That second number is what a research team in Oulu went looking for, and it turned out to carry weight. In a paper published in BMC Cardiovascular Disorders, Laura Nauha and colleagues at the University of Oulu report that middle-aged adults with erratic bedtimes were about twice as likely to suffer a serious cardiac event over the following decade. One condition applies: they also had to be sleeping less than eight hours.

What a wobbly bedtime looks like on paper

Participants came from the Northern Finland Birth Cohort 1966, a group of northern Finns who have been measured, questioned and sampled since before they were born. At age 46, some 3,231 of them wore activity monitors on their wrists, and the first seven consecutive nights were used to calculate how much each person’s bedtime drifted.

Then everyone was sorted into thirds.

Regular sleepers varied their bedtime by about half an hour. Fairly regular sleepers hovered around an hour. Irregular sleepers shifted theirs by nearly two hours, night to night, week in and week out.

Finnish hospital and death registers did the rest of the work. By the end of 2023, 128 people, four per cent of the group, had experienced a major adverse cardiac event: heart attack, unstable angina, stroke, admission for heart failure, or cardiovascular death.

Averages told no story whatsoever. People who went on to have an event and people who didn’t looked much the same on paper: roughly the same usual bedtime, the same usual wake time, much the same average night’s sleep. Risk only surfaced once the analysis was split by sleep length and adjusted for weight, blood pressure, cholesterol, blood sugar and daily activity.

Why bedtime and not the alarm clock

Wake-up time showed nothing at all.

That asymmetry is the strangest result in the paper. Bedtime variability predicted risk, variability in the midpoint of the night predicted risk, and the hour people got up predicted nothing in either half of the cohort. Nauha’s own reading, given in the University of Oulu’s announcement of the findings, is that bedtime regularity “reflects the rhythms of everyday life” and how much they fluctuate.

A duller explanation is also available. Alarms get set by employers and school runs, while a bedtime answers to whatever happened that evening.

Brian Zachariah, a psychiatrist who spoke to Medical News Today about the results, laid out why a biologist would take the numbers seriously: irregular sleep timing feeds insulin resistance and systemic inflammation, which in turn push atherosclerosis and clotting. He rated a hazard ratio near 2.0 as sitting in the same company as smoking or diabetes.

Sarathi Bhattacharyya, a sleep medicine specialist quoted in that report, put it in terms of maintenance rather than mechanism. Move the start of the night around and “the body’s period for recovery” moves with it, taking hormone signalling and blood pressure regulation along for the ride. The cost accrues quietly across years rather than nights.

Where the eight hours come in

Here the picture narrows sharply. Participants were split at their median time in bed, which came out at 7 hours and 56 minutes. Among the shorter sleepers, an irregular bedtime carried a hazard ratio of 2.01, with a 95 per cent confidence interval running from 1.00 to 4.01 and a p-value of 0.049. Among the longer sleepers, nothing measurable at all.

Removing the 527 shift workers from the analysis strengthened the association rather than weakening it, which suggests rotating rosters are not the ones doing the damage.

Those numbers sit right at the edge of statistical significance, and the researchers say as much in writing: with 128 events split across groups, precision suffers. Everyone in the cohort was the same age, from the same corner of one country, and monitored for a single week.

One cohort, one week, one finding.

What the bigger studies say

Sleep regularity as a cardiovascular risk factor has support well beyond this cohort. A device-based study of 72,269 UK Biobank adults led by Jean-Philippe Chaput found that irregular sleepers ran a 26 per cent higher risk of a major cardiovascular event across nearly eight years of follow-up, with risk dropping steadily as regularity improved.

That larger study contradicts the Finnish one on its central caveat, though. Chaput’s team reported that hitting the recommended hours failed to cancel out the risk for the most irregular sleepers, whereas the Oulu data hints that eight hours or more offered some protection whatever the schedule looked like. The two cannot both be right in full. What survives the overlap is narrower and more useful: regularity counts, and whether a decent night’s sleep buys back the damage is still an open question.

There is a practical edge to all this. On the Finnish numbers, the people who stand to gain are short sleepers, who happen to be the group least able to conjure extra hours out of a working week. Going to bed at the same time costs nothing and needs no equipment.

Whether it holds up in a bigger and messier population is the next question. And 128 events in a single Finnish birth year will not answer it.