The claim
The "sardine fast" is simple: for 72 hours you eat nothing but sardines. A Daily Mail explainer by the dietitian Johannah Katz (7 November 2026) sums up what the trend promises:
Practitioners of a three-day sardine fast claims it leads to quick weight loss. … Supporters claim it can help people enter ketosis – when the body burns fat for energy instead of glucose – improve blood sugar, reduce inflammation and break through stubborn weight-loss plateaus.
The Daily Mail piece is not promoting the fast. It is skeptical, and most of what follows draws on it. The promotion comes from the trend's originator, Dr Annette Bosworth ("Dr Boz"). LADbible reports that she "claims that abiding by this diet for short periods can help you 'reset' your metabolism". In LADbible's paraphrase, she says the fast will send your body "into advanced ketosis to rapidly burn fat for energy". Her rules, as LADbible quotes them from the Diary of a CEO podcast: "The only thing on the menu for three days is sardines. There's no eating window. You can eat as [many sardines] as you want. There's no limit to the amount."
Taken apart, the claim makes four assertions:
- Three days of sardines will make you lose weight quickly.
- That loss is fat, burned through ketosis.
- The fast resets your metabolism, or does something beyond cutting calories.
- It improves blood sugar and related markers.
The first is true. The second and third are not supported. The fourth is partly true, but not for the reason the trend suggests.
The scale will drop
There is no dispute here. Eat only sardines for three days and you will almost certainly weigh less at the end. The version promoted by Dr Dominic D'Agostino, the researcher Dr Boz credits with the idea, is "two to three cans of sardines daily for three days, along with water and electrolytes", which the Daily Mail puts at "around 380 to 575 calories" a day. Dr Boz's version has no limit, so intake varies, but nobody eats a normal amount of food living on canned fish. The dietitian Bonnie Taub-Dix told the Daily Mail: "If you eat only sardines, you'll probably lose weight, but not because they're magical."
D'Agostino says he lost "around six pounds during his own three-day sardine fasts". That is one man's self-report, and he attaches caveats to it (see below). We have no other figures worth quoting. Noom says "Anecdotally, people report losing 2 to 5 pounds over a few days", but it labels that as anecdote and gives no source, so we have left it out. LADbible's case study, a woman who "claimed to lose 25lbs", ate sardines daily for 120 days, not three. That is one testimonial and says nothing about a three-day fast.
What the weight is made of
The trend says this quick loss is fat. The evidence says most of it is water.
The body stores carbohydrate as glycogen, and glycogen is stored "in the liver, muscles, and fat cells in hydrated form (three to four parts water)", according to a 1992 conference paper by Kreitzman and colleagues. Stop eating carbohydrate and you use up the glycogen, and the water bound to it goes too. The authors found that early changes in body composition on very-low-calorie diets "primarily reflect glycogen storage". They also warn of "exaggerated regain if carbohydrate loading occurs". We read only the abstract, which gives no figure for how many pounds this accounts for. The "a kilo or two of water" figure you will see online has no source in anything we read, so we don't repeat it.
The clearest comparison is an older study. Yang and Van Itallie (1976) put six obese people through ten days on an 800-calorie ketogenic diet and ten days on an 800-calorie mixed diet. Same calories, different carbohydrate. The ketogenic diet lost weight much faster: about 467 grams a day against 278. But the composition was different too:
| 800 kcal/day for 10 days | Water | Fat | Protein |
|---|---|---|---|
| Ketogenic diet | 61.2% | 35.0% | 3.8% |
| Mixed diet | 37.1% | 59.5% | 3.4% |
At the same calories, cutting out carbohydrate made the scale move faster, and the extra came mostly from water. That is exactly what a three-day sardine fast does. It is a small study, we have only the abstract, and the figures are ten-day averages rather than a day-by-day breakdown, so we won't turn them into a grams-of-fat estimate for three days. The direction is clear, though.
None of this means the loss is all water. Even on the ketogenic diet in Yang's study, about a third of the weight lost was fat, and a deficit this large will remove some fat in three days. The issue is how much and how long it lasts.
The Daily Mail and Noom say the same thing. The Daily Mail: "much of the initial weight change may come from water rather than body fat." Noom: "The weight loss that happens during a 3-day fast is largely water weight—it returns when carbs are reintroduced."
Ketosis is real; extra fat burning is not
Ketosis is real. Cut carbohydrate hard and the body turns to fat and makes ketones, and D'Agostino says nutritional ketosis may begin within 24 to 48 hours, though timing varies. The mistake is assuming ketosis means more fat loss than the calorie deficit alone would produce.
