Eating Before Bed: How Late Dinners Affect Your Sleep

Eating close to bedtime is linked to more night waking, not to trouble falling asleep. In a survey of 793 young adults, eating within three hours of bed raised the odds of waking during the night by about 60 percent, while sleep onset and total sleep time were unchanged. Controlled work points the same direction on meal composition: high fibre predicts deeper sleep, and saturated fat and sugar predict lighter, more fragmented sleep. Timing matters, but what is on the plate matters at least as much.

Avoiding food before bed is one of the oldest pieces of sleep hygiene advice, and for a long time it rested on almost nothing. That has changed. Over the last decade researchers have looked at meal timing in survey cohorts, in national nutrition data, in night-shift nurses and in overnight sleep labs, and a reasonably consistent picture has emerged. It is also a messier picture than the advice suggests, because at least one well-controlled trial found the opposite of what everyone expected. Here is what the research supports, where it disagrees with itself, and what actually seems worth changing about your evening.

1. The three-hour window keeps showing up in the data

The clearest single finding comes from a cross-sectional study of 793 Australian university students aged 18 to 29. Researchers split participants by whether their evening meal fell inside or outside a three-hour window before bed, then looked at three outcomes: night waking, taking more than 30 minutes to fall asleep, and sleeping six hours or less.

Only one of the three moved. Eating within three hours of bed was associated with waking during the night (odds ratio 1.61, 95 percent confidence interval 1.15 to 2.27). Sleep onset latency and short sleep duration showed no significant association. The relationship survived adjustment for ethnicity and body mass index, though it narrowed considerably (OR 1.43, 95 percent CI 1.00 to 2.04), which is worth noting rather than glossing over.

Citation: Chung N, Bin YS, Cistulli PA, Chow CM. Does the Proximity of Meals to Bedtime Influence the Sleep of Young Adults? A Cross-Sectional Survey of University Students. International Journal of Environmental Research and Public Health, 2020. PubMed: 32295235

A much larger analysis backs the direction of travel. Using NHANES data from 7,023 US adults with two 24-hour dietary recalls each, researchers compared meal timing against Pittsburgh Sleep Quality Index scores. Later first meals, later midpoint meals, later final meals and more frequent eating occasions were all associated with worse sleep quality after adjustment. The effect sizes are small, but the pattern is consistent across every timing measure they looked at.

Citation: Yan LM, Li HJ, Fan Q, Xue YD, Wang T. Chronobiological perspectives: Association between meal timing and sleep quality. PLOS ONE, 2024. PubMed: 39088487

2. Why the effect lands on night waking rather than falling asleep

This is the part that gets lost in most advice. If late eating simply made you alert, you would expect it to delay sleep onset. It does not, at least not reliably. What it appears to do is make the second half of the night less stable.

Two things plausibly drive that. The first is thermoregulation. Sleep onset depends on core body temperature falling, and digesting a meal generates heat. The second is circadian. Your gut, liver and pancreas run their own peripheral clocks, and feeding them at a time your brain has already flagged as night creates a mismatch that the NHANES work above interprets as a chrono-nutrition effect rather than a simple digestion effect.

The night-shift literature is a useful stress test here, because these workers eat at times almost guaranteed to be out of phase. In 30 female nursing professionals on permanent 12-hour night shifts, every one-hour reduction in the gap between last meal and sleep onset was associated with 0.39 hours more daytime sleep duration, and higher fat and carbohydrate intake at that last meal was associated with longer sleep onset latency (0.13 hours per gram of fat, 0.02 hours per gram of carbohydrate).

Citation: Nogueira LFR, Pellegrino P, Cipolla-Neto J, Moreno CRC, Marqueze EC. Timing and Composition of Last Meal before Bedtime Affect Sleep Parameters of Night Workers. Clocks & Sleep, 2021. PubMed: 34698111

3. What you eat may matter more than when

The strongest controlled evidence in this whole area is not about timing at all. It is about composition. In a randomised crossover inpatient study, 26 normal-weight adults spent four days on a controlled diet and one day eating whatever they chose, with polysomnography throughout.

