open, network-based system that can support their existing systems and provide access to a broad spectrum of shared information. A representation of this open system environment is contained in Figure 4-3. The Open Systems Foundation has created a model of seven levels of integration, representing valid integration issues among departments within an organization or between health care enterprises. Some organizations are using that model to guide their own integration efforts.


Figure 4-3: An Open Information System


[Figure 4-3]

Source: IBM Healthcare Solutions

At the same time that the use of information to create interenterprise integrated health care delivery systems holds the promise of increasing access to medical services and lowering costs, it also raises complex legal and social questions. For example, who will have access to the information collected and for what purposes? The sensitive nature of the information flowing over electronic networks between the health care system and the consumer/patient will require the industry to urgently address issues of privacy, ownership, accuracy, and access to information.

Financial industries have had legislation to provide guidelines for the reporting and release of information. A person's medical data, however, has not enjoyed such comprehensive protection. In fact, no consistent set of legal guidelines exists defining the rights and responsibilities involved in the use and disclosure of identifiable patient information. While U.S. lawmakers have considered uniform rules in the past, they have so far failed to adopt legislation to preempt the existing multiplicity of state confidentiality laws. The European Union, by contrast, recently adopted a wide-ranging directive to harmonize the laws governing the protection of personal data, including health information, across member states. The continuing failure within the U.S. to address this issue will limit integration to an intra organizational activity, thus missing the tremendous opportunity provided by the NII to link all health care stakeholders in enterprise wide and interenterprise networks. (See the benefits chart in the "Quantifying Benefits" section later in this chapter.)



Technology is available for accomplishing information systems integration, and successful applications within the health care industry exist. Several are profiled below in case studies. For the most part, though, health care has remained a relatively slow adopter. This is due, in part, to the high cost of the technology and the complexity of the task at hand. No single product will cause the integration of the health care community.


Factors Slowing Market Adoption of Integrated Systems


The commercial health care industry may be slow in integrating information-based systems for a variety of reasons:

1. Piecemeal purchasing environment. The purchasing environment for information systems is fragmented in many health care delivery organizations. Most preexisting information systems lack open characteristics or extendable networks that allow for the merger of differing subsystems. Charles J. Singer & Co. (1995) reports that of the health care information systems in operation today, virtually all will be replaced over the next 5 to 10 years. These billing-centric systems were developed primarily under the fee-for-service paradigm and are inflexible and unable to meet the needs of newer, risk-sharing health care delivery models. Nor are they able to meet practitioners' needs for on-line clinical data.

Health care delivery organizations appear confused by the number of available choices for information systems, their rapid rate of obsolescence, the variance in and reliability of vendors, and the absence of clear market leadership. They are fearful that information system costs will exceed potential benefits. Some early adopters of departmental information systems have found that these systems are now outdated and incompatible, and that they have failed to prove their cost-benefit advantages over time. Perhaps because of the proliferation of information system vendors and products, satisfaction among end users is mixed. In a survey of more than 300 information systems managers and CIOs, only 48 percent reported being satisfied with their information system products; 11 percent responded that they were dissatisfied (Bergman 1995). As a result, the general characteristic of purchasing is risk-averse inertia.

2. Proliferation of proprietary financial and administrative systems. The proliferation of payer systems, each with its own proprietary documentation system, has contributed to increased provider costs. This has led some providers to contract with outside organizations for automated claims processing. The same problem exists for payers as they try to communicate electronically with their providers. For example, Humana, a national HMO, has contracts with more than 38,000 providers, using a variety of the 2,000 practice management systems on the market to keep track of in-house records. Humana's integration efforts are hampered by the sheer number of practice management systems, differing programming priorities of vendors, lack of data standards, and financial disincentives for upgrading or changing existing systems. To date, Humana has linked 7,000 of its physicians via an open health information network that enables the electronic exchange of mail and eligibility, referral, claims, and encounter information. Figure 4-4 illustrates part of the Humana challenge and is indicative of the systems integration task facing the entire health care market.


Figure 4-4: Humana's Integration Challenge


[Figure 4-4]

Source: Humana, Inc.

Information system managers and other health care professionals involved in systems integration report similar problems. Their current top priority is integrating systems across separate facilities, followed closely by the need to implement a computer-based patient record and upgrade network infrastructure (HIMSS/HP 1995).

3. Lack of well-defined business incentives to invest in integrated systems. Traditionally, health care organizations have not operated in the kinds of enterprises that are developing today. As a result, they have not had the business incentives to invest in the systems necessary to move financial accounting and clinical information outside of their domain. This includes the investment required to migrate from paper medical charts. Practitioners have accepted the drawbacks of manual repositories of patient information and have not had the clinical incentives to change their own work practices to adapt to computerized systems. Compounding the challenge is the fact that many executives in these organizations have not recognized the need to prepare for this kind of environment and have not made the appropriate investments to ensure that they would be equipped today with these systems. Few health care stakeholders have information system development plans clearly linked to their immediate business and long-term strategic priorities. And unless many invest similarly, it is difficult for the "early adopters" to realize value for their systems. To experience the benefits that integration can deliver, there must be a critical mass of organizations willing to adopt these tools.

These phenomena are not unusual when new technologies are being introduced. Before a critical mass of the general public drove automobiles, cities and states lacked the incentive to make significant investments in roads. Likewise, until a critical mass of the public was willing to purchase telephones, that new technology had limited value because there were so few people to call.

As a result of these lagging incentives, the health care industry's overall spending on information systems still trails that of other information-intensive sectors. A 1995 Information Week survey revealed that the top systems spenders in the banking industry invested, on average, 5 percent of revenue on information systems. The top spenders in the financial industry dedicated an average of 7.5 percent of their revenue. In contrast, the top health care companies spent only about 1.8 percent of revenue (Biggest 1995).


Establishing integrated information systems appears expensive, especially in today's cost-conscious environment. Various estimates suggest that costs can reach from $5 million to $50 million annually during the start-up phase for large hospitals, HMOs, and other provider organizations (Charles J. Singer 1995). Historically, health care capital dollars were invested in medical equipment and "bricks and mortar," not information technology. However, even at a time when most health care organizations are reducing costs, redesigning their business and clinical processes, and downsizing their work forces, indications are that the importance of systems integration is rising rapidly. More than 89 percent of health system CIOs anticipate greater than 20 percent increases in information technology budgets within the next two years (HIMSS/HP 1995). They are recognizing that basic infrastructure upgrades are required before health care organizations can effectively move to enterprise wide integration.


Promising Integration Efforts

Although market adoption has been slow, a number of systems integration initiatives are proceeding. Payers, health care delivery organizations, and employers, including the military, have begun intra organizational, as well as enterprise wide, integration of both administrative and clinical systems. In some cases, integration has been promoted by state legislatures. In 1992, the Minnesota state legislature enacted Minnesota Care to encourage the development of integrated service networks by, among other things, granting state antitrust exemptions to providers and purchasers that form business alliances. In other cases, multiple health care stakeholders are jointly pursuing systems integration solutions in combination with single or multiple vendors. These solutions involve health information networks that enable the stakeholders to access clinical information, laboratory reports, specialist examinations, insurance information, and accounts receivable data simply by using unique patient and provider identifiers.

In Ohio, for example, the Ohio Corporation for Health Information is working with IBM, TRW, Sterling, and Medpower. The Greater Dayton Area Hospital Association, with funding from 19 hospitals and one insurer, has built a patient information network that at the outset links 5 hospitals and medical centers. Eventually, all of the investors will be linked. IBM and Ameritech are developing and supporting the network infrastructure. Programs are being supported by both public and private funds. The data collected has the potential for addressing issues of access, costs, quality of care, delivery of health services, disease incidence, and public health and health planning. The Regional Health Information Network of Northern Ohio, supported by Ernst & Young and Ameritech, has similar objectives. The Greater Cleveland Hospital Association provides the community health information network services. Governance of the network is controlled by the hospital association, not the network provider. Integrated Medical Systems (IMS) has health information networks linking physicians, hospitals, independent labs, and managed care organizations. IMS networks are currently being used by Humana in Tampa and Miami, among other markets.


One of the largest health care systems integration efforts launched to date is the U.S. Department of Defense Composite Health Care System (CHCS) which is described in Case Study 4-1.


Case Study 4-1: Composite Health Care System, Department of Defense


The Composite Health Care System (CHCS) is the integrated information management system that supports U.S. military health care worldwide. The CHCS was designed, developed, and implemented by Science Applications International Corporation (SAIC) under a $1.1 billion, eight-year, contract. It provides the clinical and administrative system that serves over nine million beneficiaries at more than 167 hospitals and 583 outlying medical and dental clinics. The volume of visits in DoD medical facilities exceeds 22.8 million annually with over 464,000 hospital admissions.

This system automated many medical center operations and patient record keeping. It permits authorized users to quickly access all CHCS data from work centers in hospitals and clinics, the pharmacy, the laboratory, admissions, and other access points. Through the CHCS system, users can communicate with selected external computer systems, such as the Defense Enrollment Eligibility Reporting System to verify eligibility for care.

The clinical CHCS modules have been expressly developed for and by physicians, nurses, and allied health professionals to serve as the vehicle for entering and retrieving inpatient and outpatient data. These modules include Registration/Admitting, Patient Appointment and Scheduling, Clinical Orders and Results, Pharmacy, Laboratory, Radiological, Dietetics, Managed Care, Record Tracking Quality Assurances, and Medical Services Accounting. Over 11 million CHCS patient records are accessible online by over 19,000 physicians and 153,000 CHCS system users.





Performance and Quality Measures

Corporate health care purchasers, public purchasers (Medicare and Medicaid), health plans, and consumers are all struggling to make more informed health care choices. They are demanding that health care delivery organizations supply performance measurement information using standardized definitions, agreed-upon data sources, and similar data collection methods. Integrated information systems will assist delivery organizations in determining these outcome measures by supplying the necessary data.

The need for such measures is being addressed by the independent, non-profit, National Committee for Quality Assurance (NCQA). The NCQA Committee on Performance Measurement has created the Health Plan Employer Data and Information Set (HEDIS), an integrated set of performance measures used to document the quality and value of health plans. HEDIS requires HMOs, preferred provider organizations (PPOs), and point-of-service plans to provide quantitative data on enrollment, member satisfaction, use, clinical care, staff credentialing, and financial stability. Although the process is still new and is being revised to better measure clinical quality and productivity, it represents a promising way to monitor, measure, and improve managed care systems. Already, four large employers are using it to produce health plan report cards. By 1996, over one-half of the nation's managed care organizations will be participating in the HEDIS process.

Organizations using the integrated information from HEDIS, along with employee surveys, can select high-quality, best-practice HMOs. At least one employer, Xerox, realized substantial benefits in controlling the escalating costs of health care for its 55,000 U.S. employees, retirees, and eligible dependents while maintaining employee health care plan satisfaction through elective plan participation. (See Figure 4-5.)


Figure 4-5: Managing Costs Down at Xerox


[Figure 4-5]


Figures represent average yearly increase in medical premiums.

Source: Xerox Corporation


Although analysts differ in their definition of the market, they all agree it is set for major expansion. According to Alex Brown & Sons, the health care information technology industry will more than double in size by the end of the century, from about $9 billion a year to $20 billion (Salganik 1995). Integrated hospital information systems are expected to have a 39 percent compound growth rate from 1991 to 2001, when revenues of $53 million are projected (Frost & Sullivan 1994). Information gleaned from a 1994 survey of 137 integrated health care delivery organizations conducted by Charles J. Singer & Co. indicates that virtually every information system in operation today will be replaced in the next 5 to 10 years. As a result, these organizations are expected to spend more than $2 billion annually for integration services over the next five years (Charles J. Singer 1995). The Singer survey projects that information technology expenditures by the health care industry will rise from their norm of between 1 percent and 2 percent of revenue to an average of 5 percent to 7 percent. (See Figure 4-6.)


Figure 4-6: Information Technology Spending


[Figure 4-6]


There were few sizable new initiatives in 1994.

Source: Charles J. Singer 1995

The absolute dollars that organizations are targeting to enhance their information infrastructures are not insignificant. For example, delivery systems such as Fallon Health Plan in Worcester, Mass., and Allina in Minneapolis, Minn., are earmarking $50 million to $150 million annually in capital and operating expenditures over the next 5 to 10 years. Kaiser Permanente Northern California Region is projecting that it will invest $1.2 billion in clinical information systems over the next five years. This figure is very comprehensive as it includes money being spent on existing applications as well as all operation costs, the computer network and platform to support the new system, training and replacement costs, redesign costs, and central computing costs. Based on current forecasts, Kaiser estimates that the break-even point associated with the achievement of anticipated benefits for an integrated and comprehensive clinical information system is approximately 6 years, with a 200 percent benefit payback on investment achieved in 10 years.


