Search

Objectives: The growing use of health information technology in operational settings, along with the maturation of the discipline of biomedical informatics, requires reorganization of educational programs in the field. The objective of this paper is to provide a context and description of the biomedical informatics education program at Oregon Health & Science University.Methods: The details of the program are provided.Results: The paper describes the overall program and its component curricula.Conclusions: OHSU has developed a program that caters to the full spectrum of those who will work in the field, allowing education tailored to their career goals and needs. The maturation of Internet technologies also allow most aspects of the program to be delivered on-line. The informatics field must step up to the challenge of educating the best workforce to achieve our goals for the optimal use of HIT. [ABSTRACT FROM AUTHOR]

Background: Text-mining can assist biomedical researchers in reducing information overload by extracting useful knowledge from large collections of text. We developed a novel text-mining method based on analyzing the network structure created by symbol co-occurrences as a way to extend the capabilities of knowledge extraction. The method was applied to the task of automatic gene and protein name synonym extraction. Results: Performance was measured on a test set consisting of about 50,000 abstracts from one year of MEDLINE. Synonyms retrieved from curated genomics databases were used as a gold standard. The system obtained a maximum F-score of 22.21% (23.18% precision and 21.36% recall), with high efficiency in the use of seed pairs. Conclusion: The method performs comparably with other studied methods, does not rely on sophisticated named-entity recognition, and requires little initial seed knowledge. [ABSTRACT FROM AUTHOR]

The Canon Group is an informal organization of medical informatics researchers who are working on the problem of developing a “deeper” representation formalism for use in exchanging data and developing applications. Individuals in the group represent experts in such areas as knowledge representation and computational linguistics, as well as in a variety of medical subdisciplines. All share the view that current mechanisms for the characterization of medical phenomena are either inadequate (limited or rigid) or idiosyncratic (useful for a specific application but incapable of being generalized or extended). The Group proposes to focus on the design of a general schema for medical- language representation including the specification of the resources and associated procedures required to map language (including standard terminologies) into representations that make all implicit relations “visible,” reveal “hidden attributes,” and generally resolve ambiguous or vague references. The Group is proceeding by examining large numbers of texts (records) in medical sub-domains to identify candidate “concepts” and by attempting to develop general rules and representations for elements such as attributes and values so that all concepts may be expressed uniformly. [ABSTRACT FROM PUBLISHER]

Objective: Despite the proliferation of electronic information retrieval (IR) systems for physicians, their effectiveness has not been well assessed. The purpose of this review is to provide a conceptual framework and to apply the results of previous studies to this framework.Data Sources: All sources of medical informatics and information science literature, including MEDLINE, along with bibliographies of textbooks in these areas, were searched from 1966 to January 1998.Study Selection: All articles presenting either classifications of evaluation studies or their results, with an emphasis on those studying use by physicians.Data Extraction: A framework for evaluation was developed, consisting of frequency of use, purpose of use, user satisfaction, searching utility, search failure, and outcomes. All studies were then assessed based on the framework.Data Synthesis: Due to the heterogeneity and simplistic study designs, no meta-analysis of studies could be done. General conclusions were drawn from data where appropriate. A total of 47 articles were found to include an evaluation component and were used to develop the framework. Of these, 21 articles met the inclusion criteria for 1 or more of the categories in the framework. Most use of IR systems by physicians still occurs with bibliographic rather than full-text databases. Overall use of IR systems occurs just 0.3 to 9 times per physician per month, whereas physicians have 2 unanswered questions for every 3 patients.Conclusions: Studies comparing IR systems with different searching features have not shown that advanced searching methods are significantly more effective than simple text word methods. Most searches retrieve only one fourth to one half of the relevant articles on a given topic and, once retrieved, little is known about how these articles are interpreted or applied. These studies imply that further research and development are needed to improve system utility and performance. [ABSTRACT FROM AUTHOR]

