Mark Hyman, MD · 2025-07-30 · Mark Hyman (host), Iain Campbell

Is Bipolar Disorder Really a Diet Problem?

46 research-tied claims examined: 2 contradicted 4 overstated 4 context 28 supported 8 unverified

28 Supported by research
0:02:20Mark Hyman (host)supportedhigh

Globally, almost one billion people suffer from some type of mental health issue.

"So many people suffer so terribly from mental health issues, and globally it's almost a billion people—a billion people that suffer from some type of mental health issue." (said at 0:02:20)

Global Burden of Disease (GBD) systematic analyses and World Health Organization reports establish that approximately one billion individuals worldwide live with a mental disorder. In previous iterations (such as GBD 2019), estimates were around 970 million people ('nearly one billion'), and more recent comprehensive estimates from GBD 2023 indicate approximately 1.17 billion prevalent cases globally across 12 major classes of mental disorders.

0:02:30Mark Hyman (host)supportedhigh

Depression affects almost 4% of the global population.

"And it could be anything from depression, which affects about almost 4% of the global population, anxiety, to bipolar, which is 1 or 2%, schizophrenia, substance use, eating disorders." (said at 0:02:30)

Global epidemiological analyses from the World Health Organization (WHO) and the Global Burden of Disease study estimate that depression affects approximately 3.8% of the total global population (including roughly 5% of adults).

0:02:30Mark Hyman (host)supportedmoderate

Bipolar disorder affects 1 to 2% of the global population.

"anxiety, to bipolar, which is 1 or 2%, schizophrenia, substance use, eating disorders." (said at 0:02:30)

Large-scale cross-national epidemiological data support the claim that bipolar disorder affects approximately 1% to 2% of the population. In the World Health Organization World Mental Health Survey Initiative across 11 countries (n = 61,392), the aggregate lifetime prevalence was 0.6% for bipolar I disorder and 0.4% for bipolar II disorder (totaling 1.0% for threshold bipolar disorder), while the total bipolar spectrum prevalence (including subthreshold cases) was 2.4% for lifetime and 1.5% for 12-month prevalence.

0:19:44Iain Campbellsupportedhigh

Clinical trials for psychiatric medications last a very short period of time compared to the duration that patients actually take them.

"The trials we do on medication last a very short period of time compared to the length of time that patients take them for." (said at 0:19:44)

Evidence demonstrates a substantial disparity between the duration of psychiatric clinical trials and the length of real-world patient use. A systematic analysis comparing 52 randomized controlled trials of major antidepressants with National Health and Nutrition Examination Survey (NHANES) real-world data found that the median duration of clinical trials was 8 weeks (IQR: 6–12 weeks), with 88.5% of trials lasting 12 weeks or fewer and none exceeding 52 weeks. In contrast, the median duration of real-world antidepressant use among patients in the United States was approximately 5 years.

0:19:55Iain Campbellsupportedhigh

Many of the mainstream medications for bipolar disorder directly lead to metabolic dysfunction.

"So many of the mainstream medications for bipolar directly lead to metabolic dysfunction" (said at 0:19:55)

Mainstream pharmacological treatments for bipolar disorder—including atypical (second-generation) antipsychotics (such as olanzapine and quetiapine) and mood stabilizers (such as valproate and lithium)—are well-documented causes of metabolic dysfunction. Clinical trials and reviews consistently show these agents independently induce substantial weight gain, insulin resistance, dyslipidemia, and increased risk for metabolic syndrome and type 2 diabetes through various physiological and molecular mechanisms.

0:20:00Mark Hyman (host)supportedhigh

Antipsychotic drugs used to treat bipolar and schizophrenia cause patients to overeat and develop diabetes or insulin resistance.

"the drugs that treat bipolar and schizophrenia are often antipsychotic drugs, which cause you to overeat and become diabetic or insulin resistant, which is what's causing the brain dysfunction in the first place." (said at 0:20:00)

Extensive randomized controlled trial evidence confirms that antipsychotic medications commonly prescribed for schizophrenia and bipolar disorder cause significant weight gain, impaired glucose control, insulin resistance, and an increased risk of diabetes. A 2025 systematic review and meta-analysis of 163 randomized clinical trials encompassing over 54,000 individuals (both patients and healthy volunteers) demonstrated that antipsychotic exposure significantly increased fasting glucose, fasting insulin, HbA1c, and the odds of hyperglycemia compared to placebo. A 2018 meta-analysis in healthy volunteers further confirmed that atypical antipsychotics directly decrease insulin sensitivity and increase weight independently of underlying psychiatric disease.