Figure 4-7: Typical Three- to Five-Year Capital Budget for an Integrated Health Care Information System*


Initiative Type


Capital Budget



Planning, Staffing, and other Foundation Infrastructure

$5 million to $15 million


Communications Network

Local Area Networks (LANs), Wide Areas Networks (WANs), Electronic Data Interchange (EDI), E-mail, Video Conferencing, Telemedicine, etc.

$10 million to $30 million

Master Patient Index, Central Registration, Enterprise Appointment, and Resource Scheduling


$10 million to $20 million


Data Repositories and Decision-Support Systems


$10 million to $30 million


Replacement and Consolidated Legacy Support

Hospital Information System (HIS), Capitation, Claims, Lab, Pharmacy, Radiology

$10 million to $50 million


Physician Office Support Systems

Billing, EDI, Risk, Reimbursement, Appointment Scheduling

$10 million to $30 million



Clinical Information Systems

Longitudinal Patient Records, Order Management, Clinical Guidelines

$20 million to $100 million



$75 million to $275 million



Estimates are for capital expenditures of a health care elivery organization with one billion dollars to two billion dollars in revenue. Capital expenditures include capitalized labor costs dedicated to the development and implementation of new systems.


Source: The Charles J. Singer 1995

Figure 4-7 illustrates the investment that a health care delivery organization with $1 billion to $2 billion in revenue can be expected to make for a health care information system over a three- to five-year period. It does not include the cost of mistakes, such as choosing the wrong vendor and/or technology, or of upgrading/replacing existing legacy systems. These could add at least 20 percent, and in the worst case up to 100 percent, to the costs, particularly those related to capital expenditures.

Offsetting these industry expenditures are a number of prospective savings. Projections on the use of basic interenterprise administrative systems indicate a savings of $45 billion from reducing redundant testing and excessive paperwork and through the identification and earlier treatment of disease (AHA 1993). An HHS study estimates that a nationwide electronic health care information network could produce savings of more than $100 billion over the next eight years on these grounds. Over half of these savings are expected to come from a reduction in the number of diagnostic tests ordered, increased efficiency, and shorter hospitals stays (HHS 1992b).

Prospective savings may be greater and initial investments may decrease over time as more applications and data reside in the networks rather than in each stakeholder location. Networked computing may hold the key to reducing investment barriers by spreading the cost of development and maintenance across communities of users. Individual hospitals, plans and practitioners could avoid the cost of upgrading or replacing existing systems. This would require a shift in today's competitive model, where competition is based on access to information and not how it is creatively used.

Despite the magnitude of the projected opportunity and the associated cost-savings in all stakeholder segments, three fundamental issues must be addressed in order to achieve this market potential:

--Concerns surrounding privacy, confidentiality, and security of medical information collected and disseminated.
--The need for standards in information content and networking.
--The growing demand for ways to quantify benefits in terms of objective cost-benefit analyses.

Because of their impact on the developing health information systems market, these issues will be addressed separately.


Privacy, Confidentiality, and Security


Source: Louis Harris 1995

Creating a viable, integrated, health care system where sharing of information is considered beneficial rather than detrimental, requires that a balance be struck between the need to ensure an individual's right to privacy and the need for practitioners, health care delivery organizations, payers, and employers to share information. Currently, there is no agreed-upon definition for personally identifiable medical information. Moreover, no uniform "rules of the road" for health care stakeholders exist to ensure that they are handling this information appropriately. This social and legal ambiguity has consumers, practitioners, and health care delivery organizations concerned. Many organizations are hesitating to invest in their own integrated information capabilities, as well as to extend those capabilities beyond the immediate patient interaction. For example, employers and other payers may access an individual's health information because they are frequently paying a significant portion of the health claims. However, if individuals fear that their employers will have access to sensitive health information, it may affect whether they seek treatment within their health care plan, go outside their plan, or even seek treatment at all. (See Figure 4-8.)


Figure 4.8 - Public Concern Over Privacy of Personal Health Information



Data from a 1995 Harris/Equifax pole confirm that clarifying the "rules of the road" for protecting personal health information will help significantly to address public concern:

Of those surveyed, 80 percent would be willing to have their medical records in a computerized medical records system if privacy standards are put in place.

Of the 74 percent of respondents who were either "very" or "somewhat" concerned about the potential negative effects of a computerized medical records system, more than three-quarters said they would be willing to have their records in the system if privacy safeguards were present.




A viable, robust marketplace for health information networks and services will be slow to develop so long as public policies and legal guidelines regarding confidentiality and privacy of health care information are neither clear nor complete. The legal environment surrounding health care-related privacy, confidentiality, and security is a patchwork of uncoordinated state and federal policies. Federal legislation has established guidelines on releasing information about drug abuse and mental health treatments. There are also restrictions on releasing federally held information on individuals. Yet only 28 states currently allow individuals to access their medical records, and individual state approaches to privacy vary greatly. Often, they combine state constitutional safeguards, criminal and civil statutes, common law, and administrative adjudication. For example, California limits the disclosure of medical information by providers and third parties and guarantees individuals the right to inspect their own medical records. In addition, both civil and criminal penalties are provided for violations. In contrast, some states like Arkansas offer the individual virtually no protection at all.

Currently, a number of states provide no legal guidelines on privacy and access to a person's medical records. In addition, 34 states have conflicting laws governing confidentiality. In a society as mobile as the United States, where more than 50 percent of the population is living on or near a state border, health information regularly travels across state lines, further affirming the need for uniform federal guidelines. This patchwork of inconsistent statutes is depicted in Figure 4-9.


Figure 4-9: Conflicting Confidentiality Laws Exist in 34 States


[Figure 4-9]


Source: AHIMA 1995

Individual patients and others will not feel confident that they know how to hold the health care system accountable unless they know the information "rules of the road." Correspondingly, health care information stakeholders may refrain from making the necessary investments or properly using their information systems until their obligations and responsibilities are clearly articulated.


Standards for Administrative and Clinical Data

A standard is defined by the National Standards Policy Advisory Committee as "(a) prescribed set of rules, conditions, or requirements concerning definitions of terms; classification of components; specification of materials, performance, or operations; delineation of procedures; or measurement of quantity and quality in describing materials, products, systems, services or practices." Standards are created either informally, as solutions developed by individual companies and eventually adopted or adapted to on an industry wide scale, or formally, by recognized standards-creating bodies. The informal process evolves through market competition, while the formal process involves voluntary industry wide consensus building. (See Figure 4-10 for examples of the informal process.) Each has its drawbacks. Informal standards may be slow to develop or may never emerge if there is no clear market leader. The resulting market confusion often retards investment and slows purchasing because customers are concerned that products acquired today may soon be obsolete. The formal process is often slow and unwieldy due to the number of participants. Technology development often out paces the formal process, resulting in standards that fail to address the problems they were intended to solve or that are irrelevant because the market has changed.


Figure 4-10: Examples of the Informal Standards Development Process



The development of the banking industry's ATM network is a good example of the informal standards development process. It took more than 20 years before the nation's ATMs were fully integrated, although the technology had been available since 1970. With the industry comprising more than 45,000 distinct financial institutions at that time, no single organization had sufficient market share to force the integration of ATMs. More than a decade passed before consumer demand finally created the market pressures necessary to bring the country's banks together to begin the integration process.

The residential real estate industry provides another example of the informal standards development process. For decades, this industry has shared information using its Multiple Listing Service (MLS). The MLS has traditionally integrated listing, search, mortgage services, and geographic information into a standard format enabling consumers and agents to access greater amounts of housing information on a local or regional basis. More than 1,800 regional real estate boards use this service.

This information was first integrated into a paperbook service and later automated. In 1994, the National Association of Realtors unveiled a five-year, multi-million dollar, information initiative called the Realtor Information Network. The network uses the Internet and over 500 home pages to display roughly 600,000 listings nationwide maintained by local agencies. This initiative is driven by consumer demand for access to real estate information. The Austin, Texas Board of Realtors projects a 90 percent cost-savings from its decision to put the MLS on the Internet rather than use its traditional paperbook service. With this decision, it joins real estate boards in cities in Colorado, California, and New Jersey in moving to an online information-sharing way of doing business.




The development of universally accepted standards for the definition, collection, communication, and storage of administrative and clinical data is a prerequisite for creating an integrated health care delivery system from disparate participants. Given the "many-to-many" health care industry environment, universally accepted standards, rather than a variety of proprietary standards, are urgently required. Linking systems and enabling information to be accessed by distinct users of information does not improve health care delivery if the information cannot be readily understood or compared. Complete health information systems integration requires that all health information be communicated in a common "language," with accepted spelling, grammatical rules, and agreed-on nomenclature. The data must be structured in such a way that it can be easily reviewed by the provider of care, regardless of the source/stakeholder that created the information originally. Standards for health care information are also required to measure outcomes and measure them consistently across health care plans.

The health care industry has been trying to develop these standards through the formal standards-setting process. The process has been slow, due partly to the perplexing array of committees and organizations that are involved. At the international level, health care standards fall, to a large extent, under the purview of the International Organization of Standards (ISO). The ISO itself is composed of 92 separate national standards organizations. One of those, the American National Standards Institute (ANSI), is the private, non-profit accrediting body for standards development in the United States. Its wide-ranging efforts include the input of more than 1,300 national and international companies, 30 governmental agencies, 20 institutional members, and 250 professional, technical, trade, labor, and consumer organizations. Further complicating the situation are the myriad of subcommittees and advisory panels that compose each of the hundreds of ANSI-accredited Standards Development Organizations (SDOs). Consider, for example, the Accredited Standards Committee X3, on Information Processing Systems, which has over 75 technical subcommittees and involves no fewer than 2,500 volunteers. The Medical Records Institute has identified at least six categories of health-related organizations involved in the standards development process. These are profiled in Figure 4-11.


Figure 4-11: Categories of Organizations Involved in the Health Care Information Standards Process in the U.S.




Standards Development Organizations (SDOs)

ACR/NEMA (DICOM - American College of Radiology and National Electrical Manufacturers Association

DISA - Data Interchange Standards Association

ASTM E31 - American Society for Testing Materials Committee for Healthcare Informatics

HL7 - Health Level 7

IEEE - Institute for Electrical and Electronic Engineers

HIBCC - Health Industry Business Communications Council

NCPDP - National Council for Prescription Drug Programs

These eight organizations are among the core bodies directly involved in the creation of standards for health care.

Professional Societies

ADA - American Dental Association

AMA - American Medical Association

ANA - American Nurse Association

AIIM - Association Information and Image Management

AHIMA - American Health Information Management Association

AMIA - American Medical Informatics Association

These professional organizations are accredited by ANSI to develop health care standards.

Trade Associations

HIMA - Health Industry Manufacturers Association

This association influences the process by directly participating in standards organizations or by coordinating its members' participation.

Government Organizations

HCFA - Health Care Financing Administration

FDA - Food and Drug Administration

NLM - National Library of Medicine

AHCPR - Agency for Health Care Policy and Research

VA - Veterans Administation

DoD - Department of Defense

NIST - National Institute of Standards and Technology

These government agencies offer valuable technical and administrative support to the standards creation process and use their influence in the health care market to advance standards adoption.

National Organizations

WEDI - Workgroup for Electronic Data Interchange

CPRI - Computer-based Patient Records Institute

MRI - Medical Records Institute

NCQA - National Committee for Quality Assurance

JCAHO - Joint Commission on Accreditation of Healthcare Organizations

ANSI - American National Standards Institute

ANSI HISB - ANSI Healthcare Informatics Standards Board

These widely recognized organizations assume leading roles in coordinating standards efforts and educating the public.

(Base or Bitway) Standards Organizations

ASC X3 - Accredited Standards Committee for Information Processing Systems

ASC X9 - Accredited Standards Commmittee for Financial Services Industry

These bodies establish standards that cut across various industry sectors.

In an attempt to address the shortcomings of the formal standards process for health care, ANSI convened the Health Informatics Standards Planning Panel (HISPP). It was set up as a temporary planning panel to facilitate harmonization among health care message format standards and to provide a focal point for communications between standards development organizations in the United States and European standards activities, coordinated by the European Committee for Standardization, Technical Committee 251 (CEN TC251). HISPP performed this role well and demonstrated that these functions needed to become part of a permanent organizational entity. As a result, ANSI has decided to replace the temporary HISPP with a permanent Healthcare Informatics Standards Board (HISB).