Background: This report is a supplement to an earlier evidence report, Telemedicine for the Medicare Population, which was intended to help policymakers weigh the evidence relevant to coverage of telemedicine services under Medicare. That report focused on telemedicine programs and clinical settings that had been used with or were likely to be applied to Medicare beneficiaries. While we prepared that report, it became apparent that there are also telemedicine studies among non-Medicare beneficiaries--e.g., children and pregnant women--that could inform policymakers and provide more comprehensive evidence of the state of the science regarding telemedicine applications. In addition, the first evidence report only partially included a class of telemedicine applications (called self-monitoring/testing telemedicine) in which the beneficiary used a home computer or modern-driven telephone system to either report information or access information and support from Internet resources and indirectly interact with a clinician. Self-monitoring/testing applications in the first report required direct interaction with a clinician. The goal of this report is to systematically review the evidence in the clinical areas of pediatric and obstetric telemedicine as well as home-based telemedicine where there is indirect involvement of the health care professional. (In this report, we will refer to the latter as clinician-indirect home telemedicine.) Specifically, the report summarizes scientific evidence on the diagnostic accuracy, access, clinical outcomes, satisfaction, and cost-effectiveness of services provided by telemedicine technologies for these patient groups. It also identifies gaps in the evidence and makes recommendations for evaluating telemedicine services for these populations in the future. The evidence is clustered according to three categories of telemedicine service defined in our original report: store-and-forward, self-monitoring/testing, and clinician-interactive services. The three clinical practice areas reviewed in this report are defined as follows. The term pediatric applies to any telemedicine study in which the sample consisted wholly or partially of persons aged 18 or younger, including studies with neonatal samples. The term obstetric applies to any telemedicine study in which the sample consisted entirely of women seeking pregnancy-related care. The term clinician-indirect home telemedicine applies to home-based telemedicine (called self-monitoring/testing in our original report) where a telemedicine application used in the home has only indirect involvement by the health care professional. Interactive home telemedicine was applied in this report to all patient populations., Key Questions: The key questions that served as a guide for reviewing the literature in the evaluation of pediatric, obstetric, and clinician-indirect home telemedicine applications were derived by consensus among the evidence-review team based on the analytic framework established for the original evidence report. For the current report, the questions were applied to studies in all three practice areas as a whole group within each of the three categories of telemedicine services: store-and-forward; self-monitoring/testing; and clinician-interactive. The specific key questions were: 1. Does telemedicine result in comparable diagnosis and appropriateness of recommendations for management? 2. Does the availability of telemedicine provide comparable access to care? 3. Does telemedicine result in comparable health outcomes? 4. Does telemedicine result in comparable patient or clinician satisfaction with care? 5. Does telemedicine result in comparable costs of care and/or cost-effectiveness?, Methods: We searched for peer-reviewed literature using several bibliographic databases. In addition, we conducted hand searches of leading telemedicine journals and identified key papers from the reference lists of journal articles. For our original evidence report on telemedicine for the Medicare population, we designed a search to find any publications about telemedicine and used it to search the MEDLINE, CINAHL, and HealthSTAR databases for all years the databases were available. Through this process, we captured studies of pediatric, obstetric, and clinician-indirect home telemedicine; however, they were excluded from the original report since they were outside its scope. For this supplemental report, we reviewed our original search results and identified studies relevant to this report. We identified additional studies from the reference lists of included papers and from hand searching two peer-reviewed telemedicine publications, the Journal of Telemedicine and Telecare and Telemedicine Journal. We critically appraised the included studies for each study area and key question and discussed the strengths and limitations of the most important studies at weekly meetings of the research team. We also developed recommendations for research to address telemedicine knowledge gaps. To match these gaps with the capabilities of specific research methods, we classified the telemedicine services according to the type of evidence that would be needed to determine whether the specific goals of covering such services had been met. We emphasized the relationship between the type and level of evidence found in the systematic review of effectiveness and the types of studies that might be funded to address the gaps in knowledge in this growing field of research., Findings: We identified a total of 28 eligible studies. In the new clinical areas, we found few studies in store-and-forward telemedicine. There is some evidence of comparable diagnosis and management decisions made using store-and-forward telemedicine from the areas of pediatric dental screening, pediatric ophthalmology, and neonatalogy. In self-monitoring/testing telemedicine for the areas of pediatrics, obstetrics, and clinician-indirect home telemedicine, there is evidence that access to care can be improved when patients and families have the opportunity to receive telehealth care at home rather than in-person care in a clinic or hospital. Access is particularly enhanced when the telehealth system enables timely communication between patients or families and care providers that allows self-management and necessary adjustments that may prevent hospitalization. There is some evidence that this form of telemedicine improves health outcomes, but the study sample sizes are usually small, and even when they are not, the treatment effects are small. There is also some evidence for the efficacy of clinician-interactive telemedicine, but the studies do not clearly define which technologies provide benefit or cost-efficiency. Some promising areas for diagnosis include emergency medicine, psychiatry, and cardiology. Most of the studies measuring access to care provide evidence that it is improved. Although none of these studies were randomized controlled trials, they provide some evidence of access improvement over prior conditions. Clinician-interactive telemedicine was the only area for which any cost studies were found. The three cost studies did not adequately demonstrate that telemedicine reduces costs of care (except comparing only selected costs). No study addressed cost-effectiveness., Conclusions: This supplemental report covering the areas of pediatrics, obstetrics, and indirect-clinician home telemedicine echoes the findings of our initial report for the Medicare domain, which is that while the use of telemedicine is small but growing, the evidence for its efficacy is incomplete. Many of the studies are small and/or methodologically limited, so it cannot be determined whether telemedicine is efficacious. Future studies should focus on the use of telemedicine in conditions where burden of illness and/or barriers to access for care are significant. Use of recent innovations in the design of randomized controlled trials for emerging technologies would lead to higher quality studies. Journals publishing telemedicine evaluation studies must set high standards for methodologic quality so that evidence reports need not rely on studies with marginal methodologies.

