UNSTRUCTURED
Purpose: One-third of older inpatients experience adverse-drug events (ADEs), which increase their mortality and morbidity, and health care use and costs. In particular, antithrombotic drugs are among the most at-risk medications in this population. Reporting systems have been implemented at national, regional, and provider levels to monitor ADEs and design prevention strategies. Because of their well-known limitations, automated detection technologies based on electronic medical records (EMR) are currently being developed to detect routinely or predict ADEs.
Aim: This study aims to develop and validate an automated detection tool for the monitoring of antithrombotic-related ADEs using EMRs from four large Swiss hospitals. We aim to assess cumulative incidences of hemorrhages and thromboses in older inpatients associated with the prescription of antithrombotic drugs, identify triggering factors, and propose improvements for clinical practice.
Method: This collaborative and interdisciplinary project is a multicenter, cross-sectional study based on 2015-2016 EMR data from four large hospitals in Switzerland: Lausanne, Geneva, and Zürich University hospitals, and Baden Cantonal hospital. We have included older inpatients aged ≥65 years who stayed at one of the four hospitals during 2015 or 2016, received at least one antithrombotic drug during their stay, and signed or were not opposed to a general consent for participation in research. First, expert clinical pharmacists have selected a list of relevant antithrombotic drugs along with their side effects and risk and confounding factors. Second, administrative, clinical, prescription, and laboratory data available in the form of free texts and structured data have been extracted from study participants’ EMRs. Third, several automated rule-based and machine learning-based algorithms are currently being developed, allowing to identify hemorrhage and thromboembolic events and their triggering factors from extracted information. Finally, we plan to validate the developed detection tools (one per ADE type) through manual medical record review. Performance metrics for assessing internal validity will comprise the area under the receiver operating curve (AUC), F1-Score, sensitivity, specificity, and positive and negative predictive values.
Relevance/Application: The project will allow the introduction of measures to improve safety in the prescribing of antithrombotic drugs, which today remains among the drugs most involved in adverse drug events. The findings will be implemented in clinical practice by means of indicators of adverse events for risk management, and training for healthcare professionals; the tools and methodologies developed will be disseminated for new research in this field.
Conclusion: The increased performance of natural language processing as an important complement to structured data will bring existing tools to another level of efficiency in the detection of ADEs. Currently, such systems are not available in Switzerland and nor can “ready-made” systems from other countries be adapted since they are language dependent.