The ability to realize market potential for interenterprise information systems is dependent on providers having electronic access on demand to any health-related information that the patient has released for view, given the need to know and appropriate security. Although the formal standards-setting process has had some successes (see Figure 4-12), there is still no clearly defined set of reasonably comprehensive standards to guide health care information integration. The task is complex and progress is incremental. The various health care stakeholders need to be aware of efforts under way. These are outlined in Figure 4-13. By supporting these activities and using their market power, major participants can and should accelerate the standards-setting process.


Quantifying Benefits

Given the high costs of implementing integrated information systems, there is a tremendous need to have near-term, demonstrable benefits that justify expenditures. Health care purchasers/users must be able to:

--Distinguish between strategies that result in cost-cutting and those that shift costs or even increase costs.
--Select procedures/guidelines that result in improved outcomes and quality enhancement.
--Determine access to the most appropriate care for their employees/patients.
--Provide services that ensure appropriate use of both labor and material resources.

For example, the Wisconsin Health Information Network (WHIN) recently reported substantial savings in both cost and time. The electronic network connects practitioners, health care delivery organizations, payers, employers, and related health care entities in Wisconsin, providing an efficient alternative to expensive proprietary links. The first two hospitals to join WHIN reported a minimum annual savings of $200,000 (Atkinson 1995).

Each stakeholder has a range of potential benefits that are realizable through integrated information systems. (See Figure 4-14.) Those stakeholders that successfully manage the transition to integrated information systems have the greatest chance for solidifying or expanding their share of the health care market and realizing the various benefits that integration brings. Case Study 4-2 illustrates some of the cost-savings that could be achieved within an enterprise by using integrated systems in the management of behavioral health care.

Case Study 4-2: Benefits of Integrated Information Systems in Behavioral Health Care


Managed behavioral care can provide an excellent perspective on the potential benefits from NII-enabled information management systems within a health care enterprise. Behavioral care has been greatly influenced by the shift toward managed care. The result has been a change in focus from long-term psychotherapy to short term, problem-focused therapy, a radical reduction in utilization patterns, the formation of contacted "wholesale volume" service networks, and a massive build-up of case management overhead because of case-by-case review. Medical costs have been reduced but administrative costs have exploded, absorbing much of the savings. Administrative costs (documentation time, reporting, information management and review personnel) now comprise up to 40 percent of the total dollars spent on behavioral care when delivered through a managed care organization (MCO). As accountability and quality requirements tighten, these costs could rise. The NII is seen as a catalyst that can help health care providers develop Organized Systems of Care (OSC) based on information standards, performance requirements, patient-centered procedures, and outcomes-based decision-making.

By implementing an integrated information system, an OSC for behavioral care can target high administrative costs and offer economic, outcome, access, and service benefits to each stakeholder. Payers and providers would realize reduced costs in data entry, data retrieval, and data storage as economic benefits of an interenterprise integrated information system. Payer, employer, and provider staff would be able to use, transfer, and analyze data more quickly and more accurately, thereby becoming more efficient. Staff productivity would be increased for payers, providers, and employers.

Cost-savings for a hypothetical MCO with mental health benefits for 100,000 covered lives could be significant. At an average premium of $60 per year, the MCO would yield $6 million in gross income. If 17 percent (Gorski 1995) of the behavioral health care dollar is currently being spent to monitor providers, the cost to this hypothetical MCO would be $1.02 million per year. Conservatively estimating the rest of the MCO's administrative costs at 8 percent of total gross, the total MCO administrative cost for the year would be $1.5 million. It is here that an integrated information system could produce substantial savings. If the information system resulted in a 25 percent net decrease in general administrative costs from 8 percent to percent of gross, the MCO would realize a savings of $120,000 per year. Even greater savings could be realized if payers were able to use the new system to shift from case-by-case review to a preferred-provider system that managed the quality of care through aggregated information reporting. If this shift brought a 50 percent net decrease in provider monitoring costs, the MCO would save $510,000 annually.

There is considerable room for cost reductions for practitioners, too. On average, practitioners are spending 15 percent to 22 percent (Gorsky 1995) of the behavioral health care dollar on administration of the information links required by managed care. At an average penetration of 4 percent for outpatient cases, the MCO is paying practitioners $1.44 million for about 4,000 cases a year (at an average of six sessions/case, $60/session) The practitioner's costs for MCO reporting and review is approximately 20 percent of gross income, or $298,000. If the information system improves efficiency and reduces those costs by 25 percent (from 20 percent of gross income to 15 percent), practitioners would realize an annual savings of $72,000 per year.




Figure 4-14: Benefits to be Derived from Systems Integration





Practitioner and Health Care Delivery Organization


More effective/speedier medical management, disability/workers compensation management, and claims processing.

More fully integrated administrative benefits and eligibility systems, providing single platforms for enterprise claims processing, billing, and enrollment.

Reduced administrative costs with electronic communication of standardized health care transactions.

Reduced repeat testing.

Reduced costs associated with treating diseases.

Diminished administration and staffing levels to provide routine and often redundant information.

Reduced number of rejected, refiled claims.

Reduced number of claims paid for ineligible enrollees.

Reduced costs of manual administration.

Enhanced ability to analyze outcomes.

Reduced up-front expenses through current copays and deductibles.

Enhanced compliance with prescribed treatment through improved information and take-away instructions.

Faster transfer of paperless communications between providers and payers at less cost.

Enhanced monitoring of patient compliance with pharmacy benefits.

Faster payments.

More time with patient, less time on administration.

Improved task coordination (e.g., note in chart generates prescription.)


Improved medical management.

Increased ability to monitor provider compliance with practice guidelines.

More precise determination of and influence over physician's practices and prescribing patterns.

Enhanced data collection for basic outcomes research.

Enhanced compliance with plan and coverage rules.

Expedited care planning.

Faster access to ancillary services and specialists.

Better tracking of patients through continuum of care.

More precise determination of levels of care required and associated costs.

Better analysis of outcomes of data.

More accurate data collection for disease management contracting.

More effective intervention targeting.

Better comparability of quality across health plans.

Increased productivity: back to work sooner.

Faster test results leading to ease of mind, quick preparation for the next steps, and satisfaction with care given.

Higher comfort level with new providers as a result of provider access to more complete medical history.

Reduced risk from duplicative testing, erroneous results, or incompatible treatments.

Enhanced access to clinical trials and faster and more comprehensive trial results.

Improved health status.

Enhanced possibility of linking individual record keeping to practitioner information.

Faster physician response time and more in-depth record keeping.

Better monitoring of patient care, leading to better patient management.

Improved integration of patient information from all sources.

Increased ability to identify problems with potential to impact outcomes.

Increased speed of care delivery when referrals are involved.

Enhanced data collection for basic outcomes research.

More concise nomenclature and unique identifiers for comprehensive outcomes studies to determine best treatment methods.

Enhanced systemization of record keeping.


Reduced costs of total care.

Decreased worker absenteeism.

Increased employee productivity.

Quicker access to services with timely and accurate plan and administrative data.

Increased access times and locations.

Enhanced use and satisfaction.

More focus on care delivery and wellness.

Customer Service

More assistance in designing benefits.

Increased employee satisfaction with current and accurate plan data, copays, deductibles, and out-of-pocket expenses.

Enhanced ability to verify coverage information.

Reduced administrative hassles, erroneous claims, refiling, and multiple contacts to increase employee satisfaction.

Greater patient satisfaction with the health system.

Less disruptive participation in health care system.

Ability to provide online information taking full account of patients' clinical and coverage status.

Improved management of cost and quality of service and care.

Improved patient/membership satisfaction.

Less administrative hassle, leaving more time to practice medicine.

Source: Council on Competitiveness.


The impediments to creating viable, commercial, integrated delivery systems that rely on computerized information revolve around the fragmentation problem: many pieces of information, in many formats, on many platforms, in many stakeholder environments, to many geographic locations. These barriers have been subdivided into organizational, legal and standards-related, and implementation issues.


Organizational Barriers

Most of the issues blocking information system implementation revolve around the organizational changes associated with the move toward integrated systems.


1. Health care practitioner resistance. Health care practitioners believe they are being buffeted by the changes in health care, including, but certainly not limited to, new information demands and increased reliance on computers. They report feelings of denial, anger, bitterness, helplessness, and disillusionment as they are forced to accommodate new health plan procedures that require additional paperwork and diminution of their roles as autonomous decision-makers in the care process. With the advent of managed care, whereby health care practitioners may believe that cost takes precedence in determining a treatment protocol, and integrated information systems promote the distribution of standardized care guidelines, the physician is no longer an independent decision-maker.


2. Failure of management to purchase information systems and adopt integration strategies. Senior managers in health care delivery organizations have only relatively recently recognized the kinds of health care enterprises that will be required to succeed in an accountable managed care world. Therefore, most have not made the appropriate investments in information systems to prepare for this environment. Integrated information systems must compete for the needed investment funds with alternatives such as salaries, new medical equipment, and facilities. Large capital investments may be required and the rate of return may be perceived as more uncertain than that on other investment alternatives. Management failure to insist on a coherent information architecture and implementation plan may also cause delays in purchasing information systems.


3. Stakeholder inability to adapt to change. Without adequate user and management training, preparation, and realistic system expectations, implementation efforts are doomed to failure. Issues that can stall the transition to integrated information systems include:


--Failure to understand the range of services associated with continuum-of-care delivery systems.
--Inability to overcome individual stakeholder resistance to allowing information to flow among all entities within the health care system.
--Inability to convince major revenue generators such as hospitals to accept a role as team players in an integrated system.
--Resistance to changes in management as a result of integration.


Legal and Standards-Related Barriers


These barriers focus on privacy, confidentiality, and security, as well as the lack of standards for information systems for health care.

1. Lack of guidelines for protection of personal medical information. So long as public policy regarding the confidentiality and privacy of health care information networks and services is unclear and incomplete, the commercial market for integrated systems will be inhibited in its growth. Current medical ethics, the Federal Privacy Act of 1974, and various state laws do not provide the clear, comprehensive coverage required to enable practitioners, health care delivery organizations, patients, and payers to feel comfortable with "how" to integrate. Until a framework is created that enables individuals and patients to feel confident that sharing their health information enhances their well-being rather than jeopardizes it, there will be reluctance and even opposition to the development of health information infrastructures to increase health care productivity. Technical capabilities to ensure security are not the issue. The real barrier is concern about individual and organizational behavior that might result in the improper use of the information. These threats include erroneous administrative actions and access by unauthorized parties.

2. Difficulty and cost of developing, enforcing, and conforming to standards. The lack of comprehensive and broadly accepted standards within the health care industry for the definition, collection, communication, and storage of administrative and clinical data is retarding widespread systems integration. Another significant barrier is the inability of any group of participants in the industry to mandate standards for all stakeholders on even the simplest of transactions. Voluntary bodies have strongly promoted the use of standards in primarily administrative transactions and have even arranged for their successful demonstration, yet they are unable to enforce their use.

The cost of conforming is yet another issue. Health care stakeholders and vendors must bear the initial cost of modifying their information systems to conform to standards. But these costs may be prohibitive for the smallest vendors and providers. Large groups, too, are reluctant to make the investment without more assurance that others will do likewise so that benefits can be realized across an enterprise.

Suppliers of practice management systems and other subsystems must modify their systems to conform to standards that support basic transactions. However, these suppliers profit from charging for the development of multiple system interfaces, so they have a disincentive to change. And because they receive payment from existing clearinghouses for the transactions they pass on, these suppliers also discourage providers from linking their practices to new health information networks.


Implementation Barriers

These issues center on the problems encountered by stakeholders as they transition to integrated information systems.

1. Confusion in the purchasing environment. The less-than-satisfied attitudes of information system users with vendor and system performance create a major barrier to information system purchasing. Health information networks are emerging that do integrate stakeholders in the marketplace, but vendors face a difficult sales cycle. A few stakeholders still favor proprietary systems, believing this adds to their competitiveness, when in fact such systems may cost more over the long-term and/or fail to tackle the fundamental cost/quality issues. Rapid changes in technology that lead to obsolete systems cause hesitancy. And an orientation toward intra organizational rather than enterprise wide purchasing contributes to the chaos and inertia in the current market.

2. Uncertainty regarding the competitive advantage to sharing information. In the past, information has been considered proprietary and has been used as a tool to obtain competitive advantages, such as greater market share. With integrated systems, a key question emerges: should access to information determine competitive advantage, or should information be made available among stakeholders so that market advantage is acquired through the creative use of information and not access to it? Further, if information is to be shared, should a common electronic linkage or network capability be available to stakeholders and funded based on the value they receive by using it to exchange whatever health care information they desire? How individual stakeholders will respond to these issues is unclear, particularly since the health care market remains highly competitive and in the throes of consolidation.

