Heterogeneous but "standard" coding systems for adverse events: Issues in achieving interoperability between apples and oranges

Knowledge-guided generative artificial intelligence for automated taxonomy learning from drug labels

OBJECTIVES

To automatically construct a drug indication taxonomy from drug labels using generative Artificial Intelligence (AI) represented by the Large Language Model (LLM) GPT-4 and real-world evidence (RWE).

MATERIALS AND METHODS

We extracted indication terms from 46 421 free-text drug labels using GPT-4, iteratively and recursively generated indication concepts and inferred indication concept-to-concept and concept-to-term subsumption relations by integrating GPT-4 with RWE, and created a drug indication taxonomy. Quantitative and qualitative evaluations involving domain experts were performed for cardiovascular (CVD), Endocrine, and Genitourinary system diseases.

RESULTS

2909 drug indication terms were extracted and assigned into 24 high-level indication categories (ie, initially generated concepts), each of which was expanded into a sub-taxonomy. For example, the CVD sub-taxonomy contains 242 concepts, spanning a depth of 11, with 170 being leaf nodes. It collectively covers a total of 234 indication terms associated with 189 distinct drugs. The accuracies of GPT-4 on determining the drug indication hierarchy exceeded 0.7 with "good to very good" inter-rater reliability. However, the accuracies of the concept-to-term subsumption relation checking varied greatly, with "fair to moderate" reliability.

DISCUSSION AND CONCLUSION

We successfully used generative AI and RWE to create a taxonomy, with drug indications adequately consistent with domain expert expectations. We show that LLMs are good at deriving their own concept hierarchies but still fall short in determining the subsumption relations between concepts and terms in unregulated language from free-text drug labels, which is the same hard task for human experts.

Characterization of adverse events in injured patients at risk of hemorrhagic shock: a secondary analysis of three harmonized prehospital randomized clinical trials

Background

The reporting of adverse events (AEs) is required and well defined in the execution of clinical trials, but is poorly characterized particularly in prehospital trials focusing on traumatic injury. In the setting of prehospital traumatic injury trials, no literature currently exists analyzing the clinical implications of AEs and their associations with mortality and morbidity. We sought to analyze AEs from three prehospital hemorrhagic shock trials and characterize their time course, incidence, severity, associated clinical outcomes, and relatedness.

Methods

We performed a secondary analysis of three prehospital randomized clinical trials. We analyzed AEs at both the patient level as well as the individual AE level. We categorized patients who had no AEs, a single documented AE and those with multiple events (>1 AE). We characterized AE timing, severity, relatedness and attributable mortality outcomes.

Results

We included 1490 patients from the three harmonized clinical trials, with 299 (20.1%) individual patients having at least a single AE documented with 529 AEs documented overall as a proportion of patients had multiple events. Over 44% of patients had a death-related misclassified AE. Patients with at least a single documented AE had a significantly higher 28-day mortality (log-rank χ2=81.27, p<0.001) compared with those without an AE documented. Patients with a single AE had a significant higher mortality than those with multiple AEs, potentially due to survival bias (log-rank χ2=11.80, p=0.006). When relatedness of each individual AE was characterized, over 97% of AEs were classified as 'definitely not related' or 'probably not related' to the intervention.

Conclusions

AEs in hemorrhagic shock trials are common, occur early and are associated with mortality and survival bias. The potential for inaccurate reporting exists, and education and training remain essential for appropriate treatment arm comparison. The current results have important relevance to injury-related clinical trials.

Trial registration numbers

NCT01818427, NCT02086500 and NCT03477006.

Level of evidence

II.

Pilot randomized controlled trial of restricted versus liberal crystalloid fluid management in pediatric post-operative and trauma patients

BACKGROUND

Intravenous (IV) fluid therapy is essential in the treatment of critically ill pediatric surgery and trauma patients. Recent studies have suggested that aggressive fluids may be detrimental to patients. Prospective studies are needed to compare liberal to restricted fluid management in these patients. The primary objective of this pilot trial is to test study feasibility-recruitment and adherence to the study treatment algorithm.

