American Hospitals spend a lot of money for computing, but physicians and nurses are dissatisfied with the computing that they receive. No one is at fault. Rather, each person, acting in his or her best interest, unwittingly conspires to produce the unfavorable result. Until hospitals make major and fundamental changes in the way they purchase and manage computing, they will continue to spend large sums without commensurate return.
The American Heritage dictionary defines the word "web" as "something intricately contrived, especially something that ensnares or entangles." The wealth of medical resources on the World Wide Web is now so extensive, yet disorganized and unmonitored, that such a definition seems fitting. In emergency medicine, for example, a field in which accurate and complete information, including patients' records, is urgently needed, more than 5000 Web pages are available today, whereas fewer than 50 were available in December 1994. Most sites are static Web pages using the Internet to publish textbook material, but new technology is extending the scope of the Internet to include online medical education and secure exchange of clinical information. This article lists some of the best Web sites for use in emergency medicine and then describes a project in which the Web is used for transmission and protection of electronic medical records.
Baby CareLink is a multifaceted telemedicine application designed to provide individualized information and support to families of Very Low Birth Weight infants. We believe that this innovative use of WWW and telemedicine technologies will improve family satisfaction and clinical care. In conjunction with improvements in family involvement, discharge planning, education, and follow-up enabled by other CareLink components, this system may allow infants to transition home even earlier in their hospital stay and thereby provide a clear cost savings. This paper discusses the CareLink architecture and lessons learned in implementing a telemedicine link with families at home from an in-hospital clinical unit.
One of the deliverables of the HOLON (Health Object Library Online) project is the specification of a reference architecture for clinical information systems that facilitates the development of a variety of discrete, reusable software components. One of the challenges facing the HOLON consortium is determining what kinds of components can be made available in a library for developers of clinical information systems. To further explore the use of component architectures in the development of reusable clinical subsystems, we have incorporated ongoing work in the development of enterprise terminology services into a Problem List subsystem for the HOLON testbed. We have successfully implemented a set of components using CORBA (Common Object Request Broker Architecture) and Java distributed object technologies that provide a functional problem list application and UMLS-based "Problem Picker." Through this development, we have overcome a variety of obstacles characteristic of rapidly emerging technologies, and have identified architectural issues necessary to scale these components for use and reuse within an enterprise clinical information system.
We have proposed elsewhere a strategy for releasing medical records via the World Wide Web. The philosophical underpinnings of this proposal balanced a need for access with a need for confidentiality of medical information. Other balance points could have been chosen, and methods of stronger and weaker protection of confidentiality are presented here along with the rationale behind the selected strategy.
We are developing a set of software components--the Problem List Toolkit (PL-Tk)--to support operations on clinical problem labels. An adaptation of the National Library of Medicine's Unified Medical Language System (UMLS) provides general vocabulary services to domain-specific software components. Our initial investigation centers on the inclusion in UMLS of problem labels used in the Beth Israel Deaconess Medical Center's Online Medical Record (OMR). We also explore the semantic typing of problem labels matched in UMLS. We have operationally defined a clinical problem to derive its semantic type from classes of terms representing findings or processes typically requiring diagnostic evaluation or therapeutic management in clinical practice. Of 1262 unique OMR problem labels, 999 terms (79%) have matches in UMLS. 986 of 999 terms (99%) map to the UMLS concept of the corresponding lexical match. 952 of 999 terms (95%) have semantic types that comply with our operational definition of clinical problems. These 952 terms (75%) constitute Version 1.0 of the problem list vocabulary B196. Matching terms with inappropriate semantic types raise issues regarding requirements for PL-Tk, typing of existing UMLS terms, and the adequacy of our operational definition for clinical problems. UMLS provides a large repertoire of pre-coordinated terms that are used as problem labels in a heavily used computer-based patient record system. The semantic type hierarchy provides a framework for the consistent use of clinical concepts in problem lists such that clinical problem labels represent "good" clinical problems.
This paper presents results from a demonstration project of nationwide exchange of health data for the home care of diabetic patients. A consortium of industry, academic, and health care partners has developed reusable middleware components integrated using the HOLON architecture. Engineering approaches for multi-organization systems development, lessons learned in developing layered object-oriented systems, security and confidentiality considerations, and functionality for nationwide telemedicine applications are discussed.
The main purpose of this study was to identify nurses' requirements for information technology in the next millennium. We distributed 350 questionnaires in a teaching hospital in Brazil and in the USA. We received 86 from Brazil and 78 from USA. Besides the capabilities of computers to be used in different sectors of human actions, nurses' requirements for information technology can not be considered sophisticated. Nurses in Brazil prefer a standalone system. In contrast, in USA they prefer systems that could be integrated to the whole healthcare system. Also, in the USA, nurses were more comfortable with use of computers than nurses in Brazil. In general terms, nurses in both countries feel that computer systems could make their practice easier and more efficient.
The capture and symbolization of data from the clinical problem list facilitates the creation of high-fidelity patient resumes for use in aggregate analysis and decision support. We report on the development of a UMLS-based semantic parser and present a preliminary evaluation of the parser in the recognition and validation of disease-related clinical problems. We randomly sampled 20% of the 26,858 unique non-dictionary clinical problems entered into OMR (Online Medical Record) between 1989 and August, 1997, and eliminated a series of qualified problem labels, e.g., history-of, to obtain a dataset of 4122 problem labels. Within this dataset, the authors identified 2810 labels (68.2%) as referring to a broad range of disease-related processes. The parser correctly recognized and validated 1398 of the 2810 disease-related labels (49.8 +/- 1.9%) and correctly excluded 1220 of 1312 non-disease-related labels (93.0 +/- 1.4%). 812 of the 1181 match failures (68.8%) were caused by terms either absent from UMLS or modifiers not accepted by the parser; 369 match failures (31.2%) were caused by labels having patterns not recognized by the parser. By enriching the UMLS lexicon with terms commonly found in provider-entered labels, it appears that performance of the parser can be significantly enhanced over a few subsequent iterations. This initial evaluation provides a foundation from which to make principled additions to the UMLS lexicon locally for use in symbolizing clinical data; further research is necessary to determine applicability to other health care settings.
