Story
Meal timing · when matters as much as what
Last updated
In one pass What decides your weight and blood sugar is, first and always, how much you eat in a day and what you eat.
Educational content, not medical advice — consult a clinician.
Story path
Chapter 1
How much timing matters
What decides your weight and blood sugar is, first and always, how much you eat in a day and what you eat. When you eat comes after that: it is a nearly free small lever, not a magic switch.
The calories you eat are like water poured into a tub. Timing only changes the rhythm of the pouring; the total is still set by how much you pour in. Moving the same meals earlier and not eating too late lets you pick up a small extra benefit by working with the body clock (a review by Panda, 2016), but it cannot outrun eating too much.
So get amount and quality under control first. Once they hold steady, eating earlier and cutting back on late-night eating are the icing. Do not lose sight of the total for the sake of watching the clock. The same meal really is handled differently in the morning and at night, because the enzymes and hormones that run digestion and metabolism themselves follow a 24-hour clock.
The calories you eat are like water poured into a tub. Timing only changes the rhythm of the pouring; the total is still set by how much you pour in. Moving the same meals earlier and not eating too late lets you pick up a small extra benefit by working with the body clock (a review by Panda, 2016), but it cannot outrun eating too much.
So get amount and quality under control first. Once they hold steady, eating earlier and cutting back on late-night eating are the icing. Do not lose sight of the total for the sake of watching the clock. The same meal really is handled differently in the morning and at night, because the enzymes and hormones that run digestion and metabolism themselves follow a 24-hour clock.
In practice · Why amount and quality come first
Put simply, the calories you eat are like water poured into a tub: meal timing only sets the rhythm of the pouring, and the total in the tub is still decided by how much you pour in. For the same 2000 kcal day, eating most of it at night is indeed slightly worse for metabolism than spreading it across breakfast and lunch (the observational study by Garaulet, 2013, points that way). But that gap is far smaller than the gap between a 2000-kcal day and a 3000-kcal day. So the order is: amount, quality, and only then the clock.Why talk about meal timing at all, then? Because it is nearly free. You do not have to eat less or change foods; you just move the same meals earlier and avoid eating too late, and you pick up a small extra benefit by working with your body clock (Panda 2016). Free is not the same as big. Treat it as your only weight-loss tactic while ignoring the total, and it will most likely not work.
The practical order is therefore plain. First get how much you eat and what you eat to hold steady; while those two are still drifting, watching the clock is pointless. Once the big levers are steady, put most of your calories in daytime and make late-night eating occasional. Don't get it backwards.
Chapter 2
Metabolism is circadian
The same meal gets a different response from the body at breakfast and at dinner. Almost all of the enzymes, hormones, and transporters that digest and process food follow a 24-hour body clock (a review by Panda, 2016). By day they are set to work; at night they are set to rest. So when you eat changes what the same food does once it is inside you.
The most direct example is insulin sensitivity — how readily cells respond to insulin. It is high in the morning and low in the evening, so the same sugar eaten at night drives post-meal blood sugar higher and clears more slowly.
This does not mean the calories were miscounted, or that calories do not matter. The total is still the lead; timing only changes how efficiently the body handles the same calories.
The most direct example is insulin sensitivity — how readily cells respond to insulin. It is high in the morning and low in the evening, so the same sugar eaten at night drives post-meal blood sugar higher and clears more slowly.
This does not mean the calories were miscounted, or that calories do not matter. The total is still the lead; timing only changes how efficiently the body handles the same calories.
Mechanism · The same sugar, morning versus evening
Several metabolic patterns follow the day-night rhythm, and together they explain why the same meal turns into two different events (a review by Panda, 2016).Insulin sensitivity is higher in the morning than at noon, and higher at noon than in the evening. The same sugar eaten at breakfast gives a clearly lower post-meal blood sugar than when eaten at dinner. Fat tissue leans toward breaking fat down by day and toward storing it at night. The liver builds glycogen more actively during the day. The satiety hormone (glucagon-like peptide-1, which the gut releases after eating) responds more to breakfast than to dinner. Core body temperature peaks in the afternoon, falls in the evening, and is lowest in the early morning. Cortisol is highest around waking (the cortisol awakening response, CAR), helping to start up blood sugar and metabolism.