Sleep after the self-selected day had less slow wave sleep and a longer onset latency than sleep after the controlled days. Within that day, greater fibre intake predicted less stage 1 sleep and more slow wave sleep. Percentage of energy from saturated fat predicted less slow wave sleep. Percentage of energy from sugar and from non-fibre carbohydrates was associated with more arousals. Low fibre, high saturated fat and high sugar together produced lighter, more interrupted sleep.

Citation: St-Onge MP, Roberts A, Shechter A, Roy Choudhury A. Fiber and Saturated Fat Are Associated with Sleep Arousals and Slow Wave Sleep. Journal of Clinical Sleep Medicine, 2016. PubMed: 26156950

A meta-analysis of 11 articles covering 27 nutrition trials adds structure to this. Compared with high carbohydrate intake, low carbohydrate intake moderately increased both the duration and the proportion of N3 deep sleep (effect sizes 0.37 and 0.51). High carbohydrate intake prolonged REM sleep instead. Carbohydrate quality did not shift sleep stages, but changes in glycaemic load did explain a meaningful share of the variation in sleep efficiency and time awake after sleep onset.

Citation: Vlahoyiannis A, Giannaki CD, Sakkas GK, Aphamis G, Andreou E. A Systematic Review, Meta-Analysis and Meta-Regression on the Effects of Carbohydrates on Sleep. Nutrients, 2021. PubMed: 33919698

4. The four-hour finding that complicates the simple rule

One of the older and more specific studies in this field found that carbohydrates before bed can help, if the gap is right. Twelve healthy men ate isocaloric rice meals of either high or low glycaemic index four hours before their usual bedtime, on separate nights, in random order.

The high glycaemic index meal cut sleep onset latency from 17.5 minutes to 9.0 minutes (p = 0.009). More interesting for our purposes, the same high glycaemic index meal given one hour before bed produced a latency of 14.6 minutes, significantly worse than the same meal at four hours (p = 0.01). No other sleep variables changed.

Citation: Afaghi A, O'Connor H, Chow CM. High-glycemic-index carbohydrate meals shorten sleep onset. The American Journal of Clinical Nutrition, 2007. PubMed: 17284739

Two things follow. Carbohydrates in the evening are not automatically a problem, and the distance between eating and sleeping does real work independent of what was eaten. This is a small trial in 12 young men, so treat the eight-minute figure as a direction rather than a number to plan around.

5. The evidence that cuts the other way

Honest coverage of this topic has to include the study that did not agree. Twenty healthy adults completed a randomised crossover protocol comparing dinner five hours before bedtime against dinner one hour before bedtime, with a fixed sleep opportunity and overnight polysomnography on both visits.

Conventional sleep stages were similar between the two conditions. The late dinner produced a 2.5 percent increase in delta power early in the night, with a matching decrease in combined alpha and beta power, and those effects reversed later in the night. In plain terms, the late dinner group slept slightly more deeply at the start and slightly more lightly toward morning. The authors describe this as postprandial hypersomnia and note that it runs contrary to the existing literature.

Citation: Duan D, Gu C, Polotsky VY, Jun JC, Pham LV. Effects of Dinner Timing on Sleep Stage Distribution and EEG Power Spectrum in Healthy Volunteers. Nature and Science of Sleep, 2021. PubMed: 34017207

Why the disagreement? The most likely explanation is population and measurement. The volunteers were healthy young adults in a lab with a fixed sleep opportunity and a controlled isocaloric meal. The survey and cohort studies capture real evenings, where a late meal usually means a bigger, richer, more improvised meal, often with alcohol, and where the person reporting poor sleep already had a reason to notice it. If you are already sleeping well, a late dinner may cost you very little. If your nights are fragmented, the evidence that it is worth changing is stronger.