METHODS

We conducted a two-part pilot randomized controlled trial (RCT) comparing liberal to restricted crystalloid fluid management in 50 pediatric post-operative (1-18 years) and trauma (1-15 years) patients admitted to our pediatric intensive care unit (PICU). Patients were randomized to a high (liberal) volume or low (restricted) volume algorithm using unblinded, blocked randomization. A revised treatment algorithm was used after the 29th patient for the second part of the RCT. The goal of the trial was to determine the feasibility of conducting an RCT at a single site for recruitment and retention. We also collected data on the safety of study interventions and clinical outcomes, including pulmonary, infectious, renal, post-operative, and length of stay outcomes.

RESULTS

Fifty patients were randomized to either liberal (n = 26) or restricted (n = 24) fluid management strategy. After data was obtained on 29 patients, a first study analysis was performed. The volume of fluid administered and triggers for intervention were adapted to optimize the treatment effect and clarity of outcomes. Updated and refined fluid management algorithms were created. These were used for the second part of the RCT on patients 30-50. During this second study period, 54% (21/39, 95% CI 37-70%) of patients approached were enrolled in the study. Of the patients enrolled, 71% (15/21, 95% CI 48-89%) completed the study. This met our a priori recruitment and retention criteria for success. A data safety monitoring committee concluded that no adverse events were related to study interventions. Although the study was not powered to detect differences in outcomes, after the algorithm was revised, we observed a non-significant trend towards improved pulmonary outcomes in patients on the restricted arm, including decreased need for and time on oxygen support and decreased need for mechanical ventilation.

CONCLUSION

We demonstrated the feasibility and safety of conducting a single-site RCT comparing liberal to restricted crystalloid fluid management in critically ill pediatric post-operative and trauma patients. We observed trends in improved pulmonary outcomes in patients undergoing restricted fluid management. A definitive multicenter RCT comparing fluid management strategies in these patients is warranted.

TRIAL REGISTRATION

ClinicalTrials.gov, NCT04201704 . Registered 17 December 2019-retrospectively registered.

Benefit-risk assessment and reporting in clinical trials of chronic pain treatments: IMMPACT recommendations

ABSTRACT

Chronic pain clinical trials have historically assessed benefit and risk outcomes separately. However, a growing body of research suggests that a composite metric that accounts for benefit and risk in relation to each other can provide valuable insights into the effects of different treatments. Researchers and regulators have developed a variety of benefit-risk composite metrics, although the extent to which these methods apply to randomized clinical trials (RCTs) of chronic pain has not been evaluated in the published literature. This article was motivated by an Initiative on Methods, Measurement, and Pain Assessment in Clinical Trials consensus meeting and is based on the expert opinion of those who attended. In addition, a review of the benefit-risk assessment tools used in published chronic pain RCTs or highlighted by key professional organizations (ie, Cochrane, European Medicines Agency, Outcome Measures in Rheumatology, and U.S. Food and Drug Administration) was completed. Overall, the review found that benefit-risk metrics are not commonly used in RCTs of chronic pain despite the availability of published methods. A primary recommendation is that composite metrics of benefit-risk should be combined at the level of the individual patient, when possible, in addition to the benefit-risk assessment at the treatment group level. Both levels of analysis (individual and group) can provide valuable insights into the relationship between benefits and risks associated with specific treatments across different patient subpopulations. The systematic assessment of benefit-risk in clinical trials has the potential to enhance the clinical meaningfulness of RCT results.

The Unified Medical Language System at 30 Years and How It Is Used and Published: Systematic Review and Content Analysis

BACKGROUND

The Unified Medical Language System (UMLS) has been a critical tool in biomedical and health informatics, and the year 2021 marks its 30th anniversary. The UMLS brings together many broadly used vocabularies and standards in the biomedical field to facilitate interoperability among different computer systems and applications.

OBJECTIVE

Despite its longevity, there is no comprehensive publication analysis of the use of the UMLS. Thus, this review and analysis is conducted to provide an overview of the UMLS and its use in English-language peer-reviewed publications, with the objective of providing a comprehensive understanding of how the UMLS has been used in English-language peer-reviewed publications over the last 30 years.

METHODS

PubMed, ACM Digital Library, and the Nursing & Allied Health Database were used to search for studies. The primary search strategy was as follows: UMLS was used as a Medical Subject Headings term or a keyword or appeared in the title or abstract. Only English-language publications were considered. The publications were screened first, then coded and categorized iteratively, following the grounded theory. The review process followed the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines.

