Level of Evidence as a Future Gold Standard for the Content Quality of Health Resources on the Internet

Objective: An assessment of the quality of health information on the Internet is an absolute necessity. In this study ‘sensitive’ information was defined as information found in documents published on the Internet, which could be used in a medical decision. For sensitive information, the main criterion chosen for the quality of the information was an indication of the level of evidence. A survey was conducted using the CISMeF health catalogue to assess how often a score of the level of evidence is mentioned in the information accessible on the Internet in French-language health resources.

Methods: Since 1999, members of the CISMeF team have systematically been searching for all documents containing ‘sensitive’ information and verifying whether the level of evidence was explicitly indicated as a score at least once in the document.

Results: As of June 2001, 10,190 resources were included in CISMeF; including 2964 textual ‘sensitive’ resources (29.1%). Out of all these resources, only 4.7% (95% confidence interval: 4.0 - 5.5%) indicated the level of evidence. A statistically significant difference in the prevalence of indicating the level of evidence according to resource types (e.g., 18.1% for guidelines compared to 0.0% for teaching material), year of publication (almost three times greater in 1997-2001 compared with 1990-1996) and publishers was observed.

Conclusion: As the number of people accessing the growing amount of information on the Internet is increasing daily, publishers have an ethical obligation to inform their readers about the validity of ‘sensitive’ information their sites contain. However, the vast majority of the French language Internet resources that were surveyed do not mention a score of the level of evidence for their sensitive information.

CISMeF: A Structured Health Resource Guide

In 1999, the Internet has become a major source of health information. The objective of CISMeF is to catalogue and index the main French-speaking health resources. In September 1999, the number of indexed resources totaled over 7,100 with a mean of 75 new sites per week. CISMeF uses two standard tools for organizing information: the Medline bibliographic database MeSH thesaurus and the Dublin Core metadata format. Resources included in CISMeF are described by the following: title, author or creator, subject and keywords, description, publisher, date, resource type, format, identifier, and language. To index resources, CISMeF uses five levels of hierarchy: “meta-term”, category, keyword, subheading, and resource type. CISMeF contains a thematic index, including medical specialities and an alphabetic index. CISMeF respects the Net Scoring, criteria to assess the quality of health information on the Internet. The CISMeF project offers a valuable tool for the French-speaking health community: 2,500 computer users visit the Web site each working day.

Physician satisfaction with transition from CPOE to paper-based prescription

INTRODUCTION

In January 2015, Rouen University Hospital's information system experienced serious issues. It was necessary to rapidly switch from the computerized provider order entry (CPOE) system towards a paper-based order entry (PBOE) system. This was an opportunity to evaluate prescriber opinion on the two provider order entry (POE) systems.

METHODS

All residents were asked to fill an augmented version of the POE satisfaction and usage survey for both POE systems. The results were compared to identify the strengths and weaknesses of each system.

RESULTS

Fifty-one respondents had used the CPOE system and the PBOE system. Overall, satisfaction was higher with PBOE than CPOE (odds ratio (OR)=3.74; p<0.001). Usability (OR=4.00; p<0.001), reliability (OR=8.54; p<0.001), time consumption (OR=0.50; p<0.05 - survey statement was formulated negatively), and communication with nurses (OR=14.27; p<0.0001) reached statistically better agreement. The more experience with CPOE the more residents were disillusioned with the reliability (OR=6.55; p<0.01), the usability (OR=5.68; p<0.01) and the patient safety (OR=0.27; p<0.05 - survey statement was formulated negatively) of CPOE. Although safety issues were reported for both systems, the causes were different; PBOE imposed frequent rewriting of the order while CPOE lack of usability might be unsafe. Another important issue with both POE systems was time consumption.

CONCLUSION

Residents did not report any increase in safety issues with the rapid switch from CPOE to PBOE. They even seemed more satisfied with the rollback to paper, which remains a possible degraded mode in case of health information technology collapse.

