As noted in this review and meta-analysis, the ketogenic diet involves a metabolic shift so that the brain is no longer using sugar as its primary source of energy.

The review included studies of adults “from psychiatric or other medical populations.” Researchers included studies with dietary interventions that were high in fat, low in carbohydrates, and moderate in protein, as well as diets where carbohydrate intake was below a certain level.

All the included studies used at least one validated psychiatric scale to evaluate mental health outcomes, were written in English, and were some type of original, peer-reviewed quantitative study. Researchers also considered the risk of bias in relevant studies.

If studies reported enough information “to calculate standardized mean differences,” researchers included them in the meta-analysis.

They performed separate meta-analyses for the randomized controlled trials and quasi-experimental studies. They also cunducted a narrative synthesis to evaluate the outcomes on mental health, such as changes in participants’ symptoms.

The big difference was related to ketone monitoring. Ketones are produced when there is a lack of carbohydrate sources that can be used as energy.

Unless ketone monitoring was present, the effect of ketogenic diets on depressive symptoms was insignificant, but it was significant when ketone monitoring was present.

When a high-carbohydrate diet was not used as a control diet, there was a large and significant association between ketogenic diets and depressive symptoms. If a high-carbohydrate diet was part of the studies, there was no significant association.

While there was a lot of heterogeneity, results on depressive symptoms were also strongest for individuals without obesity.

Similarly, only following a very low-carbohydrate diet as opposed to a low-carbohydrate diet showed a significant association with improved depressive symptoms.

The presence of a low-fat control diet or intervention duration appeared to not make a difference in the association among subgroups.

When it came to studying anxiety, researchers looked at nine randomized controlled trials. They did not find an association between anxiety symptoms and ketogenic diets.

Next, researchers considered quasi-experimental studies. Nine of these studies looked at depression, and six looked at anxiety.

These results also showed ketogenic diets being associated with improved depressive symptoms, with a greater association when there was no ketone monitoring. However, this subgroup difference did not reach a level of significance.

The results for anxiety showed a possible benefit of ketogenic diets on anxiety symptoms.

Taking these differing results into account, the researchers suggested that it is possible that ketogenic diets may help with both depression and anxiety. However, the evidence suggests more pronounced benefits for depression.

Gregory Burzynski,MD, a concierge longevity physician at resTOR Longevity Clinic, who was not involved in this research, noted the following about the review’s findings to Medical News Today:

“This new JAMA meta-analysis adds real weight to what many of us see clinically: when patients achieve true nutritional ketosis, depressive symptoms often improve. The effect size in randomized trials is meaningful — comparable to what we expect from some antidepressants — and likely reflects deeper metabolic shifts that enhance mitochondrial function, reduce inflammation, and stabilize neurochemistry.”

This review and meta-analysis cannot establish causality, and the authors warned that the results should not be interpreted as uniform.

There were substantial differences between the included studies. For example, there were differences in intervention monitoring.

The researchers also noted that whether or not participants had structured support or dietary counseling could have impacted how well they followed diets, as well as the final results.

They further noted that there appears to be variation in the association between the symptom outcomes and the ketogenic diet response. The researchers suggested several reasons for this, such as how well people adhere to ketogenic diets.

There are also the limitations of the analyzed studies themselves, such as how many trials had short-term follow-up and small sample sizes. The researchers noted that they were also limited in determining the safety of ketogenic diets.

When it came to monitoring ketosis in randomized controlled trials, the differences in heterogeneity suggest “that monitoring alone did not account for variability.”

There is possible bias related to ketone monitoring and the bigger randomized controlled trial associations. Monitoring could occur with other factors that lead to inflated results, like higher levels of contact between clinicians and participants.

Self-reported outcomes were also common in the randomized controlled trials, introducing other problems like performance bias.

Researchers further note that factors like baseline depression severity “may introduce variability not fully addressed in the primary analysis.”

Finally, the criteria researchers used to find relevant studies, such as only focusing on adult participants and studies written in English, may limit the data they evaluated.

Even with all these limitations in mind, the review findings suggest that a ketogenic diet might help people who experience depression.

However, the authors did acknowledge that not all people who try this intervention will notice symptom improvement, and that it is unclear how this would all work in a real-world scenario.

It is important for people to discuss options with clinical experts to determine if this type of diet would be helpful in their specific situations.

Review author and Distinguished Professor in Psychiatry and Clinical and Translational Science at the School of Medicine, University of Pittsburgh, Mary Phillips, MD, MD (Cantab), explained to MNT that:

“The ketogenic diet can be considered as an intervention for individuals with depression. The diet can be used in combination with medications as need be. There are considerations to be taken into account, including blood test monitoring of ketones, electrolytes, and lipids, as the ketogenic diet can cause some imbalances in electrolytes and lipids in some individuals. These effects can be managed, however, with supplementation.”