With the advent of Integrated Healthcare Delivery Systems, medical records are increasingly distributed across multiple institutions. Timely access to these medical records is a critical need for healthcare providers. The CareWeb project provides an architecture for World Wide Web-based retrieval of electronic medical records from heterogeneous data sources. Using Health Level 7 (HL7), web technologies and readily available software components, we consolidated the electronic records of Boston's Beth Israel and Deaconess Hospitals. We report on the creation of CareWeb (freya.bidmc.harvard.edu/careweb.htm) and propose it as a means to electronically link Integrated Health Care Delivery Systems and geographically distant information resources.
To help clinicians care for patients with HIV infection, we developed an interactive knowledge-based electronic patient record that integrates rule-based decision support and full-text information retrieval with an online patient record. This highly interactive clinical workstation now allows the clinicians at a large primary care practice (30,000 ambulatory visits per year) to use online information resources and fully electronic patient records during all patient encounters. The resulting practice database is continually updated with outcome data on a cohort of 700 patients with HIV infection. As a byproduct of this integrated system, we have developed improved statistical methods to measure the effects of electronic alerts and reminders.
BACKGROUND: There have been numerous reports indicating a relation between psychological distress and coronary artery disease. The authors tried to determine whether psychological distress in patients hospitalized for coronary artery disease is associated with the amount of medical care required after discharge.
METHODS: Using a prospective clinical cohort, 210 patients who had been admitted for myocardial infarction (n = 67), percutaneous transluminal coronary angioplasty (n = 75), or coronary artery bypass grafting (n = 68) were followed for 6 months. Index psychological status was determined from questionnaires measuring depression and anxiety. Disease severity was assessed by the index hospitalization medical record of left ventricular ejection fraction, number of stenotic vessels, and number of noncardiac comorbidities. The amount of subsequent medical care delivered was based on the number of days of rehospitalization for cardiac-related illness and for any reason within 6 months after discharge. This was determined from a combination of computer medical record and patient self-report.
RESULTS: The authors first determined that both psychological depression and disease severity each predicted days of rehospitalization. (Anxiety was not predictive of rehospitalization.) Next, disease severity was controlled for using partial correlation, and depression was still predictive of rehospitalization. Finally, the authors combined the predictor variables using a regression model to predict rehospitalization. Depression was a significant main effect in all models predicting rehospitalization.
CONCLUSIONS: Psychological depression appears to be an important predictor of rehospitalization among persons who have been admitted with coronary artery disease.
Some observers feel that the federal government should play a more active leadership role in educating the medical community and in coordinating and encouraging a more rapid and effective implementation of clinically relevant applications of wide-area networking. Other people argue that the private sector is recognizing the importance of these issues and will, when the market demands it, adopt and enhance the telecommunications systems that are needed to produce effective uses of the National Information Infrastructure (NII) by the healthcare community. This debate identifies five areas for possible government involvement: convening groups for the development of standards; providing funding for research and development; ensuring the equitable distribution of resources, particularly to places and people considered by private enterprise to provide low opportunities for profit; protecting rights of privacy, intellectual property, and security; and overcoming the jurisdictional barriers to cooperation, particularly when states offer conflicting regulations. Arguments against government involvement include the likely emergence of an adequate infrastructure under free market forces, the often stifling effect of regulation, and the need to avoid a common-and-control mentality in an infrastructure that is best promoted collaboratively.
To meet the needs of primary care physicians caring for patients with HIV infection, we developed a knowledge-based medical record to allow the on-line patient record to play an active role in the care process. These programs integrate the on-line patient record, rule-based decision support, and full-text information retrieval into a clinical workstation for the practicing clinician. To determine whether use of a knowledge-based medical record was associated with more rapid and complete adherence to practice guidelines and improved quality of care, we performed a controlled clinical trial among physicians and nurse practitioners caring for 349 patients infected with the human immuno-deficiency virus (HIV); 191 patients were treated by 65 physicians and nurse practitioners assigned to the intervention group, and 158 patients were treated by 61 physicians and nurse practitioners assigned to the control group. During the 18-month study period, the computer generated 303 alerts in the intervention group and 388 in the control group. The median response time of clinicians to these alerts was 11 days in the intervention group and 52 days in the control group (PJJ0.0001, log-rank test). During the study, the computer generated 432 primary care reminders for the intervention group and 360 reminders for the control group. The median response time of clinicians to these alerts was 114 days in the intervention group and more than 500 days in the control group (PJJ0.0001, log-rank test). Of the 191 patients in the intervention group, 67 (35%) had one or more hospitalizations, compared with 70 (44%) of the 158 patients in the control group (PJ=J0.04, Wilcoxon test stratified for initial CD4 count). There was no difference in survival between the intervention and control groups (P = 0.18, log-rank test). We conclude that our clinical workstation significantly changed physicians' behavior in terms of their response to alerts regarding primary care interventions and that these interventions have led to fewer patients with HIV infection being admitted to the hospital.