0:25:30Iain Campbellsupportedhigh

Many major molecular targets of lithium, such as the PI cycle, GSK-3, Akt, and mTOR, are components of the insulin signaling network.

"And I point out in this paper that many of the major targets of lithium are part of the insulin signaling network... the PI cycle, GSK-3, Akt, mTOR, these are all parts of the insulin signaling network." (said at 0:25:30)

The canonical molecular targets and downstream pathways modulated by lithium—including the phosphoinositide (PI) cycle, glycogen synthase kinase-3 (GSK-3), Akt (protein kinase B), and mTOR—are well-established components of the intracellular insulin and growth factor signaling cascade (specifically the PI3K/Akt/mTOR and GSK-3 signaling axes). Review literature directly identifies these targets as core elements integrating lithium's pharmacological actions with insulin signaling networks.

0:26:00Iain Campbellsupportedvery low

Martin Alda and his research team found that lithium modulates the insulin signaling pathway in brain organoids derived from bipolar disorder patient neurons.

"There was a study in Lancet journal EBioMedicine where Martin Alda, who developed the Alda scale for lithium, for example, and a team investigated this in—what, they make these kind of organoids, they're like mini-brains they derive from neurons from bipolar patients, and they found that lithium was modulating this insulin signaling pathway." (said at 0:26:00)

A 2024 study published in EBioMedicine (part of The Lancet Discovery Science), co-authored by Martin Alda's group, investigated induced pluripotent stem cell (iPSC)-derived neurons from patients with bipolar disorder. They demonstrated that lithium altered Akt signaling (a key effector downstream of the insulin signaling pathway) to rescue neuronal hyperexcitability in lithium-responsive bipolar disorder neurons. Because these findings are derived from in vitro cellular models, certainty regarding human clinical physiology is very low.

0:06:33Iain Campbellsupportedmoderate

Since the 1970s, very few new pharmacological treatments have been developed in psychiatry compared to other fields of medicine.

"Like you say, since the 1970s, there haven't been very many new treatments in psychiatry. In other fields of medicine, there's been many iterations of new medications. And in psychiatry, there's been very few new treatments and developments in this area." (said at 0:06:33)

The statement is supported by the literature on neuropsychiatric drug discovery. Reviews in psychopharmacology widely document a prolonged innovation drought following the initial serendipitous discoveries of major psychiatric drug classes (such as antipsychotics, monoamine-targeting antidepressants, and benzodiazepines) in the mid-20th century. Most subsequent drug development has consisted of incremental modifications or refinements of existing monoaminergic mechanisms (e.g., SSRIs and SNRIs from earlier TCAs and MAOIs, or atypical antipsychotics from first-generation D2 antagonists), with very few fundamentally novel pharmacological mechanisms reaching clinical practice compared to rapid mechanistic expansions in fields like oncology, immunology, and cardiology.

0:18:23Iain Campbellsupportedhigh

Tadafumi Kato published research linking mitochondrial dysfunction to the pathophysiology of bipolar disorder.

"this thing that you said about mitochondrial function and bipolar disorder, and you linked to this paper by Kato. So I started really looking into this and I felt like this makes sense to me. It feels like an energy disorder." (said at 0:18:23)

Tadafumi Kato and colleagues originally proposed the mitochondrial dysfunction hypothesis of bipolar disorder in 2000 and have published extensive basic, genetic, and animal-model research investigating this link. Their findings suggest that mitochondrial DNA deletions, impaired mitochondrial calcium handling, and accumulation of mutant mitochondrial DNA in brain regions such as the paraventricular nucleus of the thalamus contribute to the neurobiological pathophysiology of bipolar disorder.

0:26:53Mark Hyman (host)supportedhigh

Long-term lithium treatment for bipolar disorder can cause adverse effects on thyroid function.

"And there's other medications like lithium, which I think address aspects of insulin signaling that can be in the short term helpful for someone, but in the long term can lead to damage that leads to, you know, diabetes and so forth. HOST: Thyroid function and other things. Yeah." (said at 0:26:53)

A comprehensive systematic review and meta-analysis of 385 studies examining lithium toxicity in patients with mood disorders found that lithium treatment significantly increases the risk of hypothyroidism (odds ratio 5.78, 95% CI 2.00–16.67) and raises average thyroid-stimulating hormone (TSH) levels compared to placebo.

0:37:43Iain Campbellsupportedmoderate

Bipolar disorder affects approximately 1% to 2% of the population.

