Better Sleep Starts at Sunrise: What the Research Supports in a Morning Routine

By Courtney Turrin · October 6, 2026 · 19 min read read

woman sleeping in morning light — Photo courtesy of Pexels.

Light exposure, a steady wake time, and an afternoon caffeine cutoff carry the strongest evidence for better sleep and long-term health, and your chronotype changes how to use them. Several popular habits deliver less than promised, and one has no support at all.

For Richmonders, October marks the month with the latest sunrises of the year, reaching 7:33 a.m. on October 31. For anyone whose alarm goes off earlier, the day starts in the dark.

Chronobiologist Till Roenneberg estimates that about two-thirds of people in Western societies already live with a significant gap between their internal clocks and the demands of work and school, a gap he calls social jetlag. Shorter days give the body clock less light to work with. On December 21, the shortest day of the year, Richmond gets 9 hours and 37 minutes of daylight.

The research on what helps offset the negative impact of short days sorts the usual morning-routine advice sharply. Light, a consistent wake time, and an afternoon caffeine cutoff have the strongest evidence behind them, while several popular habits (breathing exercises, meditation, a bedtime to-do list) rest on thin evidence for sleep. Many of our readers will be relieved to hear that one piece of advice (delaying caffeine intake upon waking) rests on no evidence at all.

The scorecard below grades eight common habits by strength of evidence: how confident the research allows anyone to be that a habit delivers what it promises. An A means large studies or multiple trials point in the same direction, while a C means the claim outruns the data. The time column shows what each habit costs. As it turns out, the best-supported habits are among the cheapest to implement (Figure 1).

[datawidget key="routine-scorecard"]

Your Chronotype and Social Jetlag Come Down to Two Numbers

Before grading your routine, it helps to understand the idiosyncrasies of your personal internal clock. Author Daniel Pink, drawing on research by Roenneberg, sorts people by a single number in his book When: The Scientific Secrets of Perfect Timing: the midpoint of sleep on a day with no alarm. For most readers, this means halfway between falling asleep and waking up on a Friday night/Saturday morning or Saturday night/Sunday morning.

Pink estimates that about 15% of people are larks, early chronotypes who peak in the morning, and about 20% are owls, who thrive late at night. The remaining 65% fall in between, a group he calls third birds.

Social jetlag measures the difference between mid-sleep times on nights before work days and nights before free days. Someone who sleeps from 11 p.m. to 7 a.m. on weekdays has a 3 a.m. midpoint. If the same person sleeps from 1 a.m. to 9 a.m. on weekends, the midpoint jumps to 5 a.m., and the body absorbs a two-hour time-zone change every Friday night and reverses it every Monday morning. The size of the gap depends on chronotype as much as on the alarm, since a 7 a.m. start asks more of an owl's clock than of a lark's.

That weekly commute between time zones shows up in blood work. In 447 healthy day-shift workers aged 30 to 54, researcher Patricia Wong and colleagues found that greater social jetlag was associated with lower HDL cholesterol, higher triglycerides, insulin resistance, and more body fat, though the cross-sectional study design shows correlation rather than causation.

Enter your typical sleep and wake times on work days and free days to calculate both numbers (Figure 2).

[datawidget key="chronotype-calculator"]

Six Hours or Less: Short Sleep Is Linked to Metabolic Disease and Susceptibility to Common Colds

The American Academy of Sleep Medicine and the Sleep Research Society recommend that adults sleep seven or more hours a night on a regular basis. Their 2015 joint consensus statement associates regularly sleeping less than seven hours with weight gain, diabetes, hypertension, heart disease and stroke, depression, impaired immune function, more accidents, and a higher risk of death. The panel framed these as associations. Experimental work tests causes directly (Figure 3).

Psychiatry researcher Aric Prather and colleagues measured the sleep of 164 healthy adults with wrist monitors for a week, then quarantined them in a hotel and gave them nasal drops containing a cold virus. Participants who had slept less than six hours a night were 4.2 times more likely to develop a cold than those who slept more than seven. Short sleep outperformed every other factor the team measured in predicting who got sick.