3. Changes in patterns of fraud and abuse. Systems designed to detect fraud in paper transactions may not be suitable for use in computerized environments. Automation offers an opportunity to improve control, quality, and security brought about through the understanding and redesign of underlying business processes. However, automation alone, without process redesign, offers none of the above improvements and may often result in less control. Automation changes submission patterns and eventually removes adjudicators who are sometimes better able to view claims "in context" and therefore detect abnormalities. Also, automation can inadvertently destroy the "chain of evidence" if processes are not redesigned appropriately.

4. Competing information demands. Most practitioners are being asked to comply with differing demands from different employers, health plans, and affiliated health care delivery organizations. The cost in dollars and time to conform to differing requirements is slowing the integration process.



The following recommendations advocate the use of market power rather than regulation to accelerate the adoption of de facto standards. They also address the need for guidelines to protect sensitive medical information.


1. The largest public and private sector purchasers of health care should channel their market influence to advance the development and adoption of standards in electronic health care transactions, data content, and format. They also should require that both provider and payer systems be modified to conform to these standards. Traditional standards development processes, while useful and constructive, are often slow given their consensus nature. The largest purchasers, however, can drive the process by clearly articulating their requirements as customers for data, both financial and clinical. This could include the definitions of the data they need, the format in which they expect to receive it, the way in which it should be submitted (e.g., electronically), and the frequency with which it should be submitted.

The government, including federal, state, and local entities, is the single largest purchaser of health care in the country. It manages health care purchasing and payment for all of its civilian employees, the DOD (including active duty and retired military personnel and dependents), Medicare, Medicaid, the Department of Veterans Affairs, correctional facilities, and welfare. The sheer volume of business that HCFA alone generates in its Medicare program has the potential to propel the pace of change. If all of the government health care purchasing entities could be aligned to require common standards and so exercise their purchasing power by clearly articulating their customer needs for information, the impact would be enormous.

Dollars drive the market. If both the large public and private sector purchasers work together in a neutral forum to build on progress already underway in the private sector standards-setting bodies to develop common requirements, the economic clout they wield through their tremendous purchasing power will compel the market to respond faster and more efficiently than any regulatory process could achieve. Such collaboration would force the emergence of de facto standards. These requirements should include a provision that providers (both practitioners and health care delivery organizations) be required to electronically communicate to health plans in order to increase efficiency and cost-effectiveness. Providers will influence vendors to provide the necessary information systems and services to satisfy these needs. And the public-private sector collaboration would ensure the development of a single set of standards across the market.

2. State laws should continue to support purchasing alliances so that small purchasers of health care can join together, help drive the integration process, and enhance their economic leverage. While the large public and private sector purchasers can use their purchasing power to influence payers and providers, smaller purchasers do not have that kind of economic muscle. Some states have addressed this by enacting laws that permit single buyers and small organizations to join together to increase their economic leverage when procuring health care. States that do not permit these alliances should enact laws that do, so that smaller users can combine their purchasing capacity and help drive integration. Third-party insurers, who often serve these smaller communities, can assist in the integration process by conforming to the requirements set by the large purchasers.

3. The largest private sector purchasers should require all providers to comply with ANSI standards (ASC X12) for claims and encounter information and eligibility requirements. The large private sector purchasers, through the HEDIS process, have already defined some of the health-related information that health plans should be collecting and submitting to them as well as the submission format and frequency. The swiftness with which the market responds to direction from its major customers was demonstrated when HEDIS 2.0 was introduced. Shortly thereafter, certain software vendors modified their products to conform and began advertising they had done so to gaining a competitive advantage over those who were slower to adapt. The large purchasers can continue to influence health plans/payers and providers by requiring electronic data submission in ASC X12 format. The pattern of influence will also trickle down to smaller private sector purchasers of health care who will more than likely follow the lead of the larger institutions. A series of implementation goals is suggested in Figure 4-15. Widespread adoption of a timetable of this sort would result in concerted action by the many stakeholders involved to ensure they are not "left behind."

Figure 4-15: Proposed Timetable for Implementing ANSI Standards

Target Date


January 1, 1997

Practice Management and Hospital Information Systems vendors should make system modifications to support transactions in ANSI standard format for their provider customers.

July 1, 1997

All health care providers should be able to send electronic claims and encounters to any health plan/payer in ANSI standard format. All health plans/payers should be able to receive electronic claims and encounter information in ANSI standard format.

January 1, 1998

All health care providers should be able to send electronic requests for eligibility in ANSI standard format. All health plans/payers should be able to receive and respond to electronic requests for eligibility in ANSI standard format.

Source: Council on Competitiveness

Confidentiality and Security

1. Congress should enact legislation that protects personally identifiable health information of any kind from unauthorized access and disclosure and that allows for appropriate individual patient access and correction. Ownership of patient information has been the subject of much debate. However, governing the stewardship of and access to medical information is the more urgent and practical issue. The legislation should identify the rules by which information can be shared, should permit electronic authentication of information, and should include sanctions/penalties for violations. Sanctions/penalties should be based on the extent to which the rules are not observed and/or whether the information is misused for commercial purposes for profit.

2. Given that proprietary and open networks are becoming more integrated, all electronically transmitted, personally identifiable information should be adequately encrypted. The industry must recognize the sensitivity of the data that will be accessible over the wide range of networks being built and work rapidly toward encryption of personal medical data at its source.


The Cost-Benefit of Information Systems

1. Health care organizations should work to establish clearer and more widely accepted cost-benefit models for major integrated systems investments. Leading health care participants recognize the costliness of this next phase of information system investments. However, they will need stronger justification than is often available in order to initiate these investments and successfully compete for funds in a period of increasing economic pressure.













cclogoSMALL.gif (1930 bytes)

competetextgrey.gif (3110 bytes)


Chapter 5 - Health Care Research and Education

Highway to Health: Transforming U.S. Health Care in the Information Age


Historically, research and education have been the cornerstones of American health care. The quality and economic efficiency of medical diagnosis and therapy have been, and continue to be, driven by a continual process of analysis, integration, and dissemination of the results of basic science and clinical research. The projected 1996 budget for the National Institutes of Health (NIH), which conducts the vast majority of federal health research, is more than $11 billion, with recent special focus placed on HIV/AIDS, women's health (including breast cancer), minority health, tuberculosis, and high-performance computing (CRS 1995b). The private sector, which spent an estimated $15 billion in 1993 on research, will spend even more (Woolley 1994).

Such expenditures enjoy widespread public support. In a recent poll, 66 percent of respondents favored medical scientific research over defense scientific research, while only 4 percent favored defense over medical (Research America 1994). Another poll suggests that Americans believe that 10 percent of the total health care budget should be used to support medical research compared with the actual 3 percent (Woolley 1994). Yet, despite the positive public attitude, funding for medical research is expected to decline in the coming years as a result of changes in health care delivery systems. This will affect research, as well as the education of physicians and other health care practitioners. To make up for the reduced revenues, many medical institutions are exploring strategies that rely on NII-based applications.

Today, clinical health care research is based primarily in academic institutions and pharmaceutical laboratories. University medical departments and affiliated teaching hospitals are responsible for the major research and education activities and are credited for the role the U.S. enjoys as the world leader in medicine. With academic and industry backing, a few researchers and educators have been early adopters of computer and information technologies (including databases) as research and teaching tools, resulting in the creation of medical informatics curricula. These researchers and educators also had early access to the resources of the Internet, as network use migrated from computer science departments to medical departments.

As the NII advances toward an integrated system of networked computers and applications that extend to the desktop, health care research and education can be more integrated into the actual delivery of health care as well as the ongoing education of practitioners. Recent innovations in computation and information technologies, in networked communications and databases, and in computer-based methods and tools for collecting, analyzing, and visualizing data originating from multiple sites offer unprecedented opportunities for enhancing the quality of research and the efficiency with which results can be generated, analyzed, and integrated into health care education and delivery. Networked health applications can provide the foundation for a stunning variety of advances in health care delivery: more cost-effective monitoring and follow-up of patients beyond health care centers; dynamic, optimal targeting of specific sectors of the population for special education, screening, and, where necessary, early treatment; and better feedback loops for connecting providers, policymakers, and patients with late-breaking research and discussions about clinical decision-making policy. These same tools and techniques are also altering the way traditional medical education and continuing medical education (CME) are being designed and delivered.

This chapter examines the market opportunities for networked computing in medical research and education. Examples of innovative applications of the NII usually represent standalone research projects or small groups focusing on only a few topics without a commercial emphasis, and it is difficult to size these markets based on these limited examples. However, they are indicative of the variety of opportunities that are emerging, primarily from the 126 U.S. academic institutions that, alone and in collaboration with industry, are carrying out the majority of information-based health research. This chapter focuses on these opportunities and points out the key barriers and challenges that are inhibiting widespread adoption of NII-based tools to support these activities. It concludes with recommendations that, if adopted, will promote the use of the NII to further medical research and education.



The use of networking in medical research has, until recently, been limited to the obvious: electronic file transfer and E-mail. The sudden, explosive emergence of the Web, along with software tools that collectively link information residing on it and permit users to download complex graphics, audio clips, and video clips, may change this picture dramatically. These NII-based tools may make it possible for multiple users at different institutions to access medical decision-making software and collaborate on medical research and education much more closely than was previously possible. The Internet, along with emerging higher bandwidth communications hardware, has the potential for totally changing the way computers are used in medical practice and research. However, these new tools are only now being developed and implemented, and the market is just beginning to take shape. An update to this chapter will be imperative in three to four years.



Although the market opportunity for NII-based tools to support medical research is only now emerging, the potential is becoming clearer in three areas: medical outcomes research collaboration, real-time decision support, and communications.


Enhancing the Quality and Relevance of Outcomes Research

Outcomes research provides essential feedback to health care providers and the authors of health care guidelines and policies about how well a given treatment protocol is working. In the case of a randomized, carefully controlled, prospective clinical trial, such information is used to determine which of two (or more) treatment methods provides better health outcomes. The NII is spawning networking and collaboratory software tools that will facilitate cooperation among researchers. These tools promise to reduce the costs of clinical research by significantly enhancing the medical community's ability to collect data at the point of care and distribute outcomes information in a timely manner. Collaboratories, a virtual laboratory concept emanating from early adopters of networked computing, typically consist of users working in a specialty research area who share the same scientific language and customs and are familiar with distributed work environments and computing basics (Gorry et al. 1995). While some "stovepipe" systems that connect just a few users exist and provide some assessment capability, they are primitive, not integrated, relatively unavailable, and do not provide good access, compared with the potential benefits of these emerging collaboratory efforts.

Feedback systems connecting all stakeholders in an individual's health care process can make data collection, analysis, and distribution a by-product of the services delivery process. With appropriate networking and software tools, a continuous stream of feedback from in-patient through ambulatory care can be analyzed automatically, creating an objective record. This means that data concerning the health status of the individual and the effectiveness of the health interventions can be fed into the patient's record during each health care "event" or visit. In essence, every patient could then become part of a clinical trial as software tools automatically integrated their information with that of other patients with similar conditions at other locations.

This data can then be aggregated so that an individual's health profile and treatment outcomes can be tracked over time. When entered into the database of a population of similar patients, large-scale outcomes-based practice guidelines can be generated and distributed back to providers and patients in the form of decision-support information. Such information is invaluable for helping providers and patients to construct and maintain effective plans of care.

Large, even national databases could be used for national benchmarking and feedback and to provide decision-support for public health functions such as anticipating and tracking epidemics. By including policymakers in this feedback loop, a managed care organization, a hospital or a public health clinic could more efficiently manage its resources according to the collective knowledge concerning the changing health status of a population. For managed care organizations, having such reliable data is essential for determining their costs and predicting realistic capitation rates. Such data could also enhance medical education processes like chart reviews.

The need for such systems is documented in the HIMSS/HP (1995) survey. Respondents indicated that the need for clinical and comparative outcomes data is driving increased computerization in health care. In 1989, the federal government, recognizing the potential value of outcomes research and clinical practice guidelines, created the Agency for Health Care Policy and Research (AHCPR) to promote their development. Congress specifically directed AHCPR to both conduct and support health services research on three topics--medical effectiveness, patient outcomes, and appropriate care--while establishing evidence concerning medically effective interventions (CRS 1994b). AHCPR is now one of 7 federal agencies, 23 specialty societies, and a multitude of other organizations such as the Harvard Community Health Plan (HCHP) that are developing clinical practice guidelines. HCHP, for example, has completed or has under development guidelines in internal medicine and surgery (18), pediatrics/child mental health (10), OB/GYN (8), and adult mental health (11). Case Study 5-1 illustrates how this process might be applied.


