Filtering Entities to Optimize Identification of Adverse Drug Reaction From Social Media: How Can the Number of Words Between Entities in the Messages Help?

Abdellaoui, Redhouane; Schück, Stéphane; Texier, Nathalie; Burgun, Anita

doi:10.2196/publichealth.6577

Cited by 36 publications

(30 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…An additional thesaurus containing patient language must be created in order to normalize the vocabulary found in the messages so that it can be recognized by medical reference thesauri. We will take into account the number of words between the detected drug and event as recent evidence shows that such distance can be used for identifying false positives and filter events that are likely to be ADRs [ 48 ].…”

Section: Discussionmentioning

confidence: 99%

The Adverse Drug Reactions from Patient Reports in Social Media Project: Five Major Challenges to Overcome to Operationalize Analysis and Efficiently Support Pharmacovigilance Process

Bousquet¹,

Dahamna²,

Guillemin-Lanne³

et al. 2017

JMIR Res Protoc

Self Cite

View full text Add to dashboard Cite

BackgroundAdverse drug reactions (ADRs) are an important cause of morbidity and mortality. Classical Pharmacovigilance process is limited by underreporting which justifies the current interest in new knowledge sources such as social media. The Adverse Drug Reactions from Patient Reports in Social Media (ADR-PRISM) project aims to extract ADRs reported by patients in these media. We identified 5 major challenges to overcome to operationalize the analysis of patient posts: (1) variable quality of information on social media, (2) guarantee of data privacy, (3) response to pharmacovigilance expert expectations, (4) identification of relevant information within Web pages, and (5) robust and evolutive architecture.ObjectiveThis article aims to describe the current state of advancement of the ADR-PRISM project by focusing on the solutions we have chosen to address these 5 major challenges.MethodsIn this article, we propose methods and describe the advancement of this project on several aspects: (1) a quality driven approach for selecting relevant social media for the extraction of knowledge on potential ADRs, (2) an assessment of ethical issues and French regulation for the analysis of data on social media, (3) an analysis of pharmacovigilance expert requirements when reviewing patient posts on the Internet, (4) an extraction method based on natural language processing, pattern based matching, and selection of relevant medical concepts in reference terminologies, and (5) specifications of a component-based architecture for the monitoring system.ResultsConsidering the 5 major challenges, we (1) selected a set of 21 validated criteria for selecting social media to support the extraction of potential ADRs, (2) proposed solutions to guarantee data privacy of patients posting on Internet, (3) took into account pharmacovigilance expert requirements with use case diagrams and scenarios, (4) built domain-specific knowledge resources embeding a lexicon, morphological rules, context rules, semantic rules, syntactic rules, and post-analysis processing, and (5) proposed a component-based architecture that allows storage of big data and accessibility to third-party applications through Web services.ConclusionsWe demonstrated the feasibility of implementing a component-based architecture that allows collection of patient posts on the Internet, near real-time processing of those posts including annotation, and storage in big data structures. In the next steps, we will evaluate the posts identified by the system in social media to clarify the interest and relevance of such approach to improve conventional pharmacovigilance processes based on spontaneous reporting.

show abstract

Section: Discussionmentioning

confidence: 99%

The Adverse Drug Reactions from Patient Reports in Social Media Project: Five Major Challenges to Overcome to Operationalize Analysis and Efficiently Support Pharmacovigilance Process

Bousquet¹,

Dahamna²,

Guillemin-Lanne³

et al. 2017

JMIR Res Protoc

Self Cite

View full text Add to dashboard Cite

show abstract

“…The data was extracted from the Detec’t database [ 26 ], a database developed by Kappa Santé [ 27 ] that collects messages from several French forums using a Web crawler. Detec’t extracts messages from forums based on a named entity recognition module using a drug lexicon made by Kappa Santé and a fuzzy matching algorithm.…”

Section: Methodsmentioning

confidence: 99%

Detection of Cases of Noncompliance to Drug Treatment in Patient Forum Posts: Topic Model Approach