The best test of that we have is Hall et al. (2016). Seventeen men with overweight or obesity lived on a metabolic ward, ate a high-carbohydrate diet for four weeks, then a ketogenic diet with the same calories for four weeks. Body fat was measured by DXA scan and energy use in metabolic chambers. When the ketogenic diet began, the men lost "a rapid additional 1.6 ± 0.2 kg", which the authors judged "likely primarily the result of body water loss because fat mass decreased by only 0.2 ± 0.1 kg … over the next 15 d." Energy expenditure rose only briefly, by about 100 calories a day. Their conclusion: "The isocaloric KD was not accompanied by increased body fat loss." In fact, "Body fat loss slowed during the KD and coincided with increased protein utilization and loss of fat-free mass."
The study has real limits, and the authors name them. There was no control group, so "we cannot definitively claim that the KD per se was the cause". Only men took part. The diet kept calories level rather than cutting them hard as a sardine fast does. But on the question the trend raises, whether ketosis burns extra fat, it is the most direct evidence we have, and the answer is no.
D'Agostino agrees. He told the Daily Mail that "Sardines have no special fat-burning ingredient" and that "Over time, body fat loss is governed by the energy deficit." On the trend's central promise he was blunter: "'Reset your metabolism' is a marketing phrase, not a scientific one." His six pounds, he said, "reflects a combination of body fat, glycogen, water and reduced food in the digestive tract". He also "acknowledged that changes in hydration and glycogen can distort these measurements. Some weight also returns when normal carbohydrate intake resumes." It is to his credit that the man who inspired the trend gave it such a frank account.
Blood sugar and liver fat: a real effect, overstated
Here the trend has a point. Cutting carbohydrate hard does change liver fat and blood sugar quickly. D'Agostino cites Kirk et al. (2009), a randomized trial in 22 obese, insulin-resistant adults without diabetes. All food was supplied while they stayed in hospital. Both groups ate about 1,100 calories a day, one low in carbohydrate and one high. After 48 hours, liver fat had fallen 29.6% on the low-carbohydrate diet against 8.9% on the high-carbohydrate one. The liver's glucose output fell 23.4% against 7.2%. Those are real, measured results.
Three points keep them in proportion. First, this was not a sardine fast; the Daily Mail notes that "the study did not examine the sardine fast specifically." Second, these are biomarkers: they show the body changing, not that anyone became healthier. Muscle insulin sensitivity did not change at 48 hours. Third, the gap closed. Once both groups had lost 7% of their weight, about 11 weeks later, their liver fat reductions were similar (38.0% and 44.5%), and muscle insulin sensitivity had improved by a similar amount in both. The authors' summary: "48 h of calorie restriction affects the liver … whereas moderate weight loss affects muscle." The trial was also small, 11 people per group and mostly women.
D'Agostino again: "a lower blood sugar reading after three days doesn't mean insulin resistance has been cured."
On inflammation, none of the sources we read tested whether the fast reduces it, so we have not assessed that claim. On "stubborn plateaus", no trial has tested the fast at all. D'Agostino told the Daily Mail "no clinical trials have directly tested the challenge or established whether its benefits last beyond three days", and Noom says "There are no peer-reviewed studies specifically on the sardine fast."
Sardines are not the problem
Sardines are a good food. The Daily Mail notes that a 3.75oz can has about 23g of protein plus omega-3s, vitamin B12, vitamin D and, if you eat the bones, calcium, and that they are low in mercury compared with larger fish. Taub-Dix: "I'd rather see people add sardines to their diets", but "let's not confuse the benefits of these nutrients with promises of dramatic results in just 72 hours."
For most healthy adults, three days of sardines is unlikely to do harm. The Daily Mail says it is "unlikely to cause a significant vitamin deficiency in an otherwise healthy adult". The risks fall on specific groups. People on insulin or sulfonylureas risk low blood sugar. People on SGLT2 inhibitors risk euglycemic diabetic ketoacidosis, which "requires emergency medical care". Sardines are high in purines, which matters for people with gout, and in phosphorus and sodium, which matters for people with kidney disease or high blood pressure. The Daily Mail says pregnant and breastfeeding women, children, underweight people and anyone with a serious illness should not try it. If you are in any of these groups, talk to your doctor first.
Why D
The claim is not false, so it is not E. The scale will drop, some of that is fat, ketosis happens, and severe carbohydrate restriction really does change liver fat and blood sugar within two days.
The evidence is not split, so it is not C. On what the quick drop is made of, the sources agree: mostly water, released as glycogen runs down, and partly regained when you eat carbohydrate again. On whether ketosis adds fat loss beyond the calorie deficit, the best controlled evidence says it does not.
That makes it misleading. The trend takes a real, predictable drop on the scale and sells it as rapid fat burning and a metabolic reset. The researcher it cites as its inspiration rejects both ideas. The useful lesson goes beyond sardines: any three-day plan that cuts out carbohydrate will show a fast drop on the scale, and that drop is mostly water.