Do Diet Breaks Work for Weight Loss?

21 September 202610 min read

A wooden bench on a woodland trail, a place to sit down partway through a walk. — Photo by Krzysztof Walczak on Unsplash.

A diet break is a planned stretch of eating at maintenance, usually one or two weeks, in the middle of a longer calorie deficit. Four randomised trials come closest to testing it, though one of the four is a loose fit. Three found no extra fat or weight loss compared with dieting straight through, and one found a large advantage. On whether breaks make dieting easier to bear, two trials measured it and they disagree.

In brief

In the 2021 ICECAP trial, 61 resistance trained adults dieted for 12 weeks either continuously or in blocks separated by maintenance weeks. Fat loss, muscle retention and resting metabolism showed no measurable difference between the groups.
The ICECAP group taking breaks reported lower hunger, less desire to eat and greater satisfaction, with higher levels of the fullness hormone peptide YY.
A 2023 trial of 38 resistance trained women found no difference in body composition or resting metabolic rate, and no difference on seven of its eight hunger and eating behaviour measures. It also found no effect on how easy participants said the diet was to stick to.
In the 2018 MATADOR study, 51 men with obesity were randomised and 36 finished per protocol. Among those 36, the interrupted group lost 14.1 kg against 9.1 kg, but spread the same 16 weeks of dieting across 30 weeks in total.
A 2019 trial published in the International Journal of Obesity randomised 332 adults, of whom 146 completed the year, and found no difference in weight lost between continuous dieting, week on week off dieting and a 5:2 pattern.
This post is about losing weight on purpose, because a diet break only exists inside a deficit.

What is a diet break?

A diet break is a deliberate return to maintenance calories for a set period, then a return to the deficit. It is not a week off from tracking, and it is not a cheat week. Maintenance means eating roughly the amount that holds your weight steady, which is more than your deficit target and less than you would eat without thinking about it at all.

Researchers call this intermittent energy restriction, and the name causes a lot of confusion, because much of the research filed under it is about fasting patterns instead. An eating window like 16:8 changes which hours of the day you eat in. A pattern like 5:2 cuts your intake severely on two days a week. Neither is a diet break, which holds you at maintenance rather than below it, for a week or two at a time. Evidence about one does not transfer cleanly to the others, so a study that sounds relevant often is not.

What do the trials show?

Four randomised controlled trials have compared a broken deficit against a continuous one, and they do not agree.

Trial	People	Design	Result
MATADOR, Byrne et al. 2018	51 men with obesity randomised, 36 finished per protocol	16 weeks of dieting, either continuous or in 2 week blocks separated by 2 week maintenance blocks (30 weeks total)	Interrupted group lost 14.1 kg against 9.1 kg
ICECAP, Peos et al. 2021	61 resistance trained adults, 32 of them women	12 weeks of dieting, either continuous or in 3 week blocks separated by 1 week maintenance blocks (15 weeks total)	No measurable difference in fat loss, muscle or resting metabolism
Siedler et al. 2023	38 resistance trained women	6 weeks of dieting at a 25% cut, either continuous or with a maintenance week after every 2 weeks (8 weeks total)	No measurable difference in body composition or resting metabolic rate
Headland et al. 2019	332 adults randomised, 146 completed the year	12 months of continuous dieting, week on week off dieting, or 5:2	No measurable difference between the three, around 5 to 6.6 kg lost in each

Sources: Byrne et al. 2018, International Journal of Obesity; Peos et al. 2021, Medicine and Science in Sports and Exercise; Siedler et al. 2023, Journal of Human Kinetics; Headland et al. 2019, International Journal of Obesity.

Three of the four found no advantage on the scale. That includes the only trial run entirely in women, and the only one that ran for a full year.

Headland deserves a caveat, because it is the loosest fit with what most people mean by a diet break. Its week on week off arm alternated a restricted week with a week of habitual eating rather than a prescribed maintenance target, so nobody in it was deliberately held at maintenance. Its third arm was 5:2, which cuts intake severely on two days a week rather than pausing the deficit at all. It is the only evidence here on what happens over a full year, and it is not a clean test of the intervention.

The other trials were not powered to prove that the two approaches are identical, only to detect a difference if one was there. A null result means no difference was found, not that none exists.

Why did one trial find a big difference?

MATADOR is the outlier, so what set it apart?

The first explanation people reach for is that the interrupted group simply had longer. Both groups did 16 weeks of dieting at 67% of the calories they needed to hold their weight, a cut of about a third. The continuous group did those weeks back to back. The interrupted group spread theirs across eight two week blocks with seven two week maintenance blocks in between, which took 30 weeks in total. That is nearly twice as long on the programme. But the study measured what happened during those maintenance blocks, and the average weight change across them was 0.0 kg, with a standard deviation of 0.3 kg. The extra calendar time contributed no weight loss of its own, so it cannot account for the five kilo gap.

The researchers’ own explanation was reduced metabolic adaptation, which is the subject of the next section. Two differences in the setup are worth naming alongside it. MATADOR recruited men with obesity and cut their intake by about a third, a deep cut in a group whose bodies have more to defend. ICECAP and Siedler recruited resistance trained people who were not obese, Siedler on a 25% cut and ICECAP on a restriction aimed at losing about 0.7% of body weight a week. And MATADOR’s headline figures come from the 36 participants who completed the intervention per protocol: 51 were randomised, and 47 began the intervention after a four week baseline phase. Headland’s figures are also from completers, 146 of 332, so that limitation cuts across the positive and the null results alike. ICECAP analysed everyone it randomised, which is the harder test.

Does a diet break reset your metabolism?