[datawidget key="sleep-outcomes-tiers"]

The Association Between Sleep and Dementia Remains Unsettled

The most cited evidence of a link between sleep and dementia risk comes from a cohort of British civil servants. Epidemiologist Séverine Sabia and colleagues tracked 7,959 participants whose sleep duration was recorded six times between 1985 and 2016. People who persistently slept six hours or less were about 30% more likely to develop dementia than those who slept a normal amount, and the short sleep was measured more than 25 years before the average age of dementia onset.

That long lead time is the study's main defense against reverse causation, the possibility that early dementia was disrupting sleep rather than the other way around. The defense weakens with age. The association weakened at age 70, and the researchers suspect some participants that age were already in early, undetected stages of the disease. Sabia has said she cannot claim that short sleep causes dementia, only that it may contribute.

A small experiment points to a possible pathway. Researcher Ehsan Shokri-Kojori and colleagues at the National Institutes of Health scanned 20 healthy people after a normal night and after a night without sleep. One sleepless night raised beta-amyloid, the protein that clumps into plaques in Alzheimer's disease, by about 5% in the thalamus and hippocampus. Whether recovery sleep reverses the increase is unknown.

The popular explanation for why sleep would matter for dementia risk is the glymphatic system, a proposed network that flushes waste from the brain with cerebrospinal fluid during deep sleep. In 2024, researcher Andawei Miao and colleagues measured clearance directly in mice and found that dye cleared more slowly during sleep than while awake. Another group, led by neuroscientist Maiken Nedergaard, reported that norepinephrine pulses during non-REM sleep drive fluid through the mouse brain. A third possibility sidesteps clearance entirely: people who sleep less may simply produce more amyloid during their extra waking hours.

The Lancet Commission on dementia left sleep off its list of modifiable risk factors in 2020, noting that sleep changes may be early symptoms of the disease, and its 2024 update again omitted sleep for insufficient evidence. No trial has shown that improving sleep reduces dementia risk. Current evidence supports that short sleep in midlife marks elevated dementia risk, a single night of sleep loss nudges amyloid upward, and the mechanism connecting the two remains unknown. The other outcomes in this section rest on firmer ground, and they are reason enough to protect sleep.

Indoor Light Exists in a Gray Zone: Too Dim for Day and Too Bright for Night

Sleep scientist Daniel Windred and colleagues followed 88,905 UK Biobank participants, who wore light sensors for a week, for about eight years. Compared with people in the darker half of daytime exposure, those with the brightest days had an incrementally lower risk of death, as much as a third lower at the top of the range. By contrast, people with the brightest nights had a 21% to 34% higher risk of death (Figure 4). The links held after controlling for physical activity and income, though an observational study cannot prove that light caused the difference in mortality. Participants were 40 to 69 years old, and the UK Biobank skews healthier than the general population.

[datawidget key="light-mortality-gradient"]

Sleep scientist Charles Czeisler told NPR that brighter light during the day makes people less sensitive to light at night, so a morning outside works in part by protecting the evening.

That matters most for clocks that run late. In camping experiments led by sleep researcher Kenneth Wright, volunteers who lived under only natural light for a week went to bed and rose two or more hours earlier than they had at home, and a single weekend shifted their sleep schedules by almost an hour and a half. The studies were small, but Wright's team concluded that modern indoor living pushes internal clocks later in every season.

In 2022, an international panel of circadian researchers led by Timothy Brown published the first consensus targets for light exposure: a daytime minimum of 250 lux, an evening maximum of 10 lux starting three hours before bed, and no more than 1 lux in the bedroom during sleep. When researchers later measured people in daily life against those targets, they found widespread failure to meet them, especially in the evening and at night. The targets use a newer unit that weights light by its effect on the eye's circadian receptors, which roughly matches lux for daylight but runs lower for warm indoor bulbs, so a room can look well lit while registering as dim to the body clock.

