Wearable data may help identify cognitive risk, but validation remains limited
Sleep and activity patterns were linked with cognitive outcomes, while prediction models had limited testing beyond the data used to develop them.
Based on the published abstract. The full paper may contain additional methods, results and limitations.
The 30-second takeaway
The review linked disrupted sleep, fragmented daily rhythms and irregular activity patterns with worse cognitive outcomes. Some studies also used these measurements to predict cognitive impairment. The evidence suggests a possible role for wearable data in risk assessment, but it does not establish that a consumer device can diagnose dementia or that monitoring prevents it.
The authors identified small samples, brief monitoring periods and limited external validation. Devices, outcomes and analysis methods differed enough to prevent a meta-analysis, making it hard to summarize performance with one reliable estimate.
Research devices and everyday trackers
According to the abstract, most studies used research-grade movement monitoring rather than commercial wearables. Results from specialized devices and research procedures cannot automatically be assumed to describe the performance of an everyday consumer tracker.
Association and prediction differ
An association describes measurements that vary together; prediction asks whether measurements can identify an outcome in other people. The review included both, and relatively few studies addressed early detection or prevention through longitudinal prediction.
The original publication
AI and Wearables for Early Detection of Cognitive Impairment and Dementia: Systematic Review.
Cejudo A, Arrojo M, Martín C et al.
J Med Internet Res · 2026
- PubMed ID
- 41730193
- Record checked
AI-assisted research and writing. This explains one selected publication; it is not a complete review of everything known. Our approach.
One more question, understood.
Keep track of the research you’ve explored.