Every other protocol in this section can be bought. This one cannot, which is probably why it gets skipped in favor of a colder mattress or a brighter lamp. It is also the only entry Compound labels established, because it is the rare sleep intervention with a formal consensus statement behind it rather than a promising mechanism and a hopeful supplement industry.

The protocol is two decisions and one rule. Pick a wake time. Pick a bedtime that gives you enough hours. Then hold both on Saturday. That last part is where it usually falls apart, and it is the part the research is actually about.

What it is

Regularity of sleep timing, held deliberately, measured as the day-to-day consistency of when you fall asleep and when you get up. It is a separate variable from how long you sleep. You can average a respectable seven and a half hours while going to bed anywhere between 10pm and 2am, and the research treats those as two different problems.

Sleep scientists quantify this with the Sleep Regularity Index, which scores the probability that you are in the same state — asleep or awake — at the same clock time on any two consecutive days. A perfectly regular sleeper scores 100. Someone whose schedule is effectively random scores 0. Most people land somewhere in the low 80s.

Where it came from

“Keep a regular schedule” sat on sleep hygiene checklists for decades as one bullet among a dozen, next to advice about warm milk. It was folk wisdom with a clinical veneer, and it was untestable at scale because nobody could measure it — self-reported bedtimes are unreliable, and sleep labs cannot follow you for a month.

Wrist accelerometers changed that. Once regularity could be computed from weeks of continuous movement data, researchers could separate it from duration and ask which one predicted outcomes. The Sleep Regularity Index was applied to a student cohort in 2017, then to hundreds of thousands of hours of population accelerometry, and in 2023 a National Sleep Foundation panel reviewed the accumulated literature and issued a consensus statement making regularity a formal target rather than a footnote. That arc — folk advice, measurable variable, consensus recommendation — is why the evidence label here is stronger than anything else on the sleep list.

Why people use it

The claims are that regular timing makes falling asleep easier, flattens the mid-afternoon crash, removes the Sunday-night insomnia that follows a late Saturday, and — the ambitious version — reduces long-term cardiometabolic risk. The first three are things you can verify on yourself within a few weeks. The last one is a population-level association, and the sections below keep those apart.

What happens physiologically

Two systems govern when you sleep, and irregular timing desynchronizes them.

The first is circadian: a master clock in the hypothalamus that runs on a roughly 24-hour cycle and is continuously corrected by light. That clock schedules melatonin release, the evening drop in core temperature, and the morning cortisol rise. It adjusts slowly — by under an hour per day — which means a schedule that moves three hours on the weekend is asking the clock to chase a target it cannot catch.

The second is homeostatic: sleep pressure that accumulates with every hour awake, mediated in part by adenosine, and discharges during sleep. A fixed wake time makes that pressure arrive at a predictable hour. Sleeping in until eleven on Sunday means you have banked four fewer hours of pressure by midnight, and the resulting difficulty falling asleep is not insomnia. It is arithmetic.

There is a feedback loop worth naming, because it explains why irregularity is self-sustaining. Irregular sleepers get less light during the biological day and more of it at night, which delays the clock further, which makes an early wake time harder, which produces another late night. The schedule and the light exposure degrade together, which is why this protocol pairs with morning light exposure and evening light reduction rather than substituting for them.

What the evidence supports

The 2023 National Sleep Foundation consensus statement is the reason for the label. A panel reviewed the published literature and voted formally on a set of statements, reaching agreement that consistency of sleep timing matters for health, performance, and safety, and that it should be treated as a distinct dimension of sleep health rather than folded into duration. Consensus statements are not proof, but they are the field telling you which findings it considers settled enough to recommend on.

The largest supporting dataset comes from UK Biobank. Windred and colleagues computed regularity scores from more than 10 million hours of accelerometer data across 60,977 adults, then followed mortality for about six years. The four most regular quintiles had a 20% to 48% lower risk of all-cause death than the least regular quintile, after adjustment for age, sex, ethnicity, and a long list of lifestyle and health factors. In the same dataset, regularity predicted mortality more strongly than sleep duration did — which is the finding that moved this protocol from hygiene advice to something worth ranking first.

The cardiovascular signal comes from MESA, where 1,992 adults free of cardiovascular disease wore actigraphy for a week and were followed for a median of 4.9 years. Participants whose nightly sleep duration varied by more than two hours had roughly double the risk of a cardiovascular event compared with those varying by an hour or less. Variability in sleep onset time showed a similar pattern. Both estimates were adjusted for average sleep duration, so this is not the short-sleep effect wearing a different hat, and excluding shift workers did not change the result.

For the everyday claims, the 2017 student study is the most legible. Undergraduates tracked for a month showed that more irregular sleepers had later circadian timing and poorer academic performance — and critically, regularity was largely independent of how much they slept. Two students averaging the same hours performed differently depending on how consistently they took them.