RESULTS

A total of 943 publications were included in the final analysis. Moreover, 32 publications were categorized into 2 categories; hence the total number of publications before duplicates are removed is 975. After analysis and categorization of the publications, UMLS was found to be used in the following emerging themes or areas (the number of publications and their respective percentages are given in parentheses): natural language processing (230/975, 23.6%), information retrieval (125/975, 12.8%), terminology study (90/975, 9.2%), ontology and modeling (80/975, 8.2%), medical subdomains (76/975, 7.8%), other language studies (53/975, 5.4%), artificial intelligence tools and applications (46/975, 4.7%), patient care (35/975, 3.6%), data mining and knowledge discovery (25/975, 2.6%), medical education (20/975, 2.1%), degree-related theses (13/975, 1.3%), digital library (5/975, 0.5%), and the UMLS itself (150/975, 15.4%), as well as the UMLS for other purposes (27/975, 2.8%).

CONCLUSIONS

The UMLS has been used successfully in patient care, medical education, digital libraries, and software development, as originally planned, as well as in degree-related theses, the building of artificial intelligence tools, data mining and knowledge discovery, foundational work in methodology, and middle layers that may lead to advanced products. Natural language processing, the UMLS itself, and information retrieval are the 3 most common themes that emerged among the included publications. The results, although largely related to academia, demonstrate that UMLS achieves its intended uses successfully, in addition to achieving uses broadly beyond its original intentions.

Challenges in Using a Graph Database to Represent and Analyze Mappings of Cancer Study Data Standards

While using data standards can facilitate research by making it easier to share data, manually mapping to data standards creates an obstacle to their adoption. Semi-automated mapping strategies can reduce the manual mapping burden. Machine learning approaches, such as artificial neural networks, can predict mappings between clinical data standards but are limited by the need for training data. We developed a graph database that incorporates the Biomedical Research Integrated Domain Group (BRIDG) model, Common Data Elements (CDEs) from the National Cancer Institute's (NCI) cancer Data Standards Registry and Repository, and the NCI Thesaurus. We then used a shortest path algorithm to predict mappings from CDEs to classes in the BRIDG model. The resulting graph database provides a robust semantic framework for analysis and quality assurance testing. Using the graph database to predict CDE to BRIDG class mappings was limited by the subjective nature of mapping and data quality issues.

Using an artificial neural network to map cancer common data elements to the biomedical research integrated domain group model in a semi-automated manner

BACKGROUND

The medical community uses a variety of data standards for both clinical and research reporting needs. ISO 11179 Common Data Elements (CDEs) represent one such standard that provides robust data point definitions. Another standard is the Biomedical Research Integrated Domain Group (BRIDG) model, which is a domain analysis model that provides a contextual framework for biomedical and clinical research data. Mapping the CDEs to the BRIDG model is important; in particular, it can facilitate mapping the CDEs to other standards. Unfortunately, manual mapping, which is the current method for creating the CDE mappings, is error-prone and time-consuming; this creates a significant barrier for researchers who utilize CDEs.

METHODS

In this work, we developed a semi-automated algorithm to map CDEs to likely BRIDG classes. First, we extended and improved our previously developed artificial neural network (ANN) alignment algorithm. We then used a collection of 1284 CDEs with robust mappings to BRIDG classes as the gold standard to train and obtain the appropriate weights of six attributes in CDEs. Afterward, we calculated the similarity between a CDE and each BRIDG class. Finally, the algorithm produces a list of candidate BRIDG classes to which the CDE of interest may belong.

RESULTS

For CDEs semantically similar to those used in training, a match rate of over 90% was achieved. For those partially similar, a match rate of 80% was obtained and for those with drastically different semantics, a match rate of up to 70% was achieved.

DISCUSSION

Our semi-automated mapping process reduces the burden of domain experts. The weights are all significant in six attributes. Experimental results indicate that the availability of training data is more important than the semantic similarity of the testing data to the training data. We address the overfitting problem by selecting CDEs randomly and adjusting the ratio of training and verification samples.

CONCLUSIONS

Experimental results on real-world use cases have proven the effectiveness and efficiency of our proposed methodology in mapping CDEs with BRIDG classes, both those CDEs seen before as well as new, unseen CDEs. In addition, it reduces the mapping burden and improves the mapping quality.

