William M. Detmer

Using the Internet to improve knowledge diffusion in medicine

M edical professionals are facing an information crisis. Medical knowledge is expanding and changing at an unprecedented rate, yet practitioners often do not become aware of important advances in a timely manner. While more than 360,000 articles are published in medical journals every year, knowledge diffusion to clinicians is typically slow. For instance, one study found that two years after wide publication, fewer than 50% of general practitioners knew that laser surgery could save the sight of some of their diabetic patients. Barriers to knowledge diffusion are many [12]. They include clinicians’ lack of access to up-to-date information resources, ignorance of the availability of relevant information, and lack of time for inquiry as well as poor organization of available information. Progress could be made if up-to-date information, relevant to clinicians’ information needs, were rapidly available in all work settings: office, clinic, hospital ward, library, and home. Retrieval systems, using concepts and modifiers entered directly by clinicians or automatically by electronic medical records systems, could rapidly display chunks of relevant summary information and provide links to supporting evidence and analysis. To realize this vision a combination of content, information science methods, and technology is required. In the area of content, clinicians need access to a wide variety of information including medical literature, expert summaries as found in textbooks and guidelines, information on medications and diagnostics tests, and procedural knowledge such as healthplan coverage or institutional policies. Much of this content is available today. For instance, MEDLINE, the nine-million-record bibliographic database produced by the National Library of Medicine, contains citations to the last 30 years of medical literature and has been available for three decades [10]. Knowledge-based systems that assist in diagnosis or therapy selection have, after a period of relative obscurity, become more commonplace. And fulltext journal articles, textbooks, guidelines, and drug information are increasingly available in electronic form from print publishers. But content alone will not solve the information needs of practicing clinicians. Information science methods are needed to address problems such as how to structure content to achieve optimal retrieval, how to select a resource to best

Information-Retrieval Systems

Information retrieval (IR) is the science and practice of identification and efficient use of recorded media. Although medical informatics has traditionally concentrated on the retrieval of text from the biomedical literature, the domain over which IR can be applied effectively has broadened considerably with the advent of multimedia publishing and vast storehouses of chemical structures, cartographic materials, gene and protein sequences, video clippings, and a wide range of other digital media of relevance to biomedical education, research, and patient care. As the ease with which information can be recorded and stored in digital form grows, the growth of biomedical knowledge, once spoken of primarily with reference to the biomedical literature, is now equally relevant to the retrieval of information from text within computer-based patient records, documents concerning the administration of medical care, supporting data for biomedical publications, and the literally tens of thousands of electronic-mail messages now sent and stored every day within personal and professional settings.