Does choline affect anxiety and depression?

Nobody has tested it for anxiety. We found no randomized trial that gave people choline, from food or from a supplement, with anxiety as its outcome. What exists is observational. In 5,918 Norwegian adults, those in the lowest fifth of blood choline had 33 percent higher odds of high anxiety scores, with no link to depression. In 12,906 US adults, the fifth eating the most choline had 43 percent lower odds of depressive symptoms than the fifth eating the least. Neither study can say the choline did it. The only trials are of citicoline, a drug-like choline donor, given alongside treatment for depression, and they are small.

Unsettled. The surveys and the trials point at different things. The surveys measure choline in blood or in a diet recall and find patterns. The trials test citicoline in people already being treated for depression. Nothing connects eating more choline to feeling less anxious, because that step has never been tested.

What the trials found

The strongest evidence on this page is a 2026 network meta-analysis of 163 randomized trials, in 15,757 adults with major depressive disorder, comparing supplements and plant products with placebo or antidepressants. Citicoline was associated with significantly more dropouts than placebo, and it was not among the compounds that kept an effect after quality filters. The reviewers rated confidence in all of the effects low to very low, and the larger effects came from the poorer quality trials. Anxiety was a secondary outcome in that review, and only curcumin showed an effect on it. We read the abstract only, so the size of the citicoline dropout excess is unknown to us.

Three single trials sit underneath that. In 50 people with major depression taking citalopram, adding citicoline for 6 weeks improved depression scores more than placebo at weeks 2, 4 and 6, and more people reached remission. It was small, short and run at one center, and it did not report anxiety. In 60 adults with depression and methamphetamine dependence, citicoline for 12 weeks improved depression scores more than placebo, in a narrow group that tells us little about anyone else. In a very small trial in lithium-treated rapid-cycling bipolar disorder, choline bitartrate for 12 weeks did not change depression or mania scores against placebo.

The surveys, read carefully

The Norwegian study found low blood choline linked to anxiety and not to depression. The authors measured choline once, and blood choline reflects diet only in part. An Iranian study of 533 adults found the reverse split: the highest choline intake went with lower odds of depression, an odds ratio of 0.51, while the link with anxiety did not survive adjustment. Only 5.1 percent of that group had anxiety, which leaves little power to find anything. In US adults the top fifth of intake had an odds ratio of 0.57 for depressive symptoms, from single-day diet recalls, and choline intake tracks eggs, meat and overall diet quality.

Brain scans add a hint. A 2025 meta-analysis of six prefrontal datasets, 78 patients with anxiety disorders and 89 healthy controls, found choline compounds lower in the patients, with a standardized difference of minus 0.46. That measures brain chemistry and says nothing about what anyone ate. The authors themselves wrote that whether choline supplements could help anxiety is a question for future studies.

Two groups where the pattern changes

In 949 pregnant women in Singapore, higher blood choline went with slightly more anxiety and depressive symptoms, the opposite direction to the Norwegian adults. Pregnancy changes how the body handles choline, so the blood level may not mirror intake. After an ischemic stroke, 612 patients in the top third of blood choline had about half the odds of depression at 3 months, an odds ratio of 0.54. That is a blood marker after a brain injury, not a test of eating more.

Who should be careful

Not for everyone. Choline has a real upper limit. For adults it is 3,500 mg a day, and very high intakes cause a fishy body odor, vomiting, heavy sweating and salivation, low blood pressure and liver toxicity, according to the US National Institutes of Health. The limit does not apply to people taking high doses under medical supervision.

Citicoline is the form with mood trials behind it, and it is also the form that led to more dropouts than placebo in the 2026 network meta-analysis. The two positive citicoline trials gave it alongside an antidepressant or in a specialist setting. If you are being treated for depression or anxiety, adding a choline product is a decision for the person treating you. The pregnancy data above run the opposite way to the adult surveys, and no trial has tested choline for mood in pregnancy.

What expert bodies say

The NIH Office of Dietary Supplements describes choline as needed to make acetylcholine, a messenger for memory, mood and muscle control. That is the basis of the mood idea, and the fact sheet makes no recommendation for anxiety or depression. In Europe, choline may be advertised only for the liver, fat metabolism and homocysteine. The memory claims put forward for lecithin and for citicoline were both refused. We found no agency position on choline for anxiety or mood.