Searching for rare diseases in PubMed: a blind comparison of Orphanet expert query and query based on terminological knowledge

BACKGROUND

Despite international initiatives like Orphanet, it remains difficult to find up-to-date information about rare diseases. The aim of this study is to propose an exhaustive set of queries for PubMed based on terminological knowledge and to evaluate it versus the queries based on expertise provided by the most frequently used resource in Europe: Orphanet.

METHODS

Four rare disease terminologies (MeSH, OMIM, HPO and HRDO) were manually mapped to each other permitting the automatic creation of expended terminological queries for rare diseases. For 30 rare diseases, 30 citations retrieved by Orphanet expert query and/or query based on terminological knowledge were assessed for relevance by two independent reviewers unaware of the query's origin. An adjudication procedure was used to resolve any discrepancy. Precision, relative recall and F-measure were all computed.

RESULTS

For each Orphanet rare disease (n = 8982), there was a corresponding terminological query, in contrast with only 2284 queries provided by Orphanet. Only 553 citations were evaluated due to queries with 0 or only a few hits. There were no significant differences between the Orpha query and terminological query in terms of precision, respectively 0.61 vs 0.52 (p = 0.13). Nevertheless, terminological queries retrieved more citations more often than Orpha queries (0.57 vs. 0.33; p = 0.01). Interestingly, Orpha queries seemed to retrieve older citations than terminological queries (p < 0.0001).

CONCLUSION

The terminological queries proposed in this study are now currently available for all rare diseases. They may be a useful tool for both precision or recall oriented literature search.

Knowledge Representation and Management. From Ontology to Annotation. Findings from the Yearbook 2015 Section on Knowledge Representation and Management

OBJECTIVE

To summarize the best papers in the field of Knowledge Representation and Management (KRM).

METHODS

A comprehensive review of medical informatics literature was performed to select some of the most interesting papers of KRM published in 2014.

RESULTS

Four articles were selected, two focused on annotation and information retrieval using an ontology. The two others focused mainly on ontologies, one dealing with the usage of a temporal ontology in order to analyze the content of narrative document, one describing a methodology for building multilingual ontologies.

CONCLUSION

Semantic models began to show their efficiency, coupled with annotation tools.

[Limiting a Medline/PubMed query to the "best" articles using the JCR relative impact factor]

BACKGROUND

Medline/PubMed is the most frequently used medical bibliographic research database. The aim of this study was to propose a new generic method to limit any Medline/PubMed query based on the relative impact factor and the A & B categories of the SIGAPS score.

MATERIAL AND METHODS

The entire PubMed corpus was used for the feasibility study, then ten frequent diseases in terms of PubMed indexing and the citations of four Nobel prize winners. The relative impact factor (RIF) was calculated by medical specialty defined in Journal Citation Reports. The two queries, which included all the journals in category A (or A OR B), were added to any Medline/PubMed query as a central point of the feasibility study.

RESULTS

Limitation using the SIGAPS category A was larger than the when using the Core Clinical Journals (CCJ): 15.65% of PubMed corpus vs 8.64% for CCJ. The response time of this limit applied to the entire PubMed corpus was less than two seconds. For five diseases out of ten, limiting the citations with the RIF was more effective than with the CCJ. For the four Nobel prize winners, limiting the citations with the RIF was more effective than the CCJ.

CONCLUSION

The feasibility study to apply a new filter based on the relative impact factor on any Medline/PubMed query was positive.

Managing free text for secondary use of health data

OBJECTIVE

To summarize the best papers in the field of Knowledge Representation and Management (KRM).

METHODS

A comprehensive review of medical informatics literature was performed to select some of the most interesting papers of KRM and natural language processing (NLP) published in 2013.

RESULTS

Four articles were selected, one focuses on Electronic Health Record (EHR) interoperability for clinical pathway personalization based on structured data. The other three focus on NLP (corpus creation, de-identification, and co-reference resolution) and highlight the increase in NLP tools performances.

