A dog walks into the exam room moving normally. The owner pulls out their phone: months of mobility data from a wearable device show a gradual decline in activity, and videos reveal increasing stiffness after rest.
Which patient gets diagnosed, the one in front of us, or the one in the dataset?
The plain answer: we're now diagnosing both, and there's no protocol for it
Veterinarians have historically made clinical decisions based on snapshots, physical exam findings, diagnostic results taken in the clinic on that visit. Those snapshots are increasingly being supplemented by continuous data streams collected outside the practice: smart collars tracking daily steps and heart rate variability, smart litter boxes logging urination frequency and duration, continuous glucose monitors showing post-meal spikes, and client videos capturing intermittent lameness or coughing that never shows up during the appointment.
"We're seeing a new type of client, one that is hyper-connected and much more informed, and that is generating and has access to much more information," says Adam Little, DVM, who focuses on the intersection of technology and animal health. He believes we are seeing the early phase of a reimagination of the typical veterinary visit.
Home-generated data can provide a fuller picture of a patient's health. Since it is collected when a pet feels relaxed and secure, it may also provide a more realistic glimpse into its everyday activity and physiologic parameters. Eleanor Green, DVM, a passionate advocate for veterinary innovation, puts it this way: "Veterinary data has shifted from a passive, retrospective approach to active, real-time decision-driving infrastructure. The data are no longer confined to the clinic visit."
The technology itself is no longer niche. According to a 2024 survey, most pet owners are ready to embrace pet tech. Home health monitoring provides a wealth of data that can lead to earlier detection of subtle changes, more personalized care plans, and better monitoring and management of chronic conditions.
The caveat: integration burden, liability exposure, and questions no one has answered yet
The problem is not whether the data exists, it does, and clients are bringing it. The problem is what a working DVM is supposed to do with it, in a ten-minute appointment slot.
Smart litter boxes use cameras, artificial intelligence, and app controls to collect data including frequency and duration of litter box visits, urine output, weight, and posture. Some varieties use facial recognition to distinguish among multiple cats and can send notifications about prolonged periods of inactivity, possibly alerting owners to a urinary blockage or constipation. Data collected by smart litter boxes can help veterinary teams detect changes that can be early indicators of conditions such as kidney disease, urinary tract infections, gastrointestinal disease, and diabetes.
Wearable devices, available as smart collars, harnesses, or tags, can monitor daily steps, activity level, sleep patterns, vital signs such as heart rate and respiratory rate and temperature and heart rate variability, and even excessive scratching or licking. Continuous glucose monitors, though not yet manufactured specifically for pets, have been used successfully in veterinary patients for some time using human versions. "Wearables are especially valuable when there are subtle, continuous changes that precede obvious changes," Dr. Green says. They are particularly helpful in monitoring parameters in pets with chronic diseases, such as activity levels in pets with osteoarthritis or respiratory rate in pets with heart disease. For pets with diabetes, she notes, pet owners can monitor glucose and see that after a particular treat the number goes up.
Client-shared photos and videos can be helpful in triaging pet owner concerns, evaluating abnormal behaviors such as limping or coughing or seizures or aggression, monitoring postoperative patients, and performing rechecks. Viewing photos and videos stored in the medical record allows veterinarians to monitor progress over time, such as the healing of a chronic wound or the progression of lameness. Visuals can often improve diagnostic accuracy when symptoms aren't visible in the clinic.
But here is the liability question we're hearing more often: if a client brings months of wearable data showing subclinical decline and the clinician does not review it, or reviews it and dismisses it, and the patient deteriorates, what does that chart note look like in a negligence claim? And if the data is trusted and intervention happens based on a device that has not been validated in a peer-reviewed study, is that practicing medicine or reacting to a gadget?
Dr. Little recommends first ensuring that image submissions are easy for clients to make and can be stored and accessed by veterinary teams. "Can they send it via text, or do they need to log into some weird website in order to upload a photo, but it can only be two megabytes, and half of them fail?" he adds. "Do you have the ability to actually see them, search them, and have them in your medical records?"
The same question applies to wearable datasets. The integration burden falls on the practice, and we're not seeing standard workflows yet for how to ingest this data or how much appointment time to allocate to reviewing it.