"I think bipolar disorder is an extreme version. You know, it's in 1, 2% of the population, but I think it indicates an underlying dynamic that's present in many people." (said at 0:37:43)

Large-scale international epidemiological surveys support that bipolar disorder affects approximately 1% to 2% of the general population. In the World Health Organization World Mental Health Survey Initiative across 11 countries (n = 61,392), the aggregate lifetime prevalence was estimated at 0.6% for bipolar I disorder and 0.4% for bipolar II disorder (totaling 1.0% for threshold bipolar disorder), extending to 2.4% when including subthreshold bipolar spectrum conditions.

0:41:05Iain Campbellsupportedmoderate

Systematic reviews show that bipolar depressive episodes peak around the winter solstice when photoperiod is at its lowest.

"And conversely, depression happens in the winter. So the same systematic review highlights winter depression and it occurs around the weeks of the winter solstice when photoperiod is at its lowest." (said at 0:41:05)

Systematic reviews of seasonality in bipolar disorder confirm that bipolar depressive episodes and related hospital admissions show a consistent seasonal peak in winter/early winter, coinciding with the shortest photoperiod and lowest solar insolation (around the winter solstice). Conversely, manic episodes show a peak in spring and summer.

0:36:10Mark Hyman (host)supportedhigh

The human genome contains approximately 20,000 genes and 5 to 7 million variations in the genetic code.

"We can see all your 20,000 genes, but we can see your 5 to 7 million variations in that genetic code." (said at 0:36:10)

The speaker's statement accurately reflects established human genomic data. The human genome contains roughly 19,000 to 20,000 protein-coding genes. Comprehensive sequencing initiatives, such as the 1000 Genomes Project, have shown that while tens of millions of variants exist across the global population, a typical individual's personal genome differs from the reference genome at approximately 4 to 5 million (and up to several million) variant sites, primarily single nucleotide polymorphisms and small indels.

0:48:47Iain Campbellsupportedhigh

Circadian clock genes such as CLOCK, BMAL1, PER, and CRY convert light signals via the suprachiasmatic nucleus to regulate metabolic and energy output daily and seasonally.

"And the central focus of chronopsychiatry research is CLOCK, the things you're mentioning: CLOCK, BMAL1, PER, CRY. And these are your circadian regulators and they basically take signals from the environment, the light, and they convert this in the suprachiasmatic nucleus of the brain, which is kind of like a clock system in the brain, and they use that to regulate your metabolism." (said at 0:48:47)

The speaker accurately describes the core mammalian circadian timing mechanism. The suprachiasmatic nucleus (SCN) functions as the master circadian pacemaker that receives environmental light input and coordinates peripheral clocks throughout the body to regulate metabolic homeostasis. At the molecular level, this oscillator is driven by a transcriptional-translational feedback loop consisting of the core circadian clock genes CLOCK, BMAL1, PER, and CRY.

0:50:10Iain Campbellsupportedmoderate

Metabolomic differences observed in bipolar patients involve altered insulin signaling networks, specifically the phosphatidylinositol cycle, Akt, and mTOR.

"And the things that we see again are these kind of insulin signaling type mechanisms: the phosphatidylinositol cycle, Akt, mTOR. And these are the kind of differences in bipolar patients in these metabolic signaling networks." (said at 0:50:10)

Published proteomic, transcriptomic, and lipidomic studies in bipolar disorder consistently demonstrate dysregulation across insulin-related signaling pathways, notably the phosphatidylinositol cycle, PI3K/Akt, and mTOR networks. A large systematic review of peripheral proteomics across major psychiatric disorders identified significant alterations in the PI3K-Akt and Focal Adhesion-PI3K-Akt-mTOR signaling cascades, while lipidomic profiling has shown phosphatidylinositols to be among the most prominently altered lipid classes in bipolar disorder patients.

0:53:45Iain Campbellsupportedlow

A pilot study at the University of Edinburgh showed significant reductions in blood lactate in bipolar disorder patients following a ketogenic diet.

"We saw significant reductions in blood lactate on a ketogenic diet in bipolar patients. And this has been a marker for, you know, many decades in psychiatry that's been of interest, and we saw it move in these patients." (said at 0:53:45)

A pilot study led by researchers at the University of Edinburgh evaluated a 6- to 8-week modified ketogenic diet in euthymic patients with bipolar disorder (27 recruited, 20 completing the protocol). The investigators assessed clinical, metabolic, and magnetic resonance spectroscopy (MRS) biomarkers before and after the dietary intervention. However, because this was an open-label, single-arm pilot study with a small sample size and no control group, the overall certainty of the evidence is low.

