Two papers published ten days apart this autumn have nudged an unusual idea further into the mainstream of sleep research: that when you sleep, and how consistently, may matter independently of how long you sleep.
The first, in Preventive Medicine on 30 September, followed 63,083 UK Biobank participants whose sleep was measured with wrist-worn accelerometers. Over a median of 7.8 years, 1,349 cases of osteoporosis were recorded. After adjustment, short sleep duration was associated with a 16% higher risk (hazard ratio 1.16, 95% confidence interval 1.04–1.31), irregular sleep with a 21% higher risk (1.21, 1.02–1.43) and low sleep efficiency with a 23% higher risk (1.23, 1.08–1.41). Among people who also carried a high genetic susceptibility to osteoporosis, the combination with irregular sleep carried a hazard ratio of 1.86 (1.25–2.76) compared with those at low genetic risk who slept regularly.
The second, in Sleep Health on 9 October, analysed actigraphy from 3,061 adults in the National Health and Nutrition Examination Survey who had cardiovascular-kidney-metabolic syndrome — a cluster of conditions involving the heart, kidneys and metabolism. Through 2019 there were 274 deaths. More fragmented day-to-day rest-activity patterns were associated with higher all-cause mortality (hazard ratio 1.15 per 0.1-unit increase in a fragmentation measure, 1.08–1.23), while greater rhythm amplitude was associated with lower mortality (0.74, 0.66–0.82). Spending less time at the same sleep midpoint from day to day — the regularity variable in the model — was associated with higher mortality (1.26, 1.04–1.52).
Both papers are observational, and the authors of the second say so plainly in their conclusion: the findings “do not establish causality or clinical thresholds.”
The study that started this line of work
The reference point for most of these analyses is a 2024 cohort study in the journal Sleep, which used more than 10 million hours of accelerometer data from 60,977 UK Biobank participants to compute a Sleep Regularity Index for each person.
Comparing the most regular participants with the least regular, the top four regularity quintiles had 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. The authors found that regularity predicted all-cause mortality more strongly than sleep duration did, even after adjusting for age, sex, ethnicity and a range of sociodemographic, lifestyle and health factors.
That is a striking claim, and it is the reason a metric most people have never heard of now appears in journal after journal. It is also why the meaning of the metric is under scrutiny.
The question the field is still asking
On 6 October, Sleep Health published a commentary by Jean-Philippe Chaput, Susan Redline and Emmanuel Stamatakis whose title asks the question directly: “What does sleep regularity add to sleep duration?” The piece is framed as a set of priorities for evidence and prevention rather than a conclusion.
The same researchers recur across this literature — Stamatakis and Chaput are among the authors of the larger UK Biobank wearables analyses — which is worth noting, because it means the people making the strongest case for regularity are also the ones calling for the field to pin down what it adds. Among the open items: how irregular a schedule has to be before it matters, whether duration and regularity interact, and whether the association survives in populations that differ from a volunteer research cohort.
What kind of evidence this is — and is not
Every study cited here is a prospective cohort, not a randomised trial. Cohorts can show that two things travel together; they cannot show that changing one changes the other. A person whose sleep timing swings widely may be a shift worker, a new parent, a caregiver, or someone living with depression or an untreated breathing disorder — and those circumstances carry their own health risks.
The measurement has limits too. In the mortality research the regularity metric is calculated from research-grade accelerometer recordings, not from the nightly score a consumer sleep app reports; the two are not interchangeable. And a hazard ratio describes a difference between groups over years, not a personal prediction. A hazard ratio of 1.21 for irregular sleep in the bone study, for instance, is a relative difference in a condition that remains uncommon in absolute terms.
What public guidance already says
The regularity message is not new to health agencies, even if the metrics are new to journals. CDC’s sleep pages state that adults aged 18 to 64 need seven or more hours a night and adults 65 and older seven to nine, and that quality counts alongside quantity: “Getting quality sleep is not just about how many hours you sleep, but how well you sleep.”
The National Heart, Lung, and Blood Institute goes further toward consistency in its healthy-sleep-habits guidance: go to bed and wake up at the same time every day, and try to keep the same schedule on weeknights and weekends, “limit the difference to no more than about an hour.” The same pages note that caffeine’s effects can last up to eight hours, and that a bedroom should be kept quiet, cool and dark. As for the scale of the problem, NHLBI cites CDC figures that about one in three US adults report not getting enough rest or sleep every day, and links sleep deficiency to heart disease, kidney disease, high blood pressure, diabetes, stroke, obesity and depression.
Practical guidance of that kind is what the new research is ultimately in service of. It is also several steps behind the headlines the research generates.
What we did not find
Nothing in the studies reviewed here shows that making an irregular sleeper regular extends life, prevents osteoporosis or protects the heart. No randomised trial has tested that, and no threshold has been agreed: the studies do not say whether a two-hour swing in bedtime is fine and a four-hour swing is not. The strongest claims in the literature are about prediction — how well a measure sorts people by later risk — not about what a given person should do tonight.
For anyone whose sleep is persistently disrupted, or who wakes unrefreshed despite allowing enough time in bed, the relevant next step is a conversation with a healthcare professional rather than a schedule experiment. NHLBI maintains separate pages on sleep apnea, a condition in which breathing repeatedly stops and starts during sleep, precisely because some sleep problems have a clinical cause and a clinical treatment. As with everything on this site, this article is information about a claim, not medical advice.
If the mechanics of a wind-down routine are what you are after, our earlier piece on what actually helps the hour before bed covers the everyday habits, and the hydration–sleep question — including why an “evening detox drink” is not a thing — is set out in our article on evening hydration and sleep.