CONCLUSION

NLP tools are close to being seriously concurrent to humans in some annotation tasks. Their use could increase drastically the amount of data usable for meaningful use of EHR.

French Infobutton: an academic and business perspective

Infobuttons are context-specific links between clinical information systems and other online information resources. The objective of this study is to describe a French Infobutton, which will be sold in the French-speaking Health Information market.

Cataloguing and displaying Web feeds from French language health sites: a Web 2.0 add-on to a health gateway

Among the numerous new functionalities of the Internet, commonly called Web 2.0, Web syndication illustrates the trend for better and faster information sharing. Web feeds (a.k.a RSS feeds), which were used mostly on weblogs at first, are now also widely used in academic, scientific and institutional websites such as PubMed. As very few French language feeds were listed or catalogued in the Health field by the year of 2007, it was decided to implement them in the quality-controlled health gateway CISMeF ([French] acronym for Catalogue and Index of French Language Health Resources on the Internet). Furthermore, making full use of the nature of Web syndication, a Web feed aggregator was put online in to provide a dynamic news gateway called "CISMeF actualités" (http://www.chu-rouen.fr/actualites/). This article describes the process to retrieve and implement the Web feeds in the catalogue and how its terminology was adjusted to describe this new content. It also describes how the aggregator was put online and the features of this news gateway. CISMeF actualités was built accordingly to the editorial policy of CISMeF. Only a part of the Web feeds of the catalogue were included to display the most authoritative sources. Web feeds were also grouped by medical specialties and by countries using the prior indexing of websites with MeSH terms and the so-called metaterms. CISMeF actualités now displays 131 Web feeds across 40 different medical specialities, coming from 5 different countries. It is one example, among many, that static hypertext links can now easily and beneficially be completed, or replaced, by dynamic display of Web content using syndication feeds.

Use of super-concepts to customize electronic medical records data display

Patient medical record systems (MRS) merely offer static applications, in which mostly unstructured text is linked to coded data. In these applications the more common presentation is a time oriented one, which does not allow easily for data and information retrieval. Concept oriented views based on supper-concepts (metaterms) initially defined in CISMeF to optimize Web medical search, was implemented in our MRS as specialties views. This work shows that these terminological tools are able to facilitate information retrieval.

Doc’UMVF: Two search tools to provide quality-controlled teaching resources in French to students and teachers

The UMVF project is a federation of medical teaching resources covering 32 medical schools in France. Today, the indexing of these resources is carried out manually by the CISMeF team at the University Hospital of Rouen. This indexing is based on MeSH thesaurus. We use a subset of SCORM metadata standard. This choice was defined in collaboration with the French Medical Virtual University consortium (French acronym: UMVF). Currently, with the UMVF searching tool (called Doc'UMVF), medical students can reach more than 3300 resources useful in their curriculum. Doc'UMVF is developed in close collaboration between the medical informatics laboratories of Rennes and Rouen. In this paper we present two complementary searching tools based on different methods and which are integrated and used to improve both the relevance and the coverage rate of the answers. A specific searching module has been built to retrieve specific resources concerning the National Medical Exam ENC ("Examen National Classant") is also available. Nevertheless, due to lack of time, numerous resources are not yet indexed. Therefore we have decided to use also automatic indexing method (Nomindex). This approach will be improved by further research works, resulting from Rouen and Geneva teams. After having built a searching meta-motor, our objective is to develop a meta-tool intended to index the whole set of digital pedagogical resources produced by the UMVF framework. This manual re-indexing will be carried out only for the most important resources (national references), with a more or less fine granularity.

Seal of transparency heritage in the CISMeF quality-controlled health gateway

Background

It is an absolute necessity to continually assess the quality of health information on the Internet. Quality-controlled subject gateways are Internet services which apply a selected set of targeted measures to support systematic resource discovery.

