Intermittent fasting: what the trials show and what happens when you stop

Intermittent fasting produces real results in trials. But the sustainability question and the differences between sexes have been largely ignored in coverage.

Intermittent fasting arrived with an unusually attractive proposition: not what you eat, just when — no food groups banned, no counting. The trials that followed have been more useful than the enthusiasm suggests, and their central finding is consistently underreported. When fasting is compared directly against ordinary calorie restriction, with both groups eating the same amount, the fasting adds nothing.

What the head-to-head trials show

The cleanest test was published in the New England Journal of Medicine in 2022. Participants with obesity were randomised for twelve months either to calorie restriction alone or to the same calorie restriction confined to an eight-hour daily eating window. Both groups lost meaningful weight — around eight kilograms — and there was no significant difference between them. The window contributed nothing once intake was matched.

Also read

Supplement Reviews

Ancient + Brave True Collagen review: the cleanest label in the category, the loosest claims

28 Aug 2026

Supplement Reviews

Vital Proteins Collagen Peptides review: a 20 g serving the research never asked for

28 Aug 2026

An earlier year-long trial compared alternate-day fasting against daily calorie restriction and found the same thing, plus something the headline missed: dropout was higher in the fasting group. Meta-analyses reach the same conclusion — intermittent fasting produces weight loss comparable to continuous energy restriction, not superior to it.

So why does it work for people?

Because compressing an eating window reduces how much most people eat, without requiring them to track anything. That is a genuine benefit, and it is worth being clear that “it works through calorie reduction” is not a dismissal. A rule that is easy to follow and produces a deficit is a useful tool, and for people who find counting intolerable it may be the more sustainable route.

The distinction matters for expectations. If fasting is a delivery mechanism for a deficit rather than a metabolic trick, then it will work for people it suits and fail for people it does not, and there is no reason to persist with a schedule that makes you miserable on the theory that something special is happening at hour sixteen.

Autophagy, and the claims that outrun the data

The autophagy argument — that fasting triggers a cellular clean-up process with longevity benefits — rests largely on cell and animal work. Autophagy is real and important; the Nobel-recognised research establishing it was done in yeast. Measuring it in living humans is difficult, animal fasting durations do not map cleanly onto human schedules, and no trial has shown that fasting-induced autophagy produces health outcomes in people.

One area is more promising. Small studies of early time-restricted eating — an eating window shifted toward the morning, with dinner unusually early — have found improvements in insulin sensitivity and blood pressure even without weight loss. The best-known of these involved eight men with prediabetes on a schedule almost nobody would keep. It suggests that when you eat may matter for circadian reasons, and that a late-evening window is probably the worst version. It is a hypothesis worth watching, not a finding to build a regime on.

The lean mass question

A 2020 trial comparing a sixteen-hour fasting window against three structured meals produced a result that got far less coverage than it deserved. Weight loss was modest and not significantly different from the control group — but in the subset where body composition was measured, a striking proportion of the weight lost in the fasting group was lean mass rather than fat.

The finding has been debated on methodological grounds, and it is one trial. But it is biologically plausible: protein intake often falls when eating windows compress, and the resistance training that protects muscle during a deficit was not part of the protocol. At minimum it argues for paying attention to protein and to strength work if you fast, rather than assuming the weight coming off is the weight you want to lose.

Where women are missing from the data

This is the gap your excerpt is right to flag, and it is a gap rather than a finding.

Most intermittent fasting trials are small and short. Many were conducted in men, and a substantial number of mixed-sex studies do not report results separately by sex — which means differences, if they exist, would be invisible. One older study of alternate-day fasting reported that men improved insulin sensitivity while women’s glucose response did not, and that result is frequently cited because there is so little else to cite.

The concerns most often raised — menstrual disruption, hormonal effects — are extrapolated primarily from rodent research, where female animals are notably sensitive to intermittent restriction, and from the human literature on low energy availability, where insufficient intake relative to expenditure does disrupt menstrual function. That is a reasonable basis for caution, but not evidence that a given schedule harms a given woman.

The honest summary: this has been studied far less in women than the volume of advice aimed at women would suggest, and premenopausal and postmenopausal responses have barely been compared.

The risk that gets the least coverage

Intermittent fasting involves deliberately not eating for extended periods, applying rigid rules to food timing, and overriding hunger — behaviours that are also features of restrictive eating disorders. Research has found associations between intermittent fasting practices and disordered eating attitudes, particularly in young women.

That does not make fasting an eating disorder, and plenty of people practise it without difficulty. But if you have a history of disordered eating, if food rules tend to escalate for you, or if the fasting window has started to feel like something you cannot break rather than something you chose, this is not the right tool. The same applies to pregnancy, type 1 diabetes, and several medications that require food — all of which warrant a clinician’s input before starting.

The practical takeaway

If an eating window makes eating less feel easier, it is a reasonable tool, and the trials say it works about as well as any other method of reducing intake. Keep protein up and resistance training in, because lean mass is the one substantive risk the research has flagged. Favour an earlier window over a late one where the schedule allows.

What the evidence does not support is persisting with a protocol you find miserable on the belief that fasting itself confers something a normal deficit does not. Twelve months of randomised comparison says it does not — and sustainability, as this publication keeps finding, is the variable that actually determines outcomes.