Wrist Sensor Algorithm Accurately Tracks Daytime Napping in Narcolepsy
A new actigraphy algorithm reliably detects daytime naps in narcolepsy type 1, and shows an orexin agonist restores sleep patterns to near-normal levels.
Summary
Researchers developed a wrist-actigraphy-based algorithm to objectively measure daytime napping, a hallmark of excessive daytime sleepiness in narcolepsy type 1. Trained on over 2,200 participants from a large multi-ethnic study, the algorithm achieved 93% specificity and 84% accuracy. When applied to people with narcolepsy type 1, it detected dramatically more daytime sleep compared to healthy matched controls — about 34 extra minutes per day and 12.8 fewer nap-free days over a month. Crucially, the algorithm was also used to evaluate an investigational drug, oveporexton (TAK-861), an orexin receptor 2-selective agonist. Treated patients gained 6–12 additional nap-free days and slept 12–33 fewer daytime minutes, approaching levels seen in healthy controls. This wearable-based tool offers a scalable, objective way to monitor sleep disorders and assess treatment response outside the clinic.
Detailed Summary
Excessive daytime sleepiness is one of the most debilitating features of narcolepsy type 1, a neurological condition caused by loss of orexin-producing neurons. Yet measuring daytime napping objectively has remained a challenge — diary reports are unreliable and polysomnography is impractical for long-term monitoring. This study addresses that gap with a new actigraphy-based nap detection algorithm designed for continuous, real-world sleep assessment.
The algorithm was developed using the Multi-Ethnic Study of Atherosclerosis dataset, which included 2,237 participants with 7 days of wrist-worn actigraphy alongside manually annotated nap records. Researchers trained and validated the algorithm, achieving high specificity (Tau B: 93.2%) and solid accuracy (F1 Area: 83.9%), meaning it reliably detected true naps while minimizing false positives from movement or wakefulness.
Applied to an observational study of people with narcolepsy type 1, the algorithm confirmed substantial differences from healthy age- and sex-matched controls: narcolepsy patients experienced 12.8 fewer nap-free days and slept 34 additional minutes per day over 28 days. These objective metrics clearly captured the burden of daytime sleepiness in this population.
Most importantly, the algorithm was deployed in a randomized clinical trial evaluating oveporexton (TAK-861), an orexin receptor 2-selective agonist currently under investigation. Treated patients gained 6.1 to 11.9 additional nap-free days and reduced daytime sleep by 12 to 33 minutes compared to baseline — approaching the napping profiles of healthy controls. This demonstrates the algorithm's sensitivity to clinically meaningful treatment effects.
For the broader longevity audience, sleep quality and architecture are increasingly recognized as core healthspan determinants. Objective, wearable-based tools that quantify daytime sleep disruption could prove valuable far beyond narcolepsy — in aging populations where fragmented sleep and daytime napping predict cognitive decline and mortality. The study is limited by reliance on abstract-only information, and the algorithm's generalizability to other sleep disorders or aging cohorts requires further validation.
Key Findings
- Actigraphy algorithm achieved 93.2% specificity and 83.9% accuracy in detecting daytime naps.
- Narcolepsy type 1 patients had 12.8 fewer nap-free days and slept 34 extra minutes daily vs. controls.
- Orexin receptor 2 agonist oveporexton reduced daytime sleep by 12–33 minutes, restoring near-normal patterns.
- Algorithm was validated on 2,237 participants from a large multi-ethnic real-world dataset.
- Wrist-worn actigraphy enables scalable, objective nap monitoring outside clinical settings.
Methodology
The nap detection algorithm was trained on 7-day wrist actigraphy data with manually annotated naps from 2,237 participants in the Multi-Ethnic Study of Atherosclerosis. It was then validated in an observational study (NCT04445129) and a randomized clinical trial (NCT05687903) of narcolepsy type 1 patients. Performance was assessed using Tau B for specificity and F1 Area for accuracy.
Study Limitations
Summary is based on the abstract only, as the full paper is not open access. The algorithm was developed on a general population dataset and requires further validation in other sleep disorders and aging populations. Generalizability of nap-detection accuracy across different wrist-worn devices and activity levels is not yet established.
Enjoyed this summary?
Get the latest longevity research delivered to your inbox every week.
Enter your email to subscribe: