Aller au contenu
Nutrition et métabolismeMéta-analyse d’essais9 min read

Early time-restricted eating linked to more weight loss and lower insulin than late eating

Collectif éditorial MoveVitals · Study published · Reading updated

Original paper:Effects of timing and eating duration of time restricted eating on metabolic outcomes: systematic review and network meta-analysis · BMJ Medicine · 2026 · Article original ↗

À retenir

Pooling 41 randomised trials (2287 participants, median intervention 8 weeks), time-restricted eating was associated with 2.15 kg lower body weight than usual diets (95% CI -2.56 to -1.74), and eating early beat eating late by 1.15 kg (95% CI -1.86 to -0.45). Most trials were short, only 6 study arms tested early eating, and the outcomes were body and blood measures, not heart attacks or deaths.

Study design

Méta-analyse d’essais

Reading basis

Full text

Reading focus

Design & applicability

Cette explication éditoriale existe en anglais, chinois simplifié et chinois traditionnel. Cette interface affiche la lecture en anglais.

Question de recherche

Cardiovascular disease is a leading cause of death, and modifiable metabolic risk factors such as high body mass index, high blood pressure and abnormal glucose and lipid levels account for a large share of it. Time-restricted eating confines food to a daily window (typically under 8 to 12 hours) without requiring people to cut total energy intake. Earlier research mostly compared it with usual diets or continuous energy restriction, so the questions of when in the day the window falls and how long it lasts had not been studied thoroughly, and the few direct comparisons of early versus late windows conflicted. The authors set out to estimate the overall effect of time-restricted eating on metabolic outcomes, and then to rank its subtypes by timing and duration using a network meta-analysis, which combines direct and indirect comparisons.

Population et méthode

This was a systematic review and network meta-analysis registered in PROSPERO (CRD42022302737). The authors searched PubMed, Embase and the Cochrane database from inception to 3 January 2023, found 4213 articles, and included 41 randomised controlled trials with 2287 participants. Trials had to test time-restricted eating for more than one month in humans; studies conducted during Ramadan were excluded. Participants were 42.0% women, the median of study-level mean ages was 37.2 years (range of means 19.3 to 69.5), and the median of study-level mean BMI was 27.8 (21.9 to 39.6). Five studies enrolled people with metabolic syndrome, type 2 diabetes, pre-diabetes or fatty liver disease, and 12 studies imposed calorie limits on the time-restricted group. The median intervention lasted 8 weeks (range of means 4 to 48). Timing was classified by the last meal: early if it ended before 1700 (5 pm), mid if between 1700 and 1900, late if after 1900, and self-selected if participants chose. Duration was classed as under 8 hours, 8 hours or over 8 hours. This gave 6 early, 9 mid, 19 late and 13 self-selected arms, and 10, 24 and 10 arms in the three duration groups. Comparators were usual diets, continuous energy restriction, dietary advice and other intermittent fasting regimens such as alternate-day fasting.

Outcomes were changes in body weight, BMI, fat mass, fat-free (lean) mass and waist circumference; systolic and diastolic blood pressure; fasting glucose, fasting insulin, HbA1c and HOMA-IR (an insulin resistance index); and triglycerides and total, LDL and HDL cholesterol. The authors fitted a random-effects network meta-analysis for each outcome, called a difference significant when the 95% confidence interval excluded zero, and ranked treatments with P scores (0 to 1, higher meaning more favourable). Risk of bias was judged with the Cochrane RoB 2 tool, and certainty with the CINeMA framework (for timing: 29.8% high, 32.5% moderate, 35.2% low and 2.4% very low). They checked coherence and heterogeneity (I²), ran meta-regression when I² exceeded 75%, and did sensitivity analyses in healthy populations and in interventions of 1 to 6 months. The combined timing-and-duration analyses were not prespecified and the authors label them hypothesis generating.

