Intermittent fasting: principles, benefits and methods
Intermittent fasting is about when you eat, not what you eat: it alternates periods with no calorie intake and periods of eating. The best-known methods are 16/8, 5:2 and alternate-day fasting. Does it help you lose more weight than a standard diet? The strongest studies say no: a Cochrane review published in 2026, covering 22 trials, concludes that it produces similar weight loss to usual dietary advice. It is therefore one tool among others, useful if the rhythm suits you, and best avoided in certain situations. Here are the methods, the trial results and the precautions.
👉 Free tool: Calorie deficit – calculate it in 30 seconds, no sign-up needed.
The principle
Intermittent fasting covers several ways of eating that have one thing in common: windows of time during which you consume little or no energy, alternating with periods of normal eating. During the fasting phase, the body draws on its reserves. The 2026 Cochrane review points out that weight loss comes first from reduced calorie intake, with other mechanisms still being studied (fat use, insulin sensitivity).
In other words, skipping a meal does not make you lose weight in itself: you lose weight if, overall, you eat less. Anses already stated this in 2010: total calorie intake is the major factor in the effects of a diet on weight.
The best-known methods
| Method | Principle | Example studied |
|---|---|---|
| 16/8 (time-restricted eating) | Eat within an 8-hour window, nothing with calories for 16 hours | Meals between 12pm and 8pm in the TREAT trial; between 8am and 4pm in the Liu trial |
| 5:2 (intermittent restriction) | Normal eating 5 days, very reduced intake 2 days a week | About 2,710 kJ (about 650 kcal) a day, 2 days a week, in the Harvie trial |
| Alternate-day fasting | Alternating a very low-intake day with a free or increased day | 25% of needs on fasting days, 125% on other days, in the Trepanowski trial |
| Whole-day fasting | One or two days a week with little or no intake | Studied in the BMJ network meta-analysis (2025) |
16/8 is the most widespread, because it often comes down to shifting or dropping breakfast or dinner. Our intermittent fasting tool helps you plan your eating windows.
What the studies say
Intermittent fasting has been widely studied over the past ten years. The results agree: it helps you lose weight, but no more than standard calorie restriction.
| Study | Participants and duration | Result |
|---|---|---|
| Cochrane review, Garegnani et al. (2026) | 22 trials, 1,995 participants with overweight or obesity, up to 12 months | Little or no difference in weight loss compared with usual dietary advice (low-certainty evidence) |
| Network meta-analysis, Semnani-Azad et al. (BMJ, 2025) | 99 trials, 6,582 adults | All strategies lead to weight loss compared with eating freely; only alternate-day fasting does slightly better than continuous restriction (-1.29 kg), mainly in short trials |
| TREAT, Lowe et al. (2020) | 116 adults, 16/8 for 12 weeks | -0.94 kg, with no significant difference from the three-meals group; greater loss of lean mass |
| Liu et al. (NEJM, 2022) | 139 adults with obesity, 12 months, all on calorie restriction | -8.0 kg with an 8-hour window versus -6.3 kg without, difference not significant |
| Trepanowski et al. (2017) | 100 adults with obesity, 12 months | Identical loss with alternate-day fasting and daily restriction at 6 months (-6.8%); more dropouts with fasting (38% versus 29%) |
| Harvie et al. (2011) | 107 women with overweight, 6 months | 5:2: -6.4 kg; continuous restriction: -5.6 kg, difference not significant |
Two lessons. First, intermittent fasting has no demonstrated advantage for medium-term weight loss. The authors of the Cochrane review suggest choosing it according to feasibility and sustainability for each person. Second, data beyond 12 months are lacking: the Cochrane review only covers the short term, and the BMJ meta-analysis notes that in trials of 24 weeks or more, only the advantage over eating freely persists.
The claimed benefits, put to the test
On weight
Real, but equivalent to those of standard calorie restriction. For some people, not eating during a fixed window is simpler than cutting back at every meal: that is its real appeal.
On metabolic markers
In the Liu trial (2022), blood pressure, blood sugar and blood fats changed in the same way with or without a restricted eating window. In the Harvie trial (2011), the fall in fasting insulin and insulin resistance was slightly greater with 5:2, for comparable weight loss. The BMJ meta-analysis finds no difference in glycated haemoglobin between intermittent fasting, continuous restriction and eating freely.
On everything else
Detox, longevity, autophagy: these promises are not assessed by the clinical trials presented here, which look at weight and cardiometabolic markers. Nothing makes it possible to put figures on them. On “detox”, see our guide to liver health and detox.
