SAVSNET: Developing the research potential of veterinary health informatics in the UK through growth, partnership and collaboration.

Lead Research Organisation: University of Liverpool
Department Name: Institute of Infection and Global Health

Abstract

Illness of pets (cats, dogs, rabbits etc.) impacts not just on their welfare, but also on that of humans. Despite this, we often have poor understanding of what makes individual animals unwell, and what makes their owners take them to the vet. One area of research revolutionising such understanding of human illness is "health informatics", where large volumes of electronic patient health records (EHRs) are anonymised and compiled into a "Big data" resource, allowing new research insights and discoveries. Whilst medical informatics has been seen a rapid recent expansion, catalysed by the establishment of a national research network (the Farr Institute), veterinary health informatics remains a relatively unexplored opportunity.
SAVSNET (Small Animal Veterinary Surveillance NETwork; www.savsnet.co.uk) is pioneering such data resources for pets. In just over 2 years, we have collected health records in real-time from diagnostic laboratories (~80,000 tests/day) and 166 veterinary practices (~3000 consultation records/day) from across the UK. This has allowed us to create two large data resources (>900,000 EHRs, >40million test results) that are updated and growing daily. Our novel software allows the attending veterinary surgeon or nurse with a single key stroke to indicate the main reason the animal was seen, and in return, allows the same vets and nurses to analyse their own data. These data are being used to provide freely-accessible disease information and surveillance updates, along with a growing portfolio of new and novel research collaborations including studies on population demographics, preventative health care, antibacterial use, rabbit disease, as well as kidney, respiratory and gastrointestinal diseases. SAVSNET is now recognised as an international leader in this new area of veterinary health informatics research.
The aim of this proposal is to increase the amount of research carried out using SAVSNET data. We will capitalise and build on our strong foundation, maximising the opportunities for these data to inform new research with a wide group of collaborators, all with the intention of improving both animal and human health and wellbeing.
The objectives of this application are to:
1: Enable more laboratories and veterinary practices to contribute data. The resultant increase in data volumes will help ensure our data is representative of the whole of the UK, and allow the study of even rare diseases.
2: Strengthen our databases to ensure they can cope with the large data volumes and growing number of users.
3: Build a web based "one-stop-shop" for researchers (e-Research Portal), to make the data more accessible, easier to understand and increase its use in research and publication.
4: Develop a world leading veterinary health informatics research base through strengthening collaboration with medical health informatics colleagues at the Farr Institute, University of Manchester.
5: Develop and showcase new data analyses tools, cascading these to other users via the SAVSNET e-Research Portal. Text mining, where computers are taught how to extract useful meaning from unstructured writing, will be used to unlock the research potential of the clinical narrative entered by vets and nurses. These tools will be developed in the context of two strategically important research projects; 1) factors associated with use of antibacterials and 2) seasonal tick activity in the context of climate.
6: Diversify data use, developing a business model for future SAVSNET sustainability. With BBSRC support, we will grow our user base and funding from academia and industry, to allow SAVSNET to become self-funding.
At the end of the project, we will have established a sustainable international centre of excellence for veterinary health informatics research, developing innovative software solutions, to answer some of the critical real-world animal and human health problems of the 21st century.

Technical Summary

Illness in pet animals impacts both on their welfare, and on that of their owners and the wider public. The Small Animal Veterinary Surveillance NETwork (SAVSNET) is pioneering the use of electronic health records (EHRs) at scale to better understand the real-world of clinical veterinary practice. SAVSNET has assembled a strong coalition of collaborators allowing us to collect UK data in real-time from commercial diagnostic laboratories (~80,000 test results/day) and a sentinel network of 166 veterinary practices (~3000 EHRs/day). We have a mature innovative infrastructure to collect and store these data, and to feedback results to our collaborators, in near real-time. Our use of these data has rapidly grown, supporting education, real-time disease surveillance, and a growing number of nascent research collaborations, with innovative informatics solutions already informing human as well as animal health.
In this proposal, we will consolidate and build on our recent advances to maximise the research potential of these data to the widest scientific community. Our key objectives are to:
1: Increase the statistical power and representativeness of SAVSNET by enabling more laboratories and veterinary practices to contribute data.
2: Future-proof our database for both growing data volumes and increasing numbers of users.
3: Facilitate and enrich data use, research and collaboration through an "e-Research Portal".
4: Develop a world leading veterinary health informatics research base in partnership with Farr@HeRC.
5: Through two research packages (risk factors for antibacterial use and seasonality of tick activity), develop and showcase innovative informatics resources, cascading these to other users.
6: Diversify data use, developing a business model for future sustainability.
We have made rapid progress since our core team was put in place. This gives us confidence we can establish SAVSNET as a world leading resource for research into animal and human health.

Planned Impact

The SAVSNET data has appeal and value to end a wide variety of end-users. The benefits and impacts that can be delivered are numerous and the examples highlighted below are intended to illustrate their diversity.