6. What this looks like as an actual evening routine

Pulling the studies together, four things are supported well enough to act on.

Aim for a three-hour gap, and treat four as the comfortable version. Three hours is the threshold that shows up in the survey data. Four hours is where the controlled onset-latency work landed. Anything in that range is defensible.

Make the last meal the smaller one. The night-worker data tied fat and carbohydrate load at the last meal to longer onset latency. Size and composition are the levers you control most easily.

Bias dinner toward fibre and away from saturated fat and sugar. This is the single best-controlled finding in the area, and it holds regardless of what time you eat.

Stop counting eating occasions as free. The NHANES analysis found more frequent meal occasions associated with worse sleep quality independent of timing. The 10pm snack counts.

What is not supported: skipping dinner, eating nothing after 6pm as a fixed rule, or treating any carbohydrate in the evening as a mistake. None of those follow from the evidence.

Where Lunia fits

Meal timing is a behavioural lever, and it is one of the better ones because it costs nothing. It also has a ceiling. Fixing your dinner hour does very little for the part of the evening where your nervous system is still running at daytime speed, and that is a different problem with a different literature behind it.

Lunia Restore is built around three ingredients that act on the inhibitory side of that system. Magnesium bisglycinate is the form used in the closest published trial to our own: a randomised, double-blind, placebo-controlled study in 155 adults reporting poor sleep found a greater reduction in Insomnia Severity Index scores than placebo over four weeks, with the effect concentrated in participants whose baseline dietary magnesium was lower. The effect size was small, and we would rather say that than round it up.

Citation: Schuster J, Cycelskij I, Lopresti A, Hahn A. Magnesium Bisglycinate Supplementation in Healthy Adults Reporting Poor Sleep: A Randomized, Placebo-Controlled Trial. Nature and Science of Sleep, 2025. PubMed: 40918053

L-theanine sits alongside it at 300 mg. A 2025 systematic review and meta-analysis of 19 articles covering 897 participants found significant improvements in subjective sleep onset latency, subjective daytime dysfunction and overall subjective sleep quality. The authors are clear that the effects are clearest on subjective measures rather than objective ones, and that most trials used combination products rather than L-theanine alone.

Citation: Bulman A, D'Cunha NM, Marx W, Turner M, McKune A, Naumovski N. The effects of L-theanine consumption on sleep outcomes: A systematic review and meta-analysis. Sleep Medicine Reviews, 2025. PubMed: 40056718

Apigenin, the flavone associated with chamomile's reputation, completes the formula at 50 mg. It is melatonin-free by design, because melatonin is a timing signal rather than a sleep-quality one, and most people taking it do not have a timing problem. Restore is taken 30 to 60 minutes before bed, which happens to sit neatly inside the gap this article argues you should be leaving after dinner anyway.

Learn more about Lunia Restore

Frequently Asked Questions

How long before bed should I stop eating?

Three hours is the threshold that appears most consistently in the research. In a study of 793 young adults, eating within three hours of bedtime was associated with a roughly 60 percent higher odds of waking during the night. Controlled work on sleep onset found four hours worked better than one hour. Anywhere in the three to four hour range is well supported.

Does eating late actually stop you falling asleep?

Generally no. The survey evidence found no significant association between late eating and sleep onset latency or short sleep duration. The effect showed up on night waking instead. One controlled trial even found a high glycaemic index meal four hours before bed shortened time to fall asleep.

Is a small snack before bed a problem?

Size and composition matter more than the fact of eating. Night-shift research linked higher fat and carbohydrate content at the last meal to longer sleep onset latency. NHANES data also associated more frequent eating occasions with worse sleep quality, so a habitual late snack is worth reviewing even if it is small.

What should dinner actually contain for better sleep?

Controlled inpatient research found that higher fibre intake predicted more slow wave sleep and less light stage 1 sleep, while saturated fat predicted less slow wave sleep and sugar was associated with more arousals. Fibre-forward, moderate in saturated fat and low in added sugar is the pattern the evidence supports.

