Study analysis · JAMA Psychiatry · 2026

This diet slashed depression in 6 weeks—then vanished. Here’s why ketones had nothing to do with it.

A super strict low-carb diet made people with severe depression feel a little better for 6 weeks, but then it wore off—and it wasn’t because of ketones.

Reading level
Moderate certainty
Level 1b · Individual RCT

Overview

What the study found

The study in plain English — the bottom line, every takeaway we extracted, and what to do with them.

In simple terms

This study is like a fair test where two groups of people with serious depression tried different diets. One group ate very low-carb food, and the other ate more veggies and healthy fats. The low-carb group felt a little better after 6 weeks, but not by much—and after 12 weeks, both groups felt the same. So we can't say the low-carb diet definitely caused the improvement.

What’s the bottom line?

Scientists tested if a strict low-carb diet (ketogenic diet) helps people with depression that didn't improve with medicine, compared to a healthy plant-based diet.

How strong is this study?

This study was pretty well done because it randomly assigned people to groups, made sure both groups got the same amount of help, and followed up with almost everyone. But it gave the low-carb group free meals and more money, which might have made people feel better for reasons other than the diet. So while it's one of the best types of studies, we still need to be careful about trusting the results too much.

Reporting

0 / 100

  • COI disclosureconflicts of interest not disclosed
  • Data availabilitydata not shared
  • Code availabilitycode not shared
Methodology

76 / 100

  • Randomization+20/20
  • Blinding+9/15
  • Control group+15/15
  • Sample size (n=88)+7.1/20
  • Follow-up+10/10
Publication

100 / 100

Statistical

100 / 100

  • P-values+15/15
  • Effect size+20/20
  • Confidence intervals+15/15
  • Pre-registration+15/15

Each component is scored out of 100 and then capped by the study design — a case series cannot reach the ceiling a randomised trial can, however well it is reported.

Where it sits

RCT reviews

Max 100

Randomized Trials

Max 90

Reviews of Cohort Studies

Max 85

Cohort Studies

Max 72

Reviews of Case-Control Studies

Max 63

Case-Control Studies

Max 58

Cross-Sectional & Case Series

Max 50

Expert Opinion

Max 5
StrongerWeaker
Randomized Trials
Level 1b
72

72 / 100

Probability of being correct

Participants are randomly assigned to treatment or control groups, minimizing bias. The gold standard for testing whether an intervention causes an effect.

This design can establish causation. Although this is a randomized controlled trial, the effect size is small, the primary outcome is borderline significant (p=0.05), and the effect disappears at 12 weeks and in sensitivity analyses using alternative models. The control group also received substantial support, making it difficult to isolate the specific effect of ketosis. Therefore, while causation is plausible, it is not definitively established.

No Conflicts

No conflicts of interest identified

No conflicts of interest or funding sources were disclosed in the provided text, and there is no evidence of industry involvement or author affiliations that suggest bias.

Independent Analysis Safeguards

  • The data analyst (M.G.) did not deliver the intervention.
  • Statistical analysis plan was finalized prior to database locking.
  • Analyses were conducted using Stata and R with intention-to-treat and prespecified methods.
  • Sensitivity analyses were performed to assess robustness of results.

Although the study appears methodologically rigorous with blinding, pre-specified analysis, and independent data analysis, the absence of any declared funding or conflict of interest section raises a minor transparency concern. However, without evidence of industry ties or funder influence, the overall risk of bias is assessed as low.

Key takeaways

  1. 01

    After 6 weeks, people on the low-carb diet felt slightly better (PHQ-9 score dropped 2.2 points more than the other group).

  2. 02

    But after 12 weeks, both groups felt the same.

  3. 03

    The better mood only happened in people with very severe depression (score ≥20), and it had nothing to do with ketones in their urine.

  4. 04

    A 2-point drop on the PHQ-9 scale is small — it’s not enough to feel dramatically better for most people, and it didn’t last.

Surprising findings

  • The control group improved just as much as the keto group over time—both dropped over 8 points on the PHQ-9 scale.People assume the keto group’s improvement was due to diet—but the control group (just eating more veggies and swapping fats) improved nearly as much, suggesting the real driver was attention, hope, or structure.
  • Only 20% of keto participants kept the diet after support ended—even though they had free meals and coaching.Keto advocates claim it’s sustainable—but here, even with perfect support, 80% quit within 6 weeks. This isn’t a lifestyle; it’s a short-term experiment.

Practical takeaways

If you have severe treatment-resistant depression (PHQ-9 ≥20), consider a short-term, supervised keto trial—only if you can get free meals and weekly coaching.

The benefit fades after 6 weeks, doesn’t improve anxiety or cognition, and 80% of people quit after support ends. It’s not a long-term solution.

medium confidence

Why this study matters

The 6-Week Miracle That Faded

In an 88-person trial, people on a ketogenic diet (<30g carbs/day) saw a 2.18-point greater drop in depression scores (PHQ-9) than those on a plant-rich diet at 6 weeks (P=0.05). But by 12 weeks, the difference disappeared—both groups felt the same.

People think keto fixes mental health long-term—but this study proves the effect is fleeting, even with perfect adherence and free meals. It’s not a cure, just a temporary bump.

Only Severe Cases Benefited

The diet’s benefit was almost entirely in people with the worst depression (PHQ-9 ≥20), where scores dropped 4.73 points more than the control group. Those with moderate depression (PHQ-9 15–19) saw zero improvement.

If you’re mildly depressed, keto won’t help—but if you’re in crisis, it might give you a short lifeline. This isn’t a universal fix; it’s a targeted tool for the most desperate cases.

Ketones Didn’t Cause the Mood Boost

Despite measuring urine ketones twice weekly, researchers found zero correlation between ketone levels and mood improvement. People with high ketones didn’t feel better than those with low or none.

The whole keto-for-brain theory hinges on ketones—but this study blows that up. The benefit came from something else: maybe the structure, the support, or the placebo effect.

Want the whole report?

Detailed mode opens the full scientific breakdown — every score component, the methodology, conflicts of interest, the evidence analysis behind each claim, and the raw study data.