A Text Searching Tool to Identify Patients with Idiosyncratic Drug-Induced Liver Injury

Abstract

Background Idiosyncratic drug induced liver injury (DILI) is an uncommon but important cause of liver disease that is challenging to diagnose and identify in the electronic medical record (EMR). Aim To develop an accurate, reliable, and efficient method of identifying patients with bonafide DILI in an EMR system. Methods 527,000 outpatient and ER encounters in an EPIC-based EMR were searched for potential DILI cases attributed to 8 drugs. A searching algorithm that extracted 200 characters of text around 14 liver injury terms in the EMR were extracted and collated. Physician investigators reviewed the data outputs and used standardized causality assessment methods to adjudicate the potential DILI cases. Results A total of 101 DILI cases were identified from the 2,564 potential DILI cases that included 62 probable DILI cases, 25 possible DILI cases, 9 historical DILI cases, and 5 allergy only cases. Elimination of the term “liver disease” from the search strategy improved the search recall from 4% to 19% while inclusion of the 4 highest yield liver injury terms further improved the positive predictive value to 64% but reduced the overall case detection rate by 47%. RUCAM scores of the 57 probable DILI cases were generally high and concordant with expert opinion causality assessment scores. Conclusions A novel text searching tool was developed that identified a large number of DILI cases from a widely used EMR system. A computerized extraction of dictated text followed by the manual review of text snippets can rapidly identify bonafide cases of idiosyncratic DILI.

Identifying Drug-Induced Liver Illness (DILI) with Computerized Information Extraction: No More Dilly-Dallying

“…After modifying search terms, the precision of the results was increased from 4 % to a astonishingly improved 64 %, with the required review time decreased from 29 to 5 h. The results from this study supported the hypothesis that the direct text searching method netted a nearly fivefold increase in the number of idiosyncratic DILI cases identified compared to a previous study that used ICD-9 codes only [7], indicating that information about DILI derived from narrative clinical notes was much more reliable than simply using the traditional identifiers of surrogate terms and ICD-9 codes. Heidemann et al have also developed a more robust terminology used as a standard of classification in order to detect DILI.…”

Novel methodology to measure pre-procedure antimicrobial prophylaxis: integrating text searches with structured data from the Veterans Health Administration’s electronic medical record

“…When the algorithm was applied to all VA facilities, we found substantial interfacility variation, which may cause implementation challenges if EMR documentation practices do not evolve. Applications developed for use within the VA EMR may not be directly applicable to other EMR systems; however, the programming used in our algorithm is relatively straightforward, and the most commonly used systems outside of the VA, including Cerner and EPIC, have text note searching options [19,28]. Structured variables, such as antimicrobial orders, are generally easily extracted from any electronic order entry system [10].…”

Evaluation of Named Entity Recognition Algorithms Using Clinical Text Data

“…In ADRs extraction, Heidemann et al developed a novel text searching tool to capture idiosyncratic drug-induced liver illness cases from electronic medical record system. This was based on KE approach [3]. Combination of deep recurrent neural networks and conditional random fields were used to develop a new model for ADRs extraction [13].…”

“…For example, the noun phrase of a disease is highlighted in the following statement "The patient is a 28-year-old woman who is HIV positive for two years". In many IE research, biomedical named entity recognition technique is widely considered to be one of the most important steps in the following works searching the diagnostic code of diseases [2], adverse drug reactions extraction [3], abbreviations extraction [4], de-identification [5], obesity [6], medication [7], relation extraction [8], coreference resolution [9], temporal relations extraction [10], etc. Named Entity Recognition (NER) is also called as concept extraction/recognition.…”