Nothing switches off and back on, so the phrase promises more than it can deliver.

There is a real effect underneath it. When you eat less for a long time, the energy you burn falls by more than the loss of body tissue alone would predict. MATADOR measured this and found something specific: the drop in resting energy expenditure was not statistically different between the two groups in raw terms, but once the researchers adjusted for the body composition each group had lost, the interrupted group’s drop was smaller. ICECAP measured resting energy expenditure and found no difference between its groups. Siedler measured resting metabolic rate and found no difference either.

One trial found a modest signal in an adjusted measure and two found none. That is a long way from a reset. Your calorie needs do drift downward as you lose weight, and that happens whether you take breaks or not, which is why a target set once in January stops matching you by March. We wrote about that in why your calorie target shouldn’t stay the same for months, and about the quieter half of it, the everyday movement that falls away without you noticing, in what is NEAT.

Will your weight jump during the break?

Your weight will very likely go up in the first few days, and that rise is not fat.

Eating more carbohydrate refills muscle glycogen, and every gram of stored glycogen holds water alongside it. More food in the gut adds weight directly. Higher sodium shifts fluid. None of that is stored energy, and none of the trials here tracked weight day by day inside a break, so we would rather not put a figure on the rise than invent one.

What the trials do tell you is how a break ends up. In MATADOR, the group’s weight change averaged across its seven maintenance blocks was 0.0 kg, and that figure was consistent from one block to the next. That is a group average rather than a promise about any one fortnight, but it says the fortnight does not undo the work. If you want the fuller picture of why the scale moves when your body has not, we covered it in why does your weight change so much from day to day.

The practical problem is that nobody warns you. The rise arrives early in the break, it looks like failure, and that is a bad moment to be making decisions about the plan.

How long should a diet break be?

The trials used one to two weeks at maintenance, repeated on a schedule, rather than one break taken when motivation ran out.

ICECAP used one week at maintenance after every three weeks of dieting.
Siedler used one week at maintenance after every two weeks of dieting.
MATADOR used two weeks at maintenance after every two weeks of dieting.

As far as we know, no trial has compared different break lengths against each other, so anyone giving you a precise optimal number is going beyond the evidence. What the studies share is that the breaks were scheduled in advance and the maintenance calories were a deliberate figure, not a pause from paying attention.

Maintenance is a number, and it is not the number you were eating before you started, because you weigh less now and your body burns less than it did.

Who is most likely to benefit?

Look at who was in each trial and a pattern appears. The one positive result for fat loss came from men with obesity on a deep cut. The two trials in resistance trained people on a milder cut found no advantage. That is a small evidence base to draw a rule from, but if there is a group with something to gain on the scale, it is people with more weight to lose who are cutting hard, rather than people who are already fairly lean.

On how the diet feels, the two trials that looked disagree, and the disagreement is worth setting out properly.

In ICECAP, the break group reported significantly lower hunger, less desire to eat and greater satisfaction than the continuous group, and had higher levels of peptide YY, a hormone involved in feeling full. Siedler assessed eight measures of hunger and eating behaviour and found no group difference on seven of them. The exception was disinhibition, a measure of how easily eating gets derailed by food, mood or social settings, which rose in the continuous group and fell in the break group. The two groups started from different baseline scores on that measure, so it is a signal rather than a settled finding. Siedler also asked participants directly how easy the diet was to stick to and how motivated they were to continue, and found no effect from the breaks at all.

So the appetite evidence is split, and the adherence evidence is worse than split. MATADOR did not report appetite outcomes. Headland is the only trial here that measured how many people dropped out, and it found no statistically significant difference between the three approaches, in a study where fewer than half of the 332 people who started completed the year. Nothing published so far shows that diet breaks help people stay with a diet.

That is a more useful answer than the one usually given, because it separates two claims that get bundled together. A break will probably not speed your fat loss. Whether it makes the diet more bearable seems to depend on who you are, and whether that bearability turns into actually sticking with it has not been demonstrated.

One honest limit on all of this. Every trial here scheduled its breaks in advance. Nobody has tested the version most people are likely to reach for, which is taking a single break in month four because they are hungry and fed up. If that is where you are, two weeks at maintenance is a reasonable thing to try and no trial has found that scheduled breaks cost fat loss, but applying that to an unscheduled one is an extrapolation rather than a tested finding. If you are six weeks in and it is going fine, no trial suggests you need one.

Staying with tracking at all is its own hurdle. A 2024 review of 18 studies covering 525,824 users of health and lifestyle apps found that a median of 70% stopped using them within the first 100 days, and we looked at what lies behind that in why do most people quit calorie tracking.

Does this apply if you are gaining weight?

A diet break is defined by the deficit it interrupts, so almost none of the evidence above carries over, and we would rather say so than stretch the research. If you are eating in a surplus to gain weight, we could find no equivalent trial testing planned maintenance blocks, and the mechanisms run in the opposite direction anyway.

What does carry across is that your maintenance calories are a moving figure. Whether you are gaining or losing, the number that held your weight steady three months ago is not the number that holds it steady today.

Ori is built around that. Your target is worked out again each week from what your weight has actually done, so if you decide to take a couple of weeks at maintenance, you can switch your goal to maintain in the app and switch back afterwards, and the target you come back to reflects the body you have now rather than the one you had in January. If that sounds like how you would rather do it, Ori is here.

Sources

A gentle safety note

Ori supports general wellness. It is not medical care. If you have a history of disordered eating, are pregnant or breastfeeding, take medicines that affect weight or appetite, or have a clinical nutrition need, speak with a qualified healthcare professional before changing your intake or weight goal.

Read Ori's Health Disclaimer