Abdellaoui¹,

Foulquié²,

Texier³

et al. 2018

J Med Internet Res

Self Cite

View full text Add to dashboard Cite

BackgroundMedication nonadherence is a major impediment to the management of many health conditions. A better understanding of the factors underlying noncompliance to treatment may help health professionals to address it. Patients use peer-to-peer virtual communities and social media to share their experiences regarding their treatments and diseases. Using topic models makes it possible to model themes present in a collection of posts, thus to identify cases of noncompliance.ObjectiveThe aim of this study was to detect messages describing patients’ noncompliant behaviors associated with a drug of interest. Thus, the objective was the clustering of posts featuring a homogeneous vocabulary related to nonadherent attitudes.MethodsWe focused on escitalopram and aripiprazole used to treat depression and psychotic conditions, respectively. We implemented a probabilistic topic model to identify the topics that occurred in a corpus of messages mentioning these drugs, posted from 2004 to 2013 on three of the most popular French forums. Data were collected using a Web crawler designed by Kappa Santé as part of the Detec’t project to analyze social media for drug safety. Several topics were related to noncompliance to treatment.ResultsStarting from a corpus of 3650 posts related to an antidepressant drug (escitalopram) and 2164 posts related to an antipsychotic drug (aripiprazole), the use of latent Dirichlet allocation allowed us to model several themes, including interruptions of treatment and changes in dosage.The topic model approach detected cases of noncompliance behaviors with a recall of 98.5% (272/276) and a precision of 32.6% (272/844).ConclusionsTopic models enabled us to explore patients’ discussions on community websites and to identify posts related with noncompliant behaviors. After a manual review of the messages in the noncompliance topics, we found that noncompliance to treatment was present in 6.17% (276/4469) of the posts.

show abstract

“…One of the principal challenges is the extraction medical entities from noisy patient-generated content. Given the large volume of social media posts, efforts towards the automatic text classification for ADR detection are receiving growing attention [70,[94][95][96][97][98][99][100][101]. However, lexicon-based approaches [47] for medical entity recognition and tools like MetaMap [102], developed by the US National Library of Medicine to identify medical concepts into the concept codes from the Unified Medical Language System Metathesaurus (UMLS), are not sufficient, given the informal, colloquial nature of discussions and the non-adherence to standardised terminology used by participants [103].…”

Section: Information Extraction From Social Mediamentioning

confidence: 99%

“…The need for manual data labelling is expected to drop considerably with the application of neural network-based tools [113,114]. Abdellaoui et al [98] apply distance-based filtering in order to distinguish between false positives and true ADR declarations. The framework proposed by Liu and Chen [71] employs a hybrid approach combining statistical machine learning methods and rule-based filtering with information from medical knowledge bases, and report source classification to reduce noise.…”

Section: Information Extraction From Social Mediamentioning

confidence: 99%

Harnessing social media data for pharmacovigilance: a review of current state of the art, challenges and future directions

Pappa

Stergioulas

2019

Int J Data Sci Anal

View full text Add to dashboard Cite

The ever-increasing supply of information combined with the growing knowledge elicitation capabilities of key emerging technologies presents pharmacovigilance with enormous opportunities. Currently, safety monitoring is expanding its evidence base, moving beyond traditional approaches towards sophisticated methods that can identify possible safety signals from multiple information sources, both structured and unstructured. In this context, health information posted online by patients represents a potentially valuable, yet currently left largely unexploited source of post-market safety data that could supplement data from traditional sources of drug safety information. As the use of social media data for pharmacovigilance is still in its infancy, the present paper explores the state of the art in the application of social data to adverse drug reaction detection; provides a thorough review of existing work in the field, highlighting important research efforts and achievements; and finally, discusses the current challenges and promising avenues for future work. Following a literature review methodology, a critical appraisal was conducted of carefully selected work on the use of social data in post-market surveillance, as presented in the recent scientific literature. Out of a sample of more than 1300 articles, which was the result of the literature search, the final selection of articles was made based on their relevance to the applications of social networking sites (SNS) to pharmacovigilance, and a thorough review of this corpus was completed with a total of 100 articles reviewed. The main contributions of this review include the mapping and systematisation of the current knowledge in the field by drawing comparisons of different approaches, types of social data and of relevant sources currently used in the field, and by developing new classifications of social data sources and taxonomies for social data for use in pharmacovigilance, as well as the identification of key challenges and the extraction of new insights in terms of potential for practical applications and future research directions in the area of pharmacovigilance. Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

show abstract

Filtering Entities to Optimize Identification of Adverse Drug Reaction From Social Media: How Can the Number of Words Between Entities in the Messages Help?

Cited by 36 publications

References 41 publications

The Adverse Drug Reactions from Patient Reports in Social Media Project: Five Major Challenges to Overcome to Operationalize Analysis and Efficiently Support Pharmacovigilance Process

The Adverse Drug Reactions from Patient Reports in Social Media Project: Five Major Challenges to Overcome to Operationalize Analysis and Efficiently Support Pharmacovigilance Process

Detection of Cases of Noncompliance to Drug Treatment in Patient Forum Posts: Topic Model Approach

Harnessing social media data for pharmacovigilance: a review of current state of the art, challenges and future directions

Contact Info

Product

Resources

About