Telemedicine Defined

Telemedicine uses technology to deliver medical services to the point of need. In its report on the evaluation of clinical applications of telemedicine, a committee of the Institute of Medicine defined telemedicine as "the use of electronic information and communications technologies to provide and support health care when distance separates the participants" [1]. The committee settled on this expansive definition after considering at least 10 others [2-6]. Probably the most inclusive definition cited in the Institute of Medicine report was that "telemedicine encompasses all of the health care, education, information and administrative services that can be transmitted over distances by telecommunications technologies" [7].

Broad definitions of telemedicine have complicated the discussion of telemedicine policy. Telemedicine covers a range of technologies, including telephone, radio, facsimile, modem, and video. It may be conducted in real time, as with interactive video, or asynchronously, for the transmission of text or graphic data, auditory verbal information, still images, short video clips, and full-motion video. Robotics and virtual reality interfaces have been introduced into some experimental applications [8]. These technologies may be applied in various ways (management of chronic conditions, routine consultation, preventive medicine, public health, and patient education, for example). Some definitions of telemedicine even encompass meetings of hospital administrators, access to MEDLINE, and continuing medical education. Because the term telemedicine can be defined in so many ways, discussions of such issues as the cost-effectiveness of telemedicine become meaningless. Hence, we have limited the scope of this paper to a somewhat narrower definition of telemedicine: the use of telecommunications and information technology to provide health care services to persons at a distance from the provider.

The Range of Transmission Media

Telemedicine has used various terrestrial and space-based (satellite) transmission media. The medium that is used is important in part because its bandwidth or bit rate (the amount of information sent per unit of time) limits the type of technology that may be used. Narrow-bandwidth systems, such as ordinary telephone lines, are inexpensive but lack the capacity for full-motion video. They may be adequate, however, for transmitting still images, voice, text, or data. No single technology or bandwidth is best for all telemedicine purposes; rather, each system's capacities and capabilities must be determined by the needs of the users.

Broad-bandwidth networks have transmission rates that permit interactive, full-motion video. For example, T1 lines have a relatively high bit rate of 1.544 megabits per second. They are not, however, available in many rural and frontier areas. Interactive video may be used with narrower bandwidths if data compression algorithms are also used, but the images are sometimes too jerky to permit resolution of detail or subtle movement. Broad-bandwidth networks are costly because transmission charges are directly related to bandwidth. This problem was partly addressed by rules that were developed by the U.S. Federal Communications Commission for the implementation of changes in the universal service program under the Telecommunications Act of 1996. These rules provide subsidies for telecommunications services, for which certain rural health care providers are eligible.

Clinical Uses of Telemedicine

Most of the early telemedicine programs used interactive video to bring patients, referring providers, and consultants together. From 1959 until the 1970s, telemedicine was tested in medical schools, state psychiatric hospitals, municipal airports, jails, nursing homes, Native American reservations, and other settings [9-16]. Most of these early programs proved too costly to be self-sustaining and were terminated when external funding ran out.

The clinical applications of telemedicine are even more varied than the technologies, although considerable attention has been focused on the use of interactive video for specialty and subspecialty consultation in rural areas. The generic interactive video telemedicine system typically uses fixed, studio-type video equipment to link a rural facility with an urban tertiary care center. Consultants communicate with patients and, often, with their primary care providers in an interactive situation. The precise configuration of these networks varies, ranging from a single source of referrals (for example, a rural community hospital) and a single source of consultants (such as an academic medical center) to complex "hub-and-spoke" networks involving many referring and consulting facilities.

Almost every clinical specialty has used telemedicine in some way, although some have used it more than others. Radiologists, for example, have embraced the technology on a large scale. Cardiologists, dermatologists, and psychiatrists have been the clinical specialists most actively involved in telemedicine. The reasons for this are unclear, but this distribution may represent a kind of "founder's effect" because physicians practicing these specialties were among the clinicians to first become involved with telemedicine. Nevertheless, the fact that these specialists choose to see patients through telemedicine suggests that the medium is suited to many of their consultative tasks.

A 1996 survey of almost 2400 nonfederal rural hospitals [17] found that about 17% were participating in a telemedicine network of some kind (including services as limited as facsimile) and that another 13% had definite plans to begin using telemedicine. The number of clinics and outpatient facilities participating in such networks is unknown. Despite widespread interest in telemedicine, the actual number of patients per telemedicine program who receive telemedicine services remains relatively low [18]. One recent survey of 80 programs (1032 sites on hub-and-spoke networks) estimated that about 21 000 consultations occurred in 1996 (mean, 37.4 consultations per site per year) [19].

