Skip to content
HabitSeek
Back to results

Avoid large meals shortly before bedtime

Finishing the last substantial meal of the day roughly three hours or more before going to bed, rather than eating a large meal close to bedtime.

At a glance

  • Evidence: One small randomised crossover trial found significantly more supine acid reflux after a late meal, but an earlier controlled trial found no difference and a formal review found no efficacy evidence.Nocturnal gastro-oesophageal reflux · Limited certainty · Randomized intervention
  • What’s uncertain: The supporting trial was unblinded and small, and the outcome was oesophageal acid exposure - a physiological surrogate rather than a clinical endpoint such as symptoms or mucosal healing.
  • First step: Aim to finish your last substantial meal about three hours before your usual bedtime.
  • Before you try it: Do not let this become skipped meals or an unintended long fast. All precautions
Read the full takeaway

Reflux and sleep studies

This habit has more randomised support than most meal-timing rules, but the trials are very small and the endpoints are surrogate or short-term. In a randomised unblinded crossover trial that enrolled 32 patients with typical reflux symptoms (30 completed), a standard meal two hours before bed produced significantly more supine acid reflux than the same meal six hours before bed (P = 0.002), with the difference concentrated in people who were overweight or had a hiatal hernia. A separate randomised crossover trial in 13 healthy young women found that dinner one hour before habitual bedtime shortened total sleep time (p = 0.013), reduced sleep efficiency (p = 0.040) and increased wake after sleep onset (p = 0.017) relative to dinner five hours before bed. Against this, an older controlled trial in 20 symptomatic reflux patients found no significant difference in reflux events, acid exposure or sleep-stage percentages between a meal before 19:00 and a late meal at 21:00, and a formal evidence-based review of reflux lifestyle measures concluded there was no published evidence of efficacy for late-evening meal avoidance.

Glucose and clinical outcomes

The glycaemic rationale is real but narrow: in 845 adults, moving a glucose challenge from four hours to one hour before habitual bedtime raised the glucose area under the curve by 8.3% and lowered insulin by 6.7%, with the effect concentrated in MTNR1B G-allele carriers. That is an acute physiological shift, not a demonstrated clinical benefit, and no retrieved trial shows that avoiding late meals prevents disease.

A modest evening meal of salmon, vegetables and potatoesDiscussed byRhonda PatrickView statements ↓
NutritionSleepLong-term healthUnder 10 min · Moderate · Free
Reviewed Sep 23, 2026
Save

Scientific evidence

What the research shows

Outcome assessments based on research, including its uncertainties and limitations.

Nocturnal gastro-oesophageal reflux

Finding: Mixed evidenceCertainty: Limitedstudies disagree, or the effect depends on who is studied

Evidence basis: randomised trials

One small randomised crossover trial found significantly more supine acid reflux after a late meal, but an earlier controlled trial found no difference and a formal review found no efficacy evidence.

Population: Adults with symptomatic gastro-oesophageal reflux disease. The positive crossover trial enrolled 32 and analysed 30 patients (63% male, mean age 46 years, mean BMI 28 kg/m2, esophagitis in 37% and hiatal hernia in 47%). The null trial studied 20 symptomatic reflux patients, with a second experiment in 17 patients of whom 12 were analysed. The evidence-based review covered the published literature from 1975 to 2004.

Research details

Summary

The randomised evidence points both ways and rests on small samples. In a prospective randomised unblinded crossover trial, 32 patients with typical reflux symptoms were randomised to eat a standard meal either 6 h or 2 h before going to bed for two consecutive nights, with acid exposure measured over 48 h by a Bravo wireless pH system; 30 patients completed, and following the late evening meal there was significantly more supine reflux (P = 0.002) than after the early meal, with significantly more supine reflux after the late meal in patients with hiatal hernia and in overweight individuals. An earlier controlled trial in 20 symptomatic reflux patients compared a non-provocative meal before 19:00 with a late evening meal at 21:00 and found no significant differences in the number or duration of reflux events, acid exposure (total %), or polysomnographic measures of per cent of sleep stages between the two nights. In the same paper, ranitidine significantly decreased acid contact time and the mean duration of reflux events after a late provocative meal, with improved subjective discomfort and sleep disturbance, but again no significant polysomnographic differences. An evidence-based review that screened 2,039 studies, identified 100 relevant ones and found only 16 clinical trials of lifestyle measures concluded that head-of-bed elevation and weight loss are the effective lifestyle interventions for reflux disease and that there is no evidence supporting an improvement in reflux measures after dietary interventions. No retrieved trial reports a clinical endpoint such as esophagitis healing, medication use or symptom scores over time.