1) Animal health industry. There are estimated 10 million cats and 10 million dogs in the UK with approximately 24%, 17% and 2% of UK households owning a dog, cat or rabbit respectively. This degree of human animal contact has considerable economic, social benefits, as well as occasional negative effects. SAVSNET collects large volumes of data across all health areas, from anal furunculosis to zoonosis, from Afghan hounds to Yorkshire terriers. As such we can impact on all areas of animal health management and research with recent projects on infection, kidney disease, and osteoarthritis.

2) Medical Health informatics. SAVSNET has close links with the Farr Institute through Farr@HeRC (Manchester). SAVSNET data is more accessible for method development and testing than equivalent human health data.

2) Researchers employed by pharmaceutical companies are beginning to recognise the value of SAVSNET data for market research either on specific products, or background animal health data that can inform commercial decisions. Many pharmaceutical companies are multinational and whilst SAVSNET data is currently focussed on UK companion animal populations, there is recognition that the research has commercial benefits worldwide. SAVSNET is becoming recognised as a centre of global excellence for such research, and already realising opportunities for inward investment to the UK.

3) Data providers (practices and laboratories) benefit directly from participation in SAVSNET through access to secure, web-based summaries of the data they have submitted. The portals have been released but need to be regularly updated and refined to ensure the information is useful and relevant. The veterinary profession as a whole will also benefit as SAVSNET will provide data to support evidence-based veterinary medicine and inform clinical decisions.

4) Government has responsibility for disease surveillance in animal populations. SAVSNET will enable cost-effective surveillance of diagnosed diseases or presenting syndromes in companion animals. In addition, the novel methods and tools developed by SAVSNET will be cascaded to farm animal researchers to facilitate alternative, cost-effective surveillance (see APHA LoS).

5) National Health Service and Public Health England (see PHE LoS). Studying zoonotic diseases in pets, risk factors for canine aggression, understanding how and why veterinary surgeons prescribe antibiotics, and identifying associations between disease outbreaks in pets and humans will all have a positive impact on human health. Radford leads the Risk Theme of the National Institute of Health, Research Health Protection Research Unit in Emerging and Zoonotic Infection, and this is already providing opportunities for SAVSNET data to impact human health (e.g. surveying tick removal in pets to develop a warning system to reduce Lyme borreliosis in humans).

6) Charities feature strongly in companion animal care in the UK, helping control stray animals, providing health care to lower income families, providing Assistance Animals, and our funders BSAVA, whose aim is to promote excellence in animal care through teaching and research. We have already helped PDSA change their charitable objectives (see PDSA LoS), and contribute education material to vets through BSAVA.

7) SAVSNET will impact on the general public and wider society by promoting improved health in pets, reducing disease burden, improving quality of veterinary care and improving understanding of the role of the veterinary profession in AMR. Our developing web site (SAVSNET.co.uk) provides one way to achieve this.

Publications

10 25 50

publication icon
Arsevska E (2017) Small animal disease surveillance: GI disease and salmonellosis. in The Veterinary record

publication icon
Farrell S (2020) Seasonality and risk factors for myxomatosis in pet rabbits in Great Britain in Preventive Veterinary Medicine

publication icon
Robinson N (2019) Risks of xylitol poisoning in dogs. in The Veterinary record

publication icon
Singleton DA (2020) Prolific vomiting in dogs. in The Veterinary record

publication icon
Sánchez-Vizcaíno F (2016) Small animal disease surveillance: pruritus, and coagulase-positive staphylococci. in The Veterinary record

publication icon
Sánchez-Vizcaíno F (2016) Small animal disease surveillance: respiratory disease. in The Veterinary record

 
Description Antibiotics are commonly used but in veterinary practice, and include worrying levels of critical antibiotics being used for cats. However, use is reducing. This year we have led on an interventional trial based on this grant and shown that a modest educational intervention can significantly reduce the use of critical antibiotics by those who have used them the most Ion preparation for publication). We have also identified critical risk factors for antibiotic use (in review by journal).
Ticks identified in health records have a striking seasonal and geographical pattern; these data may form the basis of an ongoing tick surveillance tool.
As we approach the end of the grant we have gained broad recognition for the research work we are doing. This will hopefully lead to further collaborations in the following years in line with the BBR ethos of becoming a data resource.
Further publications from the early work on this grant will be submitted this year including three more papers on antibiotic resistance and use, and a second paper on tick distributions modelling across the UK.
Exploitation Route The data we are collecting is available and now being used by other researchers within the UK. We have active collaborations at Universities of Lancaster, and Bristol, with specialist veterinary practitioners, commercial companies, Veterinary Medicines Directorate, Public Heath England, Health Protection Scotland and charities. Of greatest significance, recent funding from DogsTrust will allow us to establish an outbreak response capability (https://www.liverpool.ac.uk/savsnet/savsnet-agile/) which has already been tested as we lead a national response to an outbreak of severe vomiting in dogs (https://www.liverpool.ac.uk/savsnet/dog_vomiting_potential_outbreak/). All this has been facilitated by the role out of an on-line data application process available at https://www.liverpool.ac.uk/savsnet/using-savsnet-data-for-research/,
Sectors Agriculture, Food and Drink,Digital/Communication/Information Technologies (including Software),Healthcare,Pharmaceuticals and Medical Biotechnology,Other