How we searched

What would change this answer

• A randomized trial of choline, from food or a standard supplement, with anxiety as the main outcome. None exists, and the local registry index holds no active trial of choline or citicoline for anxiety or depression.
• A systematic review of choline intake and anxiety. We found none.
• Follow-up of people over time, measuring what they eat, rather than single blood tests or single-day diet recalls.
• The full results of the 2026 network meta-analysis for citicoline, which would show how large the dropout problem is.

For how much choline you need, where it comes from and who runs short, see the choline guide.

Sources

1. ev-chax-01 · Meta-analysis or systematic review · systematic review and network meta-analysis of randomized clinical trials, CINeMA · n = 163 studies, 15757 participants
   Citicoline was associated with significantly more dropouts than placebo.
   
   Who: adults with major depressive disorder in randomized trials of nutraceuticals and phytoceuticals versus placebo or antidepressants
   Effect: citicoline: significantly more all-cause dropouts than placebo; only EPA, vitamin D3 and St John's wort extract ZE117 kept an effect after quality filters; confidence low to very low
   Certainty: Abstract only (not open access); the size of the citicoline dropout excess and the number of citicoline trials are not in the abstract. Anxiety was a secondary outcome; only curcumin showed an anxiety effect.
   Molecular Psychiatry, 2026 (Fornaro et al., Nutraceuticals and phytoceuticals for major depressive disorder: network meta-analysis) · checked 2026-10-07 · we read the abstract
2. ev-chax-02 · Meta-analysis or systematic review · systematic review and network meta-analysis of randomized clinical trials · n = 163 studies, 15757 participants
   A meta-regression of efficacy effect sizes against adapted AMSTAR-Plus scores showed that larger SMDs were associated with lower AMSTAR scores, indicating lower study quality. Confidence in the evidence was low to very low.
   
   Who: adults with major depressive disorder in randomized trials
   Effect: larger SMDs associated with lower AMSTAR scores; confidence low to very low
   Certainty: Frames the small positive citicoline add-on trials below.
   Molecular Psychiatry, 2026 (Fornaro et al., Nutraceuticals and phytoceuticals for major depressive disorder: network meta-analysis) · checked 2026-10-07 · we read the abstract
3. ev-chax-03 · Observational data · cross-sectional population study · n = 5918
   The lowest choline quintile was significantly associated with high anxiety levels (odds ratio: 1.33; 95% CI: 1.06, 1.69) in the fully adjusted (age group, sex, time since last meal, educational level, and smoking habits) logistic regression model.
   
   Who: general population, Hordaland Health Study, aged 46-49 and 70-74 years
   Effect: lowest vs higher choline quintiles, high anxiety (HADS-A >= 8): OR 1.33 (95% CI 1.06 to 1.69); depression: no significant association
   Certainty: Plasma choline reflects diet only in part; single measurement; reverse causation not excluded.
   Am J Clin Nutr, 2009 · checked 2026-10-07 · we read the abstract
4. ev-chax-04 · Observational data · cross-sectional population study · n = 5918
   In this large population-based study, choline concentrations were negatively associated with anxiety symptoms but not with depression symptoms.
   
   Who: general population, Hordaland Health Study
   Effect: inverse association with anxiety, none with depression
   Certainty: Observational; the same question has not been tested in a trial.
   Am J Clin Nutr, 2009 · checked 2026-10-07 · we read the abstract
5. ev-chax-05 · Observational data · systematic review and meta-analysis of 1H-MRS case-control studies · n = 78 patients, 89 controls (6 datasets)
   Meta-analysis of six prefrontal cortex (PFC) datasets (78 Pts, 89 HCs) [31, 42, 58, 59, 61] showed significantly reduced tCho in AnxDs compared to HCs (g = −0.46, 95% CI = −0.77 to −0.15, p = 0.003) (Fig. 1a).
   
   Who: patients with social anxiety, generalized anxiety or panic disorder vs healthy controls, prefrontal datasets
   Effect: prefrontal tCho Hedges g -0.46 (95% CI -0.77 to -0.15), p 0.003; anterior cingulate no difference (g +0.01)
   Certainty: Measures brain choline compounds, not dietary intake; no link to what patients ate was tested.
   Mol Psychiatry, 2025 · checked 2026-10-07 · we read the fulltext
6. ev-chax-06 · Observational data · systematic review and meta-analysis of 1H-MRS studies · n = 370 patients, 342 controls (25 datasets)
   Future studies may clarify the clinical significance of reduced cortical tCho and the possibility that appropriate choline supplementation could have therapeutic benefit in anxiety disorders.
   
