The National Library of Medicine's MEDLINE (MEDLARS Online) database was the first database to be searched nationwide via value-added telecommunication networks. Now available on the World Wide Web free of charge from the National Library of Medicine and from many other sources, it is the world's most heavily used medical database. MEDLINE is unique in that each reference to the medical literature is indexed under a controlled vocabulary called Medical Subject Headings (MeSH). These headings are the keys that unlock the medical literature. MeSH multiplies the usefulness of the MEDLINE database and makes it possible to search the medical literature as we do today. This paper commemorates the 40th anniversary of the introduction of MeSH and salutes some of the farsighted persons who conceived and developed the MEDLINE database.

The Online Medical Record (OMR) is a full-featured shared electronic patient record in use since 1989 at Beth Israel Deaconess Medical Center in Boston. The first practice to use the OMR was a primary care practice. We observed the pattern of voluntary adoption of the OMR and the referral patterns from primary care to specialists. Adoption of the OMR among specialists has accelerated in recent years, in many cases mirroring the referral patterns from primary care to specialists. We hypothesize that referral of patients from primary care providers to specialists exposes these specialists to the benefits the electronic patient record and may promote the use of this technology. We conclude that these referral patterns provide a vector for the dissemination of electronic patient records. The important lesson is that EPR implementation in a health care network should begin with primary care to ensure the most efficient diffusion of this technology throughout the enterprise.

Proficiency in the interpretation of electrocardiograms (ECGs) is an essential skill for medical students, house officers, and attending physicians. However, resources to develop and upgrade the necessary high level of "ECG literacy" are limited. A small number of centers have attempted to address this challenge by developing "ECG of the week" internet sites. These resources are difficult to maintain and update, and many of them quickly become stagnant. We present "ECG Wave-Maven," an innovative web-based tutorial that overcomes these obstacles via a direct link to the hospital's extensive and increasing clinical ECG repository. By interfacing our educational tool to live data, we can greatly decrease the time and effort required from the time a practitioner notes an interesting case to its inclusion in the program. Users can opt to encounter the test cases sequentially or randomly, or by reviewing a list of questions or diagnoses, making this not just a quiz, but a basic educational reference. This tool may be useful in meeting the challenge of reducing serious medical errors related to ECG misinterpretation.

RATIONALE AND OBJECTIVES: The purpose of this study was to measure physicians' utilities for outcomes after ventilation-perfusion lung scanning and to explore physicians' attitudes toward misdiagnosis and the treatment of patients suspected of having pulmonary embolism (PE) in a quantitative manner by using a utility analysis. MATERIALS AND METHODS: Before ordering lung scanning for suspected PE, physicians rated five possible outcomes on a scale of 0-100 by using a computer order-entry system. These responses were rescaled and transformed to a utility measure by using the Torrance transformation. RESULTS: The mean utility for the potential outcomes after 341 lung scans were (a) no PE and no treatment (true-negative, 93 +/- 22 [mean +/- standard deviation]), (b) PE with appropriate treatment (true-positive, 84 +/- 24), (c) no PE but patient received treatment (false-positive, 54 +/- 32), (d) PE but patient did not receive treatment (false-negative, 14 +/- 23), and (e) death during pulmonary angiography (2 +/- 11). After lung scanning for acute PE, physicians placed greatest value on excluding the diagnosis (true-negative). Providing unnecessary treatment (false-positive) was valued in the midrange of utilities. The value of missing PE (false-negative) was rated almost equal to that of dying during pulmonary angiography. CONCLUSION: Physicians consider providing treatment for PE without objective confirmation of an embolus to be preferable to missing a case of PE.

PURPOSE: Pulmonary embolism (PE), an elusive diagnosis, is detected by a diagnostic work-up that is often guided by the physician's level of clinical suspicion. The ability to accurately assess PE risk on solely clinical grounds may increase with the physician's level of training. This study documented the ability of house staff practicing in an academic teaching hospital to accurately assess the clinical likelihood of PE in patients. METHOD: During a seven-month period, all 245 patients with suspected acute PE who had had lung scans ordered via a computerized order-entry system were enrolled in the study. When ordering the lung scans, all physicians (interns, residents, and attending physicians) were required to also enter their levels of clinical suspicion on a scale of 0 to 100. The physicians' levels of clinical suspicion were correlated with the final determinations of PE, and receiver operating characteristic (ROC) curves were calculated for patients' and physicians' subgroups. RESULTS: Attending physicians were most able to diagnose PE; residents were moderately able to make the diagnosis, and interns were least able to diagnose PE. The area under the ROC curve for a correct identification of patients with PE was greatest for attending physicians (0.839), intermediate for residents (0.601), and least for interns (0.594). CONCLUSION: The ability to correctly assess a patient's likelihood of PE increases with a physician's level of training, suggesting that more senior physicians should be involved in the diagnostic work-up of patients with suspected acute PE. More instruction may help medical students, interns, and residents navigate clinical scenarios in which the diagnosis is uncertain or in which sequential tests must be performed to reach the correct diagnosis.