URL https://www.liverpool.ac.uk/savsnet/
 
Description SAVSNET research data is already underpinning national surveillance of companion animal disease both through our live and real-time public-facing web site (https://www.liverpool.ac.uk/savsnet/real-time-data/) and also for the veterinary profession through surveillance publications in the Veterinary Record (three per year). This work has been recognised through its inclusion on the APHA web site http://apha.defra.gov.uk/vet-gateway/surveillance/experts/Small-animal.htm. Thi has allowed us to advise govt (Defra) on an potential outbreak of gastrointestinal disease in Norway, and more recently on an actual outbreak of gastrointestinal disease in the UK (https://www.liverpool.ac.uk/savsnet/dog_vomiting_potential_outbreak/). We launched a service based on our research data in 2018 which allows antibacterial prescription bench-marking for veterinary practitioners. This has been used by some 60 practice sites to allow them to better understand their antibiotic use. This was implemented in an Interventional trial aimed at reducing antibacterial use (which was successful), that is now being rolled out across a large practice group. The professional regulator of vets in the UK (Royal College of Veterinary Surgeons) has commissioned (and received) a project to provide evidence to a policy review they are completing on vaccination.
First Year Of Impact 2018
Sector Other
Impact Types Societal,Economic,Policy & public services

 
Description Antibiotic use by Scottish small animal veterinary clinicians
Geographic Reach National 
Policy Influence Type Citation in other policy documents
URL https://www.hps.scot.nhs.uk/web-resources-container/scottish-one-health-antimicrobial-use-and-antimi...
 
Description CItation in Parliamentary Office of Science and Technology POSTNote: Reducing UK Antibiotic Use in Animals.
Geographic Reach National 
Policy Influence Type Gave evidence to a government review
URL https://researchbriefings.parliament.uk/ResearchBriefing/Summary/POST-PN-0588
 
Description Inclusion in CVS Quality Improvement report - antibiotic use intervention trial
Geographic Reach National 
Policy Influence Type Influenced training of practitioners or researchers
 
Description Member of the British Veterinary Association Surveillance working group
Geographic Reach National 
Policy Influence Type Participation in a advisory committee
URL https://www.bva.co.uk/uploadedFiles/Content/News_campaigns_and_policy/Policy/Animal_disease_surveill...
 
Description Postgraduate research placements
Geographic Reach Local/Municipal/Regional 
Policy Influence Type Influenced training of practitioners or researchers
Impact Two postgraduate veterinary surgeons are completing research projects with SAVSNET data from this project and under our supervision.
 
Description Quality Improvement Advisory Board Task and Finish Group
Geographic Reach National 
Policy Influence Type Participation in a advisory committee
 
Description UK veterinary surveillance forum
Geographic Reach National 
Policy Influence Type Participation in a national consultation
URL https://www.gov.uk/government/groups/uk-surveillance-forum-uksf
 
Description Veterinary Antibiotic Resistance and Sales Surveillance Report (VARSS) report 2016
Geographic Reach National 
Policy Influence Type Citation in other policy documents
URL https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/7079...
 
Description Commercial funder
Amount £15,000 (GBP)
Organisation Ceva Animal Health 
Sector Private
Country United Kingdom
Start 12/2016 
End 07/2017
 
Description Commercial project
Amount £7,000 (GBP)
Organisation Zoetis 
Sector Private
Country United States
Start 05/2018 
End 06/2018
 
Description Confidential report
Amount £15,000 (GBP)
Organisation Zoetis 
Sector Private
Country United States
Start 12/2018 
End 05/2019
 
Description Direct commission
Amount £12,020 (GBP)
Organisation Dogs Trust 
Sector Charity/Non Profit
Country United Kingdom
Start 06/2017 
End 11/2017
 
Description Dogs Trust Canine Welfare Grants
Amount £509,702 (GBP)
Organisation Dogs Trust 
Sector Charity/Non Profit
Country United Kingdom
Start 08/2019 
End 07/2022
 
Description GP West Award
Amount £54,515 (GBP)
Funding ID GPWest_2017_02 
Organisation Animal Welfare Foundation 
Sector Charity/Non Profit
Country United Kingdom
Start 07/2018 
End 08/2019
 
Description Project led by external applicant - epilepsy in dogs
Amount £4,000 (GBP)
Organisation Petsavers Charitable Trust 
Sector Charity/Non Profit
Country United Kingdom
Start 10/2018 
 
Description Project led by external applicant - heatstroke in dogs
Amount £806 (GBP)
Organisation Nottingham Trent University 
Sector Academic/University
Country United Kingdom
Start 11/2017 
 
Description Project led by external applicant - proBNP
Amount £5,540 (GBP)
Organisation University of Bristol 
Sector Academic/University
Country United Kingdom
Start 01/2017 
End 01/2019
 
Description SAVSNet-Agile; responsive data intelligence for canine health
Amount £509,702 (GBP)
Organisation Dogs Trust 
Sector Charity/Non Profit
Country United Kingdom
Start 04/2019 
End 03/2022
 