Warning points revealed by the trials
- Muscle mass: in the TREAT trial, the 16/8 group lost more appendicular lean mass than the three-meals group. Make sure you get enough protein during your eating window, and do some strength training.
- LDL cholesterol: in the Trepanowski trial, it was higher at 12 months with alternate-day fasting than with daily restriction (+11.5 mg/dl).
- Dropping out: 38% dropouts in the alternate-day fasting group, versus 29% on daily restriction. Participants ate more than planned on fasting days and less on the other days.
- Side effects: the Cochrane review finds no clear difference, but judges the data very uncertain.
Who should avoid intermittent fasting
- Pregnant or breastfeeding women: Anses is clear, no energy restriction should be undertaken during pregnancy.
- People with diabetes taking glucose-lowering treatment: in a trial in people with type 2 diabetes (Corley et al., 2018), fasting two days a week doubled the rate of hypoglycaemia (relative risk 2.05), despite a reduction in medication. No fasting without your doctor’s agreement.
- People with a history of eating disorders: Anses describes cognitive restraint as a risk factor for these disorders, and the HAS (France’s national health authority, 2023) asks professionals to assess fasting practices.
- Children and teenagers: Anses points to a risk of disrupted growth with weight-loss practices. See our guide to teenage nutrition.
- Older or undernourished people: Anses stresses the risk of undernutrition, and the HAS points out that a high BMI does not rule out undernutrition. See nutrition for older adults.
- People on long-term treatment or with a chronic illness: ask for advice first.
Setting up a 16/8 in practice
- Choose your window to fit your social and working life. A family dinner at 8pm? A window from 12pm to 8pm suits that. The HAS (2023) encourages meals at regular times: keep the same times on weekdays and at weekends as far as possible.
- Start gradually: 12 hours without eating (from 8pm to 8am), then 14, then 16 if you feel fine.
- Drink during the fast: unsweetened water, tea, coffee and herbal teas. Santé publique France points out that they contribute to water intake. Anything that provides calories (milk, sugar, juice, fizzy drinks) breaks the fast.
- Take care over your meals: two proper, complete meals are better than continuous snacking over 8 hours. The PNNS guidelines still apply: vegetables, pulses, wholegrain starchy foods, protein, a small handful of nuts, water.
- Review after a few weeks: if you have cravings, sleep badly or think about food all the time, this rhythm is not for you. That is not a failure: standard restriction gives the same results.
What you can drink during the fast
| Drink (per 100 g) | Energy (kcal) | Compatible with fasting? |
|---|---|---|
| Tap water | 0 | Yes |
| Brewed tea, unsweetened | 0.4 | Yes |
| Espresso coffee, unsweetened | 6.5 | Yes, without milk or sugar |
| Orange juice, pure juice | 45.4 | No |
| Semi-skimmed milk | 47.5 | No |
| Sweetened cola | 40.1 | No |
Example day on 16/8 (12pm to 8pm window)
- Morning: water, tea or coffee with no sugar or milk.
- 12.30pm: a large mixed salad with chickpeas, quinoa and raw vegetables, a plain dairy product, a piece of fruit.
- 4.30pm: a snack, for example a piece of fruit and a few almonds.
- 7.30pm: a fillet of fish or an omelette, vegetables, a portion of wholegrain starchy food, a dairy product or a piece of fruit.
5:2 and alternate-day fasting in practice
These methods take more organisation than 16/8. In the Harvie trial (2011), the women in the 5:2 group consumed about 2,710 kJ, or about 650 kcal, on the two reduced days, compared with about 6,276 kJ, or about 1,500 kcal, every day in the continuous restriction group: the authors describe a 25% energy restriction in both cases, and both approaches produced the same weight loss. In the Trepanowski trial (2017), the alternate-day fasting participants ate more than planned on fasting days and less than planned on free days: the two types of day tended to move closer together.
If you choose 5:2, plan meals rich in vegetables and protein for the reduced days, and eat normally, without compensating, on the other days. And if you have treated diabetes, do not do it alone: it is precisely this type of fasting, at 2,092 to 2,510 kJ a day two days a week, that doubled the rate of hypoglycaemia in the Corley trial (2018).
Does the timing of the window matter?
The question is often asked, because the body follows a circadian rhythm. The large trials do not settle it: the Liu trial (2022) used a morning window, from 8am to 4pm, and the TREAT trial a window from 12pm to 8pm, and neither showed an advantage for weight compared with the control group. The most solid advice remains that of the HAS (2023): meals at regular times, getting up at a regular time and exposure to light in the morning, and not skipping dinner if that causes night-time cravings. So choose the window you can keep every day, weekends included.