Methods

The CISMeF health gateway became a contributor to the MedCIRCLE project to evaluate 270 health information providers. The transparency heritage consists of using the evaluation performed on providers that are referenced in the CISMeF catalogue for evaluating the documents they publish, thus passing on the transparency label from the publishers to their documents.

Results

Each site rated in CISMeF has a record in the CISMeF database that generates an RDF into HTML file. The search tool Doc'CISMeF displays information originating from every publisher evaluated with a specific MedCIRCLE button, which is linked to the MedCIRCLE central repository. Starting with 270 websites, this trust heritage has led to 6,480 evaluated resources in CISMeF (49.8% of the 13,012 resources included in CISMeF).

Conclusion

With the MedCIRCLE project and transparency heritage, CISMeF became an explicit third party.

Level of evidence as a future gold standard for the content quality of health resources on the internet

OBJECTIVE

An assessment of the quality of health information on the Internet is an absolute necessity. In this study 'sensitive' information was defined as information found in documents published on the Internet, which could be used in a medical decision. For sensitive information, the main criterion chosen for the quality of the information was an indication of the level of evidence. A survey was conducted using the CISMeF health catalogue to assess how often a score of the level of evidence is mentioned in the information accessible on the Internet in French-language health resources.

METHODS

Since 1999, members of the CISMeF team have systematically been searching for all documents containing 'sensitive' information and verifying whether the level of evidence was explicitly indicated as a score at least once in the document.

RESULTS

As of June 2001, 10,190 resources were included in CISMeF; including 2964 textual 'sensitive' resources (29.1%). Out of all these resources, only 4.7% (95% confidence interval: 4.0 - 5.5%) indicated the level of evidence. A statistically significant difference in the prevalence of indicating the level of evidence according to resource types (e.g., 18.1% for guidelines compared to 0.0% for teaching material), year of publication (almost three times greater in 1997-2001 compared with 1990-1996) and publishers was observed.

CONCLUSION

As the number of people accessing the growing amount of information on the Internet is increasing daily, publishers have an ethical obligation to inform their readers about the validity of 'sensitive' information their sites contain. However, the vast majority of the French language Internet resources that were surveyed do not mention a score of the level of evidence for their sensitive information.

MedCIRCLE: collaboration for Internet rating, certification, labelling and evaluation of health information on the World-Wide-Web

We describe MedCIRCLE, an EU-funded semantic web project to implement the first steps towards a global, collaborative rating and guidance system for health information proposed in the MedCERTAIN project. In MedCIRCLE, three European gateway sites for consumer health information will implement the metadata vocabulary HIDDEL (Health Information Disclosure, Description and Evaluation Language). HIDDEL allows portals and gateways to make the results of their evaluations accessible as XML/RDF. The three participating national portals are: AQUMED (Agency for Quality in Medicine) patienten-information, de, COMB (Official Medical College of Barcelona) and CISMeF, a quality-controlled health gateway developed at Rouen University Hospital. Other health subject gateways, accreditation, or rating services are invited to join the collaboration simply by implementing HIDDEL on their gateways. Widespread implementation HIDDEL will allow intelligent agents or client-side software to harvest statements and opinions about the trustworthiness of other websites, assisting users in selecting trustworthy websites. The MedCIRCLE project builds on, expands and continues work on rating health information on the Internet piloted within the MedCERTAIN project. While MedCERTAIN provided the core technologies and software for rating and "trustmarking" health information, MedCIRCLE is built around these technologies and involves a wider medical community to assess health information, demonstrating the power of collaborative and interoperable evaluations in a semantic web environment. MedCIRCLE is a project with the overall objective to develop and promote technologies able to guide consumers to trustworthy health information on the Internet, to establish a global web of trust for networked health information, and to empower consumers to positively select high quality health information on the web. Other aims include refinement and expansion of HIDDEL, to become a standard vocabulary and interchange format for self- and third-party ratings of health information.