This is not a list of unrelated facts. By day, that set of enzymes and hormones tunes cells to process food; at night, it tunes them to repair and sleep. Send a load of sugar into a machine already set to rest, and the blood-sugar curve runs higher and comes down more slowly. That is the same load at two different clock times, not a difference in how much you ate.
Evidence · Late lunch eaters lost less weight
Garaulet 2013 (published in the International Journal of Obesity) followed a roughly 20-week Mediterranean-diet weight-loss program in Spain, with about 420 people. The main meal there is lunch. The researchers split lunch time at the group's median, 15:00. People who ate lunch after 15:00 lost 7.7 kg on average; those who ate before lost 9.9 kg. Both groups ate similar total calories, but the late eaters skipped breakfast more often.This was an observational cohort analysis, not a trial that randomly assigned people to eat early or late. It cannot prove that calories do not matter. What it can say is this: with similar intake, stacking the main meal later came with a smaller weight loss, and the late eaters also went without breakfast more often. The total is still the lead; timing changes how efficiently the same calories are handled, plus a bundle of habits that travel with eating late (skipping breakfast, eating at night).
So the mechanisms for why late eating costs more should be read together with these numbers — not as if 7.7 versus 9.9 meant "calories don't count."
Mechanism · Why eating late costs more
In the evening, insulin sensitivity falls, so the same sugar load raises blood sugar higher. The gut moves more slowly in the evening, the stomach empties more slowly, and the risk of reflux rises. Energy expenditure falls in the evening, and basal metabolism is lower late at night. Eating dinner too close to bedtime may also disturb falling asleep and sleep quality. A short gap between dinner and the next breakfast squeezes the overnight fast (the stretch when metabolism recovers).In evolutionary terms, human ancestors foraged by day and fasted at night. Electric light plus food available around the clock makes it easy for metabolic rhythm to drift away from the body clock. Shift work, crossing time zones, and habitual late eating create social jetlag — a schedule pulled out of line with the body's clock by work or social life — which population studies have linked with obesity, diabetes, and cardiovascular risk. That is not a scolding of nightlife; it is that once the clocks are wrenched apart, the same food costs more. How shift workers might eat is covered in the chapter on limits and shift work.
Chapter 3
How the body clock handles sugar
What sets what the body should do at each hour is a set of clock genes found in almost every cell. They cycle once every 24 hours, and the liver, muscle, fat, and pancreas all work to their beat (a review by Panda, 2016).
For eating, two things matter most. In the morning the body is more sensitive to insulin, so the same sugar raises blood sugar gently. After dark, the brain releases melatonin to make you sleepy, and melatonin also holds down the pancreas's release of insulin. A big meal late at night means sugar coming in while insulin is being held down, so blood sugar is harder to bring back down. A crossover trial that had people eat the same dinner 1 hour or 4 hours before bed found exactly that: glucose tolerance was worse in the first case.
So keep sugary, heavy meals in the daytime where you can.
For eating, two things matter most. In the morning the body is more sensitive to insulin, so the same sugar raises blood sugar gently. After dark, the brain releases melatonin to make you sleepy, and melatonin also holds down the pancreas's release of insulin. A big meal late at night means sugar coming in while insulin is being held down, so blood sugar is harder to bring back down. A crossover trial that had people eat the same dinner 1 hour or 4 hours before bed found exactly that: glucose tolerance was worse in the first case.
So keep sugary, heavy meals in the daytime where you can.
Mechanism · What the clock genes set
One layer deeper: what governs what your body should do at each hour is a set of switches called clock genes, chiefly CLOCK, BMAL1, PER, and CRY. They sit in almost every cell, cycle once every 24 hours, and set the working rhythm of the liver, muscle, fat, and pancreas across the day (Panda 2016). A master clock in the brain follows light and keeps the time; the clocks in these outlying organs also listen to signals from when you eat. When the two sets of signals line up, food is handled smoothly.For eating, insulin works better in the morning. Around waking, insulin sensitivity is high, so the same sugar eaten at breakfast raises blood sugar gently; by evening, sensitivity has fallen, and the same sugar pushes blood sugar higher and clears more slowly. This is the same chain as the morning-evening blood-sugar gap in the earlier chapter on circadian metabolism; here the switches simply get their names.
