ADEPt, a semantically-enriched pipeline for extracting adverse drug events from free-text electronic health records

Abstract

Adverse drug events (ADEs) are unintended responses to medical treatment. They can greatly affect a patient’s quality of life and present a substantial burden on healthcare. Although Electronic health records (EHRs) document a wealth of information relating to ADEs, they are frequently stored in the unstructured or semi-structured free-text narrative requiring Natural Language Processing (NLP) techniques to mine the relevant information. Here we present a rule-based ADE detection and classification pipeline built and tested on a large Psychiatric corpus comprising 264k patients using the de-identified EHRs of four UK-based psychiatric hospitals. The pipeline uses characteristics specific to Psychiatric EHRs to guide the annotation process, and distinguishes: a) the temporal value associated with the ADE mention (whether it is historical or present), b) the categorical value of the ADE (whether it is assertive, hypothetical, retrospective or a general discussion) and c) the implicit contextual value where the status of the ADE is deduced from surrounding indicators, rather than explicitly stated. We manually created the rulebase in collaboration with clinicians and pharmacists by studying ADE mentions in various types of clinical notes. We evaluated the open-source Adverse Drug Event annotation Pipeline (ADEPt) using 19 ADEs specific to antipsychotics and antidepressants medication. The ADEs chosen vary in severity, regularity and persistence. The average F-measure and accuracy achieved by our tool across all tested ADEs were 0.83 and 0.83 respectively. In addition to annotation power, the ADEPT pipeline presents an improvement to the state of the art context-discerning algorithm, ConText.

Knowledge graph prediction of unknown adverse drug reactions and validation in electronic health records

“…Finally, the validation is dependent on reliable NLP to extract ADR mentions from the free text of EHRs. As the pipeline used for validation was developed and validated using the same EHR 16 , we are confident that the error rate is low.…”

“…The EHR at the South London and Maudsley NHS Foundation Trust was used to validate drug-ADR associations predicted by the algorithm. ADR mentions were identified from the free text of the EHR using a published NLP pipeline developed previously using the same EHR 16 . Reports were only considered a validation of a drug-ADR association when patients were prescribed a single drug and then reported the ADR within 30 days.…”

“…This is a widely used clinical database with a robust data governance structure which has received ethical approval for secondary analysis (Oxford REC C 08/H0606/71 + 5). Free text from these mental health electronic health records was processed using a Natural Language Processing pipeline described previously 16 . Briefly, ADR mentions are identified using a dictionary of related terms (including synonyms and misspellings) and further processed to identify negation or other experiencers, favouring precision over recall.…”

“…synonyms, acronyms or shorthand), or may be mentioned in a historical context. We therefore focus on a subset of ADRs for which the NLP pipeline is validated 16 .…”

Protected Health Information filter (Philter): accurately and securely de-identifying free-text clinical notes

“…This written medical narrative frequently captures patient experience and event ordering timelines. To date, there have been many studies that have successfully used data from clinical notes for discoveries, including detection of drug adverse outcomes 5 , identification of off-label drug use 6 , surveillance of disease states 7 , and identification of clinical concept relatedness 8 .…”