Sleep-wake patterns linked to dementia risk in older adults
Two disrupted sleep-wake profiles were linked to higher dementia risk, with hazard ratios of 1.94 and 1.96 versus strong rhythms, light activity and good sleep.

Our read
Populations- Studied:
- Cohort study of 3,973 dementia-free older adults, against adults with healthy sleep-wake patterns
- What changed:
- Dementia risk was higher with poor daily rhythms, low activity, and poor sleep than with healthy patterns (hazard ratio 1.94)
- The catch:
- Preclinical dementia could alter sleep and activity patterns, reversing the apparent relationship
Editor of record: Boris Djordjevic · AgingResearch.News
Among 3,973 UK adults aged 60 or older without dementia, a prospective population-based cohort study linked sleep-wake profiles to later dementia. Participants’ median age was 68.4 years; 26% were women. Movement sensors captured daily activity rhythms, daytime activity, sleep and preferred sleep timing. Health records documented 379 dementia cases over a median 12.3 years.
Compared with strong, stable daily rhythms, light activity and good sleep, two profiles had higher dementia risk: weak, unstable, fragmented rhythms with low activity and poor sleep; and high average activity with small daily swings and restless sleep. Their hazard ratios—measures of relative risk over time—were 1.94 and 1.96, respectively. Among 3,806 participants assessed for cognitive decline, the first profile and the most sedentary profile with the poorest rhythms had faster decline over 10 years.
Why it matters
Sleep-wake disruptions are common in dementia and may begin before diagnosis. The authors call for studies testing whether sleep-wake interventions can reduce dementia risk or delay its onset.
Caveats
Early, undiagnosed dementia may already have affected participants’ sleep-wake patterns, rather than those patterns raising dementia risk. The findings do not address adults younger than 60, and only 26% of participants were women.
Where the evidence stands
Circadian clock
- 91 independent labs in 30 countries
- 4 findings against
- Funders: National Institutes of Health and others
- 10 of 128 findings with company ties
From the paper
- Follow-up
A total of 379 participants (9.5%) had a dementia diagnosis (median [IQR] age at diagnosis 81.9 [77.4-85.9] years) over a median 12.3-year follow-up (IQR, 12.0-12.7 years).
- Result
Participants in profiles 7 and 8 had significantly higher dementia risk in fully adjusted analyses (HR, 1.94 [95% CI, 1.32- 2.84] for profile 7 and HR, 1.96 [95% CI, 1.27-3.03] for profile 8).
- Limitation
Second, participants are healthier than the general population, potentially limiting the generalizability, although exposure-outcome associations in WII have been found similar to that in general population studies.
Quoted word for word from the full text on PubMed Central.
Comments
Be the first to comment.