Evidence · Melatonin holds insulin down at night
After dark, the brain releases melatonin to make you sleepy, and melatonin suppresses the pancreas's release of insulin. Eating a big meal at that point, especially a sugary one, is like pressing the accelerator (sugar coming in) and the brake (melatonin holding insulin down) at the same time, so blood sugar is harder to bring down.Lopez-Minguez 2018 was a randomized crossover trial. The same people ate the same dinner once 1 hour before bed (when melatonin is already high) and once 4 hours before bed. The dinner 1 hour before bed produced clearly worse glucose tolerance, most strongly in people carrying the MTNR1B risk genotype (a common variant of a melatonin-receptor gene). So "don't feast late at night" is not folklore; the explanation this trial supports is that melatonin and insulin pull against each other at night.
Put sugary, high-carbohydrate, heavy meals in the daytime, when insulin works well. Once it is dark and you are getting sleepy, the body has already shifted into getting ready to rest, and forcing down a big meal then carries the highest blood-sugar cost.
Chapter 4
Eating within a set window
Time-restricted eating () means fitting each day's eating into a fixed window, without deliberately cutting calories or types of food. It is not fasting in the medical sense, and it is not homework everyone must do. A fixed window matters more than how narrow it is.
In one small randomized crossover trial, men with prediabetes moved their eating into the early part of the day, with food matched to keep their weight steady, and their insulin sensitivity and blood pressure still improved. That suggests timing itself has an effect that does not run through weight loss — but it is not magic, and it is not something everyone has to follow.
There is no need to chase a 16-hour fast from day one. People who are pregnant, have a history of eating disorders, or have type 1 diabetes are poor fits for a strict window; anyone on glucose-lowering drugs should adjust them with a doctor first.
In one small randomized crossover trial, men with prediabetes moved their eating into the early part of the day, with food matched to keep their weight steady, and their insulin sensitivity and blood pressure still improved. That suggests timing itself has an effect that does not run through weight loss — but it is not magic, and it is not something everyone has to follow.
There is no need to chase a 16-hour fast from day one. People who are pregnant, have a history of eating disorders, or have type 1 diabetes are poor fits for a strict window; anyone on glucose-lowering drugs should adjust them with a doctor first.
Evidence · An earlier window at matched calories
First, separate it from traditional intermittent fasting. Time-restricted eating () is one fixed window every day (commonly 12, 10, or 8 hours of eating) and mainly works with the day-night rhythm; 5:2 or alternate-day fasting cuts calories sharply on selected days and mainly works by eating less. The evidence for TRE is newer; for long-term weight loss, intermittent fasting does about as well as ordinary calorie restriction.Sutton 2018 was a supervised, controlled-feeding randomized crossover trial. Men with prediabetes rotated through two schedules, 5 weeks each. One was early TRE, a 6-hour eating window with dinner finished before 3 p.m. (15:00); the other was a 12-hour window. On both schedules they were given enough food to keep their weight stable. During the early window, insulin sensitivity and beta-cell responsiveness improved, blood pressure and oxidative stress fell, and weight did not change. The implication is that timing itself has a metabolic effect that does not go through weight loss. The sample was small and the physiological signal clean, but it is not a prescription that everyone must run a 6-hour window, and it did not compare an early window with a late window of the same length.
Evidence · A 10-hour window, no food rules
Wilkinson 2020 enrolled 19 patients with metabolic syndrome whose daily eating had spanned 14 hours or more, most of them on a statin or blood-pressure drugs. They ate within a self-chosen, fixed window of about 10 hours a day (for example, roughly 9 a.m. to 7 p.m.), with no rules on what to eat, for 12 weeks (n = 19). It was a single-arm trial with no control group, comparing people with themselves before and after. Weight fell by about 3% on average, and visceral fat, (low-density lipoprotein cholesterol), non-HDL cholesterol (total cholesterol minus , the high-density lipoprotein kind), and blood pressure all fell; in the subgroup whose blood sugar was high at the start, (a marker of average blood sugar over the past few months) fell slightly. Participants also ate about 8-10% less without being asked to, because a tighter window leaves fewer chances to snack.Here the effect of timing and the effect of eating less are stacked together, which is not the same question as Sutton's matched-calorie, no-weight-loss design. On the logic of , a window that stays fixed every day matters more than where it sits; by the mechanism of the day-night rhythm, earlier should fit the clock better, but no trial has yet compared an early and a late window of the same length directly.
