Social Media analytics for Pharmacovigilance of Anti-Epileptic Drugs (Preprint)

UNSTRUCTURED Epilepsy is a common neurological disorder worldwide and Anti-Epileptic Drugs (AEDs) therapy is the cornerstone of its treatment. It has a laudable aim of achieving seizure freedom and minimal, if any, Adverse Drug Reactions (ADRs). Too often, AEDs treatment is a long-lasting journey, in which ADRs have a crucial role in its administration. Therefore, from pharmacovigilance perspective, the detection of the ADRs of AEDs is a task of utmost importance. Typically, it is accomplished by applying data mining algorithms to a relevant data from spontaneous reporting systems. Despite their wide adoption for pharmacovigilance, the passiveness and high under-reporting ratio associated with them have encouraged considering other data source such as electronic health databases and pharmaceutical databases. Social media is the most recent alternative data source with many promising potentials to overcome the shortcomings of the traditional ones. Although, in the literature, some attempts have investigated the validity and utility of social media for ADRs detection of different groups of drugs, none of them was dedicated to the ADRs of AEDs. Hence, this paper presents a novel investigation of the validity and utility of social media as an alternative data source for the ADRs detection of AEDs. To this end, a dataset of consumers' reviews from two online health communities have been collected. The dataset is preprocessed, the unigram, bigram, and trigram are generated, and the ADRs of each AED are extracted with the aid of consumer health vocabulary and ADRs lexicon. Three widely used measures, namely proportional reporting ratio, reporting odd ratio, and information component are used to measure the association between each ADR and AED. The results, lists of signaled ADRs for each AED, are validated against Side Effect Resource (SIDER), a widely used ADRs database, in terms of precision of the ADRs detection. The validation results, 73%-74%, indicate the validity of the online health communities for the detection of AEDs ADRs. Furthermore, the lists of signaled AEDs ADRs are analyzed to answer questions regarding the common ADRs for all AEDs and the mutual similarities between AEDs in terms of their signaled ADRs. The consistency of the drawn answers with the existing pharmaceutical knowledge suggests the utility of the online health communities' data for knowledge discovery tasks of AEDs.