Case Study 5-1: Potential Clinical Benefits of Research Data Integration

A 52-year-old white male executive comes to his managed care organization for a yearly physical examination. He complains to his physician of having to get out of bed several times a night to urinate and claims this is a new problem for him. On physical examination the only abnormality noted is a slightly elevated blood pressure. The information is entered into the physician's personal computer via speech recognition software. The physician's personal computer is networked to a large knowledge base of cases. Inferential tools, taking advantage of access to this data, remind the physician that middle-aged males can have increased urination due to urinary outlet obstruction even with the finding of a normal-sized prostate on physical examination.

The physician, noting that the patient seems a bit "run down," decides to check blood sugar. The physician's personal compiler suggests getting a complete blood count as well because there is a rare condition in stressed white male executives that can cause a substantial rise in red blood counts--a condition that should be noted and possibly treated if present. The blood sugar is, indeed, found to be elevated, but the red blood count is normal. Inferential tools in the clinician's system make recommendations on testing and diagnosis, guided by cost-benefit considerations, balancing the cost of testing with the benefits of early treatment of possible disorders. The diagnosis of adult on set diabetes and coexisting essential hypertension is confirmed.

The physician is not sure about the best way to manage a patient with both of these conditions. She queries the compiler about experience in the medical community with similar cases. The physician's research information analysis software informs her that there is a protocol for treating patients with both conditions. A summary of key protocol characteristics is immediately provided. According to an outcomes study, which is updated weekly, the protocol has been 94 percent successful with 2017 patients, and more than 95 percent successful with 146 patients who fit the current patient's profile. The physician's personal compiler also reminds her that such patients should have a baseline retinal examination by an ophthalmologist, and that all 52-year-old males should be screened for prostate cancer with a PSA blood test.

The physician confers with the patient and provides him with a customized assessment of his health risks and health care options. With input from the patient about his preferences, they agree on a plan of action. Once the plan is recorded, organizational software communicates with compilers in other departments. Appointments are made automatically with the ophthalmologist. In addition, a blood draw for PSA and blood sugar tests are scheduled.

The agreed-upon plan is automatically sent to the digital outcomes library to be tracked for this individual patient and, stripped of individual identification, aggregated with other health care plan data. These data are linked to prostate cancer treatment databases used to develop guidelines for effective prevention of prostate cancer. As evidence of new, more effective intervention accumulates, alerts are sent to providers and researchers, who now have the opportunity and the information to make policy changes concerning the health care interventions they will recommend.


Case Study 5-1 illustrates potential opportunities for integrating data collection, analyses, and automated decision-support into the service delivery process. First, computers gather patient data from many sources and use this information to make recommendations about medical testing and therapy. The physician's personal computer also reminds the physician of rare conditions (in this case, stress polycythemia). After making the double diagnosis of diabetes and high blood pressure, the physician could have referred this patient to a specialist for treatment, but the computer reminded her of an in-house research protocol for which the patient was eligible and, more important, gave her information on how patients within her organization were responding with this protocol. She could then discuss this information with the patient before they made a decision to institute therapy. Finally, the computer reminded the physician about two steps needed to complete the exam and scheduled both of them, combining the PSA test, which it knew was not an emergency, with the repeat blood sugar analysis, thereby saving the patient an extra blood drawing. Most important, perhaps, all of this information was obtained and processed automatically in a way that did not require any extra time from the physician.

The importance of such medical effectiveness data has been noted by the public sector. Research on medical effectiveness issues recently funded by AHCPR through its Medical Treatment Effectiveness Program includes studies on the medical costs associated with AIDS, practice styles and outcomes, pediatric hospitalization, and elderly outpatients (CRS 1994b). Such data are not only being used to determine treatment outcomes; they are also being compiled by insurers to evaluate physicians on all aspects of delivering care and to assess health care delivery organizations' performance on complex medical procedures such as transplants.

By combining their data, insurers can undertake practice profiling; that is, they can profile individual physicians, health care delivery organizations, and health care networks to determine which providers have higher than normal complication rates for the same procedures. Those who show unusually high numbers of problems can be given additional education or can be removed from the network (Thornburg 1994). Profiling can also help identify best practice procedures and those individual physicians, health care delivery organizations, and health networks that adhere to them. One drawback of using practice profiling is the potential for excluding from health care plans physicians whose practices vary from the norm for sound reasons. Another is the potential for using such data to draw incorrect conclusions on how to best deliver patient care (CRS 1994b).


Enhancing Research Collaboration

Most health care researchers still conduct their work in relative isolation and depend on the traditional print medium of journals and annual conference proceedings to distribute their findings (Gorry et al. 1995). Thus the fast-paced and accelerating accumulation of medical knowledge has been largely relegated to distribution through the bottleneck of printed publications. A large knowledge base is stockpiled in libraries in an abundance that challenges even the most ardent practitioner to keep up with the latest medical advances. As a result, valuable medical information embedded in these large databases may be quite insulated from the delivery of health care, rather than applied during routine health care decision-making. Networked medical software could significantly enhance the ability of medical researchers and clinicians to gain access to this important data.

The communication and database possibilities associated with the evolving NII hold unprecedented opportunities for:

--Enhancing collaboration among researchers.

--Enhancing the quality of such efforts as multi-site clinical trials.

--Enhancing clinical trials via the use of such tools as electronic communication to minimize patient dropout.

--Maximizing long-term follow-up.

--Making the latest knowledge available to physicians and patients.

--Determining the future costs, both in dollars and resources, for health care.

Networked decision-making tools can also be used to design or implement new clinical trials in the context of ongoing trials, as well as to access research information for guiding treatment decisions on specific patient cases.

NIH has been a long-term leader in advancing medical informatics research. The National Library of Medicine (NLM), for example, has sponsored the Advanced Integrated Medical Systems Program and has funded several research efforts designed to help biomedical researchers work more efficiently, using advanced information and networking technologies. The National Cancer Institute conducts research and development using on-line cancer databases of cancer trials. NIH and the Department of Energy (DOE) each sponsor research in mapping the human genome, using the Internet to connect distributed researchers.

Recent strides have been made in the use of the Web protocols for multimedia data in basic science research, clinical research, and health care delivery. "Collaboratory" (NRC 1993) sites have been constructed successfully in such areas as biochemistry and genetics. And valuable resources have been collected for specialty areas of medicine. For example, in 1994, Dr. Peter Tarczy-Hornoch, a neonatology fellow at the University of Washington, constructed an international, shared neonatology Web resource that allows physicians throughout the world to share information on the treatment and outcome of difficult or rare cases. The Web site includes pointers to recent studies and reports from neonatology researchers, as well as information about evolving pharmaceuticals. (


Enhancing Networked Communications


Physicians are becoming more interested in NII computing and communications tools such as the Internet. The 1994 HIMSS/HP survey showed that nearly half of all hospitals were using Internet resources. In 70 percent of those hospitals, clinicians were using the Internet to access research databases; in more than 20 percent, to engage in two-way medical consultations; and in 81 percent, to communicate with other facilities by E-mail. In more than 30 percent, clinicians engaged in two-way consumer/provider exchanges via the Internet. In a study at Boston's Beth Israel Hospital, where E-mail has been in use for 20 years, researchers found that during a one-week period 1,247 people sent 7,482 messages to 1,302 individual recipients. An electronically administered questionnaire found that 66 percent of respondents used E-mail either daily or weekly, with 58 percent using it for issues relating to patient care (Sands et al. 1994).

But another national survey paints a different picture. It shows that overall, physicians are slow in adopting personal computers in their practices. BMI, American Medical Information, Inc. of Arlington Heights, Ill., a division of American Business Information, Inc. of Omaha, Nebr., surveyed the country's physicians on a variety of topics in 1995. Of the 550,000 respondents, only 6.9 percent use their personal computer to access databases, 7.7 percent to bill patients, and 5.0 percent to conduct research. The most popular use is word processing (15.0 percent). (See Figure 5-1.)


Figure 5-1: Physician Use of Computers





Respondents with Computers






Word Processing



Insurance Billing



Patient Billing



Database Access






Continuing Medical Education



Clinical Records



Claim Authorization



Data on Drug Interaction






Issuing Prescriptions




Source: BMI Research

Several projects point to the opportunity to enhance communications among health care stakeholders by linking them together electronically. In 1994, the Regional Medical Library at the University of Washington, Seattle began an NLM-sponsored project to interconnect several community hospitals in Idaho, Montana, Washington, Oregon, and Alaska. Discussion groups have been set up that include health care workers and administrators. Librarians are available on the network to field questions by E-mail. The system also makes available daily on-line consultations with pharmacists to answer questions about medications and drug interactions. The NLM-sponsored Pathfinder project on decision-support for surgical pathology relied heavily on the use of the Internet to connect researchers and expert pathologists at multiple sites. Physicians Online of Tarrytown, N.Y., created an on-line service that provides access to medical information like Medline and Cancerlit, as well as special forums, bulletin boards, and, as of October 1995, E-mail services. Physicians Online can provide basic services free for physicians and other health care workers because these services are subsidized by pharmaceutical companies' advertisements.

The growing ubiquity of electronic store-and-for-ward technology for sending E-mail, data, and graphics between physicians and their patients opens up opportunities for both groups to access one another more efficiently. Doctors who use E-mail have found it superior to the telephone for such activities as making referrals, handling questions about minor symptoms, and communicating test results. E-mail, with its store-and-forward nature, eliminates "phone tag." (Rubin 1995). Surveys at Stanford University and at the University of Kentucky on the value of E-mail to patients demonstrated that, overall, patients were enthusiastic about using it.


Enhancing Decision-Support Mechanisms

As the feedback loop between outcomes studies and clinical decision-making tightens, the need for decision-support tools to assist in the caregiving process will become increasingly important. Decision-making software spans several different classes of functionality. Some of the most sophisticated clinical decision-making software helps assess a patient's current or future condition, given their clinical background and symptoms. These systems diagnose the present condition or forecast the likelihood of different future outcomes based on alternative treatments. Some of these systems assess a patient's state by automatically building a list of possible diseases ranked by likelihood and then recommending cost-effective tests to narrow the list and reduce uncertainty. Methods have been developed that help a physician to select the best therapy under uncertainty.

Although automated decision-making tools typically address diagnostic dilemmas and difficult therapeutic decisions, they can also be designed to provide valuable alerts and reminders to physicians. Such memory-jogging software can alert physicians to possible diseases, relevant tests, appropriate drugs, and/or drug-drug interactions. One can imagine a computer-based reminder system prompting both patient and caregiver when either is straying from a treatment plan. Such systems would benefit day-to-day care and greatly benefit investigators who perform clinical trials. As an example, 7 percent to 10 percent of all hospitalizations for people under the age of 65 occur because a drug was prescribed (and taken) that exacerbates an underlying medical condition; for patients 65 and older, the rate is nearly 15 percent. Almost one quarter of the elderly living in the community are prescribed inappropriate medications (Wilcox et al. 1994).

Decision-support tools that are networked provide critical links between a current patient's condition and previous studies on related topics. But beyond simply sharing and distributing knowledge, networked computing can be used effectively to apply advances in computer-based data analysis, information retrieval, and automated inference methods. These computational techniques can provide unprecedented opportunities for integrating data gathered at multiple sites and creating understandable summaries of results. Networked databases of clinical information can provide the inputs needed to generate effective computer-based reasoning methods, based on principles of probability and utility, that practitioners can use in making recommendations on testing and therapy decisions. For example, evolving methods in medical informatics can take advantage of multiple sources of data on a computer network to perform ideal cost-benefit analysis on the next best tests to order or therapies to employ.

Significant advances in information-retrieval methods are allowing researchers to conduct centralized analysis of large amounts of pooled data. These include similarity-based, natural-language-processing, and Bayesian methods for information retrieval. Such methods allow physicians to efficiently identify patients and clusters of patients from very large data sets throughout the world that are most similar to a current patient and the clinical decision-making dilemma. The fast-paced developments in the application of easy-to-use information retrieval methods make standardizing and constructing patient databases ever more valuable and usable. There are, of course, important questions about who sets standards and who maintains the fidelity of these databases, that must be resolved for these techniques to flourish.

Decision-making systems can be limited by poor quality in the knowledge base they use in making inferences. Networked databases and automated learning and summarization techniques can enhance the quality and depth of experience represented in these knowledge bases. Also, the networking made possible by the NII permits the continuous feed of data to update databases to include changing information about diagnosis, testing, and therapy. Even such relatively unprocessed information as the prevalence of sets of disorders can be valuable if updated frequently to ensure that decision-making software performs as optimally as possible.

