Chest-Worn Sensor Data Links Erratic Bedtime Timing to Dementia Risk

An analysis of the long-running ARIC cohort found that sleep measures pulled from a chest-worn heart monitor, including how much bedtime shifted night to night, tracked with later dementia diagnoses in older adults who wore the device for roughly two weeks in 2016 and 2017.

Researchers analyzing the Atherosclerosis Risk in Communities (ARIC) Study found that sleep parameters derived from a chest-worn ambulatory heart monitor, rather than the wrist or hip devices typically used in sleep research, were associated with future dementia diagnosis, according to a study published online August 19, 2026, in SLEEP Advances. The team, led by Keran W. Chamberlin and colleagues at Johns Hopkins Bloomberg School of Public Health, built a method to extract time in bed from the Zio XT patch, an FDA-approved ambulatory ECG monitor whose built-in accelerometer also records movement and body posture, and calibrated it against hip-worn actigraphy in a subset of 405 ARIC participants.

The method was then applied to 2,316 ARIC participants, average age 79, who were free of dementia when they wore the patch during ARIC Visit 6 in 2016 and 2017. Over a median follow-up of 5.2 years, 326 participants, 14.1 percent, developed dementia. Both total time in bed and total sleep time showed U-shaped associations with dementia risk, meaning both unusually short and unusually long values were linked to higher risk. Participants in the lowest third for night-to-night variability in the midpoint of their time in bed, meaning the most consistent bedtime and wake timing, had a lower dementia hazard than those with more erratic timing. The midpoint itself, along with wake time after sleep onset and sleep efficiency, was not significantly associated with dementia risk.

The finding sits alongside a companion analysis from the same ARIC research group, published in January 2026 in Alzheimer's & Dementia: Translational Research & Clinical Interventions, which used the same chest-worn device in a similar ARIC subsample and found that lower day-to-day variability in the timing of peak activity, a separate marker of circadian rhythm, was also tied to lower dementia risk. Both are observational studies: they describe a statistical association measured at one point in time, not a cause, and neither can rule out that early, undiagnosed brain changes were already disrupting sleep and activity rhythms before a dementia diagnosis was made.

The authors frame the work mainly as a methods contribution: a way to pull sleep estimates out of a heart-rhythm monitor already worn by many patients for cardiac reasons, rather than requiring a separate sleep-tracking device. They say the dementia associations need to be tested in other cohorts with repeated, longer-term monitoring before anyone draws conclusions about an individual's risk from a single two-week recording.