Description Small Grant Scheme
Amount £7,062 (GBP)
Organisation Animal Welfare Foundation 
Sector Charity/Non Profit
Country United Kingdom
Start 06/2018 
End 09/2018
 
Description Small animal disease surveillance
Amount £30,000 (GBP)
Organisation Intervet-Schering Plough 
Sector Private
Country United States
Start 07/2019 
End 07/2020
 
Description Surveillance of clinical antimicrobial susceptibility test results in dogs and cats
Amount £16,500 (GBP)
Organisation The Veterinary Medicines Directorate 
Sector Hospitals
Country United Kingdom
Start 03/2019 
End 10/2019
 
Title mySAVSNET-AMR 
Description mySavsnet AMR is an initiative which was developed as a result of research looking at how antibiotics are used in veterinary practices taking part in SAVSNET. This research tool is aimed at those working in veterinary practice in the UK, whether part of SAVSNET or not. It allows us to collect data form veterinary practices across the UK, comparing their antibiotic prescription to other anonymised practices across the country. Data can be from an individual practitioner, a practice site or the whole practice. Multiple data sets can also be sent to see how prescription changes over time. These anonymised data will also be used by SAVSNET as part of ongoing research to understand antibiotic prescription and it's variation across the UK. Participants also benefit from a free antibiotic use benchmark report. This tool has now been used by approximately 20 practice sites across the UK to benchmark their antibiotic use - feedback has been very positive. 
Type Of Material Improvements to research infrastructure 
Year Produced 2017 
Provided To Others? Yes  
Impact This initiative has only just become available. Any impact will become apparent as more vets make use of it to audit their own antibacterial prescriptions. 
URL https://www.liverpool.ac.uk/savsnet/my-savsnet-amr/
 
Title SAVSNET-Datalab 
Description SAVSNET collects clinical records from veterinary practices. This data is parsed into central repository. SAVSNET datalab cleans the data (applies a text deidentifier) and stores it in a MySQL database and through a Python/Django based GUI allows researchers to filter records based on features in the clinical narratives and to annotate the data for features of interest. The software allows retrieval of resultant datasets along with automated production of case-matched controls. 
Type Of Material Data handling & control 
Year Produced 2017 
Provided To Others? No  
Impact This software has been used for multiple student and post-graduate studies, underpinning several publications and on-going consultancy work. Most recently this software has played a key role in collating data for identification and monitoring of a disease outbreak in dogs 
 
Title Small Animal Veterinary Surveillance Network 
Description Anonymised Electronic Health Records from Veterinary practices and diagnostic laboratories are cleaned and made research-ready. On 31st July 2016 (pre-BBR grant), we had 1,845,240 consults from veterinary practices. By February of 2019 we had collected 5,438,861 consults from veterinary practice. This means that during BBR funding we have collected data on 3,593,621 consults and we are collecting data form ~250 practices, which represents approximately 10% of all UK veterinary practices. For laboratory data we have over 100 million individual test results from approximately 8 laboratories. We estimate these laboratories complete the testing for over 70% of UK veterinary practices. These data are collected essentially yin real time and are increasingly available to researchers in real time. 
Type Of Material Database/Collection of data 
Year Produced 2015 
Provided To Others? Yes  
Impact These data now contribute to national disease research and surveillance. We have published 7 surveillance reports over the course of the project in the Veterinary Record, three in the last year. These are now referenced by the Animal and Plant Health Agency through their web site - http://apha.defra.gov.uk/vet-gateway/surveillance/experts/Small-animal.htm 
URL http://www.savsnet.co.uk
 
Description Collaboration with Naveda - Swedish Veterinary Surveillance 
Organisation Green Employers
Country Sweden 
Sector Charity/Non Profit 
PI Contribution a. the exchange of scientific and technical knowledge and materials in conjunction with the development of the Swedish database Naveda b. visits to the UK and to Sweden by the parties during the development of Naveda c. joint research activities when Naveda is functional d. shared supervision of research students e. student visits and exchanges at graduate levels f. other activities of mutual interest involving data from SAVSNET and Naveda
Collaborator Contribution a. the exchange of scientific and technical knowledge and materials in conjunction with the development of the Swedish database Naveda b. visits to the UK and to Sweden by the parties during the development of Naveda c. joint research activities when Naveda is functional d. shared supervision of research students e. student visits and exchanges at graduate levels f. other activities of mutual interest involving data from SAVSNET and Naveda
Impact Naveda, the Swedish surveillance system, is still in the early stages of development but it is anticipated that this collaboration will result in working on shared projects, sharing of code and technological developments and the sharing of datasets which will provide a unique insight into animal health in the UK and Sweden.
Start Year 2017
 