Computational constraints, imposed by the need to solve large, difficult problems in real-time or short-time horizons, can also limit the effectiveness of decision-support tools. Networking can bring more resources to bear on such problems. For example, identifying an appropriate therapy may require a sophisticated model of the patient involving one or more organ systems. Computations involving only approximate analysis and abstraction of detail may not require the most advanced computers available. However, detailed analyses such as exhaustive searches through multiple therapy options may require the computational power of supercomputers or the parallel efforts of many interconnected computers. This kind of computing power is rarely available at a single clinical facility. The NII, however, permits the linking of computational resources across geographies so that a clinical facility can tackle these kinds of problems.

Such shared supercomputing resources for clinical decision support are generally not available. Therefore, systems might use complex algorithms to predict drug blood levels or radiation dose distributions. These methods are used at a few institutions but still often require, for optimal therapy analysis, more computer power than can be provided at each desktop; i.e., they require networking to a more sophisticated computing system.

Networked computer-based software can also be useful in helping patients and practitioners to explore and grapple with difficult trade-offs. Often these concern the trade-off between a patient's quality of life and expected length of life. Such personal preferences about alternate therapy outcomes can be critical in defining the "best" therapy for a patient. Wennberg (1993) has found that informed patients often consider the quality of life to be more important than its length and can therefore be more prudent users of health care services. Yet medicine has often overlooked the subtleties of a patient's preferences when considering the optimum, most cost-effective treatment. Decision-support tools can help redress this imbalance by helping patients to make more informed decisions and physicians to advise their patients more wisely.

Another type of decision-support software that could both improve medical care and reduce costs is one that could vigilantly search for common and important errors in diagnosis and therapy. Intelligent error detection methods presuppose access to a computer-based record for each patient, and would have to be highly integrated into the clinical environment and possibly the patient's home as well. Networked tracking of common errors, including transcription errors, could be used to learn patterns of error from a large body of clinical interactions.

Decision-support tools can provide broader assistance by detecting costly testing and treatment strategies and redirecting physicians to more cost-effective approaches. In order to gather useful information about errors as well as the cost-effectiveness and clinical quality of actions, treatment protocols must be rigidly followed. However, given the limitations of current medical practice, strict adherence to protocols is difficult to enforce or even monitor. Networked decision-support systems could change this picture substantially by providing a continuous stream of feedback data on how the patient is performing and feeling. In addition, systems that appraise health risks and monitor health care performance can look for missing information, identify clinical protocols that match the patient's, and provide computer agents (other distributed decision-making tools) to conduct such chores as locating the most cost-effective MR/CT scan in a neighborhood or preparing a cost-benefit analysis on proposed tests or treatments. Once these types of systems are in place to facilitate research, they can be rapidly extrapolated to the whole realm of health delivery system providers.



With the rapidly expanding medical information knowledge base, health care practitioners must perfect their abilities to locate as well as use new information. Many medical institutions approach learning as a lifelong apprenticeship, teaching students how to learn, as well as how to use the tools available to find the information they need (Montague 1994). Networked health care information systems can be harnessed to offer unique, interactive learning experiences that have been difficult to create with traditional forms of media. This is possible not only because of their technological characteristics, but also because of the pedagogy upon which they are being built. Contemporary theories suggest that effective learning is personalized, participatory, collaborative, interdisciplinary, situated in real world contexts, and focused on the active construction of information rather than passive reception (Dede 1992). This kind of learning environment is particularly useful in health care.

Academic medical institutions, the traditional health care education providers, have an additional "bottom-line" incentive for considering these tools. These institutions have traditionally supplemented the funds available to support their medical educational activities from the revenue of their affiliated teaching hospitals. As a result, health care at academic medical centers and teaching hospitals costs, on average, between 28 percent and 41 percent more than care delivered by community hospitals, according to the federal Prospective Payment Assessment Commission. These hospitals, though, are under severe pressure to cut costs in response to capitation and competition with managed care organizations and can no longer support additional expenditures to supplement medical education. In this resource-constrained environment, medical education institutions are investigating alternative ways to fund their activities.



A few medical education institutions are beginning to explore the use of NII tools and technologies to support their activities, hinting that a market may emerge in this area. For example, Johns Hopkins Medical Institutions (JHMI), recognized as one of the country's leading academic medical centers (Hospital rankings 1995), provides a wide range of specialized care and is also a world leader in medical research and graduate medical education. Like other providers, JHMI is investigating the use of information management tools. Its information network (InfoNet) provides news and information electronically for patients, practitioners, educators, and students. Students at three medical schools in West Virginia are using the Internet, one of the fastest growing NII-related tools, to gain access to information while doing rotations at rural locations (Bergman 1994). They also use E-mail to communicate with faculty, to receive newsletters and clinical alerts, and to access treatment protocols. At Case Western Reserve University, entering medical students are provided with an E-mail account and a portable computer, which allows them access to the administration, faculty and other students. Educational information is provided on-line by a 6,000 page syllabus that is supplemented by commercial programs. The syllabus is indexed to provide a key word search capability. The students can also access all past interim exams and retired national board questions in an electronic flashcard format (Bright and Hall 1995).

In a similar fashion, medical students in Micronesia are using the Tripler Army Medical Center telemedicine system to participate in grand rounds at the Hawaii-based tertiary care center. The University of Iowa, with assistance from a three-year grant from the NLM, has created a Virtual Hospital. Initiated in November 1992 as the first medical site on the Web, the Virtual Hospital has grown from a simple home page to an elaborate service that permits health care practitioners, students, and patients to access multimedia textbooks, school lectures, government pamphlets, clinical practice guidelines, and school manuals. The site receives 200,000 queries a week and is growing at a rate of 5,000 hits per week. Case Study 5-2 describes the efforts of Columbia University Health Sciences Center to use NII tools to enhance its medical curricula.


Case Study 5-2: Revising the Development and Delivery of Medical School Curricula


Columbia University Health Sciences Center illustrates how NII-related technologies can enhance medical education. In a recent white paper, the medical faculty authors argued that knowledge expansion pressure on faculty, curriculum revision, varied student learning styles, variety of media, structure of information, CME, and the cost of primary medical education are all factors contributing to the re-examination of the delivery of educational curricula. These factors, coupled with the proliferation of high-speed networks, the availability of portable compilers, and the availability of so much information in electronic formats, have led Columbia to create a model for the development of an electronic curriculum called CHIPS -- Columbia Health Information PerspectiveS. Once fully operational the evolving curriculum will be accessible to a variety of students, including individuals enrolled in nursing, public health, dentistry, and medical programs. In addition, the information can be custom tailored to accommodate personal learning styles. Users will be able to track information across various courses, disciplines, instructors, and even years.

The CHIPS educational model stresses that students are capable of directing their own learning and that they learn more efficiently by doing so. Columbia envisions CHIPS as a learning resource available to each student that combines aspects of library resources, tutoring, and testing, and is accessible 24 hours a day from anywhere in the world.

The three-phase implementation of CHIPS is already under way. Columbia is gathering paper-based educational materials and analyzing them for structure, content, and overlap, essentially building an electronic version of its library collection The second phase, building curriculum models, is underway for medicine, with plans to expand to dentistry, nursing, and some aspects of public health. The modeling process consists of constructing a skeleton of knowledge to which various infomation resources can be attached. The third phase consists of building the links that will combine all of the multimedia elements within the new curricula and building the pathways by which students can access and use the combined resources (Molhoh 1995).


Electronic access is being viewed not only as a teaching aid, but also as a way to generate revenue to sustain CME. For example, the Regional Medical Library at the University of Washington in Seattle, the University of Pittsburgh through a program called Medical Trivial Pursuit, and the University of Health Sciences in Bethesda, Md., all use the Internet for providing CME. The University of Iowa plans to expand its Virtual Hospital for use in CME. The Texas Tech University Health Sciences Center's HealthNet program, which relies on the Rural Health Satellite Network, is being used to deliver telemedicine services to rural areas but also to provide continuing education programs to physicians, nurses, and allied health professionals in 75 rural hospitals and clinics. The live, one-way video, two-way interactive audio and PC-based system has been credited with reducing feelings of professional isolation as well as acquainting practitioners with the latest developments in their fields. This program has become so successful that other states have sought licenses to use its educational curricula.

The NII is also spawning new ways to incorporate existing technology into curriculum development and delivery. Neither single-user education software nor classroom television is new. Both have been under development for years. However, integrating them into a multimedia educational tool could offer new possibilities for collaborative learning in, for example, the area of medical simulation. Such a system could model clinician/patient interaction and provide a type of hands-on training to augment the traditional first two years of medical education. This kind of patient simulation model has been developed for medical student training by Interactive Drama Inc. (IDI), an educational multimedia firm in Bethesda, Md. IDI's full-motion video simulations are voice-driven, allowing students to engage video patients in direct dialogue within the context of a simulated clinical setting. Students profit from playing the role of clinician, making treatment decisions based on information they obtain from the patients and from a full clinical database. Outcomes for patients are determined by a built-in probability algorithm.

Computer simulation of medical conditions and treatments is a key linkage between education and decision support. The faster the feedback loops, the closer education is brought to decision-making. While some simulation software packages are relatively inexpensive and run on PC-based multimedia systems, others are costly and require scarce hardware and software resources. But costs could be shared if models and high-performance computingresources could be distributed over the NII.

Two recent ARPA initiatives have sought to combine multimedia approaches, including simulations to train medical students and assist surgeons. The University of San Diego is exploring an educational workstation that would present case data and allow medical students to access databases and consult with physicians to determine treatment protocols. In a related effort, Northrop Grumman is creating an advanced surgical workstation that would assist surgeons by providing access to test results and database resources in the surgical suite.

The NII can also stimulate development of related educational tools to assist researchers and educators. The University of Colorado Health Sciences Center, supported by funding from the NLM, has developed virtual cadavers through the Visible Man and Visible Woman projects. To create the cadavers, state-of-the-art photographic and radiographic techniques were used to collect thousands of images of the human body from which highly detailed anatomical data were derived. The approximately 16 gigabytes of data for the Visible Man and nearly 48 gigabytes of data for the Visible Woman are available over the Internet with a licensing agreement and on CD-ROMs (Skolnick 1995).

Similar powerful tools are being developed at the University of Washington as a part of the Digital Anatomist project, which is creating a computerized model of the human body, both inside and out, based on images assembled from resources worldwide. Using the models, students and practitioners alike will be able to access all available knowledge about every structure (Project 1994). Interactive atlases of the brain and thorax are already available over the Internet, on CD-ROMs, and on video discs. The goals of the program are threefold: to improve medical students' knowledge of human anatomy; to provide the public access to data explaining human anatomy and health; and to help surgeons plan for operations and treatment.



Several impediments are stalling the use of NII-related tools and technologies to support and enhance medical research and education in academic medical institutions and the practicing physician community.

1. Lack of final FDA regulations regarding standalone medical decision-support software. In 1987 and 1989, the FDA published draft guidelines on standalone medical decision-support software. The draft policies are problematic. They imply that a hospital cannot access software at another hospital over a network to help in diagnosing and treating its own patients unless the software has 510K approval. Adding to the complexity, the FDA requires approval whenever a change is made to a medical device--a category that they assert includes standalone medical decision-support software. For devices such as pacemakers, these changes may occur only every few years. In the case of software, however, this rule could be interpreted to mean that new FDA approval is required with each new release of the product. Such fine-tuning could occur frequently, requiring continual clearances from the FDA. Obviously this would quickly prove to be an unworkable process.

The FDA has never issued proposed or final rules on this subject. Although draft guidelines do not have the force and effect of final regulations, the FDA has been regulating this type of software on a case-by-case interpretive basis. Researchers and practitioners who have been cautious and requested guidance from the FDA have found themselves subject to these interpretations; those who have not contacted the FDA presumably are selling/sharing/using this software without regard to FDA intervention. In May 1995, the FDA published a notice announcing its intent to develop and publish a proposed rule modifying certain elements of its previously published draft policies. To date, that final rule has not been issued. Ironically, the FDA did not regulate hand-calculated analyses and printed guidelines, precursors to this software.

The national information infrastructure could make these software support tools more widely available, but uncertainty regarding FDA regulations discourages physicians from sharing them. As a result, other physicians are prevented from delivering enhanced medical care to their own patients. Case Study 5-3 illustrates an example where software developed with federal funds cannot be distributed because of this issue.


Case Study 5-3: FDA Slows Sharing of Research Software Used for Developing Cancer Treatments

The lack of FDA guidelines regarding standalone decision-support software is inhibiting the availability of dramatically effective methods developed through National Cancer Institute (NCI) grant to treat cancerous tumors using radiation therapy.