   Who: patients with anxiety disorders vs healthy controls
   Effect: authors leave the clinical significance of reduced cortical tCho and any benefit of choline supplementation to future studies
   Certainty: A hypothesis for trials, not evidence that eating more choline changes anxiety.
   Mol Psychiatry, 2025 · checked 2026-10-07 · we read the abstract
7. ev-chax-07 · Observational data · cross-sectional analysis of a national survey · n = 12906
   After adjusted all selected confounding factors and covariates, the odds ratio with the 95% confidence interval of depressive symptoms was 0.57 (95% CI:0.38-0.85) for the highest quintile versus the lowest quintile of dietary choline intake.
   
   Who: US adults aged 20 and over, NHANES 2011-2018, PHQ-9 >= 10
   Effect: highest vs lowest quintile OR 0.57 (95% CI 0.38 to 0.85); L-shaped dose-response
   Certainty: Single-day dietary recalls; choline tracks eggs, meat and overall diet quality.
   J Affect Disord, 2022 · checked 2026-10-07 · we read the abstract
8. ev-chax-08 · Observational data · cross-sectional study with food frequency questionnaire · n = 533
   After considering all confounders, such associations remained significant in the case of depression (OR = 0·51; 95 % CI: 0·26, 0·98) but not for anxiety and distress.
   
   Who: middle-aged Iranian adults, HADS and GHQ
   Effect: depression OR 0.51 (95% CI 0.26 to 0.98) adjusted; anxiety crude OR 0.38 (0.14 to 0.99), not significant after adjustment; betaine null
   Certainty: Only 5.1% had anxiety, so power for anxiety was low.
   Br J Nutr, 2025 · checked 2026-10-07 · we read the abstract
9. ev-chax-09 · Observational data · prospective mother-offspring cohort · n = 949
   Plasma choline concentrations were positively associated with antenatal depressive (β = .24 EPDS score [95% CI: 0.05-0.43] per μmol/L] and anxiety symptoms (β = .46 STAI-state score [95% CI: 0.03-0.88] per μmol/L) adjusting for covariates.
   
   Who: pregnant women at 26-28 weeks, GUSTO cohort
   Effect: per umol/L plasma choline: EPDS +0.24 (95% CI 0.05 to 0.43); STAI-state +0.46 (0.03 to 0.88)
   Certainty: Pregnancy changes choline metabolism; plasma level may not mirror intake.
   Depress Anxiety, 2017 · checked 2026-10-07 · we read the abstract
10. ev-chax-10 · Observational data · prospective analysis within a randomized trial cohort · n = 612
    Compared with tertile 1, the multivariable-adjusted odds ratios (95% CIs) for tertile 3 of choline and betaine were 0.54 (0.35-0.83) and 0.59 (0.38-0.92), respectively.
    
    Who: patients with acute ischemic stroke, CATIS trial
    Effect: tertile 3 vs 1: choline OR 0.54 (95% CI 0.35 to 0.83); betaine OR 0.59 (0.38 to 0.92)
    Certainty: Plasma marker, not intake or supplementation.
    Eur J Neurol, 2022 · checked 2026-10-07 · we read the abstract
11. ev-chax-11 · Randomized controlled trial(s) · double-blind randomized placebo-controlled trial, 6 weeks · n = 50
    Significantly greater improvement was observed in the HDRS scores of the citicoline group compared with the placebo group from baseline to weeks 2, 4, and 6 (Ps = 0.030, 0.032, and 0.021, respectively).
    
    Who: patients with major depressive disorder treated with citalopram
    Effect: greater HDRS improvement at weeks 2, 4 and 6 (P 0.030, 0.032, 0.021); higher remission rate (P 0.045)
    Certainty: Small, short, single center; citicoline is a drug-like choline donor, not dietary choline. Anxiety not reported.
    Clin Neuropharmacol, 2017 · checked 2026-10-07 · we read the abstract
12. ev-chax-12 · Randomized controlled trial(s) · double-blind randomized placebo-controlled trial, 12 weeks · n = 60
    An ANCOVA of the intent-to-treat sample showed that those receiving citicoline (n=28) had a statistically significantly greater improvement in IDS-C scores than those receiving placebo (n=20).
    
    Who: adults with bipolar or unipolar depression and methamphetamine dependence
    Effect: greater improvement in IDS-C scores (ITT, citicoline n 28 vs placebo n 20); no difference in memory or drug use
    Certainty: Proof-of-concept trial in a narrow dual-diagnosis group.
    J Affect Disord, 2012 · checked 2026-10-07 · we read the abstract
13. ev-chax-13 · Randomized controlled trial(s) · double-blind randomized placebo-controlled trial with MRS · n = 8
    There were no significant differences in change-from-baseline measures of CGIS, YMRS, and HDRS, brain choline/creatine ratios, and brain lithium levels over a 12-week assessment period between the choline and placebo groups or within each group.
    
