Developing electronic healthcare records as a source of real-world data for veterinary pharmacoepidemiology

Heather Davies ¹ Peter-John Noble ¹ Ivo Salgueiro Fins ¹ Gina Pinchbeck ¹ David Singleton ¹ Munir Pirmohamed ² David Killick ^1*

• ¹ Institute of Infection, Veterinary and Ecological Sciences, Faculty of Health and Life Sciences, University of Liverpool, Liverpool, United Kingdom
• ² Institute of Systems, Molecular and Integrative Biology, Faculty of Health and Life Sciences, University of Liverpool, Liverpool, United Kingdom

Spontaneous reporting of adverse events (AEs) by veterinary professionals and members of the public is the cornerstone of post-marketing safety surveillance for veterinary medicinal products (VMPs). However, studies suggest that most veterinary AEs are not reported. Veterinary medicines regulators including the UK Veterinary Medicines Directorate and the European Medicines Agency, have incorporated exploration of the use of big data to support pharmacovigilance efforts into their regulatory strategies. Here we describe the use of veterinary electronic healthcare records (EHRs), taken from the SAVSNET veterinary first opinion informatics system, to conduct pharmacoepidemiological analyses. Five VMP-AE pairs were chosen for exploration in a proof-of-concept study in which drug exposure was identified from semi-structured treatment data and AEs from the unstructured free-text clinical narrative. Dictionaries were developed to identify AEs based on standard terminology. Dictionary precision was increased when dictionaries were expanded using word vectorization and expert opinion. A key strength of first opinion EHR datasets is they can permit cohort studies and allow calculation of absolute incidence and relative risk. Thus, here we demonstrate that unstructured free-text clinical narratives can be used to identify outcomes for veterinary pharmacoepidemiological studies and therefore support and expand upon the pharmacovigilance efforts based on spontaneous AE reports.