Wearable seizure-detection devices can be useful, but they do not detect every seizure type and should not be presented as universal monitors.
What wearables usually measure
Depending on the device, sensors may measure movement, muscle activity, heart rate, electrodermal activity or combinations of these signals. The strongest evidence has generally been for detecting convulsive motor seizures, particularly tonic-clonic seizures.
Why focal seizures are harder
Many focal seizures produce subtle, brief or non-motor changes. A wrist or arm sensor may therefore have little to detect. Some focal seizures may alter heart rate or autonomic function, but those signals are not specific to seizures and can also occur with exercise, stress, sleep changes and other everyday events.
False alarms matter too
A detector that is very sensitive but produces frequent false alarms can create sleep disruption, anxiety and alarm fatigue. Performance therefore has to be judged using both sensitivity and false-alarm rate, not by asking only whether a device can detect a seizure in ideal conditions.
What the devices are for
Wearables may help alert a carer, document some nocturnal events or provide additional safety monitoring for selected people. They do not replace diagnosis, video-EEG, an individual care plan or clinical judgement.
Which seizures are detected best?
Current wearable technology is best validated for generalised tonic-clonic and focal-to-bilateral tonic-clonic seizures, particularly devices combining accelerometry and electrodermal or other autonomic signals. Detection of focal impaired-awareness and non-motor seizures remains substantially less reliable.
Sensitivity is only half the problem
A detector can identify many seizures yet still be impractical if it produces too many false alarms. Real-world usefulness depends on both sensitivity and false-alarm rate, as well as battery life, comfort and whether someone is available to respond.
Consumer devices are not all medical devices
Smartwatches and phone apps vary greatly in validation and regulatory status. Marketing claims should not be treated as equivalent to peer-reviewed clinical validation.
SUDEP claims require caution
Wearables may shorten time to assistance after some convulsive seizures, but there is currently no robust evidence that commercially available seizure-detection wearables prevent SUDEP. They should be described as detection and alert tools, not proven mortality-prevention devices.
Night-time use
For people with nocturnal convulsive seizures, wearables can be useful when the goal is to alert a nearby carer. They are less useful when seizures produce little movement or when nobody is available to respond.
Sources
ILAE/IFCN clinical practice guideline on automated seizure detection using wearable devices
Systematic review of wearable seizure-detection devices