0:57:02Mark Hyman (host)supportedmoderate

One in four teenagers in the United States has prediabetes or type 2 diabetes.

"I mean, that's terrifying considering, like, one in four teenagers has prediabetes or type 2 diabetes" (said at 0:57:02)

Nationally representative surveillance data from the National Health and Nutrition Examination Survey (NHANES) support the claim that approximately one in four (or more) adolescents in the United States have prediabetes or diabetes. While earlier NHANES cycles (2005–2016) reported an adolescent prediabetes prevalence of approximately 18% (nearly 1 in 5), more recent analyses demonstrate that the prevalence among adolescents aged 12–19 has risen to over 30% (approximately 1 in 3) when assessed by standard fasting plasma glucose and HbA1c criteria.

0:59:35Iain Campbellsupportedhigh

The human body stores approximately 2,000 calories of carbohydrate as glycogen, while fat tissue can store well in excess of 100,000 calories.

"you can store about 2,000 calories of glycogen as sugar in your body, but you can store well in excess of 100,000 calories as fat." (said at 0:59:35)

The speaker's statement accurately reflects established human physiological energy storage capacities. Human glycogen storage in liver and skeletal muscle is limited to approximately 400 to 500 grams (roughly 15 g/kg body weight), which corresponds to about 1,600 to 2,000 kcal of carbohydrate energy (at 4 kcal/g). In contrast, adipose tissue stores triglycerides with an energy density of approximately 9 kcal/g (or ~7,700 kcal/kg of adipose tissue), meaning an average adult with 12 to 15 kg of body fat possesses between 90,000 and 135,000 kcal of stored lipid energy, with values well exceeding 100,000 kcal in typical or overweight individuals.

1:06:13Mark Hyman (host)supportedhigh

The Mayo Clinic developed the ketogenic diet in 1921 as a treatment for epilepsy.

"Mayo Clinic in 1921 uh developed it for epilepsy, where there's elevated glutamate excitatory activity in the brain." (said at 1:06:13)

Historical medical records and neurological reviews confirm that the ketogenic diet was formulated and named in 1921 by Dr. Russell Wilder at the Mayo Clinic as a treatment for epilepsy to mimic the metabolic and anticonvulsant effects of fasting. Epilepsy is characterized by neuronal hyperexcitability and imbalances between excitatory neurotransmitters (such as glutamate) and inhibitory neurotransmitters (such as GABA), which the ketogenic diet helps modulate.

1:06:20Iain Campbellsupportedhigh

Carbohydrate restriction was used at the Joslin clinic in the 1900s and early 1920s to treat type 1 diabetes before insulin was developed.

"There's also the treatment for type 1 diabetes back in the day in Joslin in the 1900s and early uh in the early '20s before insulin was developed." (said at 1:06:20)

Historical and medical literature confirms that prior to the isolation and clinical availability of insulin in the early 1920s, carbohydrate and strict dietary caloric restriction (often referred to as the Allen–Joslin fasting or undernutrition regimen) was the primary clinical management strategy promoted by Dr. Elliott P. Joslin to prolong the lives of patients with juvenile (type 1) diabetes.

1:06:52Iain Campbellsupportedhigh

The anti-seizure medications lamotrigine, valproate, and carbamazepine are used to treat bipolar disorder.

"And the epilepsy treatments um are things like lamotrigine, valproate, carbamazepine— HOST: These are used for bipolar disease. GUEST1: These are all used for bipolar. Yeah, absolutely." (said at 1:06:52)

Lamotrigine, valproate (sodium valproate/divalproex), and carbamazepine are established antiseizure medications that are standard, guideline-endorsed treatments (mood stabilizers) for bipolar disorder. Systematic reviews and large clinical cohort studies demonstrate their routine use and efficacy in preventing relapses and hospitalizations across phases of bipolar disorder.

1:08:00Iain Campbellsupportedvery low

In a 20-patient pilot study using magnetic resonance spectroscopy, bipolar patients on a ketogenic diet showed an 11% to 13% reduction in brain glutamate in 6 to 8 weeks.

