remain.fit · Intermittent Fasting: What the Evidence Actually Shows

Intermittent fasting means confining eating to set hours or set days. A 2026 Cochrane review of 22 trials found it loses little to no more weight than standard dietary advice, but clearly more than doing nothing. For most people it is a simpler way to hold a calorie deficit, and adherence decides the outcome.

What we know, and what we don’t

We know:

Fasting produces meaningful weight loss compared with doing nothing. The 2026 Cochrane review of 22 randomised trials puts it at −3.42% of body weight (95% CI −4.95 to −1.90, moderate certainty).
Against standard dietary advice it adds little to no weight loss: −0.33% in the same review (95% CI −0.92 to 0.26, low certainty). A 12-month trial of 139 adults (Liu et al., NEJM, 2022) found a net difference of −1.8 kg that wasn’t statistically significant (P=0.11).
Demanding schedules lose people. In a 12-month trial of alternate-day fasting (Trepanowski et al., JAMA Internal Medicine, 2017), 38% of the fasting group dropped out against 29% on daily calorie restriction.
A few metabolic markers improved without any weight loss in one small trial of 8 men (Sutton et al., Cell Metabolism, 2018).

We don’t know:

Whether any fasting protocol beats another once calories match. No trial has shown it.
Anything past 12 months. The Cochrane authors note the absence of long-term data, and their searches ran to November 2024, so the newest trials sit outside the review.
Whether the mechanisms measured in animals, autophagy among them, translate into human health outcomes.

What counts as intermittent fasting?

Intermittent fasting is any eating pattern that alternates set periods of eating with set periods of not eating. It comes in three families. Time-restricted eating confines food to a daily window, such as 16:8 or one meal a day. Whole-day protocols like 5:2 and 4:3 insert very-low-calorie days into a normal week. Alternate-day fasting swaps between fasting days and free days. The protocols differ more in how they feel than in what they achieve, and the full comparison has its own article.

Does intermittent fasting work for weight loss?

Yes, when the comparison is doing nothing, and about as well as ordinary dieting when the comparison is that. The Cochrane figures above carry the summary. The three largest individual trials fill in the texture:

Trial	Participants	Protocol	12-month weight change	Attrition
Trepanowski 2017, JAMA Intern Med	100 adults with obesity	Alternate-day fasting vs daily calorie restriction vs control	−6.0% vs −5.3% of body weight	38% vs 29% vs 26%
Liu 2022, NEJM	139 adults with obesity	8 a.m. to 4 p.m. window plus calorie restriction vs calorie restriction alone	−8.0 kg vs −6.3 kg; net −1.8 kg (95% CI −4.0 to 0.4; P=0.11)	85% completed
Catenacci 2025, Ann Intern Med	165 adults with obesity	Three fast days plus four unrestricted days vs daily calorie restriction	−7.7 kg vs −4.8 kg (P=0.040)	19% vs 30%

Note that the comparators differ between trials: some measured fasting against structured calorie targets, others against dietary advice. That’s why the pooled Cochrane estimate is the number to hold onto rather than any single row.

Is intermittent fasting better than just eating less?

No, once calories match, and the trials that do find a difference find it in adherence. Two of the largest trials disagree about this, and the reason is probably adherence. Liu enforced calorie targets in both arms and found no significant difference. Catenacci compared what people actually managed to do, and the fasting group did more of it: less dropout (19% vs 30%) and a larger achieved calorie reduction at both 6 months (−14.25% vs −10.20%) and 12 months (−9.75% vs −5.64%). A 2025 network meta-analysis of 99 randomised trials (Semnani-Azad et al., BMJ) reached the same quiet conclusion: across all protocols, only alternate-day fasting separated from continuous restriction, by −1.29 kg.

There’s also a trial where the window did nothing at all. In TREAT (Lowe et al., JAMA Internal Medicine, 2020), 116 adults ate either within a noon-to-8 p.m. window or as three structured meals for 12 weeks, with no calorie targets in either arm. The weight difference was −0.26 kg and not significant.

So the honest summary is short. Fasting is not a metabolic trick. It’s a scheduling trick, and for some people a very good one, because a closed kitchen is easier to obey than a food diary. The head-to-head with calorie counting works through when each tool wins.

What changes in the body beyond weight loss?

A handful of markers can improve even when weight doesn’t move, and the evidence for that is one small, careful trial. Sutton et al. (Cell Metabolism, 2018) put men with prediabetes through a randomised crossover: a 6-hour eating window ending before 3 p.m. for 5 weeks, with food matched so nobody lost weight. In the 8 men who completed the trial, insulin sensitivity, beta-cell responsiveness, blood pressure and a marker of oxidative stress all improved. Eight men, five weeks. It’s a genuinely important result, and it is far too small to build a health claim on. The insulin and glucose article covers this territory properly, including what it means if you take diabetes medication.

What does fasting not do?

Several of the most repeated fasting claims run ahead of their evidence. There’s no established 16-hour threshold at which human autophagy switches on; the mechanism is real in animal and cell models, and nobody has a validated way to measure it in a person. Short fasts don’t shut your metabolism down. Fasted cardio changes what you burn during the session, not how much fat you lose over weeks. Each of these gets a fuller correction in the claims article.

Who should not try intermittent fasting?

Fasting schedules aren’t appropriate for everyone, and for some groups they’re genuinely unsafe. That includes pregnancy and breastfeeding, type 1 diabetes, a history of an eating disorder, being underweight, and being under 18. Anyone taking insulin, sulfonylureas or medication that has to be taken with food needs a clinician involved before changing meal timing, because hypoglycaemia is a real risk. When fasting isn’t the right tool names each group and explains why.

