Ketogenic Diet Shows Short-Term Depression Relief in Clinical Trial

A tightly supported ketogenic plan gave adults with treatment-resistant depression a modest short-term lift, but the edge weakened after the intensive coaching phase ended. In the DIME trial, 88 adults in the United Kingdom, ages 18 to 65, were randomized 44 to 44 between a ketogenic diet and a matched phytochemical diet, with the main question focused on change in PHQ-9 scores after six weeks.

The study, the first UK-based randomized controlled trial of ketogenic diet for treatment-resistant depression, ran from February 22 to June 15, 2024 and was later published in JAMA Psychiatry on February 4, 2026. Participants had PHQ-9 scores of 15 or higher and had already failed to respond adequately to at least two antidepressant trials, putting them in a group where new options are badly needed. Treatment-resistant depression is linked to poorer quality of life, higher healthcare costs and a higher suicide risk, which makes even a small signal worth watching closely.

The ketogenic arm was built for adherence. Participants received prepared meals, urine ketone strips, weekly individual dietetic support or coaching, and weekly health coach calls during the six-week intervention. Carbohydrate intake was kept below 30 grams a day, and calories were controlled so body weight stayed stable, allowing researchers to isolate the effect of diet composition rather than weight loss. The comparison group followed a weight-maintenance phytochemical diet centered on fruits, vegetables and healthier fat choices, with support designed to match the keto group.

The early numbers favored keto. Mean PHQ-9 improvement from baseline to week 6 was minus 10.5 points in the ketogenic group and minus 8.3 points in the phyto group, a between-group difference of minus 2.18 points, with a 95% confidence interval of minus 4.33 to minus 0.03 and a P value of .05. By week 12, that difference had narrowed to minus 1.85 points, with a 95% confidence interval of minus 4.04 to 0.33 and a P value of .10, no longer statistically significant. Anxiety scores at week 12 favored keto, but secondary outcomes overall did not separate the two diets.

Safety was not the issue. No serious adverse events were reported. The bigger challenge was sustainability, since adherence dropped after the structured support phase ended and participants were left to continue on their own. The NIHR Oxford Health Biomedical Research Centre called the findings cautiously optimistic while stressing that long-term durability remains unclear. For keto readers, the message is clear: this is an early clinical signal, not proof that ketosis has solved depression, but it is strong enough to keep metabolic psychiatry on the radar.