February 5, 2025 | Physiological changes in the body driven by inflammation, as happens with inflammatory bowel disease (IBD) and rheumatoid arthritis, can be “easily and passively” measured with commercially available wearable devices, according to gastroenterologist Robert Hirten, M.D., associate professor of medicine and artificial intelligence and human health at the Icahn School of Medicine at Mount Sinai and clinical director of its Hasso Plattner Institute for Digital Health. That means disease-associated markers could potentially be used for continuous monitoring of patients for flareups.
In an observational study involving 309 IBD patients from 36 states, both symptom flares and inflammation were detected by wearables several weeks before they manifested, as reported recently in Gastroenterology (DOI: 10.1053/j.gastro.2024.12.024). The results suggest the feasibility of using metrics collected by ubiquitous wearables—e.g., Apple Watch, Fitbit, Oura Ring, and Garmin smartwatches—to monitor and predict IBD activity.
The hypothesis is that changes in wearable derived metrics could ultimately be used to create a profile of worsening disease, be it a chronic disease or something more acute such as an infection, says Hirten. In the case of IBD, circadian patterns of heart rate variability as well as heart rate and resting heart rate differentiated whether participants with symptoms had inflammation. These metrics, in addition to oxygenation and daily activity, were significantly altered up to seven weeks prior to inflammatory and symptomatic flares.
Hirten and his team are now working on artificial intelligence (AI) algorithms using wearable device data to predict disease flares on an individual basis. The next step will be pragmatic trials to understand how well the monitoring approach works in clinical practice and, most importantly, whether it makes a meaningful difference in getting flares of disease under better control and improves the lives of patients, he says.
Progress to Date
Mount Sinai investigators previously published a smaller study that looked specifically at heart rate variability, a measure of the time differences between each heartbeat, which in medical settings is detected by an electrocardiogram machine. Heart rate variability indirectly measures autonomic nervous system activity, which is the part of the nervous system that operates outside of a person’s conscious control, explains Hirten.
Many studies have demonstrated the relationship between heart rate variability and autonomic nervous system function that relates to various types of disease states and outcomes, he continues. In people with IBD, heart rate variability has been shown to differ between people with and without the disease. Moreover, when inflammation is present, that value differs from people with IBD who don’t have active inflammation.
In participants in their study, Mount Sinai investigators saw a relationship between changes in both heart rate variability and inflammation over a nine-month monitoring period. This smaller study showed for the first time that wearables could be used to monitor metrics that relate to underlying inflammation that is occurring in the body, says Hirten.
When the pandemic hit, the research team pivoted to a study in several hundred healthcare workers demonstrating that wearables could predict a COVID infection since inflammation is a key feature. Algorithms developed from the outputs of the Apple Watch, anchored in heart rate variability, succeeded in making the prediction up to several days before someone was diagnosed via nasal PCR testing, he reports. “Even in people who were asymptomatic, we were able to identify COVID infections using wearables.”
Many conditions can alter these physiological metrics which are not specific to one condition, Hirten points out. But during the pandemic, they could serve as reliable signals of the likelihood of having COVID when there were relatively few other infections that were occurring.
In a newly completed study submitted for publication, Hirten says he and his colleagues will be reporting that flareups of rheumatoid arthritis can be predicted by wearables in a similar way as they were for IBD. “One thing we are interested in,” he adds, “is understanding how the changes in these metrics, and the pattern of this change, will vary depending on the disease or condition.”
Hirten is a practicing gastroenterologist who specializes in IBD, the two main types being Crohn’s disease and ulcerative colitis. “Much of the time that people are living with these chronic diseases is spent between doctors’ visits and the testing that is performed to monitor the disease,” he says. “Generally, most of our chronic disease monitoring is cross-sectional... so we don’t have insight into what’s going on during the time between this intermittent testing.”
This is precisely what prompted Hirten’s interest in consumer wearables, enabling the ongoing collection of information about how patients are doing in the comfort of their home and requiring only the wearing of a device, which they are often already doing. Custom and commercial devices can be “very good” at collecting the necessary metrics, he adds. “The most important thing when choosing a device is picking something that people will use regularly and over long periods of time.”
In the future, it may be possible for doctors to remotely monitor patients with IBD, rheumatoid arthritis, and other chronic inflammatory diseases in a continuous fashion as they now can do for individuals with conditions such as hypertension and diabetes, says Hirten, provided their use translates into improved health and disease management. Billing codes have already been established for passive monitoring with at-home blood pressure and glucose monitoring tools based on their demonstrated value.