How can you tell whether it’s working for you?

Watch the trend, not the reading. Day-to-day weight moves for reasons that have nothing to do with body fat, so a single number can’t answer the question you’re asking it; what a weight reading actually tells you explains where the noise comes from. Regular self-weighing is one of the better-supported monitoring habits: a systematic review (Zheng et al., Obesity, 2015) found it associated with better weight control, with no consistent psychological harm reported in the included trials. Give any protocol a few weeks, weigh yourself the same way each time, and judge the direction of the line. And if the plan is working but you can’t stand it, that’s the result too. Adherence is the whole game, and keeping the result is a different skill again.

Key points

Fasting works mainly by creating a calorie deficit. Against no intervention it produces −3.42% of body weight (Cochrane review, 22 trials, 2026, moderate certainty).
Against standard dieting it adds little: −0.33% in the same review (low certainty), and a net −1.8 kg (P=0.11) in the largest single trial (Liu et al., NEJM, 2022).
Where trials find a difference, adherence is the variable. In Catenacci et al. (Annals of Internal Medicine, 2025), the fasting group had lower attrition (19% vs 30%) and a larger achieved calorie reduction.
Demanding schedules shed participants: 38% dropout in the alternate-day fasting arm of Trepanowski et al. (JAMA Internal Medicine, 2017).
Some markers improved without weight loss in one small trial: 8 men, 5 weeks, food matched to hold weight steady (Sutton et al., Cell Metabolism, 2018).
No protocol beats another once calories match, and no trial runs past 12 months. A 2025 network meta-analysis of 99 trials (Semnani-Azad et al., BMJ) found only alternate-day fasting separated from continuous restriction, by −1.29 kg.

Common questions

Does intermittent fasting work without counting calories?

Often, yes, and that’s most of its appeal. A fixed eating window can cap intake without any arithmetic. But the window doesn’t cap intake by itself: in the TREAT trial, people ate freely inside an 8-hour window and lost almost nothing. If the window isn’t producing a deficit, nothing downstream happens. The comparison with counting covers where each approach breaks down.

Which fasting schedule should I start with?

The one you can hold for a year, which is a serious constraint. In a dataset of 792,692 fasting-app users (Torres et al., JMIR mHealth and uHealth, 2022), 16.7% were still logging at 13 weeks and 2.1% at 52 weeks. No protocol’s advantage survives quitting it. The types compared lays out what each schedule asks of you.

Is it better to skip breakfast or dinner?

The small trials that exist point toward earlier eating windows, with improvements in insulin measures that don’t always come with weight change. They’re short and small, so hold the conclusion loosely. Window timing covers the evidence properly.

Does fasting cause muscle loss?

Losing weight costs some lean tissue however you do it, and fasting doesn’t appear to make that worse when protein and training are in place. The protections are eating enough protein and lifting something heavy twice a week. Keeping muscle while losing fat has the numbers.

A note on safety

This is general information, not medical advice. If you take medication or have a health condition, talk to your clinician before changing when or what you eat.

References

Garegnani L, et al. Intermittent fasting for adults with overweight or obesity. Cochrane Database of Systematic Reviews. 2026;2(2):CD015610. doi:10.1002/14651858.CD015610.pub2
Trepanowski JF, et al. Effect of alternate-day fasting on weight loss, weight maintenance, and cardioprotection among metabolically healthy obese adults. JAMA Internal Medicine. 2017;177(7):930-938. doi:10.1001/jamainternmed.2017.0936
Liu D, et al. Calorie restriction with or without time-restricted eating in weight loss. New England Journal of Medicine. 2022;386(16):1495-1504. doi:10.1056/NEJMoa2114833
Catenacci VA, et al. The effect of 4:3 intermittent fasting on weight loss at 12 months. Annals of Internal Medicine. 2025;178(5):634-644. doi:10.7326/ANNALS-24-01631
Sutton EF, et al. Early time-restricted feeding improves insulin sensitivity, blood pressure, and oxidative stress even without weight loss in men with prediabetes. Cell Metabolism. 2018;27(6):1212-1221.e3. doi:10.1016/j.cmet.2018.04.010
Varady KA, et al. Clinical application of intermittent fasting for weight loss: progress and future directions. Nature Reviews Endocrinology. 2022;18(5):309-321. doi:10.1038/s41574-022-00638-x
Semnani-Azad Z, et al. Intermittent fasting strategies and their effects on body weight and other cardiometabolic risk factors: systematic review and network meta-analysis of randomised clinical trials. BMJ. 2025;389:e082007. doi:10.1136/bmj-2024-082007
Lowe DA, et al. Effects of time-restricted eating on weight loss and other metabolic parameters in women and men with overweight and obesity: the TREAT randomized clinical trial. JAMA Internal Medicine. 2020;180(11):1491-1499. doi:10.1001/jamainternmed.2020.4153
Schroor MM, et al. Effects of intermittent energy restriction compared with those of continuous energy restriction on body composition and cardiometabolic risk markers. Advances in Nutrition. 2024;15(1):100130. doi:10.1016/j.advnut.2023.10.003
Zheng Y, et al. Self-weighing in weight management: a systematic literature review. Obesity. 2015;23(2):256-265. doi:10.1002/oby.20946
Torres L, et al. Retention, fasting patterns, and weight loss with an intermittent fasting app. JMIR mHealth and uHealth. 2022;10(10):e35896. doi:10.2196/35896

remain.fit tracks a fasting window, weight and lab markers on one timeline, so you can see whether anything actually moved. The tools observe; the judging is left out.