Description Collaboration with Naveda - Swedish Veterinary Surveillance 
Organisation Swedish University of Agricultural Sciences
Country Sweden 
Sector Academic/University 
PI Contribution a. the exchange of scientific and technical knowledge and materials in conjunction with the development of the Swedish database Naveda b. visits to the UK and to Sweden by the parties during the development of Naveda c. joint research activities when Naveda is functional d. shared supervision of research students e. student visits and exchanges at graduate levels f. other activities of mutual interest involving data from SAVSNET and Naveda
Collaborator Contribution a. the exchange of scientific and technical knowledge and materials in conjunction with the development of the Swedish database Naveda b. visits to the UK and to Sweden by the parties during the development of Naveda c. joint research activities when Naveda is functional d. shared supervision of research students e. student visits and exchanges at graduate levels f. other activities of mutual interest involving data from SAVSNET and Naveda
Impact Naveda, the Swedish surveillance system, is still in the early stages of development but it is anticipated that this collaboration will result in working on shared projects, sharing of code and technological developments and the sharing of datasets which will provide a unique insight into animal health in the UK and Sweden.
Start Year 2017
 
Description National research and surveillance of companion animal disease: PORTUGAL 
Organisation University of Porto
Country Portugal 
Sector Academic/University 
PI Contribution Our research will provide knowledge and skills to Vet-Oncont (https://www.vetonconet.pt/) who are seeking to establish a similar research and surveillance network in Portugal with a focus on cancer.
Collaborator Contribution Currently, this contribution is mostly financial and for which we are awaiting contracts. We would envisage their team would bring expertise around cancer biology which is less strong in out Liverpool group.
Impact None yet
Start Year 2020
 
Description National research and surveillance of companion animal disease: USA 
Organisation University of Minnesota
Country United States 
Sector Academic/University 
PI Contribution We are providing IP generated over the course of this project to encourage a similar data collection and research capability led by academic partners at the University of Minnesota and the Minnesota Department of Health. This new project will be called CAVSNET - companion animal veterinary surveillance network.
Collaborator Contribution This partnership opens up new research funding opportunities in the USA.
Impact Internal Univeristy of Minnesota grant to establish a local pilot in the USA on which we are collaborators. Research pilot includes epidemiologists and veterinary clinicians.
Start Year 2019
 
Description UK Veterinary and Medical Health Informatics researchers meet in Manchester to launch Farr@Vet 
Organisation Animal Health Trust
Country United Kingdom 
Sector Charity/Non Profit 
PI Contribution We instigated this network
Collaborator Contribution Recognising both this growing veterinary health informatics research base in the UK, as well as the potential rewards of linking human and animal data together to help achieve 'One Health' outcomes, a group of UK Veterinary and Medical Health Informatics researchers met on the 9th and 10th of March in Manchester to discuss Farr@Vet, a new initiative to harness veterinary electronic health data and build capacity in veterinary health informatics. The meeting included veterinary researchers from the Animal Health Trust, the Animal and Plant Health Agency, Universities of Bristol (Bristol Cat and Generation Pup cohort studies), Edinburgh (Dogslife), Glasgow (equine clinical data), Liverpool (SAVSNET; Small Animal Veterinary Surveillance Network), and Nottingham (CEVM; Centre for Evidence-based Veterinary Medicine), and the Royal Veterinary College (VetCompass), together with colleagues from the Farr Institute (University of Manchester - Farr@HeRC). Farr@Vet recognises the huge opportunities for veterinary health informatics in the UK going forward, particularly if there is collaboration with the Farr Institute. Key areas discussed included the lack of denominator (population size) data for companion animals, the role of coding in data generated by practitioners and other veterinary surgeons, the value of "text mining" to extract meaning from free-text clinical data, and of course sustainability.
Impact Two text mining workshops
Start Year 2016
 
Description UK Veterinary and Medical Health Informatics researchers meet in Manchester to launch Farr@Vet 
Organisation Animal and Plant Health Agency
Country United Kingdom 
Sector Public 
PI Contribution We instigated this network
Collaborator Contribution Recognising both this growing veterinary health informatics research base in the UK, as well as the potential rewards of linking human and animal data together to help achieve 'One Health' outcomes, a group of UK Veterinary and Medical Health Informatics researchers met on the 9th and 10th of March in Manchester to discuss Farr@Vet, a new initiative to harness veterinary electronic health data and build capacity in veterinary health informatics. The meeting included veterinary researchers from the Animal Health Trust, the Animal and Plant Health Agency, Universities of Bristol (Bristol Cat and Generation Pup cohort studies), Edinburgh (Dogslife), Glasgow (equine clinical data), Liverpool (SAVSNET; Small Animal Veterinary Surveillance Network), and Nottingham (CEVM; Centre for Evidence-based Veterinary Medicine), and the Royal Veterinary College (VetCompass), together with colleagues from the Farr Institute (University of Manchester - Farr@HeRC). Farr@Vet recognises the huge opportunities for veterinary health informatics in the UK going forward, particularly if there is collaboration with the Farr Institute. Key areas discussed included the lack of denominator (population size) data for companion animals, the role of coding in data generated by practitioners and other veterinary surgeons, the value of "text mining" to extract meaning from free-text clinical data, and of course sustainability.
Impact Two text mining workshops
Start Year 2016
 
