What does the research show about atkins diet?

Atkins diet

Trials that tested Atkins by name found weight loss within a few kilograms of other popular diets, more than Zone in one trial and similar in another. Triglycerides fell more than on higher-carbohydrate diets in two trials. The modified Atkins diet reduced seizures in drug-resistant epilepsy. Thiamine, folate, vitamin C, iron and magnesium intake fell short more often.

Add to Bio HackChecked 2026-11-13

Status: A dietary pattern and a commercial program, not a regulated product.
Also called: Atkins, Dr. Atkins diet, Atkins 20, Atkins 40, modified Atkins diet, MAD, low-carbohydrate high-protein high-fat diet
Followed for: weight loss, triglycerides, heart health, blood sugar, epilepsy

Strong several reviews agree · Moderate one review or consistent trials · Limited one or a few small studies · Untested no study in people, listed with what it rests on

Shown to help2

ModerateShown to make a 50% seizure reduction about six times as likely as drug therapy alone in drug-resistant epilepsy, though less often than the classic ketogenic diet in children.Very common
ModerateShown in two trials to lower triglycerides more than higher-carbohydrate diets in adults with overweight, while a third found similar heart risk changes across four diets.

Reported risks8

LimitedLimited evidence: in one trial, 25% of women on Atkins had LDL cholesterol rise more than 10%, against 13% on a high-carbohydrate diet and 3% on Zone.Very common
LimitedLimited evidence: in epilepsy trials, constipation was reported in about 17%, lethargy in 11% and loss of appetite in 12% on the modified Atkins diet.Very common
LimitedOne trial found more women on Atkins shifted to thiamine intakes linked to risk of inadequacy after 8 weeks.How often is not known
LimitedOne trial found more women on Atkins shifted to folic acid intakes linked to risk of inadequacy after 8 weeks.How often is not known
LimitedOne trial found more women on Atkins shifted to vitamin C intakes linked to risk of inadequacy after 8 weeks.How often is not known
LimitedOne trial found more women on Atkins shifted to iron intakes linked to risk of inadequacy after 8 weeks.How often is not known
LimitedOne trial found more women on Atkins shifted to magnesium intakes linked to risk of inadequacy after 8 weeks.How often is not known
LimitedLimited evidence: in a 1-year trial, 47% of people assigned to Atkins dropped out, against 35% on Zone or Weight Watchers and 50% on Ornish.How often is not known

What people follow it for, and what was found

Shown to help2

ModerateVery common

Seizures in drug-resistant epilepsy (modified Atkins)

Shown to make a 50% seizure reduction about six times as likely as drug therapy alone in drug-resistant epilepsy, though less often than the classic ketogenic diet in children.

Compared with a usual diet with the same antiseizure drugs [1]; standard care with no diet therapy [2]; usual care [3]; standard drug therapy alone [4]; the classic ketogenic diet and a low glycemic index diet [5]; the classic ketogenic diet [6]

The numbers and sources

How common: Very common: more than 1 in 10 people.
Effect: In 6 RCTs (575 patients), the modified Atkins diet added to drug therapy made a 50% or greater seizure reduction more likely than drug therapy alone (RR 6.28, 95% CI 3.52 to 10.50), in children (RR 6.28) and adults (RR 6.14, 95% CI 1.15 to 32.66), and made seizure freedom more likely (RR 5.94, 95% CI 1.93 to 18.31) [1]. In adolescents and adults, 23% on the diet had a 50% reduction (95% CI 10% to 37%), RR 6.47 (95% CI 1.60 to 26.14) [2]. A Cochrane review rated the adult evidence very uncertain (RR 5.03, 95% CI 0.26 to 97.68, 2 studies), and found up to 25% of children seizure free and up to 60% with a 50% reduction [3]. In 160 adolescents and adults over 6 months, 26.2% had a 50% reduction against 2.5% on drugs alone [4]. In children, median seizure frequency fell 45% on the modified Atkins diet and 66% on the classic ketogenic diet, and the modified Atkins diet did not meet the trial's noninferiority margin [5]. A 2026 meta-analysis found the classic diet more likely to halve seizures (OR 0.55 for the modified Atkins diet, 95% CI 0.35 to 0.87), with no difference in seizure freedom [6].
Population: Children and adults with drug-resistant epilepsy on antiseizure medication [1][2][3][4]; 170 children aged 1 to 15 in India [5]; children in 8 studies, experimental and observational [6].
Sources: [1], [2], [3], [4], [5], [6]
Caveats: Average follow-up was 12 weeks [1]. Only 3 RCTs were in the adolescent and adult review [2], and its lead author also led trial [4]. Cochrane certainty was very low [3]. This is a medically supervised version used for epilepsy, not the weight-loss program.

Moderate

Triglycerides and HDL cholesterol

Shown in two trials to lower triglycerides more than higher-carbohydrate diets in adults with overweight, while a third found similar heart risk changes across four diets.

Compared with the Zone, LEARN and Ornish diets [7]; the Zone, Weight Watchers and Ornish diets [8]; a conventional low-calorie, high-carbohydrate, low-fat diet [9]; a high-carbohydrate, high-fibre diet and the high-protein Zone diet [11]

The numbers and sources

Effect: HDL rose and triglycerides fell more on Atkins than on a conventional diet through most of a year, with no difference in total or LDL cholesterol after 3 months [9]. Triglycerides fell 0.30 mmol/L more than on a high-carbohydrate diet in insulin-resistant women [11]. At 12 months, Atkins secondary outcomes, including lipids, were comparable with or more favorable than Zone, LEARN and Ornish [7]. In another trial, each of four diets lowered the LDL to HDL ratio by about 10%, with no significant change in blood pressure or glucose [8].
Population: Adults with obesity [9]; insulin-resistant women [11]; premenopausal women [7]; adults with cardiac risk factors [8].
Sources: [7], [8], [9], [11]
Caveats: Small trials with high dropout. Blood markers, not heart attacks or strokes.

