Nutrition, Diet & Fasting - Fasting
Alternate-day, 5:2 and 4:3 fasting
By Ryan Lowe. Last reviewed 3 August 2026. How we write this.
The version of intermittent fasting with the best trial evidence - and even here the advantage over just eating less is modest and fades as trials get longer.
Three related patterns sit in this family. Alternate-day fasting alternates a fast day with a normal eating day. The 5:2 diet eats normally for five days and heavily restricts on two. The 4:3 pattern restricts on three non-consecutive days a week. In practice "fast days" in most trials are not zero food - they usually allow around a quarter of normal intake, which makes them far more sustainable and is worth knowing, because the popular description often implies otherwise.
What the trials found
This is where intermittent fasting has its best evidence. A randomised clinical trial published in the Annals of Internal Medicine in 2025 compared 4:3 intermittent fasting with daily calorie restriction over twelve months and found greater weight loss in the fasting group - a genuine, properly conducted head-to-head result in the fasting camp's favour.
The wider picture is more measured. Meta-analyses pooling intermittent fasting trials find that alternate-day approaches outperform continuous calorie restriction mainly in shorter studies, and that when analysis is restricted to trials of roughly six months or more, no intermittent fasting strategy reliably beats continuous restriction. The advantage is real but modest, and it appears to be partly a function of trial length.
The useful comparison within the fasting family is the one people get backwards: the best-evidenced approach is alternate-day style fasting, not the 16:8 window that far more people actually do.
Why it may work slightly better
The most likely explanation is compliance arithmetic rather than metabolism. A large deficit on three days a week may be psychologically easier to sustain than a moderate deficit every single day, because most days require no restriction at all. There is also less evidence of full compensation than you might expect - people generally do not eat enough extra on normal days to erase the fast-day deficit, though some compensation does occur.
Metabolic explanations - deeper insulin suppression, better fat oxidation on fast days - are plausible and are frequently offered, but the trial evidence does not currently separate them from the simpler compliance account.
Where the evidence stops
Dropout rates in alternate-day fasting trials tend to be high, which both flatters the results (the people who could not tolerate it are gone) and tells you something important about the method. Long-term data past a year is scarce. Effects on lean mass vary between studies and depend heavily on protein intake and whether the person is resistance training.
Who should be careful
Fast days can cause hypoglycaemia in anyone on insulin or sulfonylureas, and doses may need medical adjustment. The pattern is not appropriate in pregnancy or breastfeeding, for anyone underweight, or for anyone with a history of disordered eating - a cycle of heavy restriction followed by unrestricted eating is a recognised route into binge patterns for susceptible people. See the fasting safety lesson.
Educational content only and not medical advice. If you take medication for diabetes, speak to your prescriber before adopting any fasting pattern.
Related in this classroom
The studies behind this page
- The effect of 4:3 intermittent fasting on weight loss at 12 months: a randomized clinical trial - Annals of Internal Medicine, 2025
- Calorie restriction with or without time-restricted eating in weight loss - New England Journal of Medicine, 2022
Sources for this classroom
Educational content only. Not medical advice, diagnosis or treatment. Always consult a qualified healthcare professional before changing your health regimen.