In practice · A 10–12 hour window is enough
The thing to remember: open your window early, close it early, and keep it fixed at around 10-12 hours — that is the least fuss for most people. There is no need to chase a strict 16-hour fast from day one, and is not homework you must do.The easiest start is a 12-hour window such as 8 a.m. to 8 p.m., which most people already come close to. For a more noticeable effect, you can tighten it to 10 hours, for example 9 to 7. Skipping breakfast and eating from midday into the evening is also a form of TRE, but by the mechanism of the body clock it usually fits the clock less well than finishing early. A 6-hour window is advanced; clinical evidence is still building, and it does not suit most people as a long-term routine.
Keep the safety limits firmly in mind: people who are pregnant or breastfeeding, children under 13, people who are underweight, anyone with a history of eating disorders, and people with type 1 diabetes are poor fits for a strict window. People with type 2 diabetes who want to try it must first work with a doctor to adjust glucose-lowering drugs (especially sulfonylureas and insulin) to avoid low blood sugar. On training days, keep your workout inside the eating window where you can. Gritting your teeth through hunger is not the rule; build up gradually.
Myth · Must you eat breakfast?
The key to is opening early and closing early, not eating breakfast no matter what. "Breakfast is the most important meal of the day" is closer to a popular advertising slogan than to a research finding. Sievert 2019 pooled randomized trials in adults of eating versus skipping breakfast; it looked only at body weight and total daily calories and measured no metabolic marker at all. People assigned to eat breakfast did not end up lighter (they were slightly heavier), and they ate more calories over the day; the trials were all short and mostly of low quality. But skipping breakfast and then eating heavily at night piles the calories into the hours when the body is getting ready to rest, which is worse for metabolism.So do not upgrade TRE into "you must eat breakfast," and do not treat skipping breakfast as a license to load up at night. A fixed window that closes as early as you reasonably can matters more than what any one meal is called.
Chapter 5
Easy meal-timing habits
You can use meal timing without strict time-restricted eating (). Only three small things are truly workable: put most of your calories at breakfast and lunch, leave a few hours between dinner and sleep, and keep your daily eating window roughly fixed. Once you do these three, fine-tuning the exact minutes adds very little.
Do not shift the whole window on weekends. A late-night snack now and then is fine; just do not make it a nightly habit.
If you want to lose weight, steady your blood sugar, train in the evening, or you are a born night owl, the details differ a little, but none of them gets around these three things.
Do not shift the whole window on weekends. A late-night snack now and then is fine; just do not make it a nightly habit.
If you want to lose weight, steady your blood sugar, train in the evening, or you are a born night owl, the details differ a little, but none of them gets around these three things.
In practice · Five rules that need no strict TRE
Five practical principles that work even without strict time-restricted eating (). They are common practical advice, not prescriptions each tested in a trial.Move the main meals earlier: put most of your calories at breakfast and lunch, and keep dinner fairly light. For example, about 30% at breakfast, about 40% at lunch, and about 30% at dinner, rather than half or more at dinner. You do not need to count grams; a rough split is enough.
Finish dinner by 7 p.m. where you can, at least 2–3 hours before sleep. This is the core of the relaxed version of meal timing. If you go to bed at 9 p.m., eating dinner at 6 p.m. leaves time for digestion and for insulin to settle.
An eating window of around 10-12 hours: breakfast at 7-8 a.m. and dinner at about 7 p.m. naturally make a 12-13 hour window. That is enough for most people; there is no need to force 16:8, and there is no rule that you must do TRE.
Make late-night eating occasional, not a habit. A late dinner out or a late meal after overtime now and then is normal; eating at 10 p.m. as a habit lets the metabolic burden add up. If you are hungry, you can go to bed earlier, or drink some water or a caffeine-free drink.
Keep weekends close to weekdays. Social jetlag is staying up and sleeping in on the weekend, then being yanked back on Monday — a steady, mild jetlag. One rule you can set yourself: keep the weekend eating window within about 1 hour of the weekday one.
In practice · Weight, glucose, evening training, shifts
To lose weight: moving the main meals earlier, plus a 10-12 hour window, plus less late-night eating, is often easier to keep up than counting calories alone. To steady blood sugar: eat as much as you can earlier in the day, and go easy on sugary food after 7 p.m. To train at 6-8 p.m.: have a small meal 1-2 hours before, and a proper meal within 1 hour after (which may run as late as 9 p.m.; fine now and then, but not recommended every day). For shift work: meal times will rarely be ideal; eat your main meals during your alert hours, when you need energy, and avoid the groggy stretches.Several related topics have their own pages: the metabolic side of shift work (Shift Work), trial details on fasting and time-restricted eating (Intermittent Fasting & Time-Restricted Eating), and the ultra-processed foods that often make up late-night snacks (Ultra-processed Foods (UPF)).
