Epilepsy affects millions of people worldwide, and for those living with uncontrolled tonic-clonic seizures, the risks extend far beyond the seizures themselves. One of the most serious concerns is Sudden Unexpected Death in Epilepsy (SUDEP), a condition strongly associated with frequent and unattended tonic-clonic seizures. Early detection and rapid caregiver intervention have long been considered critical strategies for reducing these risks.
A newly published Phase III clinical trial has brought encouraging news to the epilepsy community. Researchers found that EpiWatch, a seizure monitoring application designed for the Apple Watch, achieved exceptional accuracy in detecting tonic-clonic seizures while maintaining one of the lowest false alarm rates reported among wearable seizure detection technologies.
The study, published in the June 2026 issue of Neurology Clinical Practice Open Access, evaluated EpiWatch across multiple epilepsy monitoring units and included both pediatric and adult patients.
Tonic-clonic seizures are among the most dangerous seizure types because they can result in loss of consciousness, injury, breathing difficulties, and increased SUDEP risk. Research has shown that people who experience these seizures, especially during sleep, face significantly higher health risks when no caregiver is nearby.
Wearable seizure detection devices have emerged as an important tool for providing real-time alerts to caregivers. However, many existing systems struggle with frequent false alarms. Excessive false notifications can lead to alarm fatigue, causing caregivers to become less responsive over time.
The EpiWatch development team focused on solving this challenge by creating an algorithm capable of maintaining high sensitivity while dramatically reducing false alarms.
The Phase III study was conducted at six epilepsy monitoring units between September 2021 and October 2023. Researchers enrolled 242 participants aged five years and older who had a history of tonic-clonic seizures or were considered at risk of experiencing them.
Participants wore an Apple Watch running the EpiWatch application while undergoing continuous video EEG monitoring. An independent panel of epileptologists reviewed all seizure events and remained blinded to the app's detections, ensuring unbiased evaluation.
The primary goals of the study were to measure:
Secondary measures included detection speed and performance during sleep.
One of the most significant findings was EpiWatch's remarkable seizure detection performance.
The device successfully detected 46 out of 47 confirmed tonic-clonic seizures, resulting in an overall sensitivity rate of 98%.
Researchers reported that the only missed seizure occurred when a caregiver physically restrained the participant's watch-wearing arm throughout the event. Since wrist movement is an important signal used by the algorithm, this unusual circumstance likely interfered with detection.
Performance remained consistently strong across different age groups:
These results demonstrate that EpiWatch performs effectively across a broad age range.
While high sensitivity is essential, false alarms are often what determine whether patients and caregivers continue using a monitoring device long term.
The study recorded only 56 false alarms during more than 16,000 hours of monitoring.
This translated to a false alarm rate of just 0.08 per day, or approximately one false alarm every 12.4 days.
For comparison, many previously studied seizure monitoring systems reported false alarm rates ranging from 0.67 to more than 2 alarms per day.
Researchers noted that over 87% of participants experienced zero false alarms during their monitoring period.
This low false alarm rate could significantly improve user trust and reduce caregiver fatigue, two major barriers to widespread adoption of seizure detection technology.
Nighttime seizures are particularly concerning because they are closely linked to SUDEP risk.
One of the study's most encouraging findings was that EpiWatch detected every tonic-clonic seizure that occurred during sleep.
Researchers also found that all sleep-related false alarms were associated with actual seizure activity rather than normal sleep movements.
This suggests that EpiWatch may be especially valuable for monitoring patients who experience nocturnal seizures.
EpiWatch operates as an application on the Apple Watch and uses data collected through built-in sensors.
The system continuously analyzes:
A proprietary machine learning algorithm evaluates incoming sensor data in real time. When the system identifies a pattern consistent with a tonic-clonic seizure, it can send an alert to designated caregivers through a paired smartphone.
The algorithm was trained using extensive seizure and non-seizure datasets, allowing it to distinguish genuine seizure activity from everyday movements.
In addition to accurate detection, rapid response is essential.
The study reported a median detection latency of approximately 31.5 seconds after seizure onset.
This performance falls well within clinical recommendations for seizure detection systems and is comparable to or better than many competing devices.
Since tonic-clonic seizures often last around one minute, detecting them within the first half-minute provides caregivers with valuable time to intervene if necessary.
Based on the findings from this clinical trial, EpiWatch has received FDA 510(k) clearance and is available by prescription through Apple's App Store.
A major advantage of the platform is that it runs on a widely used consumer smartwatch rather than requiring patients to wear a specialized medical device.
Researchers highlighted that this design may reduce the social stigma sometimes associated with epilepsy monitoring equipment. Because millions of people already wear Apple Watches for everyday health tracking, EpiWatch can blend seamlessly into daily life.
The familiarity of the device may also encourage long-term adherence and consistent use.
Although the results are highly promising, researchers acknowledged several limitations.
The trial was conducted within epilepsy monitoring units under controlled clinical conditions. Real-world environments may present additional challenges not captured during inpatient monitoring.
The study also focused specifically on tonic-clonic seizures in individuals aged five years and older. Additional research is needed to evaluate performance in other seizure types and patient populations.
Future studies involving long-term home monitoring will help determine how closely real-world outcomes match clinical trial performance.
The EpiWatch trial represents a significant advancement in wearable seizure detection technology.
With 98% sensitivity, reliable sleep monitoring, rapid detection times, and an exceptionally low false alarm rate, the system addresses many of the shortcomings that have limited previous seizure monitoring devices.
For individuals living with epilepsy and their families, technology that can provide dependable alerts without constant false alarms may offer greater peace of mind and improved safety.
While no wearable device can eliminate the risks associated with epilepsy, innovations like EpiWatch could play an important role in helping caregivers respond more quickly to potentially dangerous seizure events.
As wearable health technologies continue to evolve, this study suggests that consumer smartwatches may become increasingly valuable tools in neurological care and epilepsy management.
Krauss GL, Elizebath R, Shah S, Wheless JW, Sperling MR, Cabacar DFD, Gaillard WD, Parikh NB, Chiacchierini R, Crone NE. Phase III Trial of EpiWatch for Tonic-Clonic Seizure Detection in Children and Adults. Neurology Clinical Practice Open Access. June 2026;2(2):e000111. DOI: 10.1212/WN9.0000000000000111.
This article is intended for informational and educational purposes only and should not be considered medical advice. The findings summarized here are based on a published clinical trial conducted in controlled epilepsy monitoring unit settings. Real-world performance may differ from clinical study results. Individuals with epilepsy or concerns about seizure monitoring should consult a qualified healthcare professional before making medical decisions or using any seizure detection device.
Most Accurate Healthcare AI designed for everything from admin workflows to clinical decision support.