OBJECTIVE: The purpose of this study is to document the impact of CT performed in the emergency department of patients presenting with nontraumatic acute abdominal pain. SUBJECTS AND METHODS: Fifty-seven patients were enrolled in this prospective study. Using a computer order entry system, emergency department physicians were required to report their most likely diagnosis, level of certainty, and management plan for their patients before ordering abdominal CT. After CT was performed, each physician was required to provide again his or her diagnosis, level of diagnostic certainty, and treatment plan. The outcome of each patient was evaluated by either surgery, other imaging studies, or clinical follow-up. RESULTS: After the abdominal CT, physicians' mean level of certainty in their diagnoses increased by 1.5 points (on a five-point scale; p < 0.0001). Patient management was changed in 33 (60.0%) of 55 patients. Planned treatment before CT was admission in 42 patients. Actual admissions after CT totaled 32 patients (excluding the two patients in whom preimaging information was not recorded). Thus, the net effect of abdominal CT scanning was to avert 10 (23.8%) of 42 hospital admissions. CONCLUSION: CT performed in the emergency department increases the physician's level of certainty, reduces hospital admission rates by 23.8%, and leads to more timely surgical intervention.

A growing of health-care organizations are in the process of modifying their clinical information systems (CIS) to support browser-based access. Consequently, care-providers are expected to modify their workflow to take advantage of the new technology. Intuitive interfaces, fast response and new functionality are few of the features used to promote endorsement of the change. In parallel, administrators are required to constantly assess user compliance and intervene where necessary to prevent rejection. Such monitoring translates to frequent surveys, analysis of logs and prudent utilization of user-groups. These methods tend to further burden users, suffer from "post-hoc" temporality and are difficult to maintain. In this paper we suggest an alternative approach to such data acquisition. "CareQuest" is an interactive Web-based service that can be woven into clinical applications without coding. It acquires information from the clinician at the relevant point in her workflow. It allows extensive interaction customization, data-driven response, real-time Web-based data-analysis, and full Web-based administration.

For many years, client-server systems were developed as the backbone of clinical computing in leading hospitals around the country. Beth Israel Deaconess Medical Center now faces the challenge of bridging the technology gap between such systems and the Internet. While developing Web interfaces to legacy clinical systems gives a taste of the future, it is clear that complete institutional migration to the Web is not imminent. Asking clinicians to utilize two different systems, Web-based and legacy, in the interim phase is just one of the difficulties in such transition. This paper describes "Mbridge", a solution that allows legacy system users to exploit the benefits of the Internet in a fashion that does not interfere with their workflow and is both simple and affordable to implement. The service allows clinicians to work on the legacy platform while context-sensitive clinical content is streamed to the browser without their intervention. Using the system, we can gradually expose clinicians to new Web-based applications and resources without forcing them to operate two computing environments simultaneously. The service achieves these goals by means of linkage and coordination rather than by code-translation, data exchange or replication.

Computing systems developed by the Center for Clinical Computing (CCC) have been in operation in Beth Israel and Brigham and Women's hospitals for over 10 years. Designed to be of direct benefit to doctors, nurses, and other clinicians in the care of their patients, the CCC systems give the results of diagnostic studies immediately upon request; offer access to the medical literature: give advice, consultation, alerts, and reminders; assist in the day-to-day practice to medicine, and participate directly in the education of medical students and house officers. The CCC systems are extensively used, even by physicians who are under no obligation to use them. Studies have shown that the systems are well received and that they help clinicians improve the quality of patient care. In addition, the CCC systems have had a beneficial impact on the finances of the two hospitals, and they have cost less than what many hospitals spend for financial computing alone.

Much of the work in the ICU revolves around information that is recorded by electronic devices. Such devices typically incorporate simple alarm functions that trigger when a value exceeds predefined limits. Depending on the parameter followed, these "boundary based" alarms tend to produce vast numbers of false alarms. Some are the result of false reading and some the result of true but clinically insignificant readings. We present a computerized module that analyzes real-time data from multiple monitoring devices using a customizable logic engine. The module was tested on 6 intensive care unit patients over 5 days, running alarm algorithms for heart rate, systolic and diastolic blood pressure as well as arterial oxygen saturation. Results show a ten-fold increase in positive predictive value of alarms from 3% using monitor alarms to 32% using the module. The module's overall sensitivity was 82%, failing to detect 18% of significant alarms as defined by the ICU staff. The results suggests that implementation of such methodology may assist in filtering false and insignificant alarms in the ICU setting.