Telemedicine has proven its feasibility in several challenging environments, including peacekeeping missions and the space shuttle (Pool SL, Stonesifer JC, Belasco N. Application of telemedicine systems in future manned space flight [Presented paper]. Second Telemedicine Workshop, 1975, Tucson, Arizona; [20-22]), and in the more prosaic settings of the home, clinic, hospital, and long-term care facility. It has been used for many years in Canada's maritime provinces ([23]; House AM, Keough EM. Distance health systems-collaboration brings success: the past, present, and future of telemedicine in Newfoundland [Presented paper]. Conference on Information Technology in Community Health, 1992, Victoria, Canada) and in Norway above the Arctic Circle [24]. The program at Memorial University of Newfoundland in Canada has used many technologies, from facsimile (transmission of electroencephalograms) to interactive video. In addition to gaining improved access to care for patients, referring physicians may benefit from increased contact with their colleagues and greater opportunities for continuing medical education. One observer described telemedicine as a means by which medical schools can provide an "extended warranty" on medical education.

Trends in Telemedicine Technology

Although the early 1990s saw the proliferation of telemedicine systems that provided real-time, broad-bandwidth, synchronous consultations, the focus has shifted toward more personal computer-based "store-and-forward" telemedicine. Desktop systems are a convenient and probably cost-effective means of providing services. Store-and-forward technology can be used to forward medical records, laboratory results, and radiographs and other diagnostic images to a consultant. By using a multimedia e-mail format, consultants can conduct an increasing number of telemedical consultations across a readily available, accessible, and inexpensive Internet platform. The consultant has convenient access to e-mailed consultative requests that accommodate his or her schedule. The data can be reviewed, and a report of diagnostic impressions can be e-mailed to the referring physician.

Government Telemedicine Policy

Since the 1960s, the federal government has supported the development of telemedicine through grants, contracts, and National Aeronautics and Space Administration and Department of Defense budget line items that total several hundred million dollars. Several agencies currently provide such support, and their representatives have been actively involved in discussions that shape both policy and directions of growth in telemedicine. Although a comprehensive discussion of those policy issues is beyond the scope of this paper, we address one especially important and problematic matter: coverage and payment for telemedical services.

With a few exceptions (teleradiology, some telepathology, and some cardiologic data and facsimile transmission applications), Medicare reimbursement of fee-for-service telemedicine is generally not available. The reasons for this are complex, but the most immediate impediment to the coverage of telemedicine is that telemedicine does not meet the requirements of the Health Care Financing Administration (HCFA) for in-person, face-to-face contact between providers and patients. Thus, most medical consultation through telecommunications technology is ineligible for payment. Moreover, HCFA has expressed concern about the lack of solid cost data but is studying the issue and is funding demonstration projects. Coverage will be extended to certain Health Professional Shortage Areas beginning in January 1999, but HCFA has not yet established a general coverage policy. Other payers have been slow to set policies of their own, although in several states, some commercial insurance companies, Blue Cross/Blue Shield, and Medicaid pay for telemedicine services.

There is an unresolved debate about whether telemedicine represents a change in the process of medical care or merely a change in access to services. In addition, unlike most technological advances in medicine, telemedicine is not used to diagnose or treat any specific conditions. Rather, it permits a wide range of services to be provided from a distance, essentially extending the reach of the provider's office. Telemedicine offers nothing more than what can currently be done by a physician in his or her office. Moreover, it has some circumscribed limitations compared with conventional medical care, partly because most telemedicine applications are diagnostic or consultative in nature. With the exception of the use of telemedicine in laparoscopic surgery (for example, for teleproctoring) and in certain psychiatric services, direct interventions generally are not yet feasible (although remote robotic surgery has been performed in animals).

Discussions of telemedicine coverage policy tend to treat telemedicine as if it were a readily identifiable, unitary, clinical phenomenon [9, 12, 14]. A major problem with this line of thinking is the protean nature of the health care that can be provided by using computer and telecommunications technology. For example, although many telemedicine programs use interactive video for specialty consultation, some organizations are using data transmission technology to remotely monitor the pulmonary function of patients with asthma [25]. Any fee-for-service reimbursement policy must consider these important differences. For some applications, HCFA is considering a bundled payment, and prospective payment under certain circumstances (for example, home health care) may prove to be a means of circumventing coverage limitations.

