Medicine in the Computer Age: The ACP and Medical Informatics
Medicine and Computers, 1940-1990
Computers entered the medical arena on a small scale in the 1950s. Computers of this era were primarily employed by research labs to develop more sophisticated medical diagnostic equipment for the processing of signals and images. By the late 1950s, however, computer technology entered the hospital setting: a few progressive hospitals began to apply computer technology to patient information systems which required the digital capture of a large array of disparate information. The systems used for medical research differed dramatically from patient information systems in that they only required the application of well-defined, unchanging data sets. Conversely, patient information systems demanded more complex programming and more sophisticated technology that was not available until the late 1950s and 60s.
By the mid 1960s, the confluence of improved computing and communications technology, an increasing awareness of computer technology in the hospital sector, and congressional approval of federal funding for biomedical applications, made the development of patient information systems feasible for more institutions. While more institutions began to develop patient information systems, they were crude creations supported by monstrous mainframe computers and proprietary programming languages and hardware. Such systems could not manage large numbers of patient records over extended periods of time. By the late 1960s, integrated circuit computers and relational database applications allowed for more robust, flexible "modular" approaches to system design. These systems, while greatly improved, were still crude, ungainly and difficult to use.
 |
Monitoring Cardiovascular Function with Digital Computer, Annals, 1966 |
Two good examples of early patient information systems are MUMPS (Massachusetts General Hospital Utility Multi-Programming System) and GEMISCH (GEneralized Medical Information System for Community Health). In 1966, Neil Pappalardo, Curtis Marble and Robert Greenes of Massachusetts General Hospital developed MUMPS, a programming language, to create the hospital's modular information system. MUMPS was widely adopted by other healthcare institutions, the U.S. Department of Defense and the U.S. Veteran's Administration.
Farther south, Duke University faculty William Ed Hammomd, William W. Stead and M.J. Straube were creating GEMISCH using minicomputers. GEMISCH was a database language characterized by its interactive qualities, its ability to support multiple users, and its sophisticated query capabilities, including text string and Boolean logic searching.
Throughout the 1960s, system designers and physicians continued to conjecture about how computer technology could be employed to create more sophisticated and complex patient information systems. The National Institutes of Health's efforts to computerize Index Medicus and the fruition of these efforts in its computer-based Medical Literature Analyses and Retrieval System (MEDLARS) in 1964, compelled innovative system developers to try their hands at creating "medical information systems" (MISs) that allowed for the synthesis of patient medical information and published medical studies or findings.
The 1970s ushered in new computing and communications technology that could better support MISs. In particular, advances in computing and communications technology enabled MISs to be more decentralized. In this distributed database environment, the need for standard exchange protocol and controlled vocabularies became apparent. It was not until the late 1980s, however, that real progress in the efficient sharing of medical data was achieved.
 |
Conversational Computer Program Capable of Searching a Stored Library of Clinical Information, Annals, 1972 |
Along with more sophisticated computers came more sophisticated concepts of medical computing. In 1974, the term "medical informatics", borrowed from the French term "informatique medicale," became an internationally accepted discipline at MEDINFO '74, a world conference on medical informatics, held in Stockholm, Sweden. At that time, medical informatics represented "the broad field of computers, communications, information science, engineering, and technology as applied to medical practice, research and education"(1). But shortly thereafter, a more concise and circumscribed definition emerged that firmly grounded medical informatics in the study of clinical information and clinical decision support systems.
A clearer understanding of medical informatics did not necessarily translate into tangible results. Throughout the 1980s, few physicians utilized digital clinical information systems; most relied on medical records transcribed on paper. Some progress was made, however, in clinical applications for medical laboratories, radiology and pharmacy. By the end of the 1980s, several hospitals were employing microcomputers and disk drives with larger storage capacities to monitor patient care.
 |
RX Study Module Design Flow (computer program for the analysis of routine, longitudinal and clinical data pertaining to the association of prednisone and cholesterol), Annals, 1986. |
Exchanging medical information among disparate information systems was an additional obstacle faced by physicians, healthcare organizations and research institutions. While visionaries, such as Donald Lindberg at the National Library of Medicine (NLM), attempted to deal with the problem of sharing medical data by leading efforts to create a standard medical lexicon, their pleas often fell on deaf ears. The Medicare payment system based on 468 diagnosis related groups (DRGS) did lead to some standardization, but wide-spread adoption of a medical metathesaurus, such as NLM's Unified Medical Language System, is needed for more efficient sharing of medical data.
The 1990's has brought great advancements in computing and communications technology. This technology, coupled with the development of healthcare conglomerates, presents new opportunities and challenges for the application of medical informatics in clinical practice.
(1) Morris Frank Collen, M.D., B.E.E., F.A.C.P. A History of Medical Informatics in the United States, 1950-1990. Indianapolis: American Medical Informatics Association, 1995.
Related Links
Electronic Health Records
-
This user-friendly guide teaches you how to evaluate, purchase, implement and use an electronic health records system (EHR).
New SEP Points Option with MKSAP 14!
MKSAP 14 subscribers now have a new option for earning Self-Evaluation of Medical Knowledge points in the American Board of Internal Medicine (ABIM) Maintenance of Certification program.