Why this certainty

Rated Limited because the direction is genuinely inconsistent between two small randomised or controlled crossover trials and a formal evidence-based review, so no pooled direction can be asserted. It is not rated Very limited because at least one randomised crossover trial with objective pH monitoring did find a significant difference in the expected direction. It cannot reach Moderate because all studies are small, unblinded, and use surrogate acid-exposure endpoints rather than clinical outcomes, and no synthesis of randomised reflux-timing trials was retrieved.

Limitations

The supporting trial was unblinded and small, and the outcome was oesophageal acid exposure - a physiological surrogate rather than a clinical endpoint such as symptoms or mucosal healing. The contradicting trial was also small, used a non-provocative rather than standardised provocative meal in its first experiment, and analysed only 12 of 17 patients in its second. Effect sizes were not reported for the main comparison in either trial, only P values, and the retrieved abstracts truncate the subgroup numbers. The evidence-based review concluded that no published evidence of efficacy existed for dietary measures in reflux disease, meaning the intervention most people are actually advised to try has never been properly tested against a control with clinical outcomes.

Effect estimate

Supine oesophageal acid exposure after a meal 2 h versus 6 h before bed in reflux patients: significantly more supine reflux after the late meal (P = 0.002, n = 30 completing a randomised crossover trial) (Piesman 2007). No significant difference in reflux events, acid exposure or sleep-stage percentages after a 21:00 versus pre-19:00 meal (n = 20) (Orr 1998).

Sources for this outcome:

Nocturnal sleep continuity

Finding: Benefit supportedCertainty: Limitedstudies found a benefit from doing this

Evidence basis: randomised trials

In a randomised crossover trial of 13 young women, dinner one hour before bed shortened total sleep time and reduced sleep efficiency compared with dinner five hours before bed.

Population: The randomised crossover trial studied 13 healthy young women aged 21.4 +/- 0.6 years in a real-life setting with overnight EEG and continuous glucose monitoring. The observational study surveyed 793 university students aged 18-29 (26% male).

Research details

Summary

The best available randomised evidence on sleep continuity comes from a single small crossover trial using overnight EEG. Thirteen healthy young women (21.4 +/- 0.6 years) completed two dinner-timing conditions, 1 h and 5 h before habitual bedtime, each over 6 days with dinner providing 709-740 kcal and consistent macronutrient composition across conditions. On Day 4, the late-dinner condition produced significantly shorter total sleep time (p = 0.013), reduced sleep efficiency (p = 0.040), and significantly higher wake after sleep onset (p = 0.017), Arousal Index (p = 0.041), number of stage-shifts (p = 0.016) and Stage-Shift Index (p = 0.003). Observational data point the same way but are weak: in a cross-sectional survey of 793 university students aged 18-29, eating within 3 h of bedtime was positively associated with nocturnal awakening (OR 1.61, 95% CI 1.15-2.27), but not with long sleep onset latency (1.24; 0.89-1.73) or short sleep duration (0.79; 0.49-1.26), and after adjustment for ethnicity and body mass index the association for nocturnal awakening attenuated to OR 1.43 (95% CI 1.00-2.04), which is borderline. Notably, the same randomised trial found no significant change in glucose tolerance: the oral glucose tolerance test after waking showed no significant changes, and postprandial glucose showed only transient changes, with no significant change in the glucose tolerance test after waking.

Why this certainty

Rated Limited rather than Moderate because the randomised evidence is a single 13-participant crossover trial in healthy young women without effect sizes, and the supporting observational estimate attenuates to borderline after adjustment. It is not rated Very limited because the randomised design used objective overnight EEG and found a consistent pattern across several sleep-continuity measures including total sleep time, sleep efficiency, wake after sleep onset and arousal index.

