Older adults whose day-to-day activity rhythms were weaker, more fragmented, or peaked later in the afternoon were more likely to be diagnosed with dementia over the following years, according to a study published December 29, 2025 in Neurology, the journal of the American Academy of Neurology. The finding is notable less for the direction of the association, which echoes earlier wrist-worn actigraphy research, than for its source: the rhythm measures were extracted from the accelerometer built into a Zio XT ambulatory ECG patch, a chest-worn heart monitor already used clinically for arrhythmia detection, rather than from a device deployed solely to track movement.
The analysis, led by Wendy Wang of UT Southwestern Medical Center and Lin Yee Chen of the University of Minnesota, drew on 2,183 participants in the Atherosclerosis Risk in Communities (ARIC) study who wore the patch for up to 14 days, an average of 12, between 2016 and 2017. Participants were on average 79 years old, 58 percent women, and 24 percent Black, with the rest White. None had dementia at enrollment. Over a median follow-up of 3.3 years, 176 participants were diagnosed with dementia. Comparing the third of participants with the weakest, most fragmented rhythms to the third with the strongest, the researchers reported roughly two and a half times the adjusted risk of dementia; a later timing of peak daily activity, past 2:15 p.m., was associated with a 45 percent higher adjusted risk than an earlier peak.
The authors and outside researchers caution that this is an observational cohort study, not a trial, and it cannot show that disrupted rhythms cause dementia or that fixing them would lower risk. The ARIC team itself wrote that reverse causation is plausible: early, undetected neurodegeneration could itself be flattening or delaying activity patterns, rather than the rhythm disruption acting as a cause. Wang has said future controlled studies would be needed to test whether interventions such as timed light exposure change dementia risk; none of that evidence exists yet from this study.
What is new and reproducible here is methodological: a widely deployed piece of clinical hardware, worn for a heart problem, generated an activity signal that tracked with later dementia diagnosis in a cohort that included substantial numbers of Black and White participants, a population mix many earlier actigraphy studies lacked.