A Corpus of Drug Usage Guidelines Annotated with Type of Advice

A Corpus of Drug Usage Guidelines Annotated with Type of Advice Sarah Masud Preum, Md. Rizwan Parvez, Kai-Wei Chang, John A. Stankovic University of Virginia; University of California, Los Angeles Charlottesville VA 22903; Los Angeles CA 90095 preum@virginia.edu, rizwan@cs.ucla.edu, kwchang@cs.ucla.edu, jas9f@virginia.edu Abstract Adherence to drug usage guidelines for prescription and over-the-counter drugs is critical for drug safety and effectiveness of treatment. Drug usage guideline documents contain advice on potential drug-drug interaction, drug-food interaction, and drug administration process. Current research on drug safety and public health indicates patients are often either unaware of such critical advice or overlook them. Categorizing advice statements from these documents according to their topics can enable the patients to find safety critical information. However, automatically categorizing drug usage guidelines based on their topic is an open challenge and there is no annotated dataset on drug usage guidelines. To address the latter issue, this paper presents (i) an annotation scheme for annotating safety critical advice from drug usage guidelines, (ii) an annotation tool for such data, and (iii) an annotated dataset containing drug usage guidelines from 90 drugs. This work is expected to accelerate further release of annotated drug usage guideline datasets and research on automatically filtering safety critical information from these textual documents. Keywords: Drug Usage Guidelines, Annotation Tool, Medical Corpora, Health Informatics 1. Introduction Drug safety is crucial for ensuring overall health safety of patients and effectiveness of treatment (Yi et al., 2015). In order to ensure drug safety, with each of their prescribed drugs, patients are often provided with drug usage guide- lines (DUG) documents 1. These documents contain essen- tial information regarding drug usage, including, but not limited to, dosage, drug administration, adverse reactions or symptoms, drug storage and disposal, contraindications, drug-drug interactions, and drug-food interactions (MDS, 2017) as illustrated in Table 1. Doctors and pharmacists are expected to educate patients about drug usage guidelines to ensure drug safety (Patel and Dowse, 2015). Such as, educating a patient who is prescribed Coumadin to avoid certain herbal products like St John’s Wort and foods high in Vitamin K due to potential drug-food interaction. But pa- tients are often unaware of such critical information / advice related to their prescription drugs due to several reasons, in- cluding, (i) lack of communication with their doctors and/or pharmacists (Liddy et al., 2014), (Patel and Dowse, 2015) (ii) low health literacy (Wolf et al., 2006), and (iii) volume of received information (Savas and Evcik, 2000). Textual analysis of drug usage guidelines and automatically filtering critical personalized advice can aid a patient to ad- here to the drug usage guidelines. Such research can benefit from annotated datasets where the information is annotated according to its topic, e.g., whether a piece of advice from a DUG document is related to interaction with food or bev- erage or interaction with other drugs. For example, the an- notated advice from DUG documents can be forwarded to a medication reminder app to present advice that indicates potential interaction with (i) daily activities, (ii) …

Content truncated to 500 words