VUMeF: extending the French involvement in the UMLS Metathesaurus

A considerable number of robust vocabularies and thesauri have been developed for the healthcare and biomedical domain. No single vocabulary, however; provides complete coverage of the information needs from a public health perspective. The results of an investigation of vocabulary sources for the development of a comprehensive controlled vocabulary for the public health domain at the Centers for Disease Control and Prevention (CDC) is presented.

Reading factor: a bibliometric tool to manage a virtual library

Among the many bibliometric criteria used to evaluate biomedical journals, the impact factor is the most commonly used. Despite its limitations, it quantifies the influence of a journal on secondary publications. It does not however evaluate the practical usefulness of primary documents. Usefulness is field-related and varies greatly among specialities. We introduce a new bibliographic criterion, the "reading factor", and define it as the ratio between the number of electronic consultations of a particular journal (i.e., number of clicks on a hyper-link) and the mean number of electronic consultations of all the journals studied (itself calculated by dividing the total number of electronic accesses by the number of journals in the database). We describe its observed distribution, relative to that of the impact factor, based on electronic consultation records from our University Hospital medical digital library, where full-text electronic versions of 45 major biomedical journals have been available since December 1997. From this analysis we found no correlation between the 1999 reading factor and the 1998 impact factor of these 45 journals, and we observed a dramatic change in the hierarchy of journals upon using the reading factor as the yardstick rather than the impact factor. Moreover, we describe how using the reading factor has helped in managing the collection of our University Hospital's virtual library. The selection of journals to be discarded from the virtual library for the year 2001 was based on journals' RF values and this process will repeated over the coming years. The reading factor also permits a cost-analysis of a virtual library.

CONCLUSION

The measurement of the reading factor is highly automated, practical and efficient. It appears as a new tool for electronic collection management by librarians, well fitting with economical data.

Doc'CISMEF: a search tool based on "encapsulated" MeSH thesaurus

In the year 2000, the Internet became a major source of health information for the health professional and the Netizen. The objective of Doc'CISMeF (D'C) was to create a powerful generic search tool based on an structured information model which â encapsulates' the MeSH thesaurus to index and retrieve quality health resources on the Internet. To index resources, D'C uses four sections in its information model: 'meta-term', keyword, subheading, and resource type. Two search options are available: simple and advanced. The simple search requires the end-user to input a single term or expression. If this term belongs to the D'C information structure model, it will be exploded. If not, a full-text search is performed. In the advanced search, complex searches are possible combining Boolean operators with meta-terms, keywords, subheadings and resource types. D'C uses two standard tools for organising information: the MeSH thesaurus and the Dublin Core metadata format. Resources included in D'C are described according to the following elements: title, author or creator, subject and keywords, description, publisher, date, resource type, format, identifier, and language.

An assessment of the visibility of MeSH-indexed medical web catalogs through search engines

Manually indexed Internet health catalogs such as CliniWeb or CISMeF provide resources for retrieving high-quality health information. Users of these quality-controlled subject gateways are most often referred to them by general search engines such as Google, AltaVista, etc. This raises several questions, among which the following: what is the relative visibility of medical Internet catalogs through search engines? This study addresses this issue by measuring and comparing the visibility of six major, MeSH-indexed health catalogs through four different search engines (AltaVista, Google, Lycos, Northern Light) in two languages (English and French). Over half a million queries were sent to the search engines; for most of these search engines, according to our measures at the time the queries were sent, the most visible catalog for English MeSH terms was CliniWeb and the most visible one for French MeSH terms was CISMeF.

A preliminary study to assess a French code of ethics for health teaching resources on the Internet

BACKGROUND

Constant assessment of the quality of health information on the Internet is an absolute necessity as peer review is often lacking in this media.

OBJECTIVE

To develop a simple and easy French Code of Ethics, which will enable medical students to judge quality of health information in teaching material available on the Internet.