Résultats

  1. 1
    Time-restricted eating overall was linked to modest improvements compared with usual diets: body weight -2.15 kg (95% CI -2.56 to -1.74), fat mass -1.32 kg (-1.61 to -1.03), waist circumference -1.63 cm (-3.02 to -0.23), systolic blood pressure 4.94 mm Hg lower (95% CI -6.92 to -2.95), fasting glucose -3.69 mg/dL (-6.19 to -1.18), fasting insulin -0.56 μIU/mL (-0.89 to -0.23) and triglycerides -10.12 mg/dL (-19.09 to -1.15).
  2. 2
    There were trade-offs and nulls. Fat-free (lean) mass also fell (-0.84 kg; -1.27 to -0.41), and diastolic blood pressure, HbA1c and HOMA-IR showed no significant change. Against alternate-day fasting, time-restricted eating was linked to 0.96 kg more fat mass (0.08 to 1.85) and 19.79 mg/dL higher total cholesterol (5.18 to 34.4).
  3. 3
    Eating early beat eating late. Compared with late time-restricted eating, early eating lowered body weight by 1.15 kg (95% CI -1.86 to -0.45; high certainty), waist circumference by 3.19 cm (-5.54 to -0.84; low certainty) and fasting insulin by 3.32 μIU/mL (-5.36 to -1.28; high certainty). Mid eating also beat late for weight (-0.94 kg; -1.63 to -0.24). Early eating had P scores of 0.62 to 0.86 for body measures (except lean mass) and 0.66 to 0.99 for glucose measures.
  4. 4
    Against usual diets, early eating gave the largest effects: weight 2.48 kg lower (-3.03 to -1.93), fasting insulin -3.67 μIU/mL (-5.67 to -1.67) and fasting glucose -6.13 mg/dL (-10.51 to -1.76). Late eating still lowered weight (-1.32 kg; -1.95 to -0.70) but had no significant effect on any glucose measure. Self-selected windows lowered weight by 1.94 kg (-2.89 to -0.99). Blood pressure did not differ significantly between subtypes.
  5. 5
    Window length gave no consistent winner. Eating for under 8 hours was better than 8 hours for waist circumference (2.65 cm lower; 95% CI -5.14 to -0.17) and fasting insulin (1.48 μIU/mL lower; -2.90 to -0.05), yet raised total cholesterol (+11.75 mg/dL versus 8 hours, 2.35 to 21.14; +15.17 versus over 8 hours, 6.73 to 23.62), and its overall P-score rankings were inconsistent. In exploratory combined analyses, early and mid windows of 8 hours or less ranked first or second for body measures and glucose measures, late windows over 8 hours generally ranked lowest, and late windows of 8 hours or less ranked first or second for blood pressure, fasting insulin, HOMA-IR and cholesterol.
  6. 6
    Adherence to time-restricted eating ranged from 66.3% to 99.0% and was generally higher than with continuous energy restriction in 7 studies. Among 14 studies with data there were no serious adverse events; mild ones included dizziness (12 people), headache (6) and constipation (7), often at the start and fading as people adapted.

Limites et incertitude

  1. 1
    The authors note that the combined timing-and-duration analysis rests on few studies, especially for late eating of 8 hours or less, and that these analyses were not prespecified, so they are hypothesis generating. They also note that the cut-points for eating duration (under 8, 8, over 8 hours) are somewhat arbitrary, and that no study reported metabolic syndrome or type 2 diabetes as outcomes.
  2. 2
    The authors note that the CINeMA certainty ratings may be optimistic because reporting bias is still possible, that network meta-analysis depends on extra assumptions, and that direct head-to-head evidence between subtypes is scarce. Heterogeneity was very high for HbA1c and HOMA-IR (I² above 90%), so those results deserve little weight. They also suggest that most studies drew blood in the morning, which means a longer fast before the draw for early windows and may distort lipid results.
  3. 3
    Our observation: the trials were short (median 8 weeks, range 4 to 48), searches ended on 3 January 2023 despite the 2026 publication date, only 6 arms tested early eating against 19 late arms, and 12 studies imposed calorie limits, so part of the weight loss comes from eating less.
  4. 4
    Our observation: subtype is defined only by the time of the last meal, comparators differed across trials (usual diet, continuous energy restriction, dietary advice, alternate-day fasting), and the methods describe no adjustment for the many outcomes and comparisons.
  5. 5
    Our observation: all outcomes are surrogate markers, not heart attacks or deaths. The authors' projection that a 4 to 6 mm Hg systolic drop could mean 8.6 to 13.2% lower cardiovascular mortality is an extrapolation from outside the trials.

