Using electronic health records to improve identification and reporting of adverse drug reactions in veterinary medicine


The opportunity

Train in the fields of big health data analysis, epidemiology and if you wish machine learning and computer science, as part of this novel and exciting project aimed at improving the health of UK pets through better detection of adverse drug reactions.


Pre-clinical testing cannot identify all possible adverse drug reactions (ADRs) and therefore ADRs continue to cause significant harm to some groups of clinical patients. Well known ADR examples include fluoroquinolone-induced retinal degeneration in cats and phenobarbital-induced bone marrow toxicity in certain breeds of dogs.

In the current system suspected ADRs identified in clinic are voluntarily reported to the regulatory authority (Veterinary Medicine Directorate) or manufacturing pharmaceutical company. In human medicine a similar scheme leads to reporting of only 10 % of suspected ADRs; there is also significant under reporting in veterinary medicine.

Electronic health records (EHRs) held within computerised practice management systems (PMS) hold a huge amount of data on prescribed treatments and outcomes for patients, including data regarding possible ADRs. Evaluating EHRs using data analysis approaches has been shown to enhance ADR identification in human medicine and has helped to reduce the adverse effects of drugs on patients.

SAVSNET is a big data informatics technology embedded within computerised veterinary practice management systems (PMS). SAVSNET captures data from the EHRs of animals visiting over 10% of UK veterinary practices and currently holds data from more than 4 million veterinary consultations. This data is providing new insights on species, breed and age susceptibility to disease and has been used in a number of successful projects.

What you will do

During this PhD project you will use SAVSNET to identify ADRs suffered by patients in general practice. You will be supported as you learn skills in the developing fields of pharmacovigilance, big data analysis and epidemiology. You will gain considerable practical experience with this skill set as you work toward the main aims of the project, which are to:

  1. Use big data analytic approaches such as free text and data mining of EHRs to identify possible ADRs.
  2. Evaluate whether prospective real-time monitoring of EHRs can facilitate faster identification of suspected ADRs due to newly licensed drugs.
  3. Investigate factors influencing the spontaneous reporting of suspected ADRs by veterinary surgeons.
  4. Assess whether novel features such as embedding ADR reporting tools into the PMS can enhance reporting of suspected ADRs by veterinary surgeons.
  5. Investigate whether SAVSNET data can be used to produce incidence data for ADRs to particular medicines.

Skills development and employability

As you work on the project and with skills development supported by the Liverpool Doctoral College you will enhance your core scientific, transferable and employability skills such as data analysis and problem solving, project management, teamwork, and collaborative and scientific communication skills (oral and written). There is also funding available to support attendance and presentation of your findings at conferences.

In addition there will also be opportunities to develop and apply programming skills (e.g Python, SQL), but these skills are not considered essential for the success of the project. Equally candidates without a clinical degree would be supported in learning clinical terminology used in EHRs.

Upon completion of this PhD you will be in an excellent position to develop a career in the growing field of pharmacovigilance within academia, pharma or government. Moreover, you will be ideally placed for jobs in the rapidly expanding field of data science in these sectors as well as the broader corporate world.

The team and environment

You will work within the friendly SAVSNET team and in collaboration with colleagues from the internationally recognised Centre for Drug Safety Science  and the Veterinary Medicines Directorate.

The University of Liverpool promotes gender equality in all activities. The University of Liverpool emphasizes the importance of the supportive nature of the working environment. The University of Liverpool has an ongoing commitment that the Athena SWAN principles are embedded in its activities and strategic initiatives. 

Due to its cross disciplinary nature this PhD would suit candidates holding a good undergraduate or Master’s degree in a wide range of clinical (veterinary medicine, veterinary nurse, pharmacy, medical, public health or epidemiology) or data science subjects. Enquiries to: Dr David Killick –

To apply:  please send your CV and a covering letter to with a copy to

Closing date for applications: 23th October 2018

Expected interview date/week: Late October / early November 2018


Open to students worldwide

Funding information

Funded studentship

This 3-year full-time studentship is fully funded by the Veterinary Medicines Directorate, which is part of the Department of Environment, Food and Rural Affairs. 

The award will cover University Fees at an EU/UK rate and the Doctoral Stipend for a period of 3 years full time study. Non EU/UK students will need to cover the cost of the remaining university fee at the international fee rate. The (tax-free) stipend is £19000. 

The candidate will require a good level of proficiency in English as the EHRs will be written in English.



Bates, D.W. et al., 2003. Detecting Adverse Events Using Information Technology. Journal of the American Medical Informatics Association, 10(2), pp.115–128. Available at: lookup/doi/10.1197/jamia.M1074

Radford, a et al., 2010. Developing a network for small animal disease surveillance. The Veterinary record, 167(13), pp.472–4. Available at:

Ribeiro-Vaz, I. et al., 2016. How to promote adverse drug reaction reports using information systems – a systematic review and meta-analysis. BMC Medical Informatics and Decision Making, pp.1–10. Available at: