Tracking nighttime scratching with a wearable device and app provided useful information on disease activity and clinical improvements in children with moderate-to-severe atopic dermatitis (AD) in a proof-of-concept study presented at the American Academy of Allergy, Asthma & Immunology (AAAI) 2026 Annual Meeting.

Use of digital health technology for disease monitoring continues to expand, but data on real-world applications in people with AD are limited, according to Fumiko Iwai, MD, of the National Hospital Organization Mie National Hospital in Tsu, Japan, and colleagues.

In a proof-of-concept study, they equipped children with moderate-to-severe AD with an Apple Watch and Itch Tracker app as they began standard-of-care treatment for AD. The children wore the watch at night for 5-14 nights, while the app recorded metrics on the scratch count rate (scratching events per hour), scratch duration rate (the percentage of sleep time spent scratching), and a measure of scratch burden.

The researchers compared the data from the app to baseline disease severity, based on the Eczema Area and Severity Index (EASI); patient-reported symptoms, using the Patient-Oriented Eczema Measure; and serum thymus and activation-regulated chemokine (TARC) levels.

In a logistic regression analysis, the mean scratching metrics were strongly associated with baseline severity for both EASI and TARC (r > 0.6; P < .001). Moreover, reductions in nighttime scratching “paralleled clinical improvement,” which the investigators said supported the use of scratching metrics “as dynamic indicators of treatment response.”

The scratch burden index also provided an accurate model for clinical improvement from baseline, with an area under the curve of 0.776.

The findings were limited by the study’s design, but the results demonstrate the potential of nighttime tracking to provide real-world insights that inform practice, the researchers wrote. “With further validation, these metrics could support personalized feedback, early intervention, and data-driven disease management,” they concluded.

Tapping Nighttime Data for Personalized Care

Although AD is better understood now than in the past, clinicians still rely on snapshots of data, such as the Investigator Global Assessment scale for AD, EASI scores, and patient recall, said Adam Friedman, MD, professor and chair of dermatology at the George Washington School of Medicine and Health Sciences, Washington, DC. However, “AD does not live in snapshots,” said Friedman, who was not involved in the study. “It lives in the day-to-day, and more importantly, the night-to-night.”

Nocturnal scratching is one of the most defining and disruptive features of AD, but no method currently exists to monitor it in the real world, he emphasized. “The current study brings us closer to closing that gap, leveraging something as ubiquitous as a smartwatch to passively and continuously capture disease activity in a way that actually reflects lived experience,” he said.

The findings were not surprising, but several elements were impressive, Friedman told Medscape Medical News. “Seeing such strong correlations with established markers like EASI and serum TARC levels, and watching those scratching metrics track so tightly with treatment response, is validating,” he said. “What really stands out is the predictive power of the scratch burden index; digital biomarkers often overpromise and underdeliver, but this feels like one of those rare moments where the signal is real and clinically meaningful,” he added.

Ultimately, consistent data tracking will empower clinicians to move from episodic care to treatment that is longitudinal and responsive, allowing them to see disease activity in real time rather than relying on patient reports, Friedman said. In practice, this could mean earlier intervention when patients are starting to flare, more precise treatment adjustments, and even better adherence conversations grounded in objective data, he said.

“It also opens the door to truly personalized care, to manage based on an individual patient’s digital phenotype of disease activity,” he said. From a clinical trials standpoint, the data identified in the current study could offer continuous, objective endpoints that could complement or even outperform traditional measures, he noted.

The current study is a proof of concept that now needs to be scaled, Friedman said. “Larger, more diverse populations, longer monitoring periods, and validation across different age groups and disease severities are key,” he said. Other questions to be addressed include how the metrics in the study perform in real-world conditions, such as patients with different sleep patterns and comorbid conditions and with inconsistent adherence to wearing devices, as well as how to integrate the barrage of data into workflows without overwhelming clinicians or patients, he added.

The study received no outside funding. The researchers disclosed no relevant financial relationships. Friedman disclosed having no relevant financial conflicts.