Telemedicine technology is also being used to provide home health care in several pilot programs throughout the United States. No data have been published, but anecdotal and preliminary reports from several programs suggest that for some patients, telemedicine may be a viable, cost-effective means of obtaining health care services in the home [25-27]. Certain nursing procedures cannot be done through telemedicine, and undoubtedly there are patients for whom in-person home care is preferable. Nevertheless, the technology may permit home health nurses to increase the number of patients they see each day.

Because of the high costs of equipping and using an interactive video telemedicine system (especially the cost of broad-bandwidth transmission media), many telemedicine providers are moving toward desktop store-and-forward systems. However, the only telemedicine systems that Medicare is considering covering are interactive video systems, which may represent a minority of telemedicine applications by the time a policy is finally promulgated.

Other Impediments to the Growth of Telemedicine

Lack of reimbursement for telemedicine services is only one of several factors impeding the expansion of telemedicine [28]. Licensure, for example, is regulated by individual states, and bills have been introduced or passed in some states that severely limit the interstate practice of medicine through telemedicine. Examinations assessing the competence of physicians are conducted using national standards, and patient outcome studies are conducted and practice standard guidelines are developed on a national, not a statewide, basis. State regulation of licensure-considered anachronistic by many-will continue to delay the spread of telemedicine technology. The European Union has faced a similar problem since the signing of the Maastricht agreement; that experience was addressed briefly at the Atlantic Rim Telemedicine Summit in 1997 [29].

A second difficulty concerns liability and malpractice [30]. Some providers are concerned that the use of telemedicine may increase their risk (for example, a technical failure could lead to an adverse patient outcome, or telemedicine could provide an image of inferior quality that hinders a physician's ability to make an accurate diagnosis). Conversely, some physicians are concerned that if telemedicine permits high-quality care, they might be liable for failure to use it. The situation is compounded by interstate variability in the handling of malpractice claims. Because no one has yet been sued for malpractice related to telemedicine, it is not possible to assess the validity of these concerns.

Finally, the issue of confidentiality remains contentious. The present system of medical records is already insecure, but there are additional concerns about the ability of electronic medical record systems to maintain an adequate level of security. In general, electronic records are probably more secure than paper-based charts, although a possible breach of security may mean that more unauthorized persons can obtain access to confidential data.

The confidentiality problems that may arise can be classified as breaches of security and as inappropriate disclosure of individual patient information to persons who are unauthorized to receive it. Disclosure of information about a specific patient may be as likely with electronic records as with conventional paper records. Certain types of disclosure-such as the sale of lists of patients with a specific diagnosis to marketers, mailing-list brokers, or insurance companies-may even be facilitated by the use of electronic databases. Access to electronic records must be carefully restricted to those who must have access to provide care. Even encryption and firewalls may be only temporary barriers to a person motivated to obtain unauthorized access [31].

The problem is being addressed at several levels, both governmental and nongovernmental. Contractual and legislative protections are needed to provide the highest degree of security that is realistically attainable [31, 32]. The availability of electronic medical records is of great importance to telemedicine; as medical information systems evolve, telemedicine may disappear as a distinct entity and be subsumed into medical information networks. Thus, technological safeguards are also mandatory.

Telemedicine Research

Despite the growth of telemedicine, limited research supports the medical effectiveness and cost-effectiveness of the technology [33]. Studies conducted in the 1970s showed that the effectiveness of interactive video telemedicine, audio-only telemedicine, and in-person care were similar [34, 35]; although this has been supported by recent research, these three methods of care delivery may not always be equivalent [36-39]. Many claims have been made about the cost-effectiveness of telemedicine, but telemedicine applications must be examined individually. Cost-effectiveness has not yet been studied for any application, although smaller-scale financial analyses have examined the use of telemedicine, especially in prisons [40].

Conclusions

Telemedicine is a general term that refers to a wide range of technologies and applications. The concept is almost as broad as that of medical care itself. It is futile to discuss such issues as the effectiveness or cost-effectiveness of medical care, and the same is true for telemedicine. Current efforts to develop a coverage and payment policy for telemedicine focus almost entirely on the use of interactive video consultation. Although this is an important application, it is likely to represent a diminishing percentage of the use of telemedicine in the next few years. If a rational coverage policy is to be developed, policymakers, payers, and legislators must recognize that one size does not fit all. Fine-grained distinctions must be made among different telemedicine applications.

Dr. Sanders: The Global Telemedicine Group, 1317 Vincent Place, McLean, VA 22101.