What 10 Lux Looks Like

Few people know what a lux is, so the targets need translating (Figure 5). The 1-lux bedroom target amounts to keeping the room as dark as possible. At 10 lux, the evening maximum, a lab study led by sleep scientist Andrew Phillips found that melatonin onset, the body's signal that night has begun, arrived 22 minutes later than in dim light. At about 30 lux, similar to a floor lamp or a bright smartphone screen, melatonin fell by more than half in the average participant.

A well-lit office provides about 200 lux, according to sleep neurologist Phyllis Zee. Outdoors, the numbers jump by orders of magnitude: researchers at Aarhus University describe 100 lux as a very dark, overcast day and 10,000 lux as full daylight, and direct midday sun reaches about 50,000 lux by one clinical estimate. Sources disagree on outdoor figures, so the ladder is best read as ranges.

Phillips and colleagues put the modern problem in one comparison. Before electric lighting, humans lived mostly under intense light above 300 lux or dim light below 30. Today people spend hours a day in the 30-to-300 range, particularly in the evening, a band too dim to anchor the day, yet bright enough to delay the night.

[datawidget key="lux-ladder"]

Living Room Lamps May Matter More Than Phones

Screens draw most of the blame for evening light, but evidence suggests that household lighting deserves a sharper look. When Sean Cain and colleagues measured lighting in real homes, 48% were bright enough in the evening to suppress melatonin by at least half, averaged across sensitivity levels. A 2024 review in Sleep Medicine Reviews by Serena Bauducco and colleagues concluded that blue light from screens delays sleep by only a few minutes, and that the larger problems lie in how screens displace sleep time and keep the mind engaged.

Sensitivity varies enormously: in the Phillips study, 50% melatonin suppression took 6 lux in the most sensitive participant and 350 lux in the least. The practical response is the same for everyone: dim the overhead lights and switch to lamps in the last few hours before bed.

A Steady Wake Time Predicts Mortality Better Than Total Hours of Sleep

Analyzing more than 10 million hours of wrist-sensor data from 60,977 UK Biobank participants, Windred and colleagues found that sleep regularity, the day-to-day consistency of sleep and wake times, predicted risk of death more strongly than sleep duration did. The most regular sleepers had a 20% to 48% lower risk of death than the least regular.

After a Bad Night, Keep Your Wake Time

The instinct after a rough night or a short week is to sleep in. A controlled trial led by integrative physiologist Christopher Depner tested that strategy. Participants slept no more than five hours a night for five days, slept as long as they wanted over a weekend, then returned to restricted sleep. They gained only about 1.1 hours of extra sleep, and after the weekend their insulin sensitivity fell 9% to 27% below baseline. "In the end, we didn't see any benefit in any metabolic outcome in the people who got to sleep in on the weekend," Depner said.

The trial tested chronic sleep loss, and coauthor Wright has said it remains unclear whether catch-up sleep helps after one or two short nights. In their review of social jetlag research, Roenneberg and colleagues suggest that a weekday schedule forced against the body clock is what disrupts health, and that choosing to sleep longer on free days might protect it.

As for what to do after a single night of bad sleep, no existing study has settled the question. The regularity and light research point toward getting up near the usual time, getting outside early, and letting the next night's sleep make up the difference. Sleeping in by two hours recreates the social jetlag problem.

Dark Winter Mornings: Light Boxes and Dawn Simulators Both Have Trial Support

For most people, the simplest fix for winter mornings is to go outside, even when cloudy. In a study of 507 University of Washington undergraduates fitted with wrist monitors, Gideon Dunster and colleagues found that students' sleep shifted later in winter, and that daytime light exposure predicted the shift more strongly than evening light did. Outdoor light on overcast days still drove the effect, because gray daylight is far brighter than indoor lighting.

For people with seasonal depression, a clinical condition rather than winter sluggishness, light therapy has a long trial record. A 2020 meta-analysis led by psychiatrist Edda Pjrek found that people treated with bright light were 42% more likely to respond than those given placebo light.