Myth · Small frequent meals, and night owls
Small, frequent meals raise your metabolic rate is a myth. With the same total calories, the metabolic rate is the same (a review by Bellisle, 1997). Small, frequent meals may help with acid reflux, gastroparesis, or a muscle-building phase, but a meal every 4-5 hours is usually enough. For most people, three meals a day with no snacks in between is simpler and more easily forms an eating window on its own. Grazing all day plus big meals leaves the metabolic rhythm at its most chaotic.Chronotype — whether you are naturally an early riser or a late sleeper — is partly genetic. Morning types wake and eat early by nature, so an early window is easy for them. Evening types forced into a very early finish can find it miserable; a window that sits a little later overall suits them better. The key is still to keep it fixed and to finish as early as you can relative to your own bedtime. Do not force a night owl to become a lark: it may work in the short run, but it tends to rebound in the long run.
16:8 or 5:2 are mainly tools for cutting calories, and in the long run they work about as well as ordinary calorie restriction; is mainly a tool for working with the body clock, and part of its benefit does not depend on weight loss. They are not mutually exclusive and can be combined, but neither is required.
Chapter 6
Limits, and shift workers
Meal timing is not a cure-all. No timing, however perfect, outruns eating too much.
The workable part is still those three small things: most calories at breakfast and lunch, a few hours between dinner and sleep, and a roughly fixed window. Eating late now and then is normal. People with a history of eating disorders especially should not police themselves with a strict time window.
Shift work is a hard case of its own. People are forced to eat during the body clock's rest phase, and in population studies shift work is linked with a somewhat higher risk of diabetes; this is not a lack of discipline. This page is health education and does not replace individualized medical advice.
The workable part is still those three small things: most calories at breakfast and lunch, a few hours between dinner and sleep, and a roughly fixed window. Eating late now and then is normal. People with a history of eating disorders especially should not police themselves with a strict time window.
Shift work is a hard case of its own. People are forced to eat during the body clock's rest phase, and in population studies shift work is linked with a somewhat higher risk of diabetes; this is not a lack of discipline. This page is health education and does not replace individualized medical advice.
In practice · Don't let the clock become stress
Total energy is still the lead. No timing, however perfect, outruns eating too much. Treat moving dinner earlier as your only weight-loss tactic while ignoring the total, and it will most likely not work.Only three small things are truly workable: put most of your calories at breakfast and lunch, finish dinner at least 2–3 hours before sleep, and keep your eating window roughly fixed at 10-12 hours. Do these three and you get nearly everything meal timing can offer; fine-tuning beyond that adds little. is not required; 10-12 hours is a relaxed window that is already enough, not one more assignment.
Do not let clock-watching turn eating into stress. A dinner out or a late meal after overtime now and then is normal, and flexibility over the long run matters more than one perfect day. People with a history of eating disorders especially should not police themselves with a strict time window. The body clock is a helper, not a supervisor (Panda 2016).
Evidence · Shift work and diabetes risk
People who work night shifts are often forced to eat during the body clock's rest phase. Gan 2015 pooled observational studies of shift work and diabetes in a : the pooled (OR, a statistic for how much higher one group's risk is relative to another's) for diabetes in shift workers was about 1.09 — a slightly higher risk. This is an association and cannot on its own show that shift work causes diabetes; one plausible explanation is that the body's rhythm and the working hours do not line up, rather than that these people lack discipline.The realistic advice: eat your main meals during your alert hours, when you need energy; keep eating during night shifts to a minimum and on the light side; and do not eat a big meal right before collapsing into bed after a shift. The usual "eat early, finish early" advice has to be discounted for shift work, and how to arrange things is best worked out with a doctor. This page is health education and does not replace individualized medical advice.
For most people whose hours are fairly regular, meal timing is a small lever that adds a little on top — worth doing along the way, but do not mythologize it. Get amount and quality right first, then think about timing.