DESIGN

Three medical informaticians selected ten criteria from previously established codes of ethics from Europe and the USA. This instrument was tested on a sample of 30 health Internet teaching resources.

RESULTS

For the panel of experts, chance corrected inter-observer agreement (kappa) for quality rating ranged from k = -0.19 and k = 0.33, demonstrating poor agreement among the raters.

CONCLUSION

If negative results of this preliminary study are confirmed by further research, this finding may detrimentally affect projects in Europe to accredit or certify Internet health resources.

Cost effectiveness of a medical digital library

The rapid increase in the price of electronic journals has made the optimization of collection management an urgent task. As there is currently no standard procedure for the evaluation of this problem, we applied the Reading Factor (RF), an electronically computed indicator used for consultation of individual articles. The aim of our study was to assess the cost effective impact of modifications in our digital library (i.e. change of access from the Intranet to the Internet or change in editorial policy). The digital OVID library at Rouen University Hospital continues to be cost-effective in comparison with the interlibrary loan costs. Moreover, when electronic versions are offered alongside a limited amount of interlibrary loans, a reduction in library costs was observed.

A search tool based on 'encapsulated' MeSH thesaurus to retrieve quality health resources on the internet

In the year 2001, the Internet has become a major source of health information for the health professional and the Netizen. The objective of Doc' CISMeF (D'C) was to create a powerful generic search tool based on a structured information model which 'encapsulates' the MeSH thesaurus to index and retrieve quality health resources on the Internet. To index resources, D'C uses four sections in its information model: 'meta-term', keyword, subheading, and resource type. Two search options are available: simple and advanced. The simple search requires the end-user to input a single term or expression. If this term belongs to the D'C information structure model, it will be exploded. If not, a full-text search is performed. In the advanced search, complex searches are possible combining Boolean operators with meta-terms, keywords, subheadings and resource types. D'C uses two standard tools for organising information: the MeSH thesaurus and the Dublin Core metadata format. Resources included in D'C are described according to the following elements: title, author or creator, subject and keywords, description, publishers, date, resource type, format, identifier, and language.

The contribution of morphological knowledge to French MeSH mapping for information retrieval

MeSH-indexed Internet health directories must provide a mapping from natural language queries to MeSH terms so that both health professionals and the general public can query their contents. We describe here the design of lexical knowledge bases for mapping French expressions to MeSH terms, and the initial evaluation of their contribution to Doc'CISMeF, the search tool of a MeSH-indexed directory of French-language medical Internet resources. The observed trend is in favor of the use of morphological knowledge as a moderate (approximately 5%) but effective factor for improving query to term mapping capabilities.

A study comparing centralized CD-ROM and decentralized intranet access to MEDLINE

OBJECTIVE

The purpose of this study was to evaluate the efficacy of a decentralized intranet access in each medical department as opposed to centralized unique MEDLINE access in the medical library.

DESIGN

A two-phase questionnaire to evaluate MEDLINE use was given to junior and senior physicians at Rouen University Hospital (RUH). Phase I (August-October 1996) corresponded to a time period when centralized access was the only means of access available and phase II (August-October 1997) to a time period following the introduction of decentralized intranet access.

RESULTS

A total of 168 physicians filled out at least one phase of the questionnaire, among whom 123 (73%) filled out both phases. Use of MEDLINE significantly increased in 1997 (average of 10.2+/-1.1 searches in three months) versus 1996 (average of 4.9+/-0.7 searches in three months, P<0.0001). The aim of searches changed, becoming significantly more care oriented in phase II (P<0.0001). The number of searches performed by the physicians alone increased (P<0.0001) and searches performed by the librarian decreased (P<0.0001) in phase II. The method of searches also changed, as searches by author (P< 0.0001), by journal (P = 0.0042), and by free word (P = 0.0027) increased in phase II. Knowledge of the following concepts of MEDLINE significantly increased: explosion (P<0.0001), scope note (P<0.0001), Abridged Index Medicus (AIM) journals (P<0.0001), Medical Subject Headings (MeSH) qualifier (P<0.0001), and focus (P<0.0001).

