Wearable Devices May Predict IBD Flares Weeks in Advance
FROM GASTROENTEROLOGY
according to investigators.
These findings suggest that widely used consumer wearables could support long-term monitoring of IBD and other chronic inflammatory conditions, lead author Robert P. Hirten, MD, of Icahn School of Medicine at Mount Sinai, New York, and colleagues reported.
“Wearable devices are an increasingly accepted tool for monitoring health and disease,” the investigators wrote in Gastroenterology. “They are frequently used in non–inflammatory-based diseases for remote patient monitoring, allowing individuals to be monitored outside of the clinical setting, which has resulted in improved outcomes in multiple disease states.”
Progress has been slower for inflammatory conditions, the investigators noted, despite interest from both providers and patients. Prior studies have explored activity and sleep tracking, or sweat-based biomarkers, as potential tools for monitoring IBD.
Hirten and colleagues took a novel approach, focusing on physiologic changes driven by autonomic nervous system dysfunction — a hallmark of chronic inflammation. Conditions like IBD are associated with reduced parasympathetic activity and increased sympathetic tone, which in turn affect heart rate and heart rate variability. Heart rate tends to rise during flares, while heart rate variability decreases.
Their prospective cohort study included 309 adults with Crohn’s disease (n = 196) or ulcerative colitis (n = 113). Participants used their own or a study-provided Apple Watch, Fitbit, or Oura Ring to passively collect physiological data, including heart rate, resting heart rate, heart rate variability, and step count. A subset of Apple Watch users also contributed oxygen saturation data.
Participants also completed daily symptom surveys using a custom smartphone app and reported laboratory values such as C-reactive protein, erythrocyte sedimentation rate, and fecal calprotectin, as part of routine care. These data were used to identify symptomatic and inflammatory flare periods.
Over a mean follow-up of about 7 months, the physiological data consistently distinguished both types of flares from periods of remission. Heart rate variability dropped significantly during flares, while heart rate and resting heart rate increased. Step counts decreased during inflammatory flares but not during symptom-only flares. Oxygen saturation stayed mostly the same, except for a slight drop seen in participants with Crohn’s disease.
These physiological changes could be detected as early as 7 weeks before a flare. Predictive models that combined multiple metrics — heart rate variability, heart rate, resting heart rate, and step count — were highly accurate, with F1 scores as high as 0.90 for predicting inflammatory flares and 0.83 for predicting symptomatic flares.
In addition, wearable data helped differentiate between flares caused by active inflammation and those driven by symptoms alone. Even when symptoms were similar, heart rate variability, heart rate, and resting heart rate were significantly higher when inflammation was present—suggesting wearable devices may help address the common mismatch between symptoms and actual disease activity in IBD.
“These findings support the further evaluation of wearable devices in the monitoring of IBD,” the investigators concluded.
The study was supported by the National Institute of Diabetes and Digestive and Kidney Diseases and Ms. Jenny Steingart. The investigators disclosed additional relationships with Agomab, Lilly, Merck, and others.