Mixed results1

Moderate

Weight loss

Mixed results: Atkins produced more weight loss than Zone in one trial and similar loss to other popular diets in another, with gaps shrinking by 12 months.

Compared with the Zone, LEARN and Ornish diets [7]; the Zone, Weight Watchers and Ornish diets [8]; a conventional low-calorie, high-carbohydrate, low-fat diet [9]; the Zone diet and other named diet programs [10]; a high-carbohydrate, high-fibre diet and the high-protein Zone diet [11]

The numbers and sources

Effect: At 12 months, women lost 4.7 kg on Atkins (95% CI 3.1 to 6.3), 1.6 kg on Zone, 2.6 kg on LEARN and 2.2 kg on Ornish, and only the Atkins and Zone difference was significant [7]. In another 1-year trial, mean loss was 2.1 kg on Atkins, 3.2 kg on Zone, 3.0 kg on Weight Watchers and 3.3 kg on Ornish, and weight loss tracked self-reported adherence (r 0.60), not diet type (r 0.07) [8]. Against a conventional low-fat diet, Atkins produced more loss at 3 months (6.8% against 2.7% of body weight) and 6 months (7.0% against 3.2%), and the 12-month difference (4.4% against 2.5%) was not significant [9]. In a network meta-analysis, Atkins produced 1.71 kg more loss than Zone at 6 months, and differences between named diets were minimal [10]. In insulin-resistant women, Atkins produced 2.8 kg more loss than a high-carbohydrate diet [11].
Population: 311 premenopausal women with overweight or obesity [7]; 160 adults aged 22 to 72 with hypertension, dyslipidemia or high fasting glucose [8]; 63 adults with obesity [9]; 7,286 adults in 48 trials [10]; 96 insulin-resistant women [11].
Sources: [7], [8], [9], [10], [11]
Caveats: Adherence was poor and dropout high in several trials [8][9]. Most trials ran 12 months or less.

Not shown to help1

Moderate

Blood sugar and insulin

Not shown to change blood sugar or insulin more than other diets in adults with overweight in three trials.

Compared with the Zone, Weight Watchers and Ornish diets [8]; a conventional low-calorie, high-carbohydrate, low-fat diet [9]; a high-carbohydrate, high-fibre diet and the high-protein Zone diet [11]

The numbers and sources

Effect: No diet, Atkins included, significantly changed glucose at 1 year [8]. The insulin response to a glucose drink fell on both Atkins and a conventional diet [9]. Insulin fell on all three diets in insulin-resistant women, and was the one marker that did not fall more on Atkins than on a high-carbohydrate diet [11].
Population: Adults with cardiac risk factors [8]; adults with obesity [9]; insulin-resistant women without diabetes [11].
Sources: [8], [9], [11]
Caveats: None of these trials studied people with diabetes.

How trials defined it

The trial abstracts describe Atkins as carbohydrate restriction and do not give gram limits [8]. In one trial, people got weekly instruction for 2 months, then 10 months of follow-up [7]. In another, they made a maximum effort for 2 months and then chose their own level of adherence [8]. One trial kept professional contact minimal to mirror how most people diet [9]. In insulin-resistant women, advice covered food choices without an energy target, over 8 weeks of weight loss, 8 weeks of maintenance and 8 weeks without contact [11]. In epilepsy, the modified Atkins diet was added to antiseizure drugs, with an average follow-up of 12 weeks [1] and up to 6 months [4].

Risks

Reported risks7

LimitedVery common

LDL cholesterol rise

Limited evidence: in one trial, 25% of women on Atkins had LDL cholesterol rise more than 10%, against 13% on a high-carbohydrate diet and 3% on Zone.

The numbers and sources

How common: Very common: more than 1 in 10 people.
Effect: LDL fell on high-carbohydrate and Zone diets but fluctuated on Atkins, ending 0.28 mmol/L higher than on Zone (95% CI 0.04 to 0.52). LDL rose more than 10% in 25% on Atkins, 13% on the high-carbohydrate diet and 3% on Zone [11]. In adults with epilepsy, one trial reported raised cholesterol on the modified Atkins diet [3]. Another trial found no difference in total or LDL cholesterol after 3 months [9].
Population: 96 insulin-resistant women [11]; adults with drug-resistant epilepsy [3]; 63 adults with obesity [9].
Sources: [3], [9], [11]
Caveats: Small trials that disagree. No trial measured heart attacks or strokes.

LimitedVery common

Digestive and other side effects (modified Atkins)

Limited evidence: in epilepsy trials, constipation was reported in about 17%, lethargy in 11% and loss of appetite in 12% on the modified Atkins diet.

The numbers and sources

How common: Very common: more than 1 in 10 people.
Effect: Across 5 RCTs reporting adverse events, constipation occurred in 17% (95% CI 5% to 44%), lethargy in 11% and anorexia in 12% [1]. In children, treatment-related adverse events occurred in 56.9% on the modified Atkins diet and 56.4% on the classic ketogenic diet [5]. Gastrointestinal and lipid side effects did not differ between the two diets in a 2026 meta-analysis [6].
Population: Children and adults with drug-resistant epilepsy [1][5][6].
Sources: [1], [5], [6]
Caveats: Pooled rates from the diet arms, not differences from control.