Extended Analysis of Topological-Pattern-Based Ontology Enrichment

Maintenance of biomedical ontologies is difficult. We have previously developed a topological-pattern-based method to deal with the problem of identifying concepts in a reference ontology that could be of interest for insertion into a target ontology. Assuming that both ontologies are parts of the Unified Medical Language System (UMLS), the method suggests approximate locations where the target ontology could be extended with new concepts from the reference ontology. However, the final decision about each concept has to be made by a human expert. In this paper, we describe the universe of cross-ontology topological patterns in quantitative terms. We then present a theoretical analysis of the number of potential placements of reference concepts in a path in a target ontology, allowing for new cross-ontology synonyms. This provides a rough estimate of what expert resources need to be allocated for the task. One insight in previous work on this topic was the large percentage of cases where importing concepts was impossible, due to a configuration called "alternative classification." In this paper, we confirm this observation. Our target ontology is the National Cancer Institute thesaurus (NCIt). However, the methods can be applied to other pairs of ontologies with hierarchical relationships from the UMLS.

Semantic Queries Expedite MedDRA Terms Selection Thanks to a Dedicated User Interface: A Pilot Study on Five Medical Conditions

Background: Searching into the MedDRA terminology is usually limited to a hierarchical search, and/or a string search. Our objective was to compare user performances when using a new kind of user interface enabling semantic queries versus classical methods, and evaluating term selection improvement in MedDRA.

Methods: We implemented a forms-based web interface: OntoADR Query Tools (OQT). It relies on OntoADR, a formal resource describing MedDRA terms using SNOMED CT concepts and corresponding semantic relations, enabling terminological reasoning. We then compared time spent on five examples of medical conditions using OQT or the MedDRA web-based browser (MWB), and precision and recall of the term selection.

Results: OntoADR Query Tools allows the user to search in MedDRA: One may enter search criteria by selecting one semantic property from a dropdown list and one or more SNOMED CT concepts related to the range of the chosen property. The user is assisted in building his query: he can add criteria and combine them. Then, the interface displays the set of MedDRA terms matching the query. Meanwhile, on average, the time spent on OQT (about 4 min 30 s) is significantly lower (−35%; p < 0.001) than time spent on MWB (about 7 min). The results of the System Usability Scale (SUS) gave a score of 62.19 for OQT (rated as good). We also demonstrated increased precision (+27%; p = 0.01) and recall (+34%; p = 0.02). Computed “performance” (correct terms found per minute) is more than three times better with OQT than with MWB.

Discussion: This pilot study establishes the feasibility of our approach based on our initial assumption: performing MedDRA queries on the five selected medical conditions, using terminological reasoning, expedites term selection, and improves search capabilities for pharmacovigilance end users. Evaluation with a larger number of users and medical conditions are required in order to establish if OQT is appropriate for the needs of different user profiles, and to check if conclusions can be extended to other kinds of medical conditions. The application is currently limited by the non-exhaustive coverage of MedDRA by OntoADR, but nevertheless shows good performance which encourages continuing in the same direction.

Evaluating the granularity balance of hierarchical relationships within large biomedical terminologies towards quality improvement

Organizing the descendants of a concept under a particular semantic relationship may be rather arbitrarily carried out during the manual creation processes of large biomedical terminologies, resulting in imbalances in relationship granularity. This work aims to propose scalable models towards systematically evaluating the granularity balance of semantic relationships. We first utilize "parallel concepts set (PCS)" and two features (the length and the strength) of the paths between PCSs to design the general evaluation models, based on which we propose eight concrete evaluation models generated by two specific types of PCSs: single concept set and symmetric concepts set. We then apply those concrete models to the IS-A relationship in FMA and SNOMED CT's Body Structure subset, as well as to the Part-Of relationship in FMA. Moreover, without loss of generality, we conduct two additional rounds of applications on the Part-Of relationship after removing length redundancies and strength redundancies sequentially. At last, we perform automatic evaluation on the imbalances detected after the final round for identifying missing concepts, misaligned relations and inconsistencies. For the IS-A relationship, 34 missing concepts, 80 misalignments and 18 redundancies in FMA as well as 28 missing concepts, 114 misalignments and 1 redundancy in SNOMED CT were uncovered. In addition, 6,801 instances of imbalances for the Part-Of relationship in FMA were also identified, including 3,246 redundancies. After removing those redundancies from FMA, the total number of Part-Of imbalances was dramatically reduced to 327, including 51 missing concepts, 294 misaligned relations, and 36 inconsistencies. Manual curation performed by the FMA project leader confirmed the effectiveness of our method in identifying curation errors. In conclusion, the granularity balance of hierarchical semantic relationship is a valuable property to check for ontology quality assurance, and the scalable evaluation models proposed in this study are effective in fulfilling this task, especially in auditing relationships with sub-hierarchies, such as the seldom evaluated Part-Of relationship.