Description UK Veterinary and Medical Health Informatics researchers meet in Manchester to launch Farr@Vet 
Organisation Royal Veterinary College (RVC)
Country United Kingdom 
Sector Academic/University 
PI Contribution We instigated this network
Collaborator Contribution Recognising both this growing veterinary health informatics research base in the UK, as well as the potential rewards of linking human and animal data together to help achieve 'One Health' outcomes, a group of UK Veterinary and Medical Health Informatics researchers met on the 9th and 10th of March in Manchester to discuss Farr@Vet, a new initiative to harness veterinary electronic health data and build capacity in veterinary health informatics. The meeting included veterinary researchers from the Animal Health Trust, the Animal and Plant Health Agency, Universities of Bristol (Bristol Cat and Generation Pup cohort studies), Edinburgh (Dogslife), Glasgow (equine clinical data), Liverpool (SAVSNET; Small Animal Veterinary Surveillance Network), and Nottingham (CEVM; Centre for Evidence-based Veterinary Medicine), and the Royal Veterinary College (VetCompass), together with colleagues from the Farr Institute (University of Manchester - Farr@HeRC). Farr@Vet recognises the huge opportunities for veterinary health informatics in the UK going forward, particularly if there is collaboration with the Farr Institute. Key areas discussed included the lack of denominator (population size) data for companion animals, the role of coding in data generated by practitioners and other veterinary surgeons, the value of "text mining" to extract meaning from free-text clinical data, and of course sustainability.
Impact Two text mining workshops
Start Year 2016
 
Description UK Veterinary and Medical Health Informatics researchers meet in Manchester to launch Farr@Vet 
Organisation University of Bristol
Country United Kingdom 
Sector Academic/University 
PI Contribution We instigated this network
Collaborator Contribution Recognising both this growing veterinary health informatics research base in the UK, as well as the potential rewards of linking human and animal data together to help achieve 'One Health' outcomes, a group of UK Veterinary and Medical Health Informatics researchers met on the 9th and 10th of March in Manchester to discuss Farr@Vet, a new initiative to harness veterinary electronic health data and build capacity in veterinary health informatics. The meeting included veterinary researchers from the Animal Health Trust, the Animal and Plant Health Agency, Universities of Bristol (Bristol Cat and Generation Pup cohort studies), Edinburgh (Dogslife), Glasgow (equine clinical data), Liverpool (SAVSNET; Small Animal Veterinary Surveillance Network), and Nottingham (CEVM; Centre for Evidence-based Veterinary Medicine), and the Royal Veterinary College (VetCompass), together with colleagues from the Farr Institute (University of Manchester - Farr@HeRC). Farr@Vet recognises the huge opportunities for veterinary health informatics in the UK going forward, particularly if there is collaboration with the Farr Institute. Key areas discussed included the lack of denominator (population size) data for companion animals, the role of coding in data generated by practitioners and other veterinary surgeons, the value of "text mining" to extract meaning from free-text clinical data, and of course sustainability.
Impact Two text mining workshops
Start Year 2016
 
Description UK Veterinary and Medical Health Informatics researchers meet in Manchester to launch Farr@Vet 
Organisation University of Manchester
Department The Kellgren Centre of Rheumatology
Country United Kingdom 
Sector Hospitals 
PI Contribution We instigated this network
Collaborator Contribution Recognising both this growing veterinary health informatics research base in the UK, as well as the potential rewards of linking human and animal data together to help achieve 'One Health' outcomes, a group of UK Veterinary and Medical Health Informatics researchers met on the 9th and 10th of March in Manchester to discuss Farr@Vet, a new initiative to harness veterinary electronic health data and build capacity in veterinary health informatics. The meeting included veterinary researchers from the Animal Health Trust, the Animal and Plant Health Agency, Universities of Bristol (Bristol Cat and Generation Pup cohort studies), Edinburgh (Dogslife), Glasgow (equine clinical data), Liverpool (SAVSNET; Small Animal Veterinary Surveillance Network), and Nottingham (CEVM; Centre for Evidence-based Veterinary Medicine), and the Royal Veterinary College (VetCompass), together with colleagues from the Farr Institute (University of Manchester - Farr@HeRC). Farr@Vet recognises the huge opportunities for veterinary health informatics in the UK going forward, particularly if there is collaboration with the Farr Institute. Key areas discussed included the lack of denominator (population size) data for companion animals, the role of coding in data generated by practitioners and other veterinary surgeons, the value of "text mining" to extract meaning from free-text clinical data, and of course sustainability.
Impact Two text mining workshops
Start Year 2016
 
Description UK Veterinary and Medical Health Informatics researchers meet in Manchester to launch Farr@Vet 
Organisation University of Nottingham
Country United Kingdom 
Sector Academic/University 
PI Contribution We instigated this network
Collaborator Contribution Recognising both this growing veterinary health informatics research base in the UK, as well as the potential rewards of linking human and animal data together to help achieve 'One Health' outcomes, a group of UK Veterinary and Medical Health Informatics researchers met on the 9th and 10th of March in Manchester to discuss Farr@Vet, a new initiative to harness veterinary electronic health data and build capacity in veterinary health informatics. The meeting included veterinary researchers from the Animal Health Trust, the Animal and Plant Health Agency, Universities of Bristol (Bristol Cat and Generation Pup cohort studies), Edinburgh (Dogslife), Glasgow (equine clinical data), Liverpool (SAVSNET; Small Animal Veterinary Surveillance Network), and Nottingham (CEVM; Centre for Evidence-based Veterinary Medicine), and the Royal Veterinary College (VetCompass), together with colleagues from the Farr Institute (University of Manchester - Farr@HeRC). Farr@Vet recognises the huge opportunities for veterinary health informatics in the UK going forward, particularly if there is collaboration with the Farr Institute. Key areas discussed included the lack of denominator (population size) data for companion animals, the role of coding in data generated by practitioners and other veterinary surgeons, the value of "text mining" to extract meaning from free-text clinical data, and of course sustainability.
Impact Two text mining workshops
Start Year 2016
 