During the early 1980's, researchers were developing software that would permit clinicians to create three dimensional models of cancerous tumors and determine the appropriate radiation dosage as well as the angles at which the radiation beams should be directed to the tumor. Such techniques have the potential to increase survival by enabling greater radiation doses to the tumor, while sparing normal tissues and thus minimizing complications and side-effects of the radiotherapy. By the late 1980's, it was clear that without better software tools, the implementation of these techniques in a clinic would be so labor-intensive that only the largest and most sophisticated cancer centers would be able to accomplish this for routine therapy. Radiotherapy software engineers were freely sharing their software but found that exchanged software rarely worked on a system different from the original architect's without extensive modification and tinkering. It was also clear that the radiation therapy community was duplicating each other's efforts and "reinventing the wheel" many times over. Recognizing the importance of these efforts, NCI awarded contracts in 1989 to three universities to collaboratively address these problems.

The objective of the project was twofold. First, the three universities were to develop a methodology that would enable radiation therapy departments to utilize software developed at other institutions that would transcend different operating systems, programming languages and different compiler hardware. Second, they were to develop software tools that would facilitate the use of computer modeling techniques to plan a radiation treatment program tailored to a patient's specific needs. The statement of work specified that any software developed would be in the public domain and available free of charge.

The project was very successful. The university contractors created the shared software tools to facilitate radiation treatment planning through computer modeling. They also developed software that would act as an interface between the new "shared" software and a user's in-house existing software. This would permit any interested medical center to take advantage of these treatment modeling techniques without modifying the basic software or systems at the individual sites.

Unfortunately, by the time the project was completed in 1994, the FDA changed its mind about guidelines on medical software and refused to let these tools be shared, even for research purposes, without FDA approval. (It had been fully understood at the beginning of the project that if a vendor incorporated the tools into a commercial treatment planning system, they would need to acquire FDA approval for its dissemination. Therefore, the tools were constructed with the FDA guidelines for 'good manufacturing practices' so that this eventuality would not present an excessive burden for the vendor.) Although FDA has allowed by nine institutions to use the tools in a limited project to support an NCI-supported clinical trial, they continue to be out of reach for the large majority of users for which they were intended.


2. Reluctance of practitioners to change their practice patterns. Despite all the excitement over computing and networking, physicians have been quite slow to incorporate these tools into their daily practices. As the BMI survey of today's physicians shows, personal computer use among physicians is limited and mostly confined to routine tasks such as word processing. Even the simplest collaborative tool, E-mail--a natural means of communication for busy researchers, educators, and practicing physicians--has not yet been widely adopted. Similarly, on-line medical information services such as MedLine are still widely used only within academic medical centers, although patients' use of these services may outstrip their use by practicing physicians. Beyond access difficulties, there may be a widespread perception in the medical community that the additional value to be gained by using on-line tools is not worth the effort required to learn how to use them. However, the medical community will likely see substantial changes in the next few years that will incorporate these tools in order to enhance cost-effective medical research and education.

3. Absence of standards in terminology and data format. Varying data formats can inhibit health outcomes research by making it difficult to take advantage of networked databases. Medical record standards must be developed to enhance collaboration among clinical researchers and to facilitate the local access and analysis of distant databases. Standards are also lacking for indexing topics in health care.

4. Lack of guidelines regarding privacy and the use of patient records for research. The confidentiality of patient records must be guaranteed in sharing data about health outcomes on large networks. The issues here are similar to those discussed in chapter 3, "Personal Health Information and Management."

5. Competition for and hoarding of information. Medical knowledge can indeed grow through collaborative research and by sharing data on networked computers. Competition among managed care organizations and third-party payers may, however, limit the sharing of data and the conclusions to be drawn from it.


6. Liability associated with inaccurate databases and inference. There is little precedent for cases where inaccurate databases or erroneous automated inference leads to incorrect clinical decision-making.



The following recommendations, if enacted, should spur the medical community to adopt NII-based tools to further research and enhance education.


1. The FDA should not regulate standalone medical decision-support software except when it both introduces substantial risk to patients and is to become a commercial product. The FDA should finalize and publish clear and appropriate regulations as soon as possible to resolve current uncertainties in the commercial and research communities. The FDA should seek input from the medical speciality societies, vendors, and relevant organizations so that the regulations meet appropriate standards for patient care and are not so restrictive that they inhibit widespread use of these technologies. These regulations should clarify when this software cannot be shared over a network. The FDA should consider regulatory options such as defining and implementing a labeling system in lieu of conducting formal, detailed software evaluation studies. This would significantly minimize the regulatory burden while still alerting software users to potential risks. For example, labels for diagnostic systems could distinguish between unevaluated software and software that can successfully diagnose illnesses from classic sets of symptoms.

A key issue that the FDA must consider when deliberating about a given piece of software is the fundamental similarity between the output generated by decision-support software and the guidelines in textbooks on clinical decision-making. Guidelines printed in clinical and consumer-oriented books are similar to the output generated by many of the decision-support software tools. A set of inputs (e.g., symptoms) and respective outputs (e.g., diagnoses) of a decision-support system could easily be printed out, organized into a book, and then distributed freely, because books are not medical devices and textbooks are not regulated by the FDA. The close relationship between books and many kinds of computer- generated clinical information must be taken into consideration before any attempt is made to regulate this software.

Noncommercial standalone medical decision-support software used in research should remain unregulated so that innovation and ongoing experimental efforts will not be stifled. Research software should be available free of charge, although developers/distributors should be permitted to recover costs associated with maintaining and upgrading it. In many cases, the distribution of research software should include commented source code. However, software being released for beta-testing prior to product commercialization should be exempted from this requirement.

2. Initiatives should be adopted to clarify legal, liability, and malpractice issues involved in using standalone medical decision-support research software. When experimental/research software is shared, it is not clear whether the software developer assumes all liability and malpractice responsibility for its use or whether this rests with the institution that accepts and uses it. Legal or legislative action may be required to clarify this.

3. Medical schools and continuing medical education programs should incorporate health informatics into their curricula. Many physicians lack basic computer skills and are therefore unable to make use of the many emerging NII-related applications. Moreover, some actually discourage the use of these tools because they themselves are not computer literate. Training in this area should be required in medical school and made available through continuing medical education programs. The traditional separation between the medical and computer science departments in many major research universities may also be hindering the development of appropriate training and education curricula. Increased cooperation between these departments can help prepare health care professionals to make adequate use of the wealth of information resources available to them.

4. Research and innovation on user-interface hardware and software should be stimulated so as to promote the development of better solutions to the problem of human-computer interaction in health care. User interfaces that have greater ease-of-use, and that are more easily integrated into the ergonomic patterns of health care, can catalyze greater acceptance and use of innovative computer-based tools in medicine. Advances are needed in embedded intelligent agents, hands-free computing environments, natural language processing, speech understanding, handwriting recognition, and notepad inputting schemes.




























cclogoSMALL.gif (1930 bytes)

competetextgrey.gif (3110 bytes)




Highway to Health: Transforming U.S. Health Care in the Information Age

Go to Locations Within the Bibliography




























Note: Citations in the text to sources not listed in the bibliography refer to information in unpublished form provided directly to the Council by the respective persons and organizations.

A -

Ad Age 300. 1990-1995. An annual survey of the nation's largest magazines. Advertising Age.

Allen, A., ed. 1995. Telemedicine program roundup chart. Telemedicine Today, spring.

Allen, A., and J. Hayes. 1995. Patient satisfaction with teleoncology: a pilot study. Telemedicine Journal, January.

American Health Information Management Association (AHIMA). 1995. Presentation on the status of privacy legislation. Washington, D.C.

American Hospital Association (AHA). 1995. 1994/95 AHA Hospital Statistics, the AHA Profile of United States Hospitals. Chicago.

Arthur D. Little Co. 1992. Telecommunications: Can It Help Solve America's Health Care Problems? Cambridge, Massachusetts.

Atkinson, B. 1995. Wisconsin cities win with WHIN. InfoCare, May.

Return to Top of Bibliography


Banks, H. 1995a. What's ahead for business? Forbes, 14 August.

Banks, H. 1995b. A White House challenge to telemedicine. Forbes, 14 August.

Bergman, R. 1994. The world at their fingetips: Rural providers turn to Internet. Hospitals and Health Networks, 20 July.

--. 1995. Stone age solutions? I.S. vendors aren't keeping pace with today's delivery needs. Hospitals & Health Networks, 5 February.

The biggest and the best. Information Week, 18 September.

Blakeslee, B. 1995. The military as a national test bed for health reform: industry perspective. Paper presented at National Forum: Military Telemedicine On-Line Today: Research, Practice and Opportunities, 27-29 March, at McLean Hilton, McLean, Virginia.

Blumenthal, S. J. 1995. A new national focus on women's health. In The Women's Complete Healthbook, edited by Epps, R. P. and S. C. Stuart. American Medical Women's Association. New York: Delacorte Press.

Bright, G. R., and P. W. Hall. 1995. Information technology in medical education: the Case Western Reserve University experience. Journal of the American Medical Association, 5 April.

Return to Top of Bibliography

C -

Charles J. Singer & Co. 1995. The Singer Report on Managed Care Systems and Technology, 21 August.

Congressional Budget Office (CBO). 1995. The Economic and Budget Outlook: An Update. Washington, D.C.

Congressional Research Service (CRS). 1994a. Welcome to Cyberia: an Internet guide. CRS Issue Brief. Washington, D.C.

--. 1994b. Outcomes research, clinical practice guidelines, and the Agency for Health Care Policy and Research. CRS Issue Brief. Washington, D.C.

--. 1995a. Long-term care for the elderly. CRS Issue Brief. Washington, D.C.

--. 1995b. Research and development funding: fiscal year 1996. CRS Issue Brief. Washington, D.C.

Council on Competitiveness. 1993. Vision for a 21st Century Information Infrastructure. Washington, D.C.

--. 1994. Breaking the Barriers to the National Information Infrastructure. Washington, D.C.

Cukor, P., and L. Baer. 1994. Human factor issues in telemedicine: A practical guide with particular attention to psychiatry. Telemedicine Today, summer.

Currents. Hospitals & Health Networks, 5 November.

Return to Top of Bibliography

D -

Dede, C. J. 1992. Education in the twenty-first century. Annals of the American Academy of Political & Social Science, July.

Deering, M. J., and J. Harris. 1995. Consumer health information demand and delivery: Implications for libraries. Bulletin of the Medical Library Association, April.

Dellecave, T. 1995. Technology: The best remedy. Information Week, 18 September.

DeNitto, E. 1995. Kiosks give OTC advice at Kroger. Advertising Age, 14 February.

Dorgan, C. A., ed. 1995. Statistical Record of Health and Medicine. New York: Gale Research Inc.

Return to Top of Bibliography

E -

Electronic Industries Association (EIA). Consumer Electronics Group. 1995a. U.S. Consumer

Electronics Sales and Forecasts 1991-1996. Washington, D.C.

--. 1995b. Electronic Market Data Book, Statistical Yearbook of the Electronics Industries.

Washington, D.C.

Return to Top of Bibliography

F -

Faltermayer, E. 1994. How to disarm health care's hidden bomb. Fortune, 22 August.

Frank N. Magid Associates. 1995. RTNDA executive summary, covering the '96 campaign. Conducted for the Radio and Television News Directors Foundation. St. Louis.

Frost & Sullivan. 1995. U.S. Hospital Information System Markets. Mountain View, California.

Future Health cuts expense of treating costliest patients, with considered care. 1995. Warfields Business Record, 10 February.

Return to Top of Bibliography

G -

Gillooly, B. 1995. Smoothing tech turbulence. Information Week, 18 September.

Goodall, W. M., and E. A. Murphy. 1995. The development of a telemedicine network within an integrated health care delivery system. Presented at the Second International Conference on the Medical Aspects of Telemedicine & the Second Annual Mayo Telemedicine Symposium, 6-9 April, at Rochester, Minnesota, Jacksonville, Florida, and Scottsdale, Arizona.

Gorry, G. A., L. M. Harris, J. Silva, and J. Eagli. 1995. Health care as teamwork: The Internet collaboratory. In Health and the New Media: Technologies Transforming Personal and Public Health, edited by Harris, L. M. Mahwah, New Jersey: Lawrence Erlbaum Associates.

Gorski, T. 1995. The evolution of managed care. Treatment Today, spring.

Grigsby, J. 1995. Lack of coverage for telemedicine services: a barrier to implementation of telemedicine. In Telemedicine Action Report. Western Governors' Association. Denver.

Grigsby, J., E. J. Sanberg, M. M. Kaehy, A. M. Kramer, R. E. Schlenker, and P. W. Shaughnessy. 1994. Report 2: Case Studies and Current Status of Telemedicine. Denver: Center for Health Policy Research.

Return to Top of Bibliography

H -

Harris, L. M., ed. 1995. Health and the New Media: Technologies Transforming Personal and Public Health. Mahwah, New Jersey: Lawrence Erlbaum Associates.