The Use of Automated SNOMED CT Clinical Coding in Clinical Decision Support Systems for Preventive Care

OBJECTIVE

The objective of this study is to discuss and analyze the use of automated SNOMED CT clinical coding in clinical decision support systems (CDSSs) for preventive care. The central question that this study seeks to answer is whether the utilization of SNOMED CT in CDSSs can improve preventive care.

METHOD

PubMed, Google Scholar, and Cochrane Library were searched for articles published in English between 2001 and 2012 on SNOMED CT, CDSS, and preventive care.

OUTCOME MEASURES

Outcome measures were the sensitivity or specificity of SNOMED CT coded data and the positive predictive value or negative predictive value of SNOMED CT coded data. Additionally, we documented the publication year, research question, study design, results, and conclusions of these studies.

RESULTS

The reviewed studies suggested that SNOMED CT successfully represents clinical terms and negated clinical terms.

CONCLUSION

The use of SNOMED CT in CDSS can be considered to provide an answer to the problem of medical errors as well as for preventive care in general. Enhancement of the modifiers and synonyms found in SNOMED CT will be necessary to improve the expected outcome of the integration of SNOMED CT with CDSS. Moreover, the application of the tree-augmented naïve (TAN) Bayesian network method can be considered the best technique to search SNOMED CT data and, consequently, to help improve preventive health services.

A comparative analysis of the density of the SNOMED CT conceptual content for semantic harmonization

OBJECTIVES

Medical terminologies vary in the amount of concept information (the "density") represented, even in the same sub-domains. This causes problems in terminology mapping, semantic harmonization and terminology integration. Moreover, complex clinical scenarios need to be encoded by a medical terminology with comprehensive content. SNOMED Clinical Terms (SNOMED CT), a leading clinical terminology, was reported to lack concepts and synonyms, problems that cannot be fully alleviated by using post-coordination. Therefore, a scalable solution is needed to enrich the conceptual content of SNOMED CT. We are developing a structure-based, algorithmic method to identify potential concepts for enriching the conceptual content of SNOMED CT and to support semantic harmonization of SNOMED CT with selected other Unified Medical Language System (UMLS) terminologies.

METHODS

We first identified a subset of English terminologies in the UMLS that have 'PAR' relationship labeled with 'IS_A' and over 10% overlap with one or more of the 19 hierarchies of SNOMED CT. We call these "reference terminologies" and we note that our use of this name is different from the standard use. Next, we defined a set of topological patterns across pairs of terminologies, with SNOMED CT being one terminology in each pair and the other being one of the reference terminologies. We then explored how often these topological patterns appear between SNOMED CT and each reference terminology, and how to interpret them.

RESULTS

Four viable reference terminologies were identified. Large density differences between terminologies were found. Expected interpretations of these differences were indeed observed, as follows. A random sample of 299 instances of special topological patterns ("2:3 and 3:2 trapezoids") showed that 39.1% and 59.5% of analyzed concepts in SNOMED CT and in a reference terminology, respectively, were deemed to be alternative classifications of the same conceptual content. In 30.5% and 17.6% of the cases, it was found that intermediate concepts could be imported into SNOMED CT or into the reference terminology, respectively, to enhance their conceptual content, if approved by a human curator. Other cases included synonymy and errors in one of the terminologies.

CONCLUSION

These results show that structure-based algorithmic methods can be used to identify potential concepts to enrich SNOMED CT and the four reference terminologies. The comparative analysis has the future potential of supporting terminology authoring by suggesting new content to improve content coverage and semantic harmonization between terminologies.