Are carbohydrates at dinner bad for sleep?

Not inherently. A meta-analysis of 27 nutrition trials found lower carbohydrate intake increased deep N3 sleep while higher carbohydrate intake extended REM sleep, which is a trade-off rather than a verdict. Separately, a high glycaemic index carbohydrate meal four hours before bed shortened sleep onset latency in healthy men.

Why do some studies say late dinner is fine?

A randomised crossover trial with overnight polysomnography compared dinner five hours before bed against one hour before bed in 20 healthy volunteers and found no significant adverse change in sleep architecture. Those participants were healthy young adults eating a controlled meal in a lab. Real-world late meals tend to be larger, richer and more variable, which is the likeliest source of the disagreement.

Will changing my dinner time fix broken sleep on its own?

Unlikely on its own. The associations in this literature are real but modest, and meal timing is one input among light exposure, caffeine, alcohol, stress and sleep timing. It is worth adjusting because it is free and easy to test, not because it is the largest lever available.

The Bottom Line

The old advice to stop eating three hours before bed turns out to be roughly right, for a reason nobody stated: it protects the back half of your night rather than helping you fall asleep. The larger and better-controlled finding is that dinner composition, particularly fibre against saturated fat and sugar, shapes sleep depth regardless of the clock. Give yourself a three to four hour gap, make the last meal the lighter one, and spend the attention you were putting on timing on what is actually on the plate.

References

1. Chung N, Bin YS, Cistulli PA, Chow CM. Does the Proximity of Meals to Bedtime Influence the Sleep of Young Adults? A Cross-Sectional Survey of University Students. Int J Environ Res Public Health. 2020;17(8):2677. PubMed: 32295235
2. Yan LM, Li HJ, Fan Q, Xue YD, Wang T. Chronobiological perspectives: Association between meal timing and sleep quality. PLoS One. 2024;19(8):e0308172. PubMed: 39088487
3. Nogueira LFR, Pellegrino P, Cipolla-Neto J, Moreno CRC, Marqueze EC. Timing and Composition of Last Meal before Bedtime Affect Sleep Parameters of Night Workers. Clocks Sleep. 2021;3(4):536-546. PubMed: 34698111
4. St-Onge MP, Roberts A, Shechter A, Roy Choudhury A. Fiber and Saturated Fat Are Associated with Sleep Arousals and Slow Wave Sleep. J Clin Sleep Med. 2016;12(1):19-24. PubMed: 26156950
5. Vlahoyiannis A, Giannaki CD, Sakkas GK, Aphamis G, Andreou E. A Systematic Review, Meta-Analysis and Meta-Regression on the Effects of Carbohydrates on Sleep. Nutrients. 2021;13(4):1283. PubMed: 33919698
6. Afaghi A, O'Connor H, Chow CM. High-glycemic-index carbohydrate meals shorten sleep onset. Am J Clin Nutr. 2007;85(2):426-430. PubMed: 17284739
7. Duan D, Gu C, Polotsky VY, Jun JC, Pham LV. Effects of Dinner Timing on Sleep Stage Distribution and EEG Power Spectrum in Healthy Volunteers. Nat Sci Sleep. 2021;13:601-612. PubMed: 34017207
8. Schuster J, Cycelskij I, Lopresti A, Hahn A. Magnesium Bisglycinate Supplementation in Healthy Adults Reporting Poor Sleep: A Randomized, Placebo-Controlled Trial. Nat Sci Sleep. 2025;17:2027-2040. PubMed: 40918053
9. Bulman A, D'Cunha NM, Marx W, Turner M, McKune A, Naumovski N. The effects of L-theanine consumption on sleep outcomes: A systematic review and meta-analysis. Sleep Med Rev. 2025;81:102076. PubMed: 40056718

These statements have not been evaluated by the Food and Drug Administration. This product is not intended to diagnose, treat, cure, or prevent any disease.