Health Insurance Association of America (HIAA). 1994. Source Book of Health Insurance Data. Washington, D.C.

The healthcare informatics 100. 1995. Healthcare Informatics, June.

Healthcare Information and Management Systems Society/Hewlett-Packard Leadership Survey (HIMSS/HP). 1995. Trends in Health Care Computing. Survey conducted at the sixth annual HIMSS Conference and Exhibition, San Antonio, Texas, 12-16 February.

Highlighting information-seeking methods. 1994. Bulletin of the Medical Library Association, July.

The hospital rankings. 1995. U.S. News and World Report, 24 July.

Return to Top of Bibliography

I -

Iglehart, J. K. 1992. Health care policy report: the American health care system. New England Journal of Medicine, 3 September.

Interactive Services Association (ISA). 1995. American Internet User Survey. Silver Spring, Maryland.

International Telecommunications Union (ITU). 1995. World Telecommunications Development Report. Geneva, Switzerland. The internet survey. 1995. The Economist, 1 July.

Return to Top of Bibliography

J -

Jossi, F. 1996. Easy way to find a doctor. Wired, January.

Return to Top of Bibliography

K -

Return to Top of Bibliography

L -

Louis Harris and Associates. Westin, A. F. 1994. Interactive Services, Consumers, and Privacy. New York.

--. 1995. Equifax-Harris Mid-Decade Consumer Privacy Survey 1995. New York.

Return to Top of Bibliography

M -

Mahmud, K, and J. Lenz. 1995. The personal telemedicine system, a new tool for the delivery of health care. Journal of Telemedicine and Telecare. Merrill Lynch. 1995. Adapted from materials presented at the Merrill Lynch Healthcare Research Conference, 18 April. New York.

Molholt, P. 1994. The CHIPS project: A new model for teaching and learning. In Emerging Communities: Integrating Networked Information into Library Services, edited by A. P. Bishop. University of Illinois at Urbana-Champaign.

Montague, J. 1994. CME, a school for survival? Hospitals & Health Networks, 24 June.

Return to Top of Bibliography

N -

National Association of Home Care (NAHC). 1994. Basic Statistics About Home Care. Washington, D.C.

National Research Council (NRC). 1993. Committee Toward a National Collaboratory. National Collaboratories: Applying Information Technology for Scientific Research. Washington, D.C.: National Academy Press.

New Jersey Institute of Technology and Thomas Edison State College (NJIT and Edison). 1994. Electronic Network Solution for Rising Healthcare Costs, Briefing Paper. Report of the Healthcare Information Networks and Technologies Project, 28 November.

Return to Top of Bibliography

O -

Oxbridge Communications. 1978-1995. Oxbridge Directory of Newsletters. New York.

--. 1994b. The 1994 Standard Periodical Directory. New York.

Return to Top of Bibliography

P -

Perednia, D. A., and A. Allen. 1995. Telemedicine technology and clinical applications. Journal of the American Medical Association, 8 February.

Princeton Survey Research Associates. 1995. Local Television News and Health Care Survey. Conducted for the Radio and Television News Directors Foundation, sponsored by the Robert Wood Johnson Foundation. Princeton, New Jersey.

Project to create a complete model of the body. 1994. Chronicle of Higher Education, 20 April.

Return to Top of Bibliography

Q -

Return to Top of Bibliography

R -

Research America. 1995. Medical research and health care concerns. Highlights from a survey of the American public conducted by Louis Harris and Associates. Alexandria, Virginia.

Roan, S. 1995. An on-line medical mishmash. Los Angeles Times, 15 February.

Rubin, R. 1995. Can't reach your doctor? Try E-mail. U.S. News and World Report, 13 February.

Return to Top of Bibliography

S -

Sachs Group. 1995. Health Care 1999: A National Bellwether. Evanston, Illinois.

Salganik, W. 1995. Managed care no cure for paperwork. Pittsburgh Post-Gazette, 27 December.

Sands, D. Z., C. Safran, W. V. Slack, and H. L. Bleich. 1994. Use of electronic mail in a teaching hospital. Reprinted from the 17th Annual Symposium on Computer Applications in Medical Care. Chicago: American Medical Informatics Association, McGraw-Hill.

Schiller, A. E. 1995. Redefining home health through technology, markets, communities and consumers. The Remington Report, June/July.

Scott, J., and N. Neuberger. 1994. Telemedicine and information technologies in health care, project tracking document. With partial funding by the National Aeronautics and Space Administration. Arlington, Virginia: Center for Public Service Communications.

Scott, L. 1994. Home-care revenues soar to $5.1 billion via mergers. Modern Healthcare, 23 May.

Siwicki, B. 1995. Managed care could create a technology boom. Health Data Management, July/August.

Skolnick, A. 1995. Radiologists display powerful new tools. Journal of the American Medical Association, 25 January.

Return to Top of Bibliography

T -

Technology Marketing Group (TMG). 1995. TMG research reveals PACS market moves higher on push from provider networks, managed care and telemedicine. Press release, 24 February. Des Plaines, Illinois.

Teenagers and the madness of drugs. 1995. U.S. News and World Report, 13 November.

Telemedicine Research Center (TRC). 1995. Telemedicine information exchange, 15 May. Portland, Oregon.

Terry, K. 1995. Cover story. Business and Health, April.

Thornburg, L. 1994. Taking the measure of health care providers. HR Magazine, November.

Times Mirror Center for The People and The Press. 1993. The public, their doctors, and health care reform. Press release, 14 April. Washington, D.C.

Towers Perrin. 1995. Navigating the Changing Health Care System: The Towers Perrin Survey of What Americans Know and Need to Know. New York.

Return to Top of Bibliography

U -

U.S. Department of Commerce (DOC). 1994a. Technology Administration. National Institute of Technology. Advanced Technology Program. Advanced Technology Program Proposal Preparation Kit. Washington, D.C.

--. 1994b. Technology Administration. National Institute of Technology. Advanced Technology Program. Information Infrastructure for Healthcare (95-10). Washington, D.C.

--. 1995a. Technology Administration. National Institute of Technology. Advanced Technology Program. A Guide for Program Ideas. Washington, D.C.

--. 1995b. Technology Administration. National Institute of Technology. Advanced Technology Program. Press release and program overview, 1 September. Washington, D.C.

U.S. Department of Defense (DOD). 1993. Advanced Research Projects Agency. Presentation from the Second National Health Information Infrastructure Plenary, December. San Francisco.

U.S. Department of Health and Human Services (HHS). 1992a. United States and Healthy People 2000 Review. Washington, D.C.

--. 1992b. Savings from the Medical and Health Insurance Information Reform Act of 1992. Washington, D.C.

--. 1995a. Health Care Financing Administration. Medicare: A Profile. Washington, D.C.

--. 1995b. Office on Women's Health. Older women's health. Washington, D.C.

Return to Top of Bibliography

V -

W -

Waegemann, C. P., ed. 1994. International Directory of Organizations, Standards and Developments in the Creation of Electronic Health Records. Newton, Massachusetts: Medical Records Institute. Wennberg, J. 1993. Giving patients a bigger say in choosing treatment [interview by Berkeley Rice]. Medical Economics, 11 October.

Wilcox, S. M., D. U. Himmelstein, and S. Woolhandler. 1994. Inappropriate drug prescribing for the community-dwelling elderly. Journal of the American Medical Association, 27 July.

Woolhandler, S., D. U. Himmelstein, and J. P. Lewontin. 1993. Administrative costs in U.S. hospitals. New England Journal of Medicine, 5 August.

Woolley, M. 1994. Taking research public. SRA Journal, summer.

Workgroup for Electronic Data Interchange (WEDI). 1993. Workgroup for Electronic Data Interchange Report. Washington, D.C.

X -


Y -


Z -

Return to Top of Bibliography



























cclogoSMALL.gif (1930 bytes)

competetextgrey.gif (3110 bytes)




Highway to Health: Transforming U.S. Health Care in the Information Age

The Council on Competitiveness would like to thank all of the individuals and organizations that made this report possible. Special thanks go to Project Co-Chairmen Robert Hattery, M.D. and John Rollwagen, who deftly led a diverse Advisory Committee comprised of many leaders in the health care industry. This report was developed with considerable input from that committee and the Council thanks them for the time, energy and expertise that they devoted to this effort. The Council is also greatly appreciative of the leadership provided by the Advisory Committee working group leaders: Richard Barker, Jerry Fleming, Julian Rosenman, M.D., and Jay Sanders, M.D. During the last twelve months, they chaired numerous meetings and conference calls, and critiqued several drafts of this report to ensure its accuracy. Without their assistance and attentiveness to detail, this report would not have been possible. A complete list of the Advisory Committee is provided on the following pages.

We would also like to thank the following physicians for their advice regarding this report: Michael Kienzle, M.D., University of Iowa, Hospitals and Clinics; Ron P. Lesser, M.D., The Johns Hopkins School of Medicine; Harold P. Lehmann, M.D., The Johns Hopkins School of Medicine; Max E. Stachura, M.D., Medical College of Georgia, and Eric Tangalos, M.D., Mayo Foundation.

This report was written by Council Vice President Suzy Tichenor, with expert advice and assistance from Barbara Lindatier. Council Associate Heming Nelson deserves special recognition for researching the myriad of facts and original market data that provide the foundation for much of this document. Program Associate, Terri Watson, expertly managed all the logistical details required to keep this project running smoothly. Special thanks are also due Carol Weinhaus, who provided many creative thoughts regarding some of the charts and graphs we have included. Carolyn Van Damme also deserves special recognition for managing the layout process. Without her assistance, we would not have this final report.

Advisory Committee

Health Care Infrastructure Project




Robert Hattery John A. Rollwagen

Board of Governors and CEO Senior Advisor

Mayo Clinic, Rochester, Minn. St. Paul Venture Capital

Corporation & Chairman

Plasma & Materials Technology, Inc.


Project Director Project Associate


Suzanne Tichenor Heming Nelson

Vice President Council Associate

Council on Competitiveness Council on Competitiveness




Marc Baldwin Alan Dowling

Assistant Director Partner

AFL-CIO Ernst & Young


Richard Barker Ron Edgerton

General Manager, IBM Healthcare Solutions General Manager and Vice President

IBM Corporation Kodak Health Imaging Division


Gordon Bell Jerry Fleming

President Senior VP & Director of Project Development

Community Sector Systems Kaiser Permanente


John Brasch John D. Forsyth

President Executive Director

Senior Technologies, Inc. University of Michigan Hospitals


George Buchanan Edmund A. Franken

Director of Hughes Business Communications Professor, Department of Radiology

Hughes Electronics Company University of Iowa, Hospitals and Clinics


Helen Darling John J. Gannon

Manager, Healthcare Strategies and Programs National Partner in Charge

Xerox Corporation KPMG Peat Marwick


Peter DiCicco Paul Greenspan

Secretary Treasurer, Industrial Union Industry Sales Manager: Healthcare

Department, AFL-CIO Lotus Development


William R. Hambrecht Howard Reis

Chairman and Co-Chief Executive Officer Director, Health Care Market Development

Hambrecht and Quist, Incorporated NYNEX Corporation



Linda Harris Julian Rosenman

Director, Health Practice Professor, Department of Radiation Oncology

Center for Information Technology and Services University of North Carolina

Mitretek Systems, Inc.


Eric Horvitz Robert L. Ryan

Senior Researcher, Decision Theory Group Senior Vice President and CFO

Microsoft Medtronic, Inc.


Ira Kaufman Jay Sanders

Director, Information for State Health Professor of Medicine and Surgery, and

Policy Program, University of Medicine Director of the Telemedicine Center

and Dentistry, New Jersey Medical College of Georgia


C. Everett Koop Scott Sherman

Senior Scholar Assistant Dean

Koop Institute The Johns Hopkins University School Of Medicine


Gail H. Knopf William Smith

Vice President and Chief Information Vice President, BellSouth Business Systems

Officer, Humana Inc. BellSouth


Anthony Lombardo Dan Sudnick

President, Loral Medical Imaging Systems Program Director, Health & Telemedicine

Loral Corporation Communications Systems

AT&T Bell Laboratories


Khalid Mahmud Jan Suwninski

President and Chief Executive Officer Executive Vice President, Opto-Electronics

American Telecare, Inc. Group, Corning, Incorporated


Kim McMann John H. Warner

President, State Healthcare Executive Vice President



Reynie Ortiz William W. Young

CEO, Jones Educational Networks, Inc. Vice President, Health and Operations Systems

Jones International Aetna Health Plans


Jerry Redmon

Senior Advisor to Senior Vice President Research,

Engineering and Manufacturing Processes

Rockwell International


Jack E. Reich

President, Ameritech Customized Business

Services Unit

Ameritech Corporation