Adverse drug reactions and drug safety

Exploring whether the use of information technology and informatics systems can facilitate the collection and analysis of pharmacovigilance data, as well as the assessment of interventions designed to improve reporting.

Primarily funded by Veterinary Medicines Directorate, with additional support from The Wellcome Trust and the British Small Animal Veterinary Association's (BSAVA) PetSavers, our research in adverse drug reactions is focused on:

• Understanding and facilitating spontaneous adverse drug reaction reporting
• Extracting drug safety information from electronic health records.

Spontaneous adverse drug reaction reporting

We're working to establish the facilitators of, and barriers to, spontaneous adverse drug reaction reporting.

An initial survey identified practice management system (PMS) reporting as a potential facilitator . We then used the Small Animal Veterinary Surveillance Network (SAVSNET) system to facilitate reporting direct from the PMS for SAVSNET users. The reports generated were of similar high quality to those received by the Veterinary Medicines Directorate via other routes.

In partnership with the CVS Group, we are now expanding on this work through an ongoing clinical trial, assessing whether having in-practice reporting champions will increase reporting rates. 

Extracting drug safety information

SAVSNET holds a wealth of free text clinical notes which provide a unique opportunity to explore more active surveillance methods. In particular, SAVSNET holds outcome data for each patient, enabling the calculation of absolute incidences for specific drug events. This gives the opportunity to gain new insights into drug safety, not available through use of spontaneous reports made to regulators.

To date, we have:

• Demonstrated that adverse drug reactions are rarely reported, even when suspected, a similar finding to human medicine
• Demonstrated that medication error information is commonly included in electronic health records
• Developed methodologies for using SAVSNET data to calculate absolute incidence rates for drug event pairs.

We are now exploring the use of artificial intelligence (AI) and machine learning tools to expand this work.