Learning to detect and understand drug discontinuation events from clinical narratives

OBJECTIVE Identifying drug discontinuation (DDC) events and understanding their reasons are important for medication management and drug safety surveillance. Structured data resources are often incomplete and lack reason information. In this article, we assessed the ability of natural language processing (NLP) systems to unlock DDC information from clinical narratives automatically. MATERIALS AND METHODS We collected 1867 de-identified providers' notes from the University of Massachusetts Medical School hospital electronic health record system. Then 2 human experts chart reviewed those clinical notes to annotate DDC events and their reasons. Using the annotated data, we developed and evaluated NLP systems to automatically identify drug discontinuations and reasons at the sentence level using a novel semantic enrichment-based vector representation (SEVR) method for enhanced feature representation. RESULTS Our SEVR-based NLP system achieved the best performance of 0.785 (AUC-ROC) for detecting discontinuation events and 0.745 (AUC-ROC) for identifying reasons when testing this highly imbalanced data, outperforming 2 state-of-the-art non-SEVR-based models. Compared with a rule-based baseline system for discontinuation detection, our system improved the sensitivity significantly (57.75% vs 18.31%, absolute value) while retaining a high specificity of 99.25%, leading to a significant improvement in AUC-ROC by 32.83% (absolute value). CONCLUSION Experiments have shown that a high-performance NLP system can be developed to automatically identify DDCs and their reasons from providers' notes. The SEVR model effectively improved the system performance showing better generalization and robustness on unseen test data. Our work is an important step toward identifying reasons for drug discontinuation that will inform drug safety surveillance and pharmacovigilance.

[1]  Alexander Turchin,et al.  Reasons for Discontinuation of Lipid-Lowering Medications in Patients with Chronic Kidney Disease , 2014, Cardiorenal Medicine.

[2]  Yoon Kim,et al.  Convolutional Neural Networks for Sentence Classification , 2014, EMNLP.

[3]  Jürgen Schmidhuber,et al.  Long Short-Term Memory , 1997, Neural Computation.

[4]  Bradley C Martin,et al.  Concordance among three self-reported measures of medication adherence and pharmacy refill records. , 2005, Journal of the American Pharmacists Association : JAPhA.

[5]  Jeffrey Dean,et al.  Efficient Estimation of Word Representations in Vector Space , 2013, ICLR.

[6]  Walter Daelemans,et al.  Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP) , 2014, EMNLP 2014.

[7]  C. Gabay,et al.  Comparison of drug retention rates and causes of drug discontinuation between anti-tumor necrosis factor agents in rheumatoid arthritis. , 2009, Arthritis and rheumatism.

[8]  Sergei Vassilvitskii,et al.  k-means++: the advantages of careful seeding , 2007, SODA '07.

[9]  Jonathan S Einbinder,et al.  Open Access Research Article Prevalence and Factors Affecting Home Blood Pressure Documentation in Routine Clinical Care: a Retrospective Study , 2022 .

[10]  H. T. Kung,et al.  Language Modeling by Clustering with Word Embeddings for Text Readability Assessment , 2017, CIKM.

[11]  Pavel Blagoveston Bochev,et al.  A vector space model for information retrieval with generalized similarity measures. , 2012 .

[12]  Alexander Turchin,et al.  Identification of Documented Medication Non-Adherence in Physician Notes , 2008, AMIA.

[13]  Gerard Salton,et al.  A vector space model for automatic indexing , 1975, CACM.

[14]  Robert J Glynn,et al.  How well do patients report noncompliance with antihypertensive medications?: a comparison of self‐report versus filled prescriptions , 2004, Pharmacoepidemiology and drug safety.

[15]  A W REID,et al.  The use of cocoa syrups for masking the taste of quinine hydrochloride. , 1956, Journal of the American Pharmaceutical Association. American Pharmaceutical Association.

[16]  K. Gleason,et al.  Reconciliation of discrepancies in medication histories and admission orders of newly hospitalized patients. , 2004, American journal of health-system pharmacy : AJHP : official journal of the American Society of Health-System Pharmacists.

[17]  A. Blaes,et al.  Shared Risk Factors in Cardiovascular Disease and Cancer , 2016, Circulation.

[18]  Gaël Varoquaux,et al.  Scikit-learn: Machine Learning in Python , 2011, J. Mach. Learn. Res..

[19]  Wanzhu Tu,et al.  Adherence: Comparison of Methods to Assess Medication Adherence and Classify Nonadherence , 2009, The Annals of pharmacotherapy.

[20]  G. Criner,et al.  Pirfenidone and nintedanib for pulmonary fibrosis in clinical practice: Tolerability and adverse drug reactions , 2017, Respirology.

[21]  Yiming Yang,et al.  A re-examination of text categorization methods , 1999, SIGIR '99.

[22]  Nello Cristianini,et al.  Advances in Kernel Methods - Support Vector Learning , 1999 .

[23]  Thorsten Joachims,et al.  Making large scale SVM learning practical , 1998 .

[24]  Colin R Simpson,et al.  Using primary care prescribing databases for pharmacovigilance. , 2011, British journal of clinical pharmacology.

[25]  Baharin Bin Ahmad,et al.  Comparison of two Classification methods (MLC and SVM) to extract land use and land cover in Johor Malaysia , 2014 .

[26]  G Savova,et al.  Capturing the Patient’s Perspective: a Review of Advances in Natural Language Processing of Health-Related Text , 2017, Yearbook of Medical Informatics.

[27]  Todd A. Lee,et al.  Exposure Definition and Measurement , 2013 .

[28]  A Cozzi-Lepri,et al.  Insights into reasons for discontinuation according to year of starting first regimen of highly active antiretroviral therapy in a cohort of antiretroviral‐naïve patients , 2010, HIV medicine.

[29]  Alexander Turchin,et al.  Comparison of information content of structured and narrative text data sources on the example of medication intensification. , 2009, Journal of the American Medical Informatics Association : JAMIA.

[30]  Jianhua Li,et al.  Medication Reconciliation Using Natural Language Processing and Controlled Terminologies , 2007, MedInfo.

[31]  Zhiyuan Liu,et al.  A C-LSTM Neural Network for Text Classification , 2015, ArXiv.