Mise en perspective

For adults like those in these trials, a daily eating window was linked to modest gains in weight, waist, systolic blood pressure, and fasting glucose and insulin compared with usual eating, and finishing the last meal early (before 1700, or 5 pm) looked better than finishing after 1900. If you try a window, moving the last meal earlier is more consistently supported than squeezing the window shorter, and a self-selected window also lowered weight, which may suit people who need flexibility.

It does not show that the shortest window is best (under 8 hours was linked to higher total cholesterol), that early eating prevents heart disease, that the benefit is independent of eating less, or that effects last beyond the 8-week median length of the trials. Lean mass fell by 0.84 kg, which the authors suggest future research try to offset, for example with resistance training. Confidence is moderate: the ratings are often moderate to high but the authors say they may be optimistic. People with diabetes, or on glucose or blood pressure medicines, should discuss any change in meal timing with their clinician.

Vos données

Log the time of your first and last bite for about 8 weeks, the median length of these trials, and note whether your last meal falls before 1700 or after 1900. Compare trends in morning weight, waist and a morning blood pressure taken with the same cuff, arm and time of day, and use lab or fasting-glucose-meter values rather than a single reading. Cholesterol depends on how long you have fasted before the blood draw, and smart-scale lean-mass estimates are too rough for small changes. Do not change medication timing based on this review.

Comprendre ce type de preuve

Pooling trials can improve precision, but compare populations, controls, bias and uncertainty. Network rankings include indirect evidence and do not identify a universal best treatment.

Sources et attribution

Effects of timing and eating duration of time restricted eating on metabolic outcomes: systematic review and network meta-analysis

BMJ Medicine · 2026-01-21

DOI: 10.1136/bmjmed-2024-001071 ↗PubMed · 41586347 ↗

Interprétation éditoriale MoveVitals · contenu éducatif · aucune évaluation clinique individuelle enregistrée

Sources vérifiées 2026-11-01

Guides pratiques

À lire ensemble

·Essai randomisé

An 8 a.m. to 4 p.m. eating window added to calorie cutting gave no clear extra weight loss

Among 139 patients with obesity on a calorie-restricted diet, adding an eating window of 8:00 a.m. to 4:00 p.m. led to a mean 12-month weight change of -8.0 kg, versus -6.3 kg with calorie restriction alone. The net difference of -1.8 kg (95% CI -4.0 to 0.4; P = 0.11) was not statistically significant, and the interval cannot rule out either a real benefit or none.

The New England Journal of Medicine

Explications approfondiesLire l’étude →
·Essai randomisé

In a 20-person ward trial, an ultra-processed menu led to more eating and weight gain

In 20 adults living in a hospital research unit for 28 days, an ultra-processed menu led them to eat 508 kcal/day more (± 106 SE, P = 0.0001) and gain 0.9 kg, while a menu of unprocessed foods led to a 0.9 kg loss, each over 2 weeks and with eating left to appetite. The study was small, short and fully provisioned, so it cannot show what happens in daily life or which feature of the foods drove the extra eating.

Cell Metabolism

Explications approfondiesLire l’étude →
·Essai randomisé

Mediterranean diet with olive oil or nuts linked to fewer heart events (corrected trial)

Among 7447 Spanish adults aged 55 to 80 at high cardiovascular risk, those assigned to a Mediterranean diet with extra-virgin olive oil (HR 0.69, 95% CI 0.53 to 0.91) or with nuts (HR 0.72, 0.54 to 0.95) had fewer major cardiovascular events than those told to cut dietary fat. The trial had randomisation errors, which is why the 2013 report was withdrawn and the estimates recalculated, so treat this as strong but imperfect evidence.

The New England Journal of Medicine

Explications approfondiesLire l’étude →
·Essai randomisé

Low-fat vs low-carb: similar weight loss at 12 months, and no genotype or insulin match

In 609 adults without diabetes, 12-month weight change was -5.3 kg on a healthy low-fat diet and -6.0 kg on a healthy low-carbohydrate diet, a 0.7 kg gap (95% CI -0.2 to 1.6 kg) that was not statistically significant. A tested genotype pattern and baseline insulin secretion did not predict who did better on which diet. Both arms received 22 coaching sessions and 79% finished, so this describes coached dieting in adults aged 18 to 50.

JAMA

Explications approfondiesLire l’étude →
Nutrition et métabolisme →