Sleep Regularity Beats Sleep Duration for Predicting Mortality

The night you “got enough sleep”… but still felt off

Picture a familiar scenario: you sleep a full eight hours on paper, wake up after your alarm, and yet your body feels like it’s dragging. The next day might be fine. Or it might be the start of a streak of headaches, foggy thinking, and low energy.

For years, the public health conversation has treated sleep as a simple quantity problem: get enough hours and you’re covered. But a growing line of research is making a more nuanced claim. Not just how long we sleep, but how consistently we sleep—night after night—may track health risk more closely.

That idea takes center stage in a new prospective study in SLEEP, which compares sleep regularity and sleep duration as predictors of mortality risk, using objective sleep data from wrist-worn activity monitors.

Sleep duration vs. sleep regularity: two different measurements

Before getting into the results, it helps to make the terms concrete.

Sleep duration is the total amount of sleep you get during the night, typically measured as hours of sleep.

Sleep regularity is the day-to-day consistency of your sleep–wake timing. If your bedtime and wake time drift around by an hour or more across many days, your sleep is “irregular.” If bedtime and wake time stay tightly clustered, your sleep is “regular.”

This distinction matters because biology doesn’t run only on “total rest time.” It also runs on timing. Your body clock—commonly called your circadian rhythm—coordinates things like hormone release, body temperature, and metabolism on a roughly 24-hour schedule.

A key question that researchers are asking is: Could going to bed at the same time matter more than getting exactly eight hours?

How the study measured sleep (without asking participants to guess)

One reason this research is persuasive is the measurement method.

Instead of relying on questionnaires like “How many hours did you sleep last month?” the researchers used actigraphy.

Actigraphy is sleep tracking using movement data, usually from a wrist-worn accelerometer (a sensor that measures motion). When you’re still for long stretches, the system infers sleep; when you move more, it infers wake.

The study population

The researchers analyzed data from 60,977 people in the UK Biobank who wore accelerometers for one week. That’s an important design choice: it means sleep was captured objectively across days, not remembered later.

The Sleep Regularity Index (SRI)

To quantify regularity, the study used a metric called the Sleep Regularity Index (SRI). Conceptually, SRI measures how consistently sleep and wake occur across the circadian timescale—capturing irregularities from things like fragmented sleep, variable sleep timing, and inconsistent duration.

Even without memorizing the formula, the intuition is simple: SRI increases when your sleep pattern resembles a stable routine, and decreases when your nights “shift around.”

Mortality risk and hazard ratios, translated into plain language

When researchers ask whether sleep predicts mortality, they usually need a model that links risk over time.

A common choice is the Cox proportional hazards model. In plain terms, this model estimates how the “hazard”—the instantaneous risk of an event happening at a specific time—changes across groups.

The output is often a hazard ratio (HR):
- HR < 1 means lower risk than the reference group.
- HR = 1 means no difference.
- HR > 1 means higher risk.

So when the paper reports HRs for sleep regularity percentiles, it’s comparing the risk of death across people with more vs. less regular sleep patterns, while adjusting for other factors.

The headline result: sleep regularity predicts mortality better than duration

Here’s the core finding.

In fully adjusted analyses, sleep regularity was a stronger predictor of all-cause mortality than sleep duration.

The study reports a monotonic pattern for regularity: as regularity increased, mortality risk decreased. In the most regular group compared with the least regular group, the fully adjusted model corresponded to an HR around 0.70, meaning about a 30% lower hazard for the most regular sleepers versus the least regular.

For sleep duration, the relationship with all-cause mortality was different: it looked more non-linear in the minimally adjusted model (a curved, “U-shaped” pattern), while the fully adjusted model showed a more restrained, roughly linear trend.

But the comparison that matters is not “which curve looks nicer.” It’s whether sleep duration explains additional risk beyond what regularity already captures.

The study tested model comparisons that effectively ask: if you already know someone’s sleep regularity, does knowing their average sleep duration still improve predictions?

The result: adding sleep duration did not meaningfully outperform models that relied on sleep regularity alone.

In other words, duration and regularity aren’t redundant, but regularity appears to carry the stronger signal for mortality prediction.

Cause-specific patterns: regularity matters most for some death categories

The paper also separates mortality into different causes, including:
- Cardiometabolic mortality (deaths tied to cardiovascular and metabolic disease)
- Cancer mortality
- Other-cause mortality

To handle multiple possible causes of death, the researchers used a competing-risks framework (a method that acknowledges that the occurrence of one type of event can “prevent” observing another type later).

Broadly:
- Sleep regularity predicted lower risk for multiple categories of death.
- Sleep duration showed weaker and more selective predictive value for specific causes.
- In the fully adjusted combined models (regularity plus duration), regularity remained significant for cancer and other-cause mortality, while the association for cardiometabolic mortality attenuated.

That pattern fits a biological story: disrupted circadian timing can ripple into many systems—immune function, metabolic regulation, inflammation, and behavior rhythms—so timing irregularity may map onto broader long-term risk.

Why regularity might beat duration: the circadian disruption theory

A persuasive part of the discussion is the mechanism.

If irregular sleep is the product of shifting light exposure, changing meal timing, inconsistent activity schedules, or other “timing noise,” then it acts like a proxy for circadian disruption.

Even when average sleep duration is adequate, your body still receives mismatched timing cues. Over weeks and months, that mismatch can influence physiology in ways that don’t show up immediately on a “total hours slept” metric.

It’s not that sleep duration is unimportant. The study’s contribution is sharper: duration helps, but regularity appears to be a more sensitive marker of the kind of internal timing instability that tracks long-term outcomes.

What “better sleep” looks like in practice when you care about regularity

Translating a regularity-first result into daily life is where things get interesting—and also a little tricky. People often focus on bedtime alone, but regularity is really about both anchors: bedtime and wake time.

A practical way to improve sleep regularity usually involves:
- Keeping a consistent wake time, even on days when bedtime might shift.
- Reducing big swings in bedtime, especially later in the evening.
- Managing naps so they don’t create a cycle of “catch-up” sleep that throws nighttime timing off.
- Being cautious with schedules that repeatedly shift circadian timing, such as frequent late nights or rotating shifts.

Regularity tends to be built through repeated exposure to stable cues. Your brain learns the routine the way a city learns a traffic pattern: slowly, then all at once.

And this also reframes a common frustration. Someone can follow a “sleep 8 hours” rule and still fail the regularity test if their schedule drifts by hours most days.

The takeaway: sleep isn’t only a duration story anymore

This study pushes the field toward a timing-first view of sleep health.

By using objective actigraphy data, a dedicated sleep regularity metric (SRI), and statistical models designed to compare competing predictors, the findings support a clear message: sleep regularity is a stronger predictor of mortality risk than sleep duration.

So the next time sleep advice says “just get more hours,” it’s worth remembering what your biology seems to prefer. Not only rest, but rhythm. Not only sleep quantity, but sleep timing that stays steady enough for your circadian system to keep its footing.