Detection of Pharmacovigilance-Related Adverse Events Using Electronic Health Records and Automated Methods

Abstract

Electronic health records (EHRs) are an important source of data for detection of adverse drug reactions (ADRs). However, adverse events are frequently due not to medications but to the patients’ underlying conditions. Mining to detect ADRs from EHR data must account for confounders. We developed an automated method using natural-language processing (NLP) and a knowledge source to differentiate cases in which the patient’s disease is responsible for the event rather than a drug. Our method was applied to 199,920 hospitalization records, concentrating on two serious ADRs: rhabdomyolysis (n = 687) and agranulocytosis (n = 772). Our method automatically identified 75% of the cases, those with disease etiology. The sensitivity and specificity were 93.8% (confidence interval: 88.9-96.7%) and 91.8% (confidence interval: 84.0-96.2%), respectively. The method resulted in considerable saving of time: for every 1 h spent in development, there was a saving of at least 20 h in manual review. The review of the remaining 25% of the cases therefore became more feasible, allowing us to identify the medications that had caused the ADRs.

Application of electronic medical record data for health outcomes research: a review of recent literature

“…While NLP is still evolving in healthcare, it has the potential to greatly expand the utility of EMRs for outcomes research. Researchers are clearly making progress in this domain, as a post-hoc search for NLP-based outcomes studies published since the review cutoff date, 1 January 2012, identified six additional outcomes methodology-based studies that report on the use of NLP for identifying/classifying patients and outcomes measures [114,115,[117][118][119][120].…”

Clinical Data Reuse or Secondary Use: Current Status and Potential Future Progress

“…Examples of the combined use of standard NLP and text-and data-mining are found in [139][140][141] where cTAKES is used with Boolean logic to perform phenotyping and to extract drug-side effects. MedLEE was applied for: 1) adverse drug reaction (ADR) signaling, where the association between a drug and an ADR was obtained by using disproportionality analysis [142,143] or Boolean logic [144], or by building and analyzing statistical distributions of concepts (i.e., diseases, symptoms, medications) extracted from the narrative text [145]; 2) EHR-data driven phenotyping using Boolean logic on MedLEE-extracted concepts [136,146]; 3) automated classification of outcomes from the analysis of emergency department computed tomography imaging reports using machine learning methods, such as decision trees [147]. MetaMap has been used with logistic regression in [148] to discover inappropriate use of emergency room based on information on drugs, psychological characteristics, diagnoses, and symptoms.…”

A methodology pathway to develop a Computer Drug Safety program in Primary health care setting based on clinical audit

“…[11] Drug-ADEs association rules have been used to assess associations between drugs and symptoms. [9] Medications are prescribed mostly in PHC, yet information on ADEs from this setting is scarce. In this paper, we want to bridge the gap and to highlight the potential of IT in facing this challenge.…”

“…[8] What makes the use of this method difficult is the fact that ADEs are usually contextdependent, that means that they are caused by disease-related and patient-related, rather than drug-related conditions. [9] In order to make detection of ADEs from eHRs much easier, new IT approaches have emerged, putting attention on the precision and strength of ADEs signals generated from medical data. [8] To further strenghten the efforts on ADEs information collection, some supplement data sources have been used to complement the standard ones, including drug labels, biomedical literature and biomedical knowledge.…”