The American Academy of Family Physicians recommends a light box of 2,500 to 10,000 lux for 30 to 60 minutes each morning, ideally just after waking. Dawn simulators, which brighten a bedside lamp gradually before waking, need no time commitment once set, and the AAFP finds them similarly effective for mild to moderate symptoms of seasonal depression, with bright light possibly better for severe cases. Head-to-head trials disagree: one randomized trial favored dawn simulation, and a later trial found no advantage over bright light after waking.

People with bipolar disorder should use light therapy only under clinical supervision, since guidance lists it as contraindicated for anyone with current or recent mania, hypomania, or rapid cycling.

Popular Habits, Ranked From Best Supported to Unsupported

The trials behind these habits vary widely in size and rigor.

Caffeine Consumption Too Close to Bedtime Costs 45 Minutes of Sleep

The strongest evidence on caffeine concerns the afternoon. In a placebo-controlled study, sleep researcher Christopher Drake and colleagues gave 12 healthy sleepers 400 milligrams of caffeine, roughly two to three cups of coffee, at bedtime, three hours before bed, or six hours before bed. Even the six-hour dose cut total sleep time by more than an hour, as measured by an in-home sleep monitor.

A 2023 meta-analysis led by researcher Carissa Gardiner pooled 24 controlled studies and found that caffeine reduced total sleep time by 45 minutes on average and sleep efficiency by 7%, while adding 9 minutes to the time needed to fall asleep and 12 minutes of waking during the night. To avoid losing any sleep, the authors estimated that a cup of coffee should be consumed at least 8.8 hours before bedtime, which works out to about 1 p.m. for a 10 p.m. bedtime. Because the cutoff is counted back from bedtime, it shifts with chronotype: about 4 p.m. for someone who goes to bed at 1 a.m.

A 10- to 20-Minute Outdoor Walk Combines Light, Steps, and a Drop in Cortisol

The best-known study of walking and the brain set a high bar. Environmental scientist Gregory Bratman and colleagues sent 38 healthy adults on a 90-minute walk through either grassland or a busy street. Only the nature walkers reported less rumination afterward and showed reduced activity in the subgenual prefrontal cortex, a brain region linked to repetitive negative thought. While ninety minutes in nature sounds delightful, few people have that much time on a workday.

A more attainable dose of outdoor exposure comes from researcher Mary Carol Hunter and colleagues, who asked 36 city residents to spend at least 10 minutes, three times a week, somewhere that gave them a sense of contact with nature. Each session lowered cortisol 21.3% per hour beyond its normal daily decline, and the benefit per minute was greatest between 20 and 30 minutes. Sitting worked as well as walking. Participants were asked to skip social media, phone calls, and reading, which makes leaving the phone put away in a pocket part of the protocol.

The steps pay off independently. A 2025 meta-analysis in The Lancet Public Health led by Ding Ding found that 7,000 steps a day was associated with 47% lower risk of death compared with 2,000 steps, and nearly matched the benefit of 10,000 steps. A dog helps: in a matched study of adults 65 and older, dog owners walked an extra 22 minutes and 2,760 steps a day compared with non-owners.

When to walk matters less than whether to do it at all. Separate UK Biobank analyses disagree: one found morning activity associated with lower cardiovascular risk, another found higher cardiovascular mortality among early-morning exercisers who were evening chronotypes, and a third found midday and afternoon activity outperformed morning activity. Owls who dread the dawn walk can take theirs later and still collect the steps, while getting some outdoor time earlier in the day for the light benefits.

A Five-Minute Bedtime To-Do List Helped Young Adults Fall Asleep Faster

The habit with the most direct link between intention-setting and sleep takes place at night. Sleep researcher Michael Scullin and colleagues brought 57 healthy young adults into a sleep lab and measured their brain activity overnight. Before lights out, half spent five minutes writing a to-do list for the coming days, and half wrote about tasks they had already completed. The to-do group fell asleep in about 16 minutes on average, compared with 25 for the completed task list, and the more specific their to-do lists were, the faster they fell asleep.

