Bipolar disorder (BD) is a chronic, often disabling illness, and many individuals remain symptomatic despite pharmacotherapy. Significant mood variability often persists throughout the lifespan and predicts relapse, leading to functional impairment. Metabolism of dietary essential polyunsaturated fatty acids has been shown to be upstream of the neuroinflammatory processes that may lead to neurotoxicity and chronicity of illness in BD. The investigators hypothesize that an intervention diet designed to alter intake of polyunsaturated fatty acids that augments mood stabilizing medications will reduce inflammation; and that the reduction of inflammation will reduce mood variability in bipolar disorder. After a two-the investigatorsek baseline-monitoring period, the investigators will randomize individuals with BD to an intervention or a control diet. Mood will be measured daily using a smartphone. Phase 2 will consist of 12 the investigatorseks of a less intense intervention. Follow-up will then be completed at 6, 9, and 12 months post-baseline to assess for recurrence of mood episodes. By maintaining a certain diet in addition to taking mood-stabilizing medication, researchers hope to see whether specific dietary plans have any bearing on mood variability.

The investigators will conduct a 2-arm, parallel group, randomized, controlled 24-week dietary intervention to evaluate the therapeutic efficacy of two dietary interventions in patients with BD. After a two-week baseline-monitoring period, the investigators will randomize 84 patients with chronic BD to augment usual treatment with either an experimentally-altered omega-3, omega-6 intervention or a control diet with average US intakes of n-3 and n-6 fatty acids. Subjects in both groups will remain under the care of their physician for the full duration of the trial. During the first phase of the trial (12 weeks), the intervention will be intense, during the second phase (12 weeks), a less intense intervention will be delivered, and after the two phases of intervention there will be a 24-week follow-up period.

The study is designed to achieve the following specific aims and obtain support for the following hypotheses:

Specific Aim 1 (primary mood outcome): To compare the efficacy of the experimental dietary intervention to the control diet in reducing variability of mood symptoms and reducing general psychological distress.

Hypothesis 1: Compared to the control diet, the experimental intervention will produce significant improvement in (1a) variability of daily mood symptoms, energy, and impulsivity using a Visual Analogue Scale measured using ecological momentary analysis (1); and (1b) psychological distress and general functioning using the PROMIS-29.

Specific Aim 2 (secondary mood outcome): To compare the efficacy of the experimental dietary intervention to the control diet in reducing recurrence.

Hypothesis 2: Compared to the control diet, the experimental intervention will produce significant reduction in recurrence of mood episodes over a 12-month period. To adequately power this study for measurement of recurrence, the investigators would need a sample size 2-3 times what the investigators estimate for this study, therefore the primary mood outcome is not recurrence of episode, but is measurement of mood variability, which in turn predicts recurrence. The investigators will measure recurrence to gather data for future studies.

Specific Aim 3 (primary biochemical outcome): To evaluate whether the experimental dietary intervention can alter n-6 AA and its bioactive metabolites, n-3 DHA and its bioactive metabolites in patients with BD.

Hypothesis 3: Compared to the control diet, the experimental intervention will significantly (1a) alter n-6 AA and n-3 DHA in erythrocytes; and (1b) alter 17-hydroxy DHA and reduce PGE2 in plasma.

Exploratory Aims: In an exploratory manner, the investigators will also (1) compare the efficacy of the experimental dietary intervention to the control diet in improving headache-related clinical endpoints in the subset of BP patients with comorbid migraines; (2) test the effects of the dietary interventions on a wide array of bioactive derivatives of n-3 and n-6 fatty acids and other inflammatory mediators (e.g. cytokines, CRP); and (3) evaluate whether biochemical alterations in n-3 DHA, n-6 AA and their bioactive metabolites are related to clinical improvements (4) compare the efficacy of the experimental dietary intervention to the control diet in preventing recurrence of illness in follow-up; (5) store samples for exploratory biomarker analysis.

Inclusion Criteria:

1. BD - history of at least one manic or mixed episode
2. Clinically significant hypomania or depression
3. Current psychiatric treatment
4. Over 18

Exclusion Criteria:

1. Current hospitalization for BD
2. Active substance dependence or eating disorder
3. Active suicidal/homicidal ideation
4. Pregnancy
5. Active treatment for major medical illness
6. History of specific food allergies such as, but not limited to, fish, gluten, dairy products
7. Strong aversion to fish
8. Any condition or attitude which, in the opinion of the PI, would prevent full cooperation and commitment to the study