CONCLUSION

A decentralized intranet access to MEDLINE increased the number of searches and knowledge of this bibliographic database. MEDLINE intranet access modified the purpose and the methods of searching.

CISMeF: cataloque and index of French speaking health resources

In 1999, the Internet has become a major source of health information. The objective of CISMeF is to catalogue and index the main French-speaking sites and documents concerning health. Currently, the number of resources already totalled over 6,100 with a mean of 75 new sites each week. CISMeF contains a thematic index, including medical specialities and an alphabetic index. CISMeF uses two standard tools for organising information: the MeSH (Medical Subject Heading) thesaurus from the Medline bibliographic database (National Library of Medicine) and the Dublin Core metadata format. A brief description of the site is systematically added. CISMeF respects the Net Scoring, criteria to assess the quality of health information on the Internet. The CISMeF project fulfils a valuable tool for the French-speaking health community: 2,500 machines visit the Web site each working day.

CISMeF: a structured health resource guide

In 1999, the Internet has become a major source of health information. The objective of CISMeF is to catalogue and index the main French-speaking health resources. In September 1999, the number of indexed resources totaled over 7,100 with a mean of 75 new sites per week. CISMeF uses two standard tools for organizing information: the Medline bibliographic database MeSH thesaurus and the Dublin Core metadata format. Resources included in CISMeF are described by the following: title, author or creator, subject and keywords, description, publisher, date, resource type, format, identifier, and language. To index resources, CISMeF uses five levels of hierarchy: "meta-term", category, keyword, subheading, and resource type. CISMeF contains a thematic index, including medical specialities and an alphabetic index. CISMeF respects the Net Scoring, criteria to assess the quality of health information on the Internet. The CISMeF project offers a valuable tool for the French-speaking health community: 2,500 computer users visit the Web site each working day.

[CISMeF: catalog and index of French-speaking medical sites]

The Internet has now become a major source of health information. The aim of CISMeF is to catalogue and index the main French-speaking sites and documents concerning health. This project was initiated by Rouen University Hospital. Its URL is http://www.chu-rouen.fr/cismef. CISMeF covers all areas of health care and medical sciences, and is indexed both alphabetically and according to subject. It was set up on a Sun workstation under the Sun UNIX operating system and is entirely based on static HTML. By May 1999, the number of sites and documents indexed was already over 6,500, with a mean of 75 new sites added each week. CISMeF is updated via a five-step process: resource collection, filtering, description, classification, and indexing. The Net Scoring criteria are used to assess the quality of health information on the Internet. These criteria concern eight categories: credibility, content, links, design, interactivity, quantitative aspects, ethics and accessibility. CISMeF uses two standard tools to organize information: the MeSH (medical subject heading) thesaurus from the Medline reference database (National Library of Medicine, USA) and the Dublin core metadata format. The sites and documents included in CISMeF are described using the following elements from the Dublin core project: title, author or creator, subject and keywords, description, publisher, date, resource type, format, identifier, and language.

SETH: an expert system for the management on acute drug poisoning in adults

The aim of SETH is to give end-users specific advice concerning treatment and monitoring of adult drug poisoning. SETH is developed with an off the shelf expert system shell (KBMS) and runs on a microcomputer. Technical choices were done according to this analysis, financial considerations and portability. Currently, the database contains 1000 French drugs from 75 different toxicological classes. The SETH expert system simulates the expert reasoning, taking into account for each toxicological class delay, signs and dose. Two phases of evaluation were performed. The experimental implementation of Seth began in April 1992 in our Poison Control Centre. Since then, 1100 cases inputted by residents were analysed by SETH. The extension of the knowledge base to child poisoning began in March 1993.

