“Smart” Data for the Betterment of People and Pets

Over the past 20 years, the world has seen a rapid shift in its ability to collect and store data. As our ability to capture data has expanded, our focus has shifted from the acquisition of data to translating data into meaningful insights. While this change has brought about new opportunities as well as new obstacles, the intention of veterinary analytics has remained the same:??

• Leverage data to understand pet health risks?
• Minimize the impact of adverse animal health issues?
• Monitor animal health trends?
• Detect emerging health threats?

Although data has become easier to generate and store, we still face challenges. Today, one of the biggest obstacles we face is translating available data (big data) into meaningful information (smart data). From the sporadic bits of data our systems continuously compile, we must create value – for the betterment of pets, pet owners, veterinary practitioners, and all other constituents in the pet health ecosystem.?

One example of big data being put to “smart” use is BluePearl Specialty and Emergency’ Pet Hospital’s annual Pet Health Trends Report . Released for the first time this year, the report leverages medical data from more than 100 BluePearl hospitals, and more than 1.29 million pets seen annually. Of the pets seen, approximately 184,000 were cats, 662,000 were dogs, and 11,500 were other species such as birds, rabbits, and reptiles.?

Revealing the year’s most popular pet names and breeds, top pet diagnoses and toxicities, and pet obesity and pet blood donations trends, the report delves into the most pressing health issues facing pets in the U.S. The report also shows that not only is pet ownership continuing to rise (specifically among cats and dogs less than two years old), but each year more people are opting for pet insurance. In fact, the number of BluePearl clients with pet insurance rose from 18% in 2020 to 20% in 2021. Furthermore, the percentage of people with pet insurance used at BluePearl was highest among 25- to 44-year-olds (Millennials).??

While BluePearl’s Pet Health Trends Report speaks to the current state of animal health in the U.S., offering unique insights for the benefit of pets, pet owners, and veterinary professionals, there are still improvements that can be made in the collection of such data. Putting data to work (translating big data into smart data) requires proactive (vs retroactive) accessibility of clinical data or digitized health records or databases, and automation of IT systems.??

Improvements to Data Systems?

The advent of more advanced IT systems created a revolutionary opportunity for veterinary analytics. Efficiencies in these systems and the wide use of diagnoses coding guides, especially those used by pet hospitals, with hundreds or thousands of practices that can collect data over geographically diverse areas, have allowed us to harness big data. Today, our hospitals’ electronic record systems serve as “data pipelines,” funneling information on animal diseases and issues to a master dataset from which we can extract unique insights on disease spread. These records include information such as laboratory test results, physical examination findings, diagnoses, treatments, procedures performed, and demographic information about the patients and their owners. However, the surveillance of this data is primarily retrospective.??

While these databases have produced valuable pet health insights, the true promise of big data lies largely in the potential to target specific populations and anticipate trends . As we introduce new IT capabilities and move from manual data input and exfiltration to a state of automation, where emerging animal health concerns can be identified in near real-time, we can significantly improve our ability to respond; thus, minimizing the impact of adverse health issues. This can be done by increasing hospital protocols to prevent further disease spread or awareness building on serious health threats among pet owning populations.?

With new capabilities of new or existing capabilities, we not only increase the effectiveness of big data, but we are able to:?

• Analyze when and where animal health diseases or issues are occurring – and the movement of these diseases or issues as they propagate among populations or geographical regions?
• Combine disparate data to cultivate better approaches to medical treatment and prevention?
• Monitor animal health trends for early detection and possible prevention of emerging health threats?

Data and Translational Medicine?

Health information collected from dogs continues to be used to inform healthcare decisions in dogs as well as in other species, such as humans.??

Authors of the research paper, Dogs really are man's best friend--canine genomics has applications in veterinary and human medicine! , argue that common diseases in companion animals are similar and sometimes identical to human diseases. In terms of disease causes, progression, and response to treatment, dogs provide an ideal model for translational medicine. The reason for this is dogs’ phenotypic diversity and the fact that they – out of all land mammals other than humans – have the most known naturally occurring diseases. In fact, there are 400 inherited disorders in dogs that are relevant to humans, according to the journal article, Veterinary Big Data: When Data Goes to the Dogs. Furthermore, canines share disease outcomes of cancer, osteoarthritis, spinal cord disorders, eye disorders, cardiomyopathies, and infections.?

The largest direct access dataset in the U.S. today is a part of the Dog Aging Project. This dataset includes information regarding the participant’s physical and chemical environments, diet, exercise, behavior, and health history. In the journal article, Veterinary Big Data: When Data Goes to the Dogs , the authors describe several other veterinary medical datasets that have implications for use in translational medicine. These datasets include but are not limited to:?

• Banfield Pet Hospital - In 2019, Banfield had collected clinical data from more than 2.5 million dogs from 43 states in the U.S. to pair electronic record data with existing ecological data to monitor infectious disease spread in animals in the U.S.?
• The Veterinary Medical Database (VMDB) - VMDB is the oldest companion animal health database in the United States, which provides veterinary medical datasets to researchers at little or no cost.?
• The Small Animal Veterinary Surveillance Network (SAVSNET) - SAVSNET is a database of electronic health and environmental data on companion dogs in the United Kingdom. SAVSNET dedicated to enhancing understanding of the impact of antimicrobial resistance, climate, environmental risk factors, and infections on overall health.?
• The Veterinary Companion Animal Surveillance System (VetCompass) - VetCompass began collecting clinical data from primary practices in the United Kingdom in 2009, and now holds data on more than 15 million animals collected from over 1,800 veterinary practices across the UK, a third of all UK veterinary practices.?
• Golden Retriever Lifetime Study (GRLS) - GRLS is a prospective study restricted to a single cohort of approximately 3,000 Golden retriever dogs located throughout the U.S. This data was longitudinally collected and at regular intervals. The dataset not only contains information on the health of animals collected from their veterinarian, including laboratory values, but also extensive environmental and behavioral health data collected from their owners.?
• Pet Protect from the UK and Agria from Sweden – These two well-established insurance databases have been utilized for epidemiological research on the causes of morbidity and mortality in insured dog populations.?

Large-scale datasets from pet hospitals, individual research projects, or pet insuring companies such as these are and will continue to be utilized to enhance people's understanding of both dog and human health, disease outcomes, and how to increase healthy lifespan. Thus, we, as consumers of human medicine, may begin to see more treatments and medical interventions that have been influenced by “big dog data.”?

Preparing for the Future?

In addition to automation, data requires the inclusion of new skills into veterinary professional training – whether among veterinary technicians, veterinarians, and/or hospital staff. This training of new skills may include machine learning and coding, with the objective of preparing this new generation to proactively engage with and understand big data.??

Establishing these digital and human pipelines to analyze big data in near real-time is not only essential to the future of smart data, but overall pet health. When we collectively use big data to understand health risks, we can bring more efficiency to issue management, make more informed treatment decisions, and minimize the impact of adverse animal health issues. The future state of veterinary data is one where all parts of the pet ecosystem benefit – and together, we as leaders in pet health can begin to build the frameworks needed to support this.?