A summary of published research, not medical advice. See the Disclaimer.
Studies: every indexed trial and review on this topic, scored
Time-restricted eating confines all daily food intake to a fixed window, usually 6 to 10 hours. Total calories may or may not change.
Best-supported effect: Weight loss in adults with overweight or obesity, compared with unrestricted eating. Probably helps a little (moderate evidence).
Most-cited weak claim: Weight loss beyond what the same calorie cut achieves without a time window. Probably makes little or no difference (moderate evidence).
How we grade evidence · Last reviewed: October 2026
What you need to know
In adults with overweight or obesity, eating within a 6 to 8 hour window led to a little more weight loss than unrestricted eating, with moderate evidence. When both groups cut calories by the same amount, the time window made little or no difference to weight loss, also with moderate evidence. Small, short trials found small improvements in insulin sensitivity and glucose markers (weak evidence), and whether an early window beats a later one is unknown. One trial found a small loss of lean mass, with weak evidence. Trials enrolled only adults, and only four completed trials included people with type 2 diabetes, too few to establish safety in that group. Trials used windows of 6 to 8 hours for 5 weeks to 12 months.
| Effect | Verdict | What was measured | Studied in |
|---|---|---|---|
| Weight loss compared with unrestricted eating | Probably helps a little (moderate evidence) Most trials under 24 weeks, results disagreed between trials |
Measured | Adults with overweight or obesity, including type 2 diabetes |
A 2025 network meta-analysis of 99 randomised trials found that time-restricted eating reduced body weight compared with unrestricted eating, by about as much as plain calorie restriction [1]. Individual trials disagreed. In a 12-week trial of a noon-to-8 pm window with no calorie targets, weight loss did not differ significantly from a group eating three structured meals a day (about 0.9 kg against 0.7 kg) [2]. A 6-month trial in adults with obesity and type 2 diabetes found more weight loss with an 8-hour window than in the control group [3]. Probably helps a little (moderate evidence): downgraded because most trials ran for under 24 weeks, and results disagreed between trials.
| Effect | Verdict | What was measured | Studied in |
|---|---|---|---|
| Extra weight loss beyond equal calorie restriction | Probably makes little or no difference (moderate evidence) Pooled analysis comparing diets found no meaningful gap, one 14-week trial disagreed |
Measured | Adults with obesity |
More on this result under Claimed but unsupported or contested.
| Claim | Verdict | What was measured | Studied in |
|---|---|---|---|
| Better insulin sensitivity and glucose markers independent of weight loss | May help a little (weak evidence) Mainly one 5-week trial in men with prediabetes, pooled trials did not separate out weight loss |
Lab marker | Men with prediabetes; adults without obesity |
| Early-day window works better than a later one | Unknown (very weak evidence) Rests on two small trials |
Lab marker | Adults without obesity; adults with overweight or obesity |
The glucose claim rests mainly on one 5-week crossover trial in men with prediabetes. On a 6-hour window with dinner before 3 pm, and enough food supplied to keep weight stable, insulin sensitivity and blood pressure improved without weight loss [6]. A 5-week trial in 90 adults without obesity found that an early window improved insulin sensitivity more than a midday window [7]. A meta-analysis of glycaemic outcomes reported small reductions in HbA1c and fasting insulin, while fasting glucose fell only in early-window trials [8]. The claim is about these markers, so the lab-marker rule does not apply. Only the crossover trial kept weight stable, and the meta-analysis did not separate out weight loss. May help a little (weak evidence): a single trial that has not been repeated.
In a 12-month trial of 139 adults with obesity, all given the same calorie targets, the group eating from 8 am to 4 pm lost 8.0 kg and the group with no time window lost 6.3 kg. The difference was not significant [4]. The network meta-analysis also found no meaningful difference between time-restricted eating and calorie restriction [1]. One 14-week trial found the opposite. An early window (7 am to 3 pm) with calorie restriction produced about 2.3 kg more weight loss than eating over 12 or more hours with the same weight-loss treatment [5]. Probably makes little or no difference (moderate evidence): downgraded for that inconsistency. Whether an early window beats a later one rests on two small trials [5][7], so that claim is unknown (very weak evidence).
