SLEEP REGULARITY AFTER 50: WHAT THE COHORT DATA MEASURED

Most sleep advice after 50 is about hours: get seven, get eight, stop cutting it short. A newer line of research measures something different, which is how consistent the timing of your sleep is from one day to the next. Several large cohorts with wrist-worn accelerometers have now reported on it. Here's what they measured, how large the associations were, where two analyses of the same cohort disagree, and what no trial has tested yet.

What "Sleep Regularity" Means in These Studies

Most of the studies below use the Sleep Regularity Index (SRI). Cribb and colleagues describe it in their 2023 eLife paper as the probability of being in the same state, asleep or awake, at any two time points 24 hours apart, averaged over a week of accelerometer wear. It runs from 0 to 100, where 100 means perfectly regular.

One caution before comparing numbers across papers: two teams analyzed accelerometer data from the same UK Biobank cohort and reported very different typical scores. Windred and colleagues reported a median SRI of 81; Cribb and colleagues reported a median of 60. The index is calculated differently depending on the method, so an SRI cutoff from one paper does not transfer to another.

Regularity and Mortality

Windred and colleagues, publishing in Sleep in 2024, calculated SRI scores for 60,977 UK Biobank participants with a mean age of 62.8 years, then followed them for a mean of 6.3 years, during which 1,859 died. Compared with the least regular fifth of participants, the top four fifths were associated with a 20% to 48% lower risk of all-cause mortality, a 16% to 39% lower risk of cancer mortality, and a 22% to 57% lower risk of cardiometabolic mortality, after adjustment for age, sex, ethnicity, and sociodemographic, lifestyle, and health factors.

The authors also compared regularity against sleep duration head to head and reported that regularity was the stronger predictor of all-cause mortality. That is the finding the paper's title leads with, and it is a comparison of predictors within one cohort, not a test of what happens if someone changes either one.

Cribb and colleagues ran a separate analysis of 88,975 participants in eLife (2023), with 3,010 deaths over a mean 7.1 years. The relationship was non-linear. Relative to the median score, participants at the 5th percentile of regularity had a hazard ratio of 1.53 (95% CI 1.41 to 1.66) for all-cause mortality, while those at the 95th percentile had 0.90 (95% CI 0.81 to 1.00). The upper confidence bound at the regular end reaches 1.00, while the interval at the irregular end sits well above it.

Regularity and Cardiovascular Events

Chaput and colleagues (Journal of Epidemiology and Community Health, 2025) followed 72,269 UK Biobank adults aged 40 to 79 for 8 years. Irregular sleepers (SRI below 71.6) had a hazard ratio of 1.26 (95% CI 1.16 to 1.37) for major adverse cardiovascular events compared with regular sleepers (SRI above 87.3).

The part that bears directly on the "just get your hours" advice: among moderately irregular sleepers, meeting the age-specific sleep-duration recommendation was associated with no significant excess risk (HR 1.07, 95% CI 0.96 to 1.18). Among the most irregular sleepers, it was not (HR 1.19, 95% CI 1.06 to 1.35). The authors' own disclosures list industry relationships for two co-authors, including ResMed funding tied to one author's academic chair.

A smaller Finnish cohort points the same direction with a narrower result. Nauha and colleagues (BMC Cardiovascular Disorders, 2026) followed 3,231 adults from age 46 for about 10 years. Irregular bedtimes were associated with roughly double the risk of cardiac events (HR 2.01, 95% CI 1.00 to 4.01), but only among people sleeping less than the group median of 7 hours 56 minutes. Variability in wake-up time was not associated with risk. The study recorded 128 events in total, and the lower confidence bound for that result is 1.00.

Where Two Dementia Analyses Disagree

Two groups analyzed dementia in the same UK Biobank accelerometer sample and reported different shapes.

• Yiallourou et al., Neurology 2024: 88,094 participants, 480 dementia cases. The association was U-shaped. The least regular sleepers had a hazard ratio of 1.53 (95% CI 1.24 to 1.89) relative to the median, but the most regular also trended higher, at 1.16 (95% CI 0.89 to 1.50, not significant).
• Bian et al., BMC Public Health 2025: 82,391 participants, 694 cases. The association was near-linear. Higher regularity (SRI 70 or above) was associated with a hazard ratio of 0.74 (95% CI 0.63 to 0.87), and that association appeared only among people sleeping under 7 hours or 8 hours or more.

Both are observational analyses of largely the same people, using different methods and different case definitions. Bian's authors frame their own conclusion conditionally: assuming the associations are causal, regular sleep might offset some of the risk of short sleep.

