"Have you in the past year had an unexplained skin rash?" We will say "No." And this is a general question regarding the family history of allergy which may be of use in determining the significance of a person's allergy history. It says, "Has any one of your parents, brothers or sisters had asthma, drug allergy, hay fever, and so forth?" And if the patient says "No" to this, the computer says, "This is the end, and thank you very much for your participation." Now, upon completion of this computer-based history, a printout is generated by means of a teletype machine and the printout is in a traditional form with the exception that it is legible. Our aim is to provide this to the physician for use in patient care. He will have this complete summary available to him prior to his initial interview with the patient. At the same time, all of the patient's responses have been entered directly into the computer and stored on magnetic tape where they can be used, both for care of his patient in the future and also for purposes of clinical research. The printing process takes a bit of time. The system has been designed to be economical. This computer is being leased by the University of Wisconsin for approximately $600 a month, which is about the price of one laboratory technician, and so even on a research basis it is economically feasible to use this as an interviewing machine. Senator YARBOROUGH. Madam Chairwoman, may I ask a question at this point? Doctor, since only a relatively small percentage of the people know how to type and most are not in that category, what do you do with the large number of people who don't know what to do with the keys on the typewriter? Dr. SLACK. Before the patient begins the medical history a teaching machine program comes on instructing this person how to use the computer. And most of the responses are made by just pressing buttons, numbered 1, 2, 3, and 4. The 1 button is pressed for a "yes" answer, the 2 for a "no," the 3 for a "don't know," and the 4 for a "don't understand." We have found that the person with no typing experience takes longer, but as long as they are literate and this is one of the limitations of this system now. A person has to be able to see, and they have to be literate. But we are working to overcome this. We are planning on the spoken word as a means of presenting questions. But we found that patients can hunt and peck their name, and they never mistype their name. They never misspell their name. They see it on the screen, and they can change it; whereas to become a patient at the University of Wisconsin, to be admitted to the University Hospital one has to give his name 15 times and it is not uncommon to have the patient discharged with a different name than he came in with. We are working to overcome this. Before the patient is permitted to go on with the medical history there is a little test, testing to see if they have familiarized themselves with the operation. So by the time they answer the first medical question, we have determined that they are good at operating the computer. With an eighth-grade education, people have been able to use our system quite well. And I should say that the patient reaction to this program has been quite gratifying. They have all enjoyed it, practically all have enjoyed it, found it interesting and have not in any way been threatened by being interfaced with a computer. As a matter of fact, the churning tapes and the flashing lights, I think, have added to it. The CHAIRMAN. Doctor, what is the machine now doing? What are those sounds? Dr. SLACK. What the machine is doing now is scanning the response tape for positive responses, and then checking to see if there are printout statements associated with those responses, and if there are, it will print them out on paper by teletype machine. I will present you with a copy of the printout. But what it does is give the patient's name and age and it reads "Allergy history, computer-based interview." 1. Asthma syndrome: "Wheezing not understood." 2. "Allergic rhinitis-none by history." Allergic rhinitis is the pedantic phrase for hay fever. 3. Urticaria-none by history. 4. Drug allergies and then it describes the penicillin allergy. "Penicillin therapy-adverse reaction experienced manifested by skin rash, joint pain, lymphadenopathy, facial swelling: one reaction; occurred in 1965"-and so forth, giving a summary of all of the clinically significant responses made by the patient, and this can go directly into the chart. But if it is lost, all of these responses are still on magnetic tape. (The print-out statement follows:) OCCURRED AFTER PENICILLIN INJECTION DURATION ABOUT 52 HOURS-- LONGEST OR ONLY REACTION ASPIRIN TAKEN NO ADVERSE REACTION TO ASPIRIN OTHER DRUGS-- NO ADVERSE REACTIONS 5. ECZEMA NONE BY HISTORY 6. UNEXPLAINED SKIN RASH NONE IN PAST YEAR 7. FAMILY HISTORY NEGATIVE FOR ASTHMA, HAY FEVER, HIVES AND DRUG ALLERGY END SUMMARY Senator NEUBERGER. Thank you very much for that demonstration. I should think this would make going to the doctor fun. Maybe if you had a little coin slot in there, you could help amortize that machine. (Statement of Dr. Slack follows:) PREPARED STATEMENT BY WARNER V. SLACK, M.D., ASSISTANT PROFESSOR OF MEDICINE AND COMPUTER SCIENCES, UNIVERSITY OF WISCONSIN The clinical transaction between doctor and patient involves data obtained from 3 sources—the medical history, the physical examination and the laboratory examination. The medical history consists of information of potential clinical significance collected during the course of interview between doctor and patient. Of the 3 sources of clinical data, relatively little research has been done on the medical history. This is true in spite of the fact that many clinicians consider medical history to be the most important aspect of the patient's examination. Apart from theoretical interest of research on the medical history, there are practical reasons why such effort is urgently needed. History taking is very time consuming and incompleteness often results from time limitations beyond the physicians control. Further, the lack of standardization from interviewer to interviewer, together with the traditional illegibility of hand-recorded patient records, makes information retrieval for patient care and clinical research difficult and often impossible. Improved methods of collecting and recording detailed medical histories are needed. A computer-based medical history system is being developed at the University of Wisconsin Medical School in which a small, digital computer collects the information of clinical histories directly from patients, prints out summaries in a form of immediate use to physicians and stores all