Breaking the fast well
Avoid rushing at sugary or ultra-processed foods: in the Hall et al. trial (2019), an ultra-processed diet led people to eat 508 kcal more per day than a minimally processed one. A meal rich in protein, fibre and good fats is more filling: quinoa and chickpea buddha bowl, spinach and avocado omelette, chicken and quinoa salad or lentil dahl. For other approaches, see our weight-loss guides and our slimming recipes.
Intermittent fasting or standard restriction: how to choose
Since the weight results are equivalent, the right choice is the one you will stick to. Intermittent fasting may suit you if you are not hungry in the morning, if you prefer two proper meals to several small ones, or if counting calories wears you down. Continuous restriction, described in our guide to the calorie deficit, suits you better if you need regular meals, if you exercise early in the morning or if you tend to get cravings. In both cases, the basics are those of a balanced diet, and physical activity remains, according to Anses, the best factor for not regaining the weight you have lost.
Myths about intermittent fasting
“During the window, I can eat whatever I want”
Intermittent fasting only helps you lose weight if it reduces total intake. If the meals in the window compensate, no weight is lost. And the quality of what you eat matters for health, whatever the timing.
“16/8 burns more fat than a standard diet”
The TREAT (2020) and Liu (2022) trials find no significant difference from eating across the whole day. The Cochrane review (2026) reaches the same conclusion.
“Fasting detoxifies the body”
The clinical trials cited in this guide measure weight and cardiometabolic markers: none shows a “detox” effect. The term has no precise scientific definition.
“The longer I fast, the more I lose”
Diets under 800 kcal a day must be supervised by a doctor specialising in nutrition, the HAS reminds us. Prolonged fasts without follow-up carry the same risks as very restrictive diets.
Frequently asked questions
Does intermittent fasting really help you lose weight?
Yes, as much as a standard low-calorie diet, but no more. The 2026 Cochrane review finds little or no difference in weight loss compared with usual dietary advice.
Does coffee break the fast?
An unsweetened espresso provides about 6.5 kcal per 100 g according to the Ciqual table, and Santé publique France points out that unsweetened coffee contributes to water intake. With milk or sugar, it provides energy and breaks the fast.
Which method should you choose to start?
16/8 is the simplest to set up, starting with 12 hours without eating. Alternate-day fasting showed more dropouts in the Trepanowski trial (2017).
Can you exercise while fasting?
The trials presented here did not assess exercise in a fasted state. As a precaution, if you feel unwell, have treated diabetes or are doing intense exercise, schedule your session within your eating window instead and ask your doctor for advice.
Does intermittent fasting cause muscle loss?
In the TREAT trial, the loss of lean mass was greater with 16/8. Enough protein and strength training help limit it, as with any diet.
This article is for information only and does not replace the advice of your doctor or a dietitian.
Our sources
- Garegnani LI, Oltra G, Ivaldi D et al. Intermittent fasting for adults with overweight or obesity. Cochrane Database Syst Rev. 2026. PMID 41692034.
- Semnani-Azad Z, Khan TA, Chiavaroli L 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. PMID 40533200.
- Lowe DA, Wu N, Rohdin-Bibby L 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 Intern Med. 2020. PMID 32986097.
- Liu D, Huang Y, Huang C et al. Calorie restriction with or without time-restricted eating in weight loss. N Engl J Med. 2022. PMID 35443107.
- Trepanowski JF, Kroeger CM, Barnosky A et al. Effect of alternate-day fasting on weight loss, weight maintenance, and cardioprotection among metabolically healthy obese adults: a randomized clinical trial. JAMA Intern Med. 2017. PMID 28459931.
- Harvie MN, Pegington M, Mattson MP et al. The effects of intermittent or continuous energy restriction on weight loss and metabolic disease risk markers: a randomized trial in young overweight women. Int J Obes. 2011. PMID 20921964.
- Corley BT, Carroll RW, Hall RM et al. Intermittent fasting in type 2 diabetes mellitus and the risk of hypoglycaemia: a randomized controlled trial. Diabet Med. 2018. PMID 29405359.
- Hall KD, Ayuketah A, Brychta R et al. Ultra-processed diets cause excess calorie intake and weight gain. Cell Metab. 2019. PMID 31105044.
- Anses. Risk assessment of weight-loss dietary practices (in French). Collective expert report. November 2010.
- Anses. Ciqual food composition table (in French). 2025.
- Haute Autorité de santé. Care pathway guide: overweight and obesity in adults (in French). January 2023, updated in February 2024.
- Haute Autorité de santé. Overweight and obesity in adults: first-line medical management (in French). Supporting report. September 2011.
- Santé publique France. Recommendations on diet, physical activity and sedentary behaviour for adults (in French). 2019.