ADReCS: an ontology database for aiding standardization and hierarchical classification of adverse drug reaction terms

Adverse drug reactions (ADRs) are noxious and unexpected effects during normal drug therapy. They have caused significant clinical burden and been responsible for a large portion of new drug development failure. Molecular understanding and in silico evaluation of drug (or candidate) safety in laboratory is thus so desired, and unfortunately has been largely hindered by misuse of ADR terms. The growing impact of bioinformatics and systems biology in toxicological research also requires a specialized ADR term system that works beyond a simple glossary. Adverse Drug Reaction Classification System (ADReCS; http://bioinf.xmu.edu.cn/ADReCS) is a comprehensive ADR ontology database that provides not only ADR standardization but also hierarchical classification of ADR terms. The ADR terms were pre-assigned with unique digital IDs and at the same time were well organized into a four-level ADR hierarchy tree for building an ADR–ADR relation. Currently, the database covers 6544 standard ADR terms and 34 796 synonyms. It also incorporates information of 1355 single active ingredient drugs and 134 022 drug–ADR pairs. In summary, ADReCS offers an opportunity for direct computation on ADR terms and also provides clues to mining common features underlying ADRs.

Formalizing MedDRA to support semantic reasoning on adverse drug reaction terms

Although MedDRA has obvious advantages over previous terminologies for coding adverse drug reactions and discovering potential signals using data mining techniques, its terminological organization constrains users to search terms according to predefined categories. Adding formal definitions to MedDRA would allow retrieval of terms according to a case definition that may correspond to novel categories that are not currently available in the terminology. To achieve semantic reasoning with MedDRA, we have associated formal definitions to MedDRA terms in an OWL file named OntoADR that is the result of our first step for providing an "ontologized" version of MedDRA. MedDRA five-levels original hierarchy was converted into a subsumption tree and formal definitions of MedDRA terms were designed using several methods: mappings to SNOMED-CT, semi-automatic definition algorithms or a fully manual way. This article presents the main steps of OntoADR conception process, its structure and content, and discusses problems and limits raised by this attempt to "ontologize" MedDRA.

Drug safety surveillance using de-identified EMR and claims data: issues and challenges

The author discusses the challenges of pharmacovigilance using electronic medical record and claims data. Use of ICD-9 encoded data has low sensitivity for detection of adverse drug events (ADEs), because it requires that an ADE escalate to major-complaint level before it can be identified, and because clinical symptomatology is relatively under-represented in ICD-9. A more appropriate vocabulary for ADE identification, SNOMED CT, awaits wider deployment. The narrative-text record of progress notes can potentially be used for more sensitive ADE detection. More effective surveillance will require the ability to grade ADEs by severity. Finally, access to online drug information that includes both a reliable hierarchy of drug families as well as structured information on existing ADEs can improve the focus and predictive ability of surveillance efforts.

Determining correspondences between high-frequency MedDRA concepts and SNOMED: a case study

BACKGROUND

The Systematic Nomenclature of Medicine Clinical Terms (SNOMED CT) is being advocated as the foundation for encoding clinical documentation. While the electronic medical record is likely to play a critical role in pharmacovigilance - the detection of adverse events due to medications - classification and reporting of Adverse Events is currently based on the Medical Dictionary of Regulatory Activities (MedDRA). Complete and high-quality MedDRA-to-SNOMED CT mappings can therefore facilitate pharmacovigilance. The existing mappings, as determined through the Unified Medical Language System (UMLS), are partial, and record only one-to-one correspondences even though SNOMED CT can be used compositionally. Efforts to map previously unmapped MedDRA concepts would be most productive if focused on concepts that occur frequently in actual adverse event data. We aimed to identify aspects of MedDRA that complicate mapping to SNOMED CT, determine pattern in unmapped high-frequency MedDRA concepts, and to identify types of integration errors in the mapping of MedDRA to UMLS.

METHODS

Using one years' data from the US Federal Drug Administrations Adverse Event Reporting System, we identified MedDRA preferred terms that collectively accounted for 95% of both Adverse Events and Therapeutic Indications records. After eliminating those already mapping to SNOMED CT, we attempted to map the remaining 645 Adverse-Event and 141 Therapeutic-Indications preferred terms with software assistance.

RESULTS

All but 46 Adverse-Event and 7 Therapeutic-Indications preferred terms could be composed using SNOMED CT concepts: none of these required more than 3 SNOMED CT concepts to compose. We describe the common composition patterns in the paper. About 30% of both Adverse-Event and Therapeutic-Indications Preferred Terms corresponded to single SNOMED CT concepts: the correspondence was detectable by human inspection but had been missed during the integration process, which had created duplicated concepts in UMLS.

