T H E  N I H   C A T A L Y S T S E P T E M B E R   -   O C T O B E R  1 9 9 6


by David Ehrenstein, Ph.D., NIDCD

A doctor in a patient's room in the new NIH Clinical Center points to the computer monitor on the wall to review the patient's progress. She shows a summary of the treatment to date, along with X-rays, blood test results, and histopathology micrographs of the patient's diseased tissues over the course of the treatment regime. In response to a question about an X-ray, she calls the radiologist, who answers using a remotely activated pointer on the screen.

This scenario is now closer to fact than fantasy and may be commonplace when the new Clinical Center is built in a few years. "Telemedicine" is coming to NIH thanks in part to strong support from the head of the Clinical Center, John Gallin. "We would like the Clinical Center to be the pace-setter for using telemedicine in research," he says. According to David Henderson, the Clinical Center's deputy director for clinical care, telemedicine will "improve both the quality and efficiency" of care at NIH and "make better physicians of us all."

One element of telemedicine being developed at NIH is video conferencing. Gallin and Henderson foresee patients being interviewed for possible inclusion in NIH protocols via video link, reducing both travel costs and the hardships of travel on sick patients. Follow-up visits could also be reduced by video conference meetings with patients and collaborating doctors. Such telemedicine encounters will certainly be improved by the use of "virtual" exam tools, such as a stethoscope and an otoscope that will allow doctors at remote sites to see and hear right along with the physician doing the exam. Two telemedicine "suites," complete with such tools, are now under development at the Clinical Center.

Henderson and Gallin see these advances as especially appropriate for NIH, which draws patients from every state and many other countries. Previously unreachable patients, such as those in emergency rooms or too infectious to travel, could become subjects of NIH research. "When we conduct clinical trials, we'll be able to broaden the number of people who can be in the study - both patients and clinical collaborators," says Gallin.

Steve Holland
Photo: David Ehrenstein
Steve Holland
Of course, clinical researchers at distant sites have been collaborating for many years, but one of their major difficulties has been sharing patient records. Steve Holland of NIAID is involved in a telemedicine project that should help in this area. Holland notes that keeping track of patient records and data and getting access to them can be challenging just within one hospital, but for collaborations between doctors across the country - as with Holland's work with collaborators at the National Jewish Hospital in Denver, with whom he is studying multi-drug resistant tuberculosis (TB) - the recordkeeping is even more complicated. The solution, developed by programmers at Los Alamos National Laboratory in New Mexico and doctors at National Jewish, is software that integrates the entire set of information on each patient into a single organized entity called the Graphical Patient Record (GPR). From his office in the Clinical Center, Holland can call up the GPR for a TB patient at National Jewish and view histories, lab test results, physician comments, X-rays, and any other data, all with the click of a mouse. Without this technology, there would be a lot more correspondence to keep track of, he says. "I would get X-rays in one pile, slides in another pile, records in another pile, and then it would be up to me to keep all that straight when each one came in."

This type of software should also allow patients and their doctors to view their records efficiently and in detail without leaving the hospital ward. Holland, who is troubleshooting the new software, says it is ideal for his research on multidrug-resistant TB because a detailed account of every drug used (and its effects) is essential for proper treatment of his study patients, who are located in both Bethesda and Denver. Along with the physician reports on each treatment, the GPR gives immediate access to test results, with one icon representing each type of analysis performed - from X-rays to sputum analysis. With all the data centralized, Holland can look at a CAT scan on his computer screen at the same moment as a Denver physician who is treating the patient and discuss the prognosis by phone. He hopes the software will thus put an end to the frustrating calls in which his opinion is sought based on an X-ray described over the phone.

In addition to making data more accessible, the GPR allows new kinds of analyses to be performed on the data. For example, looking at a specific patient's X-ray, a doctor can request all similar X-rays in the entire library of data from all patients. The images are then arranged in order of the degree to which they match the original X-ray, allowing direct comparison of a new patient's progress with outcomes from previous patients. "That's got enormous implications [for diagnosis]," says Holland. The software also allows many manipulations of CAT-scans, such as viewing of a particular slice of a patient's scans over the time course of his or her treatment.

Another benefit of these new methods is educational. A joint Telemedicine project between NIH, the National Naval Medical Center, and Walter Reed Army Hospital is primarily aimed at teaching and collaboration in the D.C. area, but it has the potential to benefit doctors nationwide. Rather than simply reading reports of NIH treatments of their patients, local doctors can be involved in their patients' treatments all along through telemedicine. "People at the NIH, in general, are experts [in specific fields], and the diseases we treat tend to be uncommon," says Holland. "So the chance for us to collaborate with local physicians in the care of their patients . . . has tremendous implications for improving the knowledge about some of these diseases and, therefore, improving knowledge about the things that we study overall."

Alan Graeff, chief of information systems for the Clinical Center, has been involved in the technical side of NIH telemedicine, including Holland's computer project and the development of the telemedicine suites. He cautions that, although the dreams of these technologies are certainly achievable, progress will be incremental. "We're not going to be jumping into virtual surgery overnight," he says. His office, the Clinical Center's Information Systems Office, is already helping several NIH researchers set up video links, however, and stands ready to support new telemedicine projects.

What difficulties lie in the path to telemedicine technologies? Video conferencing in its most basic form is well established from its use in the business community, but technical standards for telemedicine video links are needed, especially for virtual exam tools. Graeff is now working with consultants to develop these standards at NIH. Down the road, if telemedicine becomes widespread, a series of complex legal issues, such as liability, billing, and licensing, may need resolution.

There is also the issue of cost. Although the equipment requires major up-front investment, Holland thinks the money may be recovered in savings. "You may be able to save millions in the travel budget," he says, noting that both the state of Georgia and the Department of Defense have made major telemedicine investments in recent years partly for that reason. Also, certain specialists, such as radiologists (who read and interpret X-rays), may not be necessary at every hospital if they can work online. "I think that telemedicine and these sorts of interactive approaches will be necessary economies in the not very distant future," says Holland.

Gallin is enthusiastic about the opportunities that telemedicine offers for improving care and research at NIH and encourages clinical researchers to start giving serious thought to its applications. "My goal is to get people oriented so that when we move to the new hospital, we'll be ready for telemedicine."

Return to the Table of Contents