Limitations

The randomised trial is very small (n = 13), enrolled only healthy young women, and ran for six days per condition, so it cannot speak to men, older adults, people with insomnia, or longer-term effects, and it reported P values without effect sizes or confidence intervals for the sleep outcomes. The observational study was cross-sectional, used self-reported sleep and self-reported meal timing, and its adjusted estimate crossed 1.0, so the association is compatible with no effect; sleep onset latency and sleep duration were both null in the same analysis. Neither source measured a clinical sleep outcome such as insomnia diagnosis, and no retrieved trial tested whether changing dinner timing improves sleep in people who already have insomnia. The mechanistic trial of 845 participants measured glucose and insulin responses to timed oral glucose tolerance tests rather than sleep, so it supports the melatonin explanation only by analogy and not as evidence about sleep. The counterweight quoted in this entry's precautions comes from a cross-sectional study of 7,701 rural Chinese adults in which each hour of later dinner timing was associated with 0.482 mmol/L lower fasting insulin and a 0.122 lower HOMA-IR index, which is a single adjusted cross-sectional association and cannot be read as a metabolic benefit of late eating (Zhang 2023). The diabetes-team advice in this entry's precautions rests on a cross-sectional study of 189 adults with type 1 diabetes on automated or non-automated insulin delivery, in which poor sleepers had higher time below range and higher glucose variability and dinner time was a main predictor of HbA1c; it is observational and does not test a timing change (Corrado 2025).

Effect estimate

Dinner 1 h versus 5 h before habitual bedtime, overnight EEG on Day 4, n = 13 healthy young women: shorter total sleep time (p = 0.013), reduced sleep efficiency (p = 0.040), higher wake after sleep onset (p = 0.017) (Enomoto 2026). Eating within 3 h of bedtime and nocturnal awakening in 793 students: OR 1.61 (95% CI 1.15-2.27), adjusted OR 1.43 (95% CI 1.00-2.04) (Chung 2020).

Sources for this outcome:

What people say

Who discusses this?

Public statements, with positions specific to each source and topic. These views do not determine the scientific evidence assessment.

What the research shows ↑

Practical steps and precautions

How to try it

  1. Aim to finish your last substantial meal about three hours before your usual bedtime. The three-hour figure is the threshold used in the largest observational study retrieved (eating within 3 h of bedtime versus more than 3 h before), and the randomised crossover trials compared roughly 2 versus 6 hours and 1 versus 5 hours; no trial established an optimal interval.
  2. Shift the meal earlier rather than shrinking it. The randomised trials matched the meal between conditions (709-740 kcal with consistent macronutrient composition in one trial), so the comparison tested timing rather than size or content, and the evidence does not show that skipping the meal is better than moving it.
  3. If an evening meal close to bedtime is unavoidable, keep it smaller and lower in fat, and remember that in the reflux trials the late meal was specifically a large standardised meal rather than a light snack.
  4. If reflux is the reason you are doing this, treat it as one part of management alongside the measures that carry better evidence, notably weight loss and elevation of the head of the bed.

Precautions

  • Do not let this become skipped meals or an unintended long fast. In one randomised crossover trial the late-dinner condition produced transiently lower postprandial glucose rather than higher, and a large cross-sectional study in 7,701 rural Chinese adults found delayed dinner timing was associated with lower fasting insulin, so the metabolic direction is not uniformly unfavourable to late eating.
  • If you take insulin or a sulfonylurea, changing meal timing alters hypoglycaemia risk. In adults with type 1 diabetes using advanced insulin delivery systems, time below range and glycaemic variability were higher in poor sleepers, and dinner timing was reported as a key factor affecting both sleep quality and glycaemic control; discuss timing changes with your diabetes team.
  • Involuntary weight loss, early satiety, difficulty swallowing, or worsening reflux despite these measures are not lifestyle problems; seek medical assessment.
  • People doing shift work cannot align meals to a conventional bedtime, and no retrieved trial tested late-meal avoidance in shift workers; the evidence base does not address this group.

References

Nocturnal reflux episodes following the administration of a standardized meal. Does timing matter?

Piesman M, Hwang I, Maydonovitch C, Wong RK · The American journal of gastroenterology · 2007

Prospective randomised unblinded crossover trialVerified from abstract
Open original source

Sleep-related gastro-oesophageal reflux: provocation with a late evening meal and treatment with acid suppression.

Orr WC, Harnish MJ · Alimentary pharmacology & therapeutics · 1998

Controlled clinical trial with two experiments using pH and polysomnographic monitoringVerified from abstract
Open original source

Are lifestyle measures effective in patients with gastroesophageal reflux disease? An evidence-based approach.

Kaltenbach T, Crockett S, Gerson LB · Archives of internal medicine · 2006

Evidence-based systematic review with formal grading of lifestyle measures for GERDVerified from abstract
Open original source

Effects of later dinner timing on subsequent metabolic function and nocturnal sleep in healthy young women.