Background: The use of telemedicine is growing, but its efficacy for achieving comparable or improved clinical outcomes has not been established in many medical specialties. The objective of this systematic review was to evaluate the efficacy of telemedicine interventions for health outcomes in two classes of application: home-based and office/hospital-based., Methods: Data sources for the study included deports of studies from the MEDLINE, EMBASE, CINAHL, and HealthSTAR databases; searching of bibliographies of review and other articles; and consultation of printed resources as well as investigators in the field. We included studies that were relevant to at least one of the two classes of telemedicine and addressed the assessment of efficacy for clinical outcomes with data of reported results. We excluded studies where the service did not historically require face-to-face encounters (e.g., radiology or pathology diagnosis). All included articles were abstracted and graded for quality and direction of the evidence., Results: A total of 25 articles met inclusion criteria and were assessed. The strongest evidence for the efficacy of telemedicine in clinical outcomes comes from home-based telemedicine in the areas of chronic disease management, hypertension, and AIDS. The value of home glucose monitoring in diabetes mellitus is conflicting. There is also reasonable evidence that telemedicine is comparable to face-to-face care in emergency medicine and is beneficial in surgical and neonatal intensive care units as well as patient transfer in neurosurgery., Conclusions: Despite the widespread use of telemedicine in virtually all major areas of health care, evidence concerning the benefits of its use exists in only a small number of them. Further randomized controlled trials must be done to determine where its use is most effective.

Objectives: Despite the growing use of online databases by clinicians, there has been very little research documenting how effectively they are used. This study assessed the ability of medical and nurse-practitioner students to answer clinical questions using an information retrieval system. It also attempted to identify the demographic, experience, cognitive, personality, search mechanics, and user-satisfaction factors associated with successful use of a retrieval system., Methods: Twenty-nine students completed questionnaires of clinical and computer experience as well as tests of cognitive abilities and personality type. They were then administered three clinical questions to answer in a medical library setting using the MEDLINE database and electronic and print full-text resources., Results: Medical students were able to answer more questions correctly than nurse-practitioner students before and after searching, but both had comparable improvements in the number of correct questions before and after searching. Successful ability to answer questions was also associated with having experience in literature searching and higher standardized test-score percentiles., Conclusions: Medical and nurse-practitioner students obtained comparable benefits in the ability to answer clinical questions from use of the information retrieval system. Future research must examine strategies that improve successful search and retrieval of clinical questions posed by clinicians in practice.

