Accuracy of an automated knowledge base for identifying drug adverse reactions.

INTRODUCTION: Drug safety researchers seek to know the degree of certainty with which a particular drug is associated with an adverse drug reaction. There are different sources of information used in pharmacovigilance to identify, evaluate, and disseminate medical product safety evidence including spontaneous reports, published peer-reviewed literature, and product labels. Automated data processing and classification using these evidence sources can greatly reduce the manual curation currently required to develop reference sets of positive and negative controls (i.e. drugs that cause adverse drug events and those that do not) to be used in drug safety research. METHODS: In this paper we explore a method for automatically aggregating disparate sources of information together into a single repository, developing a predictive model to classify drug-adverse event relationships, and applying those predictions to a real world problem of identifying negative controls for statistical method calibration. RESULTS: Our results showed high predictive accuracy for the models combining all available evidence, with an area under the receiver-operator curve of ⩾0.92 when tested on three manually generated lists of drugs and conditions that are known to either have or not have an association with an adverse drug event. CONCLUSIONS: Results from a pilot implementation of the method suggests that it is feasible to develop a scalable alternative to the time-and-resource-intensive, manual curation exercise previously applied to develop reference sets of positive and negative controls to be used in drug safety research.

Similarity-based modeling applied to signal detection in pharmacovigilance.

One of the main objectives in pharmacovigilance is the detection of adverse drug events (ADEs) through mining of healthcare databases, such as electronic health records or administrative claims data. Although different approaches have been shown to be of great value, research is still focusing on the enhancement of signal detection to gain efficiency in further assessment and follow-up. We applied similarity-based modeling techniques, using 2D and 3D molecular structure, ADE, target, and ATC (anatomical therapeutic chemical) similarity measures, to the candidate associations selected previously in a medication-wide association study for four ADE outcomes. Our results showed an improvement in the precision when we ranked the subset of ADE candidates using similarity scorings. This method is simple, useful to strengthen or prioritize signals generated from healthcare databases, and facilitates ADE detection through the identification of the most similar drugs for which ADE information is available.

Discussion of "Biomedical informatics: we are what we publish".

This article is part of a For-Discussion-Section of Methods of Information in Medicine about the paper "Biomedical Informatics: We Are What We Publish", written by Peter L. Elkin, Steven H. Brown, and Graham Wright. It is introduced by an editorial. This article contains the combined commentaries invited to independently comment on the Elkin et al. paper. In subsequent issues the discussion can continue through letters to the editor.

Medication-wide association studies.

Undiscovered side effects of drugs can have a profound effect on the health of the nation, and electronic health-care databases offer opportunities to speed up the discovery of these side effects. We applied a "medication-wide association study" approach that combined multivariate analysis with exploratory visualization to study four health outcomes of interest in an administrative claims database of 46 million patients and a clinical database of 11 million patients. The technique had good predictive value, but there was no threshold high enough to eliminate false-positive findings. The visualization not only highlighted the class effects that strengthened the review of specific products but also underscored the challenges in confounding. These findings suggest that observational databases are useful for identifying potential associations that warrant further consideration but are unlikely to provide definitive evidence of causal effects.

An inventory of publications on electronic medical records revisited.

OBJECTIVES: In this short review we provide an update of our earlier inventories of publications indexed in MedLine with the MeSH term 'Medical Records Systems, Computerized'. METHODS: We retrieved and analyzed all references to English articles published before January 1, 2008, and indexed in PubMed with the MeSH term 'Medical Records Systems, Computerized'. RESULTS: We retrieved a total of 11,924 publications, of which 3937 (33%) appeared in a journal with an impact factor. Since 2002 the number of yearly publications, and the number of journals in which those publications appeared, increased. A cluster analysis revealed three clusters: an organizational issues cluster, a technically oriented cluster and a cluster about order-entry and research. CONCLUSIONS: Although our previous inventory in 2003 suggested a constant yearly production of publications on electronic medical records since 1998, the current inventory shows another rise in production since 2002. In addition, many new journals and countries have shown interest during the last five years. In the last 15 years, interest in organizational issues remained fairly constant, order entry and research with systems gained attention, while interest in technical issues relatively decreased.

Literature-based priors for gene regulatory networks.