The study had no group that skipped writing entirely, so listing completed tasks may have slowed people down rather than the to-do list speeding up sleep relative to baseline. Furthermore, none of the study participants had sleep problems. Scullin's proposed explanation is offloading: writing tasks down relieves the mind of holding them overnight. The list also does double duty as tomorrow morning's plan, the kind of specific, written intention that motivation research links to follow-through.

Breathing and Meditation Lift Mood, but the Sleep Evidence Is Mixed

In a month-long randomized trial, researcher Melis Yilmaz Balban and colleagues assigned participants to five minutes a day of one of three breathing exercises or mindfulness meditation. All four groups reported better mood, and cyclic sighing, which emphasizes long, slow exhalations, produced the largest gains in positive mood and the largest drop in resting breathing rate. None of the groups showed a change in sleep, as measured by a wearable device or a daytime sleepiness questionnaire.

A 2019 meta-analysis led by researcher Heather Rusch found a different pattern in people who already had disturbed sleep. Pooling 18 randomized trials with 1,654 participants, mindfulness meditation improved sleep quality compared with time-matched control activities, a moderate effect with moderate-quality evidence, though it showed no clear advantage over established sleep treatments. Those trials mostly tested multi-week programs.

Taken together, the two bodies of evidence support a narrow reading: five minutes a day lifted mood without changing sleep, and the sleep benefit came from longer meditation programs in people whose sleep was already poor.

The 90-Minute Coffee Delay Has No Supporting Evidence

The advice to wait 90 minutes after waking before drinking coffee, meant to prevent an afternoon crash, has circulated widely. A 2024 evidence review in the Journal of the International Society of Sports Nutrition, led by exercise scientist Jose Antonio, found no evidence that caffeine on waking causes an afternoon crash or that delaying it would prevent one. The research on caffeine timing points to the end of the day and finds nothing for the start of it.

The same review found that moderate daily doses of 250 to 300 milligrams, roughly two to three cups, do not appear to increase urine output in habitual coffee drinkers. A glass of water before coffee is harmless and may help people who habitually drink little: a review of hydration research found that increasing water intake improved mood in low-volume drinkers, though nothing supports a particular order of operations.

A Realistic Daily Routine Takes Under 30 Minutes

Pulling the top-graded habits together produces a routine short enough for a workday (Figure 6). Wake at a consistent time, as close to your free-day wake time as your schedule allows. Spend 10 to 20 minutes outside in the first part of the morning, walking if possible, with the phone in a pocket. On mornings that start before sunrise, a dawn simulator adds light without adding minutes. Finish caffeine about nine hours before bedtime. In the last few hours before bed, switch from overhead lights to lamps, then write a specific to-do list for the next day.

No study has tested this combination as a package, but most of it costs nothing, and the parts with the strongest evidence take the least time.

[datawidget key="daily-routine-30-minutes"]

When Habits Are Not Enough: Chronic Insomnia and Sleep Apnea

Routines can improve ordinary sleep. They are not treatments for sleep disorders. About 6% to 10% of adults have insomnia that meets diagnostic criteria, and the American College of Physicians recommends cognitive behavioral therapy for insomnia, known as CBT-I, as the initial treatment for all adults with chronic insomnia. The recommendation carries the group's strongest grade. CBT-I combines changes to sleep-related thinking with behavioral techniques, and it is available through individual or group sessions, online programs, and self-help books.

Sleep apnea is the other condition worth ruling out, as research suggests a possible association between sleep apnea and dementia. Anyone whose sleep problems persist despite good habits should raise them with a physician.

What to Do Before and After a 7:33 a.m. Sunrise

The sun will rise over Richmond at 7:33 a.m. on October 31. Most of what the scorecard rewards is built around that moment: a consistent wake time that puts the body clock on a schedule, light on arrival (outdoors if possible, or a dawn simulator if the alarm comes first), and dim evenings that let the clock register the difference.

What the research cannot yet say is whether changing these habits changes outcomes. The strongest findings on light and wake time are associations from large cohorts, the trials are small and short, no study has tested the combination, and no trial has shown that improving sleep protects the aging brain. Until those trials are run, the evidence favors the habits that cost the least.