Objective: Describe the perceptions of housestaff physicians about their experience using computerized physician order entry (POE) in hospitals., Methods: Qualitative study using data from participant observation, focus groups, and both formal and informal interviews. Data were analyzed by three researchers using a grounded approach to identify patterns and themes in the texts., Results: Six themes were identified, including housestaff education, benefits of POE, problems with POE, feelings about POE, implementation strategies, and the future of POE., Conclusion: House officers felt that POE assists patient care but may undermine education. They found that POE works best when tailored to fit local and individual workflow. Implementation strategies should include mechanisms for engaging housestaff in the decision process.

The World Wide Web is an increasingly popular source for consumer health information, but many authors have expressed concerns about the quality of health information present on the Internet. We have developed a prototype system that responds to a consumer health query by returning a list of Web pages that are ranked according to the likely quality of the page contents. A computer program identifies some of the criteria that have been suggested for assessing the quality of health information on the Internet. It also identifies characteristics that may serve as proxies for desirable (or undesirable) qualities that are difficult to assess directly using an algorithm. Intervening in the search process and automatically analyzing the contents of each page returned by a general search engine may facilitate the search for high quality consumer health information on the Web.

Objective: Determine the percent of U.S. hospitals where computerized physician order entry (POE) is available and the extent of its use., Methods: A survey was sent to a systematic sample of 1,000 U.S. hospitals asking about availability of POE, whether usage is required, percent of physicians using it, and percent of orders entered by computer., Results: About 66% do not have POE available. Of the 32.1% that have it completely or partially available, 4.9% require its usage, over half report usage by under 10% of physicians, and over half report that fewer than 10% of orders are entered this way. Analysis of comments showed that many hospitals have POE available for use by non-physicians only, but that they hope to offer it to physicians after careful planning., Conclusion: Most U.S. hospitals have not yet implemented POE. Complete availability throughout the hospital is rare, very few require its use, low percentages of physicians are actual users, and low percentages of orders are entered this way. On a national basis, computerized order entry by physicians is not yet widespread.

The world's foremost medical literature is written in English, yet much of the world does not speak English as a primary language. This has led to increasing research interest in cross-language information retrieval, where textual databases are queried in languages other than the one in which they are written. We describe enhancements to the SAPHIRE concept-retrieval system, which maps free-text documents and queries to concepts in the UMLS Metathesaurus, that allow it to accept text input and provide Metathesaurus concept output in any of six languages: English, German, French, Russian, Spanish, and Portuguese. An example of the use of SAPHIRE International is shown in the CliniWeb catalogue of clinically-oriented Web pages. A formative evaluation of German terms shows that additional work is required in handling plural and other suffix variants as well as expanding the breadth of synonyms in the UMLS Metathesaurus.

Objective: To identify search strategies for retrieving high quality review studies about etiology, prognosis, therapy, and diagnosis from World Wide Web (WWW) medical documents., Design: Observational study of the performance of search strategies based on terms found in high quality review articles in a collection of hypertext medical documents from the WWW., Measurements: The sensitivity and specificity of search strategies for review articles in general and with a specific focus were determined by comparison to a manual review of a collection of hypertext medical documents., Results: A total of 1058 hypertext medical documents from seven governmental and academic WWW sites were included in the study collection. About 16% of the documents in the collection met the criteria for high quality review documents. Search strategies for review documents were identified that had 87% sensitivity and 95% specificity. Compared to simple strategies combining the term "review" and the article focus, more complex strategies based on terms found in high quality studies were more sensitive in identifying review articles of a given focus. These more complex strategies had a sensitivity of 83% for diagnosis, 85% for therapy, 79% for prognosis, and 88% for etiology, while the simple strategies had a sensitivity of 88%, 74%, 38%, and 46%, respectively. In addition, the more complex strategies were more specific for high quality review articles on diagnosis and therapy., Conclusion: Search strategies can be identified that enhance retrieval of review documents and review documents of specific focus from a collection of WWW hypertext medical documents.

Current natural language processing techniques for recognition of concepts in the electronic medical record have been insufficient to allow their broad use for coding information automatically. We have undertaken a preliminary investigation into the use of machine learning methods to recognize procedure codes from emergency room dictations for a trauma registry. Our preliminary results indicate moderate success, and we believe future enhancements with additional learning techniques and selected natural language processing approaches will be fruitful.