    Who: lithium-treated patients with rapid-cycling bipolar disorder
    Effect: no significant differences in CGIS, YMRS or HDRS change over 12 weeks
    Certainty: Very small; designed for brain chemistry, not powered for mood.
    Bipolar Disord, 2003 · checked 2026-10-07 · we read the abstract
14. ev-chax-14 · Position of an expert body · agency fact sheet
    High intakes of choline are associated with a fishy body odor, vomiting, excessive sweating and salivation, hypotension, and liver toxicity [1,2].
    
    Who: healthy children and adults
    Effect: UL for adults 3500 mg/day (ODS Table 3, table not quoted); the 2000 mg/day citicoline trial dose contains less choline than that
    Certainty: ULs do not apply to people taking high doses under medical supervision. The ODS sheet does not discuss anxiety or depression.
    NIH Office of Dietary Supplements, Choline Fact Sheet for Health Professionals · checked 2026-10-07 · we read the section
15. ev-chax-15 · Position of an expert body · agency fact sheet
    In addition, choline is needed to produce acetylcholine, an important neurotransmitter for memory, mood, muscle control, and other brain and nervous system functions.
    
    Who: general
    Effect: physiological role only; no recommendation for mood or anxiety
    Certainty: A role statement, not evidence that extra choline improves mood.
    NIH Office of Dietary Supplements, Choline Fact Sheet for Health Professionals · checked 2026-10-07 · we read the section
16. chol-09 · Position of an expert body
    High intakes of choline are associated with a fishy body odor, vomiting, excessive sweating and salivation, hypotension, and liver toxicity.
    
    Who: adults
    Effect: UL 3,500 mg/day
    NIH Office of Dietary Supplements, Health Professional Fact Sheet: Choline · checked 2026-09-15
17. ch-d-12 · Position of an expert body · entries in the EU Register on nutrition and health claims, authorized list · n = 3 authorized choline claims against 25 non-authorized ones, which is the widest refusal ratio of the five nutrients in this run
    POL-HC-6354 Choline Choline contributes to the maintenance of normal liver function Authorised
    
    Who: general population, EU food labeling
    Effect: claim POL-HC-6354, substance Choline, wording 'Choline contributes to the maintenance of normal liver function', status Authorised; the other two cover normal homocysteine metabolism and normal lipid metabolism
    Certainty: the three survivors map exactly onto the derivation described in card ch-d-06. The recommendation was set from liver damage in depleted adults, choline works as a methyl donor alongside folate which is the homocysteine link, and phosphatidylcholine is needed to export fat from the liver. In other words Europe authorized the claims that follow from the established biochemistry and refused the rest. The register carries no conditions of use, so the amount a food must contain comes from Regulation (EU) No 432/2012 on EUR-Lex before any of this reaches a label or a page.
    EU Register on nutrition and health claims (European Commission) · checked 2026-09-21 · we read the dataset
18. ch-d-14 · Position of an expert body · entries in the EU Register on nutrition and health claims, non-authorized list, for two commercial choline-bearing substances · n = 2 of the 25 non-authorized choline-related claims, covering the two forms most commonly sold for memory
    POL-HC-7500 Lecithin Lecithin (a phospholipid containing choline) supports memory and concentration. Non-authorised
    
    Who: EU food supplement labeling
    Effect: claim POL-HC-7500, substance Lecithin, wording 'Lecithin (a phospholipid containing choline) supports memory and concentration.', status Non-authorized; claim POL-HC-14782, substance Citicoline (CDP-Choline) inner salt, wording 'Citicoline supports improvement, maintenance or reduced loss of memory in healthy middle-aged or elderly adults encountering age-associated subjective memory impairment', status Non-authorized
    Certainty: these two refusals and the Cochrane review in card ch-d-01 come from completely independent processes and point the same way for citicoline. The citicoline claim is notable for how narrowly it was drafted, naming healthy middle-aged and elderly adults with subjective memory complaints, and it was still refused. As always, refused means the evidence did not reach the bar, not that the effect is disproved, and the registry does not carry the reasoning. The United States authorizes no health claim for choline either, though FDA does permit a nutrient content claim notification for choline-containing foods, which is a different instrument about how much is present rather than what it does.
    EU Register on nutrition and health claims (European Commission) · checked 2026-09-21 · we read the dataset