Computer-aided decision support in hepatology

The aim of this study was to describe and to evaluate the publications of the last 30 years devoted to computer-aided decision support in clinical hepatology. The search used Medlars and references of articles. Computer-aided decision support (CADS) was classified in two categories: statistical systems and knowledge-based systems. Two specific questionnaires were used for methodologic evaluation, one for statistical systems and one for knowledge-based systems. They were filled out independently by two observers. A total of 31 papers were selected among 55 identified between 1960 and 1991. The maximum possible for the two scores was 24. The methodologic quality ranged from 4 to 22 (median, 12) for statistical systems and from 8 to 12 (median, 9) for knowledge-based systems. The poor level of methodology could explain in part the lack of utilization of computer-aided decision support in the daily clinical practice of hepatologists.

[Computer-assisted model of weight gain during pregnancy]

The goal of this study was to modelize the evolution of "ideal" weight gain during pregnancy and to generate automatically the appropriate diet. This computerized model has been developed on a microcomputer and has two units: the first unit calculates the "ideal" weight gain during pregnancy, based on the curves of Rosso which show that weight gain is not linear with term and depends of the prepregnancy weight. The second unit calculates the appropriate diet which is depending for the first visit on height, prepregnancy weight and weight gain and for the followed visits on weight gain and the diet situation of the previous visit. The next step will be the medical evaluation of this computer-aided modelization of weight gain during pregnancy.

[Development of an expert system as a teaching aid in the serological diagnosis of acute viral hepatitis]

The serological diagnosis of acute viral hepatitis is complex for both practitioners and students. We have developed an expert system to teach the serological diagnosis of acute viral hepatitis in order to help non-specialists in deciding on the diagnosis. Three levels of increasing difficulty have been devised. The procedure consists of the following steps: choice of first-line examinations; discontinuation of the examinations in cases where the diagnosis is ascertained; choice of the most discriminant examination in cases where the diagnosis is doubtful. The value of this computer-assisted teaching must be confirmed by randomized studies involving a group of students trained by "classical" methods and a group of students trained by a computer-assisted method.

[Value of digestive endoscopic examination in acquired immunodeficiency syndrome (45 cases)]

45 patients with acquired immunodeficiency syndrome (AIDS) were studied with gastrointestinal endoscopy (42 upper endoscopies and 12 colonoscopies). 28 patients had skin or buccal Kaposi's sarcoma with or without opportunistic infections and 17 had opportunistic infections. 12 patients out of 45 (27 per cent) had 1 or several Kaposi's sarcoma macroscopic gastrointestinal localisations, 12 documented by upper endoscopy and 4 by colonoscopy. Endoscopic biopsies confirmed the diagnosis 5 times out of 16 (31 per cent). 16 patients (38 per cent) had candidosis oesophagitis, 1 had ulcerative antritis, 2 had a erythematosus duodenitis, 6 had diffuse inflammatory mucosal colonic changes. The specific abnormalities documented by histology were 1 case of total villous atrophy and 2 cases of cytomegalovirus colitis. The patients with cutaneous Kaposi's sarcoma have more often shown gastrointestinal Kaposi's sarcoma than the patients without Kaposi's sarcoma. The prevalence of tumoral or major (diffuse candidosis oesophagitis, ulcerative antritis, active colitis) endoscopic abnormalities documented by upper endoscopy was 40 per cent (10 cases out of 25) in asymptomatic patients, 43 per cent in diarrheic patients (3 cases out of 7) and 60 per cent (6 cases out of 10) in patients with dysphagia, epigastralgic pain or vomiting and that of documented by colonoscopy was 75 per cent (3 cases out of 4) in patients with mucus or bloody stools and 14 per cent (1 case out of 7) in diarrheic patients. The main result of the endoscopy on AIDS patients has therefore been the diagnosis of visceral localisations of Kaposi's sarcoma. The appearance of the secondary major lesions related to opportunistic infections is unusual, especially in asymptomatic patients. A systematic endoscopy on these patients remains a disputed question.