responses for future use in patient care and clinical research. The LINC (Laboratory INstrument Computer) has been used in the medical history system because of its low cost, high flexibility and operational ease. The LINC was developed and first constructed at the Massachusetts Institute of Technology in 1962 with the support of The National Institutes of Health. The patient sits in front of the computer and questions are displayed on a cathode-ray screen. Responses are made by keyboard entries. The responses are stored on magnetic tape for future analysis and significant responses are printed by teletype for immediate use. The teletyped phrases are in a legible but otherwise traditional format. Thus far, results with the program have been encouraging. Early efforts have dealt primarily with histories of allergy symptoms. In comparison with physician-interviewers the program is most effective when dealing with patients whose presenting or primary problem is not allergy and who might not otherwise have the advantage of a detailed allergy history. In addition to the allergy system a medical history dealing with uterine cancer is now being tried clinically at the University of Wisconsin Hospitals and gastro-enterology, neurology, cardiology, psychiatry and pediatric histories are being developed. The goal is to obtain at least as much detail about potentially significant phenomena occurring in patients' medical histories as would satisfy conscientious clinicians were they to have taken the medical histories themselves. With proper technological development (using such means as mobile medical history units and regional health centers) low cost, high quality computer-based medical interviewing can be made available to large groups of people who might not otherwise seek medical care as well as to those patients whose physicians need help in their task of medical history taking. SUPPLEMENTAL STATEMENT SUBMITTED BY DR. SLACK Computer-based medical interviewing is needed to supplement health screening programs. When properly developed, computer interviewing systems will be less expensive, more complete and available to larger populations than are physician interviewers. Complete medical histories taken regularly from all citizens which should be our goal-cannot be done by physicians. There does not exist enough physician time for such a large scale medical interviewing project. Physicians must spend their time as judiciously as possible and providing physicians with information reliably obtained by computer interview will enable them to better use their time in making diagnostic and therapeutic decisions. To realize the goal of a good medical history screening project, the following are needed. The complete medical history (including questions relevant to all major disease processes) must be developed, tried clinically and improved on the basis of experience. The system design should be such that abnormalities, when elicited, will be qualified in detail by further questioning. Programmed learning and explanatory sequences should be incorporated into the computer history to increase the validity of the information collected. To present the medical questions to patients, receive and store patient responses and print out history summaries for physician use, computer hardware must be designed specifically for computer-based interviewing. This will entail the production of computer processing units capable of controlling multiple, individual interviewing terminals-with terminals equipped with large screens for displaying questions (in an easy-to-read manner), audio mechanisms for presentation of questions by spoken voice (as a means of overcoming illiteracy and vision difficulties) and keyboards made for patients' responses. Such computer interviewing equipment should be designed for use in regional, multiphasic screening centers. The cost can be kept low-well below one dollar per interviewing hour. The computer equipment described should also be suitable for the collection of physical examination and clinical laboratory data. Research on the use of a computer to obtain physical examination information by interviewing examining physicians is currently being conducted at the University of Wisconsin Medical Center. Dr. G. Phillip Hicks and associates at the University of Wisconsin have developed a computer-based clinical laboratory system using a small, inexpensive, general purpose computer (Laboratory INstrument Computer). Laboratory technologists are in direct communication with the computer-the computer communicates by cathode-ray scope and the technologists enter laboratory data by keyboard. Dr. Hicks has also interfaced laboratory machines such as the autoanalyzer directly with the computer. By using the computer as an active participant in the process of clinical data collection, laboratory, physical examination and medical history data will be obtained immediately in computer processable form (reliably and economically) and thus made available for purposes of multiphasic screening, individual patient care and clinical research. Small computers similar to the Laboratory Instrument Computer and capable of interviewing and being interfaced with laboratory machines should be available for use in areas removed from large medical centers and for use in mobile multiphasic screening units. Senator NEUBERGER. All right. We have next Dr. Caceres, who is the Instrumentation Activities Chief, Heart Disease Control Branch, Division of Chronic Diseases, Bureau of State Services, Public Health Service, Washington, D.C. Dr. CACERES. Thank you. A great variety of preventive, diagnostic, and care services are available today, in theory at least, to the elderly, and the battery of services is bound to increase. The resources available for the elderly to pay for existing services are inadequate, and we can't expect that the resources of the elderly will be sufficient pay for the more diverse services of the future. The problem then is: What can we do to so organize the delivery of health and medical services that we can economically provide all of the elderly with the quality service of today as well as with the better quality service that is going to be introduced in the future? The solution depends on doing three things. One, we must bring systems analysis to health services for the elderly, just as it has successfully been focused on the military and industrial problems of today. The need, the supply, and the delivery of health services must be viewed broadly as a total system and not as a jumble of unrelated bits and pieces. Second, in the use of health manpower, every medical and health test should be done by the person with the least formal education, |