Title Identification of parasitism and infectious diseases 
Description Our research findings are being used by a commercial company to promote infectious disease awareness and prophylaxis 
IP Reference  
Protection Protection not required
Year Protection Granted 2018
Licensed Commercial In Confidence
Impact Our research findings are being used by a commercial company to promote infectious disease awareness and prophylaxis
 
Title Deidentification of veterinary clinical narratives 
Description The clinical narratives contained within veterinary electronic health records often contain unintentional personal details such as the name of the attending veterinary surgeon and the name and contact details of owners and their relatives. A tool was developed by a HeRC-funded PGR at University of Liverpool to automatically recognise such details and replace the text with context-specific placeholders that enable the clinical narrative to be read without loss of meaning. 
Type Of Technology Software 
Year Produced 2016 
Impact The inclusion of personal data in veterinary clinical narratives is commonly encountered by many groups working to develop veterinary health informatics. The tool described is currently being used by several collaborator groups in the UK. 
 
Title SAVSNET Datalab 
Description SAVSNET Datalab is a web front end to interact with the large SAVSNET database. Datalab allows complex rule-based searching of clinical records for specific features in addition to manual tagging or records for use in epidemiological studies. The software is written in Python using the DJango framework, and builds on command-line Python tools already developed previously in 2016. Datalab allows rapid screening of candidate cohorts within the SAVSNET dataset prior to more focused analyses. It is currently only accessible to those working within the University of Liverpool firewall. Datalab is a prototype for an externally accessible yet secure health informatics suite which will be made available through SAVSNET in coming years. 
Type Of Technology Webtool/Application 
Year Produced 2017 
Impact Datalab is a health informatics tool that has enabled students and researchers to carry out various studies, some of which are now reaching publication. 
 
Title The use of text-mining methods to automatically extract clinically-relevant information from veterinary clinical narratives and to classify veterinary consultations based on presenting signs 
Description Veterinary clinical narratives are often unstructured, unpunctuated and contain many abbreviations and colloquialisms. A software tool was developed by a HeRC-funded PGR at the University of Liverpool that used a series of nested regular expressions to automatically extract clinically-relevant information from the clinical narratives, including temperature, pulse and respiration rates, presenting signs (taking account of negation) and prescribed treatments. 
Type Of Technology Software 
Year Produced 2016 
Impact This tool is starting to be used within the SAVSNET team to rapidly extract valuable clinical information and to classify individual consultations based on presenting signs without the need for the attending veterinary surgeon to enter clinical codes. Clinical coding systems are available in the veterinary field but they are utilised in irregular and unpredictable ways. The use of the described tool enables the large number of consultations in a database to be rapidly and consistently classified. 
 
Title Veterinary Practice Portal / Dashboard 
Description Practices that send us their data benefit form a free dashboard that shows how their data compares to their anonymised peers. This includes maps, KPIs, syndrome data and treatments, which we have recently upgraded to allow better comparison of antibiotic use between a practice's sites and at the species levels. 
Type Of Technology Webtool/Application 
Year Produced 2018 
Impact Better support for veterinary practices in assessing their antibiotic usage which is not available through any other means. 
 
Description Edinburgh International Science Festival 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach Regional
Primary Audience Public/other audiences
Results and Impact To introduce members of the public attending the Edinburgh International Science Festival to a new area of science called Health Informatics, which reuses electronic health records for research and surveillance. Specifically, we will showcase the power of health informatics research using anonymised data collected from a large sentinel network of veterinary practitioners by SAVSNET (the Small Animal Veterinary Surveillance Network - SAVSNET.co.uk). The data used will specifically address "One Health" issues, including leading research on ticks.
Our stand will introduce members of the public to Health Informatics research. We will show how data is collected, and what type of data is collected. We will create age appropriate opportunities for members of the public to interact with real data, and interpret it and present their results. These will range from using coloured blocks to build towers representing how common two states are in a simple data set - for example dogs and cats. The towers when viewed side-on are actually histograms. At the other end of the scale we will involve members of the public in text mining, a methodology to extract clinical meaning from free text, to create a map of a particular disease - for example ticks. The latter will require electronic tablet devices (or similar), and a "sticky Velcro map" of the UK, to allow results to be presented. Over the course of the week, this science could build up into a meaningful piece of real research that could perhaps contribute to a publication.
Year(s) Of Engagement Activity 2017
 
Description Healtex Launch Event 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Professional Practitioners
Results and Impact This was the official launch of the network that gathered over 70 people for a two day event. It provided an opportunity for establishing links and networking.