CONCLUSIONS

Identification of composite mapping patterns, and the types of errors that occur in the MedDRA content within UMLS, can focus larger-scale efforts on improving the quality of such mappings, which may assist in the creation of an adverse-events ontology.

Achieving standardized medication data in clinical research studies: two approaches and applications for implementing RxNorm

The National Institutes of Health has proposed a roadmap for clinical research. Test projects of this roadmap include centralized data management for distributed research, the harmonization of clinical and research data, and the use of data standards throughout the research process. In 2003, RxNorm was named as a standard for codifying clinical drugs. Clinical researchers looking to implement RxNorm have few template implementation plans. Epidemiological studies and clinical trials (types of clinical research) have different requirements for model standards and best implementation tools. This paper highlights two different (epidemiological and intervention) clinical research projects, their unique requirements for a medication standard, the suitability of RxNorm as a standard for each, and application and process requirements for implementation. It is hoped that our experience of selecting and implementing the RxNorm standard to address varying study requirements in both domestic and international settings will be of value to other efforts.

Automated summaries of serious adverse events in the hepatitis C antiviral long-term treatment against cirrhosis trial

BACKGROUND

Even though adverse event (AE) collection and official accounting are mandatory for clinical trials, there are limited detailed guidelines specifying how to summarize the event for reporting in a timely and expeditious manner. This article details the AE and serious adverse event (SAE) reporting summary developed for a large multi-center National Institutes of Health (NIH)-sponsored clinical trial.

PURPOSE

To review and analyze the large volume of AE data reported by 10 sites (806 SAEs and 19,034 AEs from August 2000 to May 2007) the automated SAE summary was developed. It was designed to ensure timeliness and clarity in the complex process of AE review and reporting.

METHODS

The AE and SAE case report forms (CRFs) as well as the automated SAE summary were developed within a database management system developed by the Data Coordinating Center (DCC) which allowed for web-based data entry at the DCC and 10 sites and offered immediate overall and site-specific reports accessible by the DCC, site, and NIH project staff.

RESULTS

The automated SAE summary pulled data from multiple CRFs to create a succinct and informative summary and allowed for prompt and easy reporting to the regulatory agencies. The summary was adaptable to the needs of reviewers because of the availability of multiple search options.

Using SNOMED CT in combination with MedDRA for reporting signal detection and adverse drug reactions reporting

OBJECTIVE

To investigate the feasibility of using SNOMED CT as an entry point for coding adverse drug reactions and map them automatically to MedDRA for reporting purposes and interoperability with legacy repositories.

METHODS

On the one hand, we attempt to map SNOMED CT concepts to MedDRA concepts through the UMLS, using synonymy and explicit mapping relations. On the other, we compute the set of all fine-grained concepts that can be reached from concepts having a mapping to MedDRA.

RESULTS

58% of the Preferred Terms in MedDRA have a mapping to SNOMED CT. Through the descendants in SNOMED CT, 108,305 additional SNOMED CT concepts can be linked to MedDRA.

CONCLUSIONS

Fine-grained SNOMED CT concepts can be mapped automatically to MedDRA. This approach has the potential to enable the collection of adverse events related to drugs directly from clinical repositories. The quality of the mapping needs to be evaluated.

An automated standardized system for managing adverse events in clinical research networks

Multi-site clinical protocols and clinical research networks require tools to manage and monitor adverse events (AEs). To be successful, these tools must be designed to comply with applicable regulatory requirements, reflect current data standards, international directives and advances in pharmacovigilance, and be convenient and adaptable to multiple needs. We describe an Adverse Event Data Management System (AEDAMS) that is used across multiple study designs in the various clinical research networks and multi-site studies for which we provide data and technological support. Investigators enter AE data using a standardized and structured web-based data collection form. The automated AEDAMS forwards the AE information to individuals in designated roles (investigators, sponsors, Data Safety and Monitoring Boards) and manages subsequent communications in real time, as the entire reporting, review and notification is done by automatically generated emails. The system was designed to adhere to timelines and data requirements in compliance with Good Clinical Practice (International Conference on Harmonisation E6) reporting standards and US federal regulations, and can be configured to support AE management for many types of study designs and adhere to various domestic or international reporting requirements. This tool allows AEs to be collected in a standard way by multiple distributed users, facilitates accurate and timely AE reporting and reviews, and allows the centralized management of AEs. Our design justification and experience with the system are described.