Enomoto M, Kitamura S, Kunieda T, Eto T · Journal of physiological anthropology · 2026

Randomised crossover trial with overnight EEG and continuous glucose monitoringVerified from full text
Open original source

Does the Proximity of Meals to Bedtime Influence the Sleep of Young Adults? A Cross-Sectional Survey of University Students.

Chung N, Bin YS, Cistulli PA, Chow CM · International journal of environmental research and public health · 2020

Cross-sectional survey with logistic regressionVerified from full text
Open original source

Interplay of Dinner Timing and MTNR1B Type 2 Diabetes Risk Variant on Glucose Tolerance and Insulin Secretion: A Randomized Crossover Trial.

Garaulet M, Lopez-Minguez J, Dashti HS, Vetter C, Hernández-Martínez AM, Pérez-Ayala M, Baraza JC, Wang W, Florez JC, Scheer FAJL, Saxena R · Diabetes care · 2022

Randomised crossover trial with two evening oral glucose tolerance tests per participantVerified from full text
Open original source

Association between dinner timing and glucose metabolism in rural China: A large-scale cross-sectional study.

Zhang K, Feng Y, Chai Y, Wang C, Yu S · Nutrition (Burbank, Los Angeles County, Calif.) · 2023

Large cross-sectional study within a rural cohortVerified from abstract
Open original source

Interplay among sleep quality, dinner timing, and blood glucose control in users of advanced technologies: A study in a cohort of adults with type 1 diabetes.

Corrado A, Scidà G, Abuqwider J, Annuzzi E, Giosuè A, Pisano F, Annuzzi G, Bozzetto L · Diabetes research and clinical practice · 2025

Cross-sectional cohort study in adults with type 1 diabetesVerified from abstract
Open original source

Community

Discussion

Have you tried this? What helped or made it difficult?

Reader experiences are separate from the scientific evidence. Explore all discussions

No discussions yet.

Be the first to ask a question or share your experience.

Sign in to vote or join the discussion.

Related habits

Explore other habits in a shared category. Each has its own evidence and precautions.

Editorial history and review method

Revision 3 · Approved for publication · Codex (publication approval explicitly authorized by site owner; category migration)

Prepared and checked with AI. Publication approval does not imply independent expert review. Request a correction.

Applied the group-2 audit. The verdict no longer says the Piesman trial had 30 patients (32 enrolled, 30 completed) and no longer lists esophagitis as a subgroup specific to the late meal. The two uncited precaution sentences were sourced rather than deleted: Zhang 2023 (7,701 rural Chinese adults, cross-sectional) and Corrado 2025 (adults with type 1 diabetes on automated or non-automated insulin delivery) were retrieved and both are now cited from nocturnal-sleep-continuity, which already carries the glucose side of this outcome; the metabolic counterweight and the diabetes-team advice are now both backed. The sleep-continuity summary no longer claims postprandial glucose iAUC was lower on late-dinner days, which the retrieved abstract does not support; it now says glucose showed only transient changes with no significant change on the post-breakfast glucose tolerance test. The card summary no longer reads as endorsement of a Mixed-evidence outcome. Second reviewer must re-check: (a) the direction of the Enomoto glucose finding, which could not be resolved because Europe PMC full text returned HTTP 500 and the publisher redirects to an identity provider; (b) whether the card should lead with nocturnal-sleep-continuity (Benefit supported) instead of the Mixed reflux outcome; (c) that the glucose and type 1 diabetes sources are acceptable inside a sleep outcome, since the schema gives this Standard-coverage habit only two outcomes. Publication release 2026-10-04: site owner authorized a bounded AI audit and release of this exact revision. See docs/release-audit-2026-10-04.md for source-check findings and limits. No independent health-professional or peer review is claimed. Category migration 2026-10-04: removed the separate goals field; categories: Nutrition, Sleep → Nutrition, Sleep, Long-term health. Research assessments, sources, guidance, precautions, and evidence review dates are unchanged. This revision awaits publication approval; no new evidence review is claimed. Publication approval 2026-10-04: site owner explicitly accepted all current category-migration revisions for deployment and requested that remaining limitations be retained in docs/content-review-backlog.md. Approval applies to this exact content revision. No new evidence review, independent expert review, or clinical validation is claimed; existing precautions and disclosures remain in effect.