We have created a clinical performance support system that transforms surgical informed consent into an interactive process capable of evolving in response to institution-specified, provider-specified and patient-specified needs. The system functions in several capacities, including: (1) a source of standardized and comprehensive content and format the transmission of procedure-related risk and complications; (2) as expert critique, providing cues in an effort to reduce the effects of biased risk appraisal; (3) captures and archives clinician behavior relating to use, modification and disclosure of standardized knowledge sources; (4) provides just-in-time access to procedural descriptions information relating to risks and complications; (5) captures, archives and makes available to the clinician patient use of procedure-related knowledge resources. By design, the system will be used to assess the relationship between clinician perception and heuristics surrounding risk appraisal and disclosure and patient perceptions based on response to the disclosure process. The system prototype is currently being deployed in a breast surgery unit at the Beth Israel Deaconess Medical Center.

To test the hypothesis that educational information provided via the web would not only be accessed by our patients, but could also reduce postoperative pain following ambulatory surgery, we enrolled 195 patients into a randomized controlled study. Fifty-two percent of our ambulatory surgery patients already knew how to use and had access to the Internet. Eighty-five percent of our study patients accessed the resources made available to them on the web site. Patients who had access to the pain management information on the ambulatory surgery web site reported significantly less postoperative pain on arrival to their home after surgery (p < 0.016) and into the night after surgery (p < 0.013). These patients also reported significantly less postoperative pain for the day immediately following surgery (p < 0.037). We conclude that using the Internet to provide just-in-time patient education can significantly effect the clinical outcome of care.

The electronic patient record at the Beth Israel Deaconess Medical Center has fundamentally changed the practice of medicine in ways that its developers never foresaw. This type of highly interactive and work flow enabled program is creating new collaborative roles for computers in complex organizations [4]. With the system able to supervise and monitor care, computers are able to perform many care coordination and documentation functions, freeing people to concentrate more on interpersonal interactions and provision of health care services. One of the challenges in the design of electronic patient records to assist health care providers is how to support collaboration while not requiring that people meet face-to-face. Moreover, a greater challenge for each of us as clinicians is to use this technology as a bridge (rather than a barrier) towards better patient-doctor relationships.

This paper describes an intelligent information filtering system to assist users to be notified of updates to new and relevant medical information. Among the major problems users face is the large volume of medical information that is generated each day, and the need to filter and retrieve relevant information. The Internet has dramatically increased the amount of electronically accessible medical information and reduced the cost and time needed to publish. The opportunity of the Internet for the medical profession and consumers is to have more information to make decisions and this could potentially lead to better medical decisions and outcomes. However, without the assistance from professional medical librarians, retrieving new and relevant information from databases and the Internet remains a challenge. Many physicians do not have access to the services of a medical librarian. Most physicians indicate on surveys that they do not prefer to retrieve the literature themselves, or visit libraries because of the lack of recent materials, poor organisation and indexing of materials, lack of appropriate and available material, and lack of time. The information filtering system described in this paper records the online web browsing behaviour of each user and creates a user profile of the index terms found on the web pages visited by the user. A relevance-ranking algorithm then matches the user profiles to the index terms of new health care web pages that are added each day. The system creates customised summaries of new information for each user. A user can then connect to the web site to read the new information. Relevance feedback buttons on each page ask the user to rate the usefulness of the page to their immediate information needs. Errors in relevance ranking are reduced in this system by having both the user profile and medical information represented in the same representation language using a controlled vocabulary. This system also updates the user profiles, automatically relieving this burden from the user, but also allowing the user to explicitly state preferences. An initial evaluation of this system was done with health consumers using a web site on consumer health. It was found that users often modified their criteria for what they considered relevant not only between browsing sessions but also during a session. A user's criteria for what is relevant is constantly changing as they interact with the information. New revised metrics of recall and precision are needed to account for the partially relevant judgements and the dynamically changing criteria of users. Future research, development, and evaluation of interactive information retrieval systems will need to take into account the users' dynamically changing criteria of relevance.

Since 1989, Beth Israel Hospital has been deploying an extensive online patient record (the OMR), which augmented a heavily used integrated hospital information system. Initially begun in a large primary care practice, the system is now used to share patient records among 36 practices on three campuses. Although the system was intended to eliminate the need for paper, we have found that it has, in the short term, increased the amount of paper produced. Elimination of paper record in ambulatory care has saved us $56,000, but we have yet to realize the savings of an additional $200,000 per year. We explore the factors that contribute to this "paper paradox" and discuss the costs associated with increased paper production, areas in which we have reduced paper handling, and strategies for reducing our reliance on paper.

Guidelines regarding patient-provider electronic mail are presented. The intent is to provide guidance concerning computer-based communications between clinicians and patients within a contractual relationship in which the health-care provider has taken on an explicit measure of responsibility for the client's care. The guidelines address two interrelated aspects: effective interaction between the clinician and patient, and observance of medicolegal prudence. Recommendations for site-specific policy formulation are included.