What the evidence does not support

Every health outcome above is observational. Nobody has randomized irregular sleepers into a regular schedule and measured whether they live longer, and it is not obvious anyone will. That leaves the causal direction open in both a boring way and an important way: illness, pain, depression, caregiving, and shift work all fragment sleep timing, so some of the association is sick people sleeping irregularly rather than irregular sleep making people sick. The adjustments in these studies reduce that concern. They do not eliminate it.

The effect sizes also describe extremes. The MESA hazard ratios compare people whose sleep varies by more than two hours a night against people who vary by less than one. That is not the difference between a 6:30 and a 7:00 wake time. If your schedule already holds within half an hour, the marginal return on tightening it further is probably small, and there is no evidence supporting the perfectionist version of this protocol.

Regularity is also not a substitute for sufficiency, and the two can pull against each other. A consistent five hours is still five hours. If you are chronically short on sleep, rigidly defending the wake time locks the deficit in place; the fix is an earlier bedtime, not a stricter alarm. Anyone genuinely underslept should solve the hours first and the consistency second.

Finally, the protocol does nothing about the reasons people cannot sleep. Apnea, anxiety, pain, a hot room, a bright room, a noisy street: a fixed schedule does not touch any of them, and holding one while lying awake for an hour is actively counterproductive. See the note on insomnia below.

How to use it

  1. 1Set the wake time by your least flexible morning of the week, not your ideal one. If one day requires 6:00, the anchor is 6:00 — an anchor you can only hold five days out of seven is not an anchor.
  2. 2Count back your sleep need, seven to nine hours for most adults, to get lights-out. If that number is uncomfortably early, that is the actual finding.
  3. 3Set an alarm for bedtime as well as morning. Almost nobody drifts late on the wake time; everyone drifts late on the bedtime.
  4. 4Start the wind-down 60 to 90 minutes before lights-out. A bedtime with no runway is a suggestion.
  5. 5Hold weekend timing within about 30 minutes of weekday timing. This is the whole protocol. Everything else is scheduling.
  6. 6After a bad night, get up at the normal time anyway and skip the recovery sleep-in. Avoid naps after early afternoon. Let that night's higher sleep pressure do the correction for you.
  7. 7Reinforce the schedule with light: bright light within an hour of waking, dim light for the last few hours of the evening. Timing cues without light cues are half a protocol.
  8. 8Give it three weeks before judging it, and judge it on how long you take to fall asleep and how you feel at 3pm — not on a sleep score.

How the home enables it

This is the protocol with no equipment, which makes it easy to assume the house is irrelevant. The opposite is true: a fixed wake time only survives if the bedroom lets you sleep through to it. Most broken schedules are broken from outside the body.

A bedroom that goes bright at 5:12am in June has already chosen your wake time for you, and it is not the one you picked. Blackout that genuinely blacks out — side channels sealed, not just fabric hung — is what makes the anchor yours. Street noise at dawn does the same thing more subtly, fragmenting the last sleep cycle without fully waking you, which is the case for treating the bedroom's noise floor as infrastructure. A room that holds afternoon heat into the night pushes sleep onset later and takes the bedtime with it; see sleep temperature.

Automation is where lighting control earns its cost on this specific protocol. Shades that open at your wake time convert the schedule from an act of willpower into an environmental cue, and they deliver the morning light dose at the same moment. Lights that ramp down on a timer in the evening do the same work at the other end. The house holds the schedule so you do not have to.

Some of this is decided long before move-in. Bedroom placement relative to the street and the mechanical room, glazing orientation, and wall assemblies are effectively fixed. Shades, bulbs, seals, and schedules are not, and they cover most of the gap. If you share a bed with someone on a different schedule, separate zoning — thermal, and where possible light — is worth more than either of you compromising.

Risks and contraindications

Low risk, with one real caveat. Do not lie in bed awake enforcing a bedtime. Extended wakefulness in bed is how conditioned insomnia forms, and the established treatment for chronic insomnia — cognitive behavioral therapy for insomnia — deliberately does the opposite, restricting time in bed and getting you up when sleep does not come. Anyone with persistent insomnia should follow that guidance, where a fixed wake time is part of the protocol but a fixed bedtime is not.

Shift workers cannot apply this as written and should not try to force it; the relevant approaches are different and worth discussing with a clinician. And if tracking the metric is making you anxious about sleep, stop tracking. Sleep anxiety driven by device data is a documented problem, and it costs more than the regularity gains.

The compressed version

Pick the wake time your worst weekday requires. Count back seven to nine hours for lights-out. Alarm at both ends. Weekend drift under thirty minutes. After a bad night, get up anyway. Bright light on waking, dim light at night, and a bedroom dark, quiet, and cool enough that the schedule is yours to keep rather than the room's to break.