Agenda (Monday, November 14th):

12:30-2:00 Registration and Lunch
2:00-2:15 Goran Nenadic: "The UK healthcare text analytics network"
2:15-2:45 Iain Buchan: "Mining text for health discovery and actionable analytics"
2:45-3:15 Robert Stewart: "Unmet needs and immediate challenges in healthcare text processing - the CRIS experience"

3:15-4:00 Break, posters, networking

4:00-4:30 Nigel Collier: "NLP capabilities and challenges in the health arena"
4:30-5:00 Elizabeth Ford: "Issues of Privacy and Governance when Using Medical Free Text for Research"

5:00-6:30 Posters, networking, wine/nibbles reception
Year(s) Of Engagement Activity 2016
URL http://healtex.org/event/healtex-launch/
 
Description Meet the scientists 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach Local
Primary Audience Schools
Results and Impact Workshop for school children as part of the Institute of Infection and Global Health Christmas event - focused on infection control. Children got involved in practical activities and discussion around how infection can be controlled.
Year(s) Of Engagement Activity 2018
 
Description SAVSNET Summer School 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach Local
Primary Audience Professional Practitioners
Results and Impact The SAVSNET Summer School enables A-Level students to gain experience of a research environment and investigate a range of topics using SAVSNET data. The Summer School involves a series of seminars including data collection, introduction to epidemiology and a focus on a clinical condition (this year the focus was on obesity). Students were given a topic to investigate and at the end of the Summer School, they presented to the SAVSNET. The Summer School is useful for testing ideas and this year we are pleased to say that one topic, chocolate toxicity, looked very promising - this has been further developed by the SAVSNET team and will be submitted for publication.
Year(s) Of Engagement Activity 2017
 
Description SAVSNET research summer school 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach Local
Primary Audience Schools
Results and Impact The SAVSNET Summer School enables A-Level students to gain experience of a research environment and investigate a range of topics using SAVSNET data. The Summer School involves a series of seminars including data collection, introduction to epidemiology and a focus on a clinical condition (this year the focus was on obesity). Students were given a topic to investigate and at the end of the Summer School, they presented to the SAVSNET. The Summer School is useful for testing ideas and this year we are pleased to say that one topic, chocolate toxicity, looked very promising - this has been further developed by the SAVSNET team and will be submitted for publication.
Year(s) Of Engagement Activity 2018
 
Description SAVSNET's 10th birthday celebration 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Professional Practitioners
Results and Impact To mark ten years of the Small Animal Veterinary Surveillance Network (SAVSNET), we held an event to showcase the research being done using electronic health data from veterinary practices and laboratories, and how it is informing veterinary pratice.

Some forty delegates came including from the USA and Portugal. Day One was a series of talks ranging from an update on vaccination including parvovirus, antimicrobial resistance and antibacterial use, and the role of ticks in disease.

On Day Two there was a practical demonstration of practice based research using big data. Delegates worked on projects of their own choice including otitis, fleas and muzzling, experiencing the real power of health informatics research first hand.
Year(s) Of Engagement Activity 2018
URL https://www.liverpool.ac.uk/savsnet/news/stories/title,1078472,en.html
 
Description Science Jamboree hosted by University of Liverpool 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach Local
Primary Audience Schools
Results and Impact Dr Elena Arsevska spent a day in this workshop. In total there were about 100 children who were split throughout the day into five workshops. Activities included
1.A presentation on why animals go to the vet
- where to find the nearest vet
- also the savsnet network with the map of the vets if they can find one
- what to do if they find a stray animal
2.Watching a video of how to put a bandage on a dog
3. Teaching children how to bandage an animal
4. the others in the meanwhile did masks of animals and we sicked bandages on the masks like they were hurt
Year(s) Of Engagement Activity 2017
URL https://www.liverpool.ac.uk/central-teaching-hub/science-jamboree-2017/
 
Description Veterinary Text Mining Workshop 
Form Of Engagement Activity A formal working group, expert panel or dialogue
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact This workshop had two aims:
1) present the state of the art in processing free-text in veterinary electronic health records, and
2) establish a special interest group that will work together on improving the capabilities for making sense of veterinary free-text.

The event followed the two veterinary text mining bootcamp events. The workshop was held in Nottingham and was co-organised by the Centre for Evidence-based Veterinary Medicine @ Nottingham, SAVSNET and Healtex.
Year(s) Of Engagement Activity 2016
URL http://healtex.org/event/veterinary-text-mining-workshop/
 
Description Veterinary text-mining bootcamp 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Postgraduate students
Results and Impact This was a hands on informal event aimed at increasing the familiarity with the potential of text mining in their datasets, and equipping a cross-section of researchers with the fundamental skills to extract information from narrative data. The participants have been given tasks to develop clinical text analytics methods to extract specific information from veterinary records. By the end of the day most participants had got a practical skill to extract information from narrative.

Venue/date: Leahurst on the 15th September 2016.
Year(s) Of Engagement Activity 2016
URL http://healtex.org/event/second-veterinary-text-mining-bootcamp/