"Um so in the patients in our pilot trial, we saw significant reductions in brain glutamate, about 10 um about 11 to 13% reductions... and so this was um um magnetic resonance spectroscopy looking at um uh metabolites that change in the brain... and the reduction we saw was about, like I say, 11 to 13%. And the reductions you typically see in, like, three months, for example, on lamotrigine, but we were seeing much larger reductions in six to eight weeks of a ketogenic diet." (said at 1:08:00)

A 6- to 8-week open pilot study of a modified ketogenic diet in individuals with bipolar disorder (27 recruited, 20 completers) evaluated brain neurometabolites using magnetic resonance spectroscopy (MRS). The study reported that brain glutamate plus glutamine (Glx) concentrations decreased significantly by 11.6% in the anterior cingulate cortex (P = 0.025) and by 13.6% in the posterior cingulate cortex (P < 0.001). Because this was a small, open-label, single-arm pilot trial without a randomized control group, the GRADE certainty of the evidence is very low.

1:10:02Iain Campbellsupportedhigh

Glutamate is present in the brain in millimolar concentrations, whereas monoamine neurotransmitters are in micromolar concentrations.

"It's in millimolar concentrations, and all many of the other monoamines are only in micromolar concentrations. And so it's a really abundant neurotransmitter" (said at 1:10:02)

Glutamate is the most abundant excitatory neurotransmitter in the central nervous system. Total brain tissue/intracellular glutamate concentrations are in the millimolar range (typically 5 to 15 mM), whereas classic monoamine neurotransmitters (such as dopamine, serotonin, and norepinephrine) are present at orders-of-magnitude lower concentrations (micromolar or nanomolar range). Extracellular glutamate is tightly cleared by transporters to maintain resting extracellular levels in the low micromolar range to prevent excitotoxicity.

  • supports: Chronic Glutamate Toxicity in Neurodegenerative Diseases-What is the Evidence? (Frontiers in neuroscience 2015) · cited 725x in the literature
    "Together with aspartate, glutamate is the major excitatory neurotransmitter in the brain. Glutamate binds and activates both ligand-gated ion channels (ionotropic glutamate receptors) and a class of G-protein coupled receptors (metabotropic glutamate receptors). Although the intracellular glutamate concentration in the brain is in the millimolar range, the extracellular glutamate concentration is kept in the low micromolar range by the action of excitatory amino acid transporters that import glutamate and aspartate into astrocytes and neurons." (abstract, background, passage verified)
    pubmedfull study (doi)
1:15:40Iain Campbellsupportedmoderate

The Medical Research Council and Baszucki foundation co-funded a £4 million program to establish a metabolic psychiatry hub in the UK.

"And so we were funded um uh about £4 million, and the Baszuckis also provided co-funding for it, um to really start a large government-funded program to investigate uh metabolic health and mental illness" (said at 1:15:40)

Published study protocols from the UK confirm the establishment and operation of the Hub for Metabolic Psychiatry (such as the METPSY prospective observational study investigating metabolic biomarkers and clinical outcomes in young adults with severe mental illness) dedicated to evaluating the links between metabolic dysfunction and psychiatric conditions.

1:12:04Mark Hyman (host)supportedmoderate

Psychiatric drugs are the second or third most prescribed class of medications, following statins and gastrointestinal drugs.

"It's it's in fact, I think it's the third leading class of drugs that are prescribed out there after statins and GI drugs, or maybe a second—it goes back and forth between GI drugs and psychiatric drugs." (said at 1:12:04)

Nationally representative prescription drug monitoring data in the United States show that lipid-lowering medications (antihyperlipidemics/statins), gastrointestinal agents (such as proton-pump inhibitors), and psychiatric medications (predominantly antidepressants) consistently rank among the top prescribed drug classes among adults alongside antihypertensive agents.

1:09:13Mark Hyman (host)supportedhigh

Glutamate signaling is a therapeutic target in the medical treatment of Alzheimer's disease.

"Glutamate is a very excitatory neurotransmitter. It it's it's a It's also um the target of of treatment in Alzheimer's." (said at 1:09:13)

Glutamate neurotransmission is an established therapeutic target in the pharmacological management of Alzheimer's disease. Memantine, an uncompetitive NMDA (N-methyl-D-aspartate) glutamate receptor antagonist designed to mitigate excitotoxicity, is an approved and widely utilized medical treatment for moderate-to-severe Alzheimer's disease.

1:09:53Mark Hyman (host)supportedhigh

Clinical trials have evaluated intranasal insulin for the treatment of depression and cognitive impairment.

"It's pretty amazing when you see these trials, like even using things like intranasal insulin for depression and cognitive impairment. Right?" (said at 1:09:53)

Multiple randomized controlled clinical trials have evaluated intranasal insulin for cognitive impairment (including amnestic mild cognitive impairment and Alzheimer's disease) as well as for major depressive disorder.

Unverified means no publication matching the claim was located; it does not prove the claim false. Spotted an error? See the corrections policy - disputes from the people quoted are prioritized.