Keto diet helps you lose more weight than other diets

People lose weight on keto, and in trials that compare it with other weight-loss diets it comes out ahead: by about 1.5 to 3.5 kg (roughly 3 to 8 pounds) in pooled analyses of strict ketogenic diets, and by about 1 kg (roughly two pounds) in analyses limited to a year or more of follow-up. The largest, a 2025 analysis of 174 randomized trials of carbohydrate-restricted diets, found 3.36 kg more weight loss across its 28 ketogenic comparisons than on higher-carbohydrate diets. A 2022 Cochrane analysis of 61 randomized trials in 6,925 adults with overweight or obesity, most of them testing moderate carbohydrate restriction, found that low-carbohydrate diets produced 1.07 kg more weight loss than balanced-carbohydrate diets over three to about eight months and 0.93 kg more over one to two years, which the reviewers described as probably little to no difference. A 2013 meta-analysis limited to strict ketogenic diets (50 grams of carbohydrate a day or less) found 0.91 kg more over a year or longer compared with low-fat diets, about 1% of body weight by its authors' own arithmetic. These figures leave out very-low-calorie ketogenic programmes of roughly 600 to 800 calories a day, which this claim does not cover.

The sources also report safety findings. The prescribing label for empagliflozin (Jardiance), one of the SGLT2 inhibitor drugs used for type 2 diabetes, lists a ketogenic diet among the conditions that can precipitate ketoacidosis, a medical emergency that has been fatal in people taking these drugs. Several trials also found LDL cholesterol rose more on keto than on the comparison diet. The 2025 pooled analysis put that rise at about 13 mg/dL across its ketogenic comparisons, and found that lean mass fell about 0.7 kg more on keto than on comparison diets. The label's warning applies to people taking SGLT2 inhibitors; the sources here do not test keto in people on these drugs.

Some of the early edge reflects how long people stay on the diet. In the 2013 review, carbohydrate intake at the end of most trials was higher than the ketogenic limit allowed, and adherence had been better early on. When longer follow-up data from one trial were substituted, the advantage fell to 0.73 kg and was no longer statistically clear. The 2025 pooled analysis, looking across all the carbohydrate-restricted diets it covered, did not find that longer trials showed smaller weight differences.

When calories are matched on paper, a small gap remains in short trials. A 2026 meta-analysis of six trials in 259 adults, each prescribing keto and a higher-carbohydrate diet at comparable energy intake, found 1.49 kg more loss on keto, rated low certainty. In the three trials that also matched protein, the difference was compatible with none, and the review did not pool body-composition data, so it cannot say whether the extra loss was fat. The authors of the 2025 pooled analysis reported similar body-composition results whether or not the diets being compared were matched for calories.

Two tightly controlled NIH studies tested what the carbohydrate-insulin model predicts. In a National Institutes of Health inpatient study, 20 adults ate as much as they wanted of a ketogenic diet and of a low-fat, plant-based diet for two weeks each, and they ate 689 calories a day more on keto. In an earlier metabolic-ward study of 17 men, switching to keto at the same calorie intake raised daily energy expenditure by roughly 57 to 151 calories depending on how it was measured, while body fat loss slowed rather than sped up.

Against a Mediterranean-style diet the results differ from one comparison to the next. In a 12-week crossover trial of 33 adults with prediabetes or type 2 diabetes, weight fell 8% on a well-formulated ketogenic diet and 7% on a Mediterranean-style diet, with the result depending on which diet came first. LDL cholesterol rose 10% on keto and fell 5% on the Mediterranean diet, and follow-up data suggested the Mediterranean diet was easier to sustain. In a 2025 trial in Spain that assigned 160 adults with obesity to five calorie-restricted diets for three months, the 32 on keto lost 11.9 kg and the 32 on a Mediterranean diet lost 8.4 kg, a difference of 3.78 kg. Bioimpedance measurements did not show significantly more fat loss on keto, and lean mass and LDL cholesterol did not differ. That trial was registered about three months after it began, with gut bacteria as its main outcome and weight as a secondary one. A 2026 pooled analysis of randomized trials comparing the two diets found 3.56 kg more weight loss on keto and no difference in LDL cholesterol, and its authors described the advantage as short term.

The UK's national obesity guideline committee reviewed trials comparing low-carbohydrate and conventional, typically low-fat diets in 2025, found the evidence generally could not tell them apart, and did not recommend any particular macronutrient pattern, noting that any approach has to lower energy intake. All of this comes from adults with overweight or obesity, mostly in high-income countries. None of these sources speaks to pregnancy, children or type 1 diabetes.