Objective: Identify the lexical content of a large corpus of ordinary medical records to assess the feasibility of large-scale natural language processing., Methods: A corpus of 560 megabytes of medical record text from an academic medical center was broken into individual words and compared with the words in six medical vocabularies, a common word list, and a database of patient names. Unrecognized words were assessed for algorithmic and contextual approaches to identifying more words, while the remainder were analyzed for spelling correctness., Results: About 60% of the words occurred in the medical vocabularies, common word list, or names database. Of the remainder, one-third were recognizable by other means. Of the remaining unrecognizable words, over three-fourths represented correctly spelled real words and the rest were misspellings., Conclusions: Large-scale generalized natural language processing methods for the medical record will require expansion of existing vocabularies, spelling error correction, and other algorithmic approaches to map words into those from clinical vocabularies.

Integrating functions from disparate and widely-distributed information systems has been an interest of the medical informatics community for some time. Barriers to progress have included the lack of network-accessible information sources, inadequate methods for inter-system messaging, and lack of vocabulary translation services. With the advent of the World Wide Web (WWW) and the evolution of the National Library of Medicine's Unified Medical Language System (UMLS), it is now possible to develop applications that integrate functions from diverse, distributed systems. In this paper we describe one such system, MedWeaver, a WWW application that integrates functions from a decision support application (DXplain), a literature searching system (WebMedline), and a clinical Web searching system (CliniWeb) using the UMLS Metathesaurus for vocabulary translation. This system demonstrates how application developers can design systems around anticipated clinical information needs and then draw together the needed content and functionality from diverse sources.

We evaluated the ability of two natural language parsers, CLARIT and the Xerox Tagger, to identify simple, noun phrases in medical discharge summaries. In twenty randomly selected discharge summaries, there were 1909 unique simple noun phrases. CLARIT and the Xerox Tagger exactly identified 77.0% and 68.7% of the phrases, respectively, and partially identified 85.7% and 80.8% of the phrases. Neither system had been specially modified or tuned to the medical domain. These results suggest that it is possible to apply existing natural language processing (NLP) techniques to large bodies of medical text, in order to empirically identify the terminology used in medicine. Virtually all the noun phrases could be regarded as having special medical connotation and would be candidates for entry into a controlled medical vocabulary.

Clinical practice guidelines will be more beneficial when they are integrated with the electronic medical record. We applied natural language processing (NLP) techniques to extract clinical findings from outpatient progress notes in an attempt to: 1) select outpatient chart notes appropriate for evaluation against the National Heart, Lung, and Blood/National Asthma Education Program clinical practice guidelines for the diagnosis and management of asthma, 2) determine patient need for inhaled anti-inflammatory agents, and 3) quantify the severity of asthma. Our results were compared to judgements by an expert panel of practitioners. We were able to correctly identify the need for inhaled anti-inflammatory agents 76% of the time. The success of this pilot project could have broad implications for the application of clinical practice guidelines.

A major impediment to the full benefit of electronic medical records is the lack of a comprehensive clinical vocabulary. Most existing vocabularies do not allow the full expressiveness of clinical diagnoses and findings that are often qualified by modifiers relating to severity, acuity, and temporal factors. One reason for the lack of expressivity is the inability of traditional manual construction techniques to identify the diversity of language used by clinicians. This study used advanced natural language processing tools to identify terminology in a clinical findings domain, compare its coverage with the UMLS Metathesaurus, and quantify the effort required to discover the additional terminology. It was found that substantial amounts of phrases and individual modifiers were not present in the UMLS Metathesaurus and that modest effort in human time and computer processing were needed to obtain the larger quantity of terms.

We discuss the development and evaluation of an automated procedure for extracting drug-dosage information from clinical narratives. The process was developed rapidly using existing technology and resources, including categories of terms from UMLS96. Evaluations over a large training and smaller test set of medical records demonstrate an approximately 80% rate of exact and partial matches' on target phrases, with few false positives and a modest rate of false negatives. The results suggest a strategy for automating general concept identification in electronic medical records.