References · 8
- Panda, S. (2016). Circadian physiology of metabolism. Science, 354(6315), 1008-1015. 10.1126/science.aah4967
- Garaulet, M., Gómez-Abellán, P., Alburquerque-Béjar, J. J., Lee, Y. C., Ordovás, J. M., & Scheer, F. A. J. L. (2013). Timing of food intake predicts weight loss effectiveness. International Journal of Obesity, 37(4), 604-611. 20-week Mediterranean weight-loss programme, n≈420. Early vs late lunch split at the population median (15:00), not 13:30. Late lunch eaters lost 7.7 vs 9.9 kg (P=0.008); weekly rate 360 vs 454 g. Energy intake similar; late eaters skipped breakfast more often. 10.1038/ijo.2012.229
- Lopez-Minguez, J., Saxena, R., Bandín, C., Scheer, F. A. J. L., & Garaulet, M. (2018). Late dinner impairs glucose tolerance in MTNR1B risk allele carriers: a randomized, cross-over study. Clinical Nutrition, 37(4), 1133-1140. Eating dinner ~1 h before bed (high endogenous melatonin) vs ~4 h before worsened glucose tolerance, especially in MTNR1B risk-allele carriers — supports a causal melatonin-glucose interaction at night. pubmed.ncbi.nlm.nih.gov/28455106
- Sutton, E. F., Beyl, R., Early, K. S., Cefalu, W. T., Ravussin, E., & Peterson, C. M. (2018). Early time-restricted feeding improves insulin sensitivity, blood pressure, and oxidative stress even without weight loss in men with prediabetes. Cell Metabolism, 27(6), 1212-1221. Supervised controlled-feeding randomized crossover in men with prediabetes: 6-h early window (dinner before 3 p.m.) vs a 12-h window, 5 weeks each, fed enough to keep weight stable. eTRF improved insulin sensitivity, beta-cell responsiveness, blood pressure, oxidative stress and appetite. It did not compare early vs late windows of equal length. 10.1016/j.cmet.2018.04.010
- Wilkinson, M. J., Manoogian, E. N. C., Zadourian, A., Lo, H., Fakhouri, S., Shoghi, A., et al. (2020). Ten-hour time-restricted eating reduces weight, blood pressure, and atherogenic lipids in patients with metabolic syndrome. Cell Metabolism, 31(1), 92-104. Single-arm, paired-sample trial, not randomised and without a control group: 19 people with metabolic syndrome whose baseline eating window was 14 h or more, most on statins and/or antihypertensives, ate within a self-selected 10-h window for 12 weeks. Body weight -3.3 kg (-3%), waist -4.5 cm, systolic/diastolic blood pressure -5.1/-6.5 mmHg, LDL-C -11.9 mg/dL (-11%), non-HDL-C -9%; glucose, insulin, triglycerides and hs-CRP did not change significantly in the whole group; HbA1c fell (-0.22%) only in the 12 participants with elevated baseline glucose or HbA1c (abstract, PMID 31813824; full text, PMC6953486). 10.1016/j.cmet.2019.11.004
- Sievert, K., Hussain, S. M., Page, M. J., Wang, Y., Hughes, H. J., Malek, M., et al. (2019). Effect of breakfast on weight and energy intake: systematic review and meta-analysis of randomised controlled trials. BMJ, l42. 13 RCTs; pooled only body weight and daily energy intake, no metabolic markers. Breakfast groups gained slightly more weight (mean difference 0.44 kg, 0.07-0.82) and ate more energy per day (260 kcal, 79-441) than skippers; all trials were at high or unclear risk of bias with short follow-up (about 7 weeks for weight, 2 for energy), so the authors ask for caution (abstract, PMID 30700403). 10.1136/bmj.l42
- Bellisle, F., McDevitt, R., & Prentice, A. M. (1997). Meal frequency and energy balance. British Journal of Nutrition, 77(S1), S57-S70. Classic review showing no metabolic-rate advantage of small frequent meals over fewer larger meals at matched calories. 10.1079/BJN19970104
- Gan, Y., Yang, C., Tong, X., Sun, H., Cong, Y., Yin, X., et al. (2015). Shift work and diabetes mellitus: a meta-analysis of observational studies. Occupational and Environmental Medicine, 72(1), 72-78. Pooled OR for diabetes with shift work approximately 1.09 (95% CI 1.05-1.12). 10.1136/oemed-2014-102150