MOTIVATION: The use of prior knowledge to improve gene regulatory network modelling has often been proposed. In this article we present the first research on the massive incorporation of prior knowledge from literature for Bayesian network learning of gene networks. As the publication rate of scientific papers grows, updating online databases, which have been proposed as potential prior knowledge in past research, becomes increasingly challenging. The novelty of our approach lies in the use of gene-pair association scores that describe the overlap in the contexts in which the genes are mentioned, generated from a large database of scientific literature, harnessing the information contained in a huge number of documents into a simple, clear format. RESULTS: We present a method to transform such literature-based gene association scores to network prior probabilities, and apply it to learn gene sub-networks for yeast, Escherichia coli and Human organisms. We also investigate the effect of weighting the influence of the prior knowledge. Our findings show that literature-based priors can improve both the number of true regulatory interactions present in the network and the accuracy of expression value prediction on genes, in comparison to a network learnt solely from expression data. Networks learnt with priors also show an improved biological interpretation, with identified subnetworks that coincide with known biological pathways.

Mapping the domain of medical informatics.

OBJECTIVES: The domain of medical informatics (MI) is not well defined. It covers a wide range of research topics. Our objective is to characterize the field of MI by means of the scientific literature in this domain. METHODS: We used titles and abstracts from MEDLINE records of papers published between July 1993 and July 2008, and extracted uni-, bi- and trigrams as features. Starting with the ISI category of medical informatics, we applied a semi-automated procedure to identify the set of journals and proceedings pertaining to MI. A clustering algorithm was subsequently applied to the articles from this set of publications. RESULTS: MI literature can be divided into three subdomains: 1) the organization, application, and evaluation of health information systems, 2) medical knowledge representation, and 3) signal and data analysis. Over the last fifteen years, the field has remained relatively stable, although most journals have shifted their focus somewhat. CONCLUSIONS: We identified the scientific literature pertaining to the field of MI, and the main areas of research. We were able to show trends in the field, and the positioning of different journals within this field.

Distribution of information in biomedical abstracts and full-text publications.

MOTIVATION: Full-text documents potentially hold more information than their abstracts, but require more resources for processing. We investigated the added value of full text over abstracts in terms of information content and occurrences of gene symbol--gene name combinations that can resolve gene-symbol ambiguity. RESULTS: We analyzed a set of 3902 biomedical full-text articles. Different keyword measures indicate that information density is highest in abstracts, but that the information coverage in full texts is much greater than in abstracts. Analysis of five different standard sections of articles shows that the highest information coverage is located in the results section. Still, 30-40% of the information mentioned in each section is unique to that section. Only 30% of the gene symbols in the abstract are accompanied by their corresponding names, and a further 8% of the gene names are found in the full text. In the full text, only 18% of the gene symbols are accompanied by their gene names.

Virtual reality treatment versus exposure in vivo: a comparative evaluation in acrophobia.

The aim of the present study was to evaluate the effectiveness of low-budget virtual reality (VR) exposure versus exposure in vivo in a between-group design in 33 patients suffering from acrophobia. The virtual environments used in treatment were exactly copied from the real environments used in the exposure in vivo program. VR exposure was found to be as effective as exposure in vivo on anxiety and avoidance as measured with the Acrophobia Questionnaire (AQ), the Attitude Towards Heights Questionnaire (ATHQ) and the Behavioral Avoidance Test (BAT). Results were maintained up to six months follow-up. The present study shows that VR exposure can be effective with relatively cheap hardware and software on stand-alone computers currently on the market. Further studies into the effectiveness of VR exposure are recommended in other clinical groups as agoraphobics and social phobics and studies in which VR exposure is compared with more emerging virtual worlds as presented in CAVE-type systems.

Research on presence in virtual reality: a survey.

Virtual Reality (VR) is starting to be used in psychological therapy around the world. However, a thorough understanding of the reason why VR is effective and what effect it has on the human psyche is still missing. Most research on this subject is related to the concept of presence. This paper gives an up-to-date overview of research in this diverse field. It starts with the most prevailing definitions and theories on presence, most of which attribute special roles for the mental process of attention and for mental models of the virtual space. A review of the phenomena thought to be effected by presence shows that there is still a strong need for research on this subject because little conclusive evidence exists regarding the relationship between presence and phenoma such as emotional responses to virtual stimuli. An investigation shows there has been substantial research for developing methods for measuring presence and research regarding factors that contribute to presence. Knowledge of these contributing factors can play a vital role in development of new VR applications, but key knowledge elements in this area are still missing.