What the Evidence Base Looks Like as a Whole

Kalkanis and colleagues published a systematic review in Sleep Medicine Reviews (2025) covering 59 primary studies. Measurement methods and outcomes varied too much to pool into a meta-analysis, so the synthesis was narrative. The authors graded the evidence linking irregular sleep timing to depressive and anxiety symptoms, higher body mass index, insulin resistance, hypertension, and cardiovascular events as consistent and moderate-certainty. They called for randomized intervention trials, because the base is almost entirely observational.

That matters because irregular sleep travels with a lot of other things: shift work, illness, pain, depression, caregiving, and unstable routines. The cohorts adjust for many of these, but adjustment cannot rule out that poor health causes irregular sleep rather than the reverse.

What an Intervention Has Tested So Far

The only intervention I found that directly targeted regularity was a proof-of-concept study by Thosar and colleagues (Sleep Advances, 2025). Eleven adults aged 45 to 62 with hypertension were asked to go to bed at the same time every night for two weeks. Bedtime variability fell from about 32 minutes to about 7, and 24-hour systolic and diastolic blood pressure both dropped by an average of 4 mmHg. There was no control group, and the authors themselves say it needs a larger randomized trial.

For athletes specifically, a 2026 scoping review by Fox and colleagues (Behavioral Sleep Medicine) found 10 studies and described the evidence on regularity, performance, and wellbeing as scarce and inconsistent. I could not find a trial in masters athletes testing whether a steadier sleep schedule changes training, recovery, or health outcomes.

The Protocol

• What was measured: day-to-day consistency of sleep and wake timing, from about a week of wrist accelerometer data, not self-report.
• What it was associated with: lower all-cause and cardiovascular mortality and fewer cardiovascular events, in UK Biobank cohorts averaging about 62 years old.
• Duration versus timing: in the largest cardiovascular analysis, meeting the recommended hours was not associated with offsetting the risk among the most irregular sleepers.
• Bedtime versus wake time: in the Finnish cohort, bedtime variability was associated with risk and wake-time variability was not, and only among people sleeping under about 8 hours.
• What hasn't been tested: whether deliberately regularizing sleep changes any of these hard outcomes. The one intervention so far was 11 people, two weeks, no control group.
• Tracking it yourself: many wearables report bedtime and wake-time consistency. Their scores are not the same calculation as the SRI used in these papers.

For the book chapter WPP built around sleep as a training foundation, see Win the Long War's sleep chapter. If you have insomnia, sleep apnea symptoms, or a shift schedule you can't change, talk with your physician rather than adjusting your schedule based on population data.

Battle Hard. — Will Power

Sources

1. Windred DP, Burns AC, Lane JM, Saxena R, Rutter MK, Cain SW, et al. Sleep regularity is a stronger predictor of mortality risk than sleep duration: a prospective cohort study. Sleep. 2024;47(1):zsad253. PMID 37738616
2. Cribb L, Sha R, Yiallourou S, Grima NA, Cavuoto M, Baril AA, et al. Sleep regularity and mortality: a prospective analysis in the UK Biobank. eLife. 2023;12:RP88359. PMID 37995126
3. Chaput JP, Biswas RK, Ahmadi M, Cistulli PA, Rajaratnam SMW, Bian W, et al. Sleep regularity and major adverse cardiovascular events: a device-based prospective study in 72 269 UK adults. J Epidemiol Community Health. 2025;79(4):257-264. PMID 39603689
4. Nauha L, Niemelä M, Azadifar S, Korpelainen R, Farrahi V. Sleep timing irregularity in midlife: association with incident major adverse cardiac events and cardiovascular disease mortality over a 10-year follow-up. BMC Cardiovasc Disord. 2026;26(1):299. PMID 41877022
5. Yiallourou SR, Cribb L, Cavuoto MG, Rowsthorn E, Nicolazzo J, Gibson M, et al. Association of the Sleep Regularity Index with incident dementia and brain volume. Neurology. 2024;102(2):e208029. PMID 38165323
6. Bian W, Biswas RK, Ahmadi MN, Bin YS, Postnova S, Phillips AJK, et al. Dose-response associations of device-measured sleep regularity and duration with incident dementia in 82391 UK adults. BMC Public Health. 2025;25(1):516. PMID 39920677
7. Kalkanis A, Lenkens D, Steiropoulos P, Testelmans D. Sleep regularity as an important component of sleep hygiene: a systematic review. Sleep Med Rev. 2025;84:102203. PMID 41259946
8. Thosar SS, Sreeramadas AM, Jones M, Chaudhary N, Floyd-Driscoll C, McHill AW, et al. Bedtime regularization as a potential adjunct therapy for hypertension: a proof-of-concept study. Sleep Adv. 2025;6(4):zpaf082. PMID 41357358
9. Fox JL, Power CJ, Miller DJ, Scanlan AT. Does sleep regularity impact other sleep variables, performance and wellbeing in athletes? A scoping review and future research directions. Behav Sleep Med. 2026:1-17. PMID 42563357