All of the methods currently used to assess information retrieval (IR) systems have limitations in their ability to measure how well users are able to acquire information. We utilized a new approach to assessing information obtained, based on a short-answer test given to senior medical students. Students took the ten-question test and then searched one of two IR systems on the five questions for which they were least certain of their answer Our results showed that pre-searching scores on the test were low but that searching yielded a high proportion of answers with both systems. These methods are able to measure information obtained, and will be used in subsequent studies to assess differences among IR systems.

The objective of this study was to compare how well medical professionals are able to retrieve relevant literature references using two computerized literature searching systems that provide automated (non-human) indexing of content. The first program was SAPHIRE, which features concept-based indexing, free-text input of queries, and ranking of retrieved references for relevance. The second program was SWORD, which provides single-word searching using Boolean operators (AND, OR). Sixteen fourth-year medical students participated in the study. The database for searching was six volumes from the 1989 Yearbook series. The queries were ten questions generated on teaching rounds. All subjects searched half the queries with each program. After the searching, each subject was given a questionnaire about prior experience and preferences about the two programs. Recall (proportion of relevant articles retrieved from the database) and precision (proportion of relevant articles in the retrieved set) were measured for each search done by each participant. Mean recall was 57.6% with SAPHIRE; it was 58.6% with SWORD. Precision was 48.1% with SAPHIRE vs 57.6% with SWORD. Each program was rated easier to use than the other by half of the searchers, and preferences were associated with better searching performance for that program. Both systems achieved recall and precision comparable to existing systems and may represent effective alternatives to MEDLINE and other retrieval systems based on human indexing for searching medical literature.

Conventional approaches to indexing medical literature include the human assignment of terms from a controlled vocabulary, such as MeSH, or the computer assignment of all words in the title and abstract as indexing terms. Human indexing suffers from inconsistency, while word-based indexing suffers from the multiple meanings of words. SAPHIRE is a computer program designed to provide indexing using controlled terms that are assigned by computer, based on their occurrence in the title and abstract. In this first evaluation of SAPHIRE, the authors compared the retrieval performance of the three indexing approaches--human-based MEDLINE with text words; machine-based SAPHIRE with text words; and text words only--for searches by measuring recall and precision for each search using a test collection of 200 abstracts. The abstracts were judged by human reviewers for relevance as applied to 12 literature queries. The results suggest that text word indexing is more effective than indexing with MeSH terms. SAPHIRE's indexing performance was slightly inferior but the program has other advantageous features.

What is the best way to represent the content of documents in an information retrieval system? This study compares the retrieval effectiveness of five different methods for automated (machine-assigned) indexing using three test collections. The consistently best methods are those that use indexing based on the words that occur in the available text of each document. Methods used to map text into concepts from a controlled vocabulary showed no advantage over the word-based methods. This study also looked at an approach to relevance feedback which showed benefit for both word-based and concept-based methods.

The authors describe CLARIT processing as an approach to automatic indexing. They also explore two elements of the indexing process, identifying concepts in text and selecting concepts to reflect the perspective of a domain.

SAPHIRE is a concept-based approach to information retrieval in the biomedical domain. Indexing and retrieval are based on a concept-matching algorithm that processes free text to identify concepts and map them to their canonical form. This process requires a large vocabulary containing a breadth of medical concepts and a diversity of synonym forms, which is provided by the Meta-1 vocabulary from the Unified Medical Language System Project of the National Library of Medicine. This paper describes the use of Meta-1 in SAPHIRE and an evaluation of both entities in the context of an information retrieval study.

SAPHIRE (Semantic and Probabilistic Heuristic Information Retrieval Environment) is an experimental computer program designed to test new techniques in automated information retrieval in the biomedical domain. A main feature of the program is a concept-finding algorithm that processes free text to find canonical concepts. The algorithm is designed to handle a wide variety of synonyms and convert them to canonical form. This allows natural language to be used for query input and also serves as the basis for a new approach to automatic indexing based on a combination of probabilistic and linguistic methods.

Conventional information retrieval systems usually involve searching by terms from controlled vocabularies or by individual words in the text. These systems have been commercially successful but are limited by several problems, including cumbersome interfaces and inconsistency with human indexing. Research on methods that automate indexing and retrieval has been performed to address these problems. The three major types of automated systems are vector-based, probabilistic, and linguistic. This article describes these systems and provides an overview of the field of information retrieval in medicine.