No trials have tested lifespan, cardiovascular events or diabetes incidence.
Shown in humans
Shown only in animals or cells
Theory only
| Side effect | Verdict | How often vs placebo | Source | Studied in |
|---|---|---|---|---|
| Lean mass loss | May happen (weak evidence) One trial's in-person subgroup, lean mass not a main outcome |
Appendicular lean mass index 0.16 kg/m² lower than control | [2] | Women and men with overweight and obesity |
| Lower bone mineral density | May not increase it (weak evidence) Seven short trials pooled, few large enough to detect bone changes |
No significant difference | [9] | |
| Adverse events overall | May not increase it (weak evidence) Similar event numbers across groups, no serious events in two trials |
Similar numbers across groups; no serious events in two trials | [3][4][7] | Adults with type 2 diabetes; Healthy volunteers without obesity |
| Fatigue | May not increase it (weak evidence) No added fatigue across 15 trials of intermittent fasting generally |
Risk difference 0% (95% CI -1% to 2%) | [14] | Adults with overweight or obesity |
| Headache | May not increase it (weak evidence) No added headache across 15 trials of intermittent fasting generally |
Risk difference 0% (95% CI -1% to 2%) | [14] | Adults with overweight or obesity |
| Dizziness | May not increase it (weak evidence) About 3 percentage points higher with eating windows not set early, could be chance |
About 3 percentage points higher with non-early windows alone, not significant | [14] | Adults with overweight or obesity |
| Dropping out | May not increase it (weak evidence) Similar dropout across 15 trials of intermittent fasting generally |
Risk difference 1% (95% CI -2% to 4%) | [14] | Adults with overweight or obesity |
The lean mass finding comes from the in-person subgroup of one trial, where lean mass was a secondary outcome [2]. The bone meta-analysis pooled seven short trials, few powered for bone outcomes [9]. The fatigue, headache, dizziness and dropout figures pool 15 trials of intermittent fasting in general, not only time-restricted eating [14].
This lists what studies and labels report. A pharmacist or prescriber can say whether it applies to a specific person.
This lists what studies and labels report. A pharmacist or prescriber can say whether it applies to a specific person.
| Combined with | What happened | Verdict | Source | Studied in |
|---|---|---|---|---|
| Glucose-lowering medicines (metformin, insulin, sulfonylureas and others) in type 2 diabetes | A review of four trials found no major safety issues when medication was monitored and adjusted. A 6-month trial reported no serious adverse events and no change in medication effect score | May not increase it (weak evidence) Four trials reviewed, medication monitored and adjusted, no major safety issues |
[3][15] | Adults with type 2 diabetes |
No interaction studies with other drugs or supplements were found.
These are the amounts used in the studies. They are not recommendations.
Trials used eating windows of 6 to 8 hours for 5 weeks to 12 months. Some set no calorie targets; others combined the window with calorie restriction [2][3][4][5][6][7].
| Trial | Eating window | Duration | Population |
|---|---|---|---|
| [6] | 6 hours, dinner before 3 pm, weight-maintaining food supplied | 5 weeks per phase, crossover | Men with prediabetes |
| [7] | Early-day vs midday window | 5 weeks | Adults without obesity (n=90) |
| [2] | Noon to 8 pm, no calorie targets | 12 weeks | Adults with overweight or obesity (n=116) |
| [5] | 7 am to 3 pm, with calorie restriction | 14 weeks | Adults with obesity (n=90) |
| [3] | Noon to 8 pm, no calorie counting | 6 months | Adults with obesity and type 2 diabetes (n=75) |
| [4] | 8 am to 4 pm, with calorie restriction | 12 months | Adults with obesity (n=139) |