Folate (folic acid)

What it is

Folate is vitamin B9, a nutrient the body needs to make DNA and RNA, to process amino acids and to attach methyl groups that help control gene activity [1][2]. "Folate" refers to the forms found naturally in foods, while folic acid is the synthetic form used in most supplements and added to fortified foods; folic acid fortification of cereal products has been mandatory in the United States since 1 January 1998 to reduce neural tube defects [3][1]. Supplements also come as 5-methyltetrahydrofolate (methylfolate) and folinic acid; folic acid is the form that has been shown to prevent neural tube defects in the large randomised trials [2]. Added folate is counted as more potent than food folate: regulators treat folic acid and methylfolate as 1.7 times natural food folate for intakes under 400 micrograms a day, and methylfolate as 2 times at higher intakes [4]. Folate deficiency, often together with vitamin B12 deficiency, causes megaloblastic anaemia - large, poorly formed red blood cells - and is treated by replacing the vitamin [5]. In the United States, women who are planning a pregnancy or could become pregnant are advised to take 0.4 to 0.8 mg (400 to 800 micrograms) of folic acid a day [6].

How it might work

Folate supplies one-carbon units for building DNA and for converting homocysteine into methionine, so blood homocysteine is used as a marker of how much folate and vitamin B12 the body is using [1]. Folic acid supplements reliably lower blood homocysteine - in a dose-finding trial of 2,163 adults, the reduction grew up to about 1.2 mg a day and then levelled off - and this biomarker finding was the rationale for large trials testing whether lowering homocysteine prevents heart attacks and strokes [7][8]. In the developing embryo, folate is needed for the rapid cell division and gene-regulating methylation that shape the neural tube, the proposed mechanism behind neural tube defect prevention [1][2].

What the research says

Neural tube defects (birth defects): this is the best-supported benefit. A Cochrane review of five randomised trials in 7,391 women found that folic acid around conception (400 micrograms a day or more, alone or with other vitamins) prevented neural tube defects, with a relative risk of 0.31 (95% CI 0.17 to 0.58; 6,708 births; high-quality evidence); among women who had already had an affected pregnancy the reduction was similar (RR 0.34, 95% CI 0.18 to 0.64) [9]. A 2025 umbrella review of ten systematic reviews (296,816 participants) found 57% lower pooled risk (RR 0.43, 95% CI 0.29 to 0.58; high heterogeneity), with a matching estimate when only randomised trials were pooled (RR 0.43, 95% CI 0.32 to 0.54) [10]. Newer observational evidence points the same way: a US review found folic acid started before pregnancy was linked to a 46% lower adjusted risk of an affected pregnancy (adjusted RR 0.54, 95% CI 0.31 to 0.91) [11]. Country-level data agree: spina bifida prevalence is 33.9 per 100,000 live births where fortification is mandatory versus 48.4 per 100,000 where it is voluntary [12].

Heart attacks and strokes: folic acid (usually with vitamins B6 and B12) lowered homocysteine in the large trials - by 27% in NORVIT and by 2.4 micromol per litre on average in HOPE-2 - yet heart attacks did not fall [13][14]. A Cochrane review of 15 randomised trials (71,422 people) found no effect of homocysteine-lowering treatment on heart attacks (RR 1.02, 95% CI 0.95 to 1.10; 12 trials, 46,699 people; high-quality evidence) or on death from any cause (RR 1.01, 95% CI 0.96 to 1.06) [8]. The three largest trials agree: heart attack risk was unchanged in NORVIT (RR 1.08, 95% CI 0.93 to 1.25 for its combined endpoint), HOPE-2 (RR 0.98, 95% CI 0.85 to 1.14) and the 20,702-person CSPPT (hazard ratio 1.04, 95% CI 0.60 to 1.82) [13][14][15]. Strokes behaved differently: Cochrane found about 10% fewer (RR 0.90, 95% CI 0.82 to 0.99; 10 trials, 44,224 people; high-quality evidence), and meta-analyses of 30 and 22 trials found the same 10-11% reduction [8][16][17]. But the benefit was concentrated where folate intakes are low: in the 22-trial analysis, stroke fell in Asian trials (RR 0.78, 95% CI 0.67 to 0.90) and did not fall in trials from the Americas (RR 1.05, 95% CI 0.90 to 1.23); trials using 0.8 mg a day or less showed the larger benefit (RR 0.78, 95% CI 0.69 to 0.88) [17]. CSPPT, the largest trial, gave Chinese adults with hypertension enalapril plus 0.8 mg folic acid or enalapril alone for a median 4.5 years: first strokes fell from 3.4% to 2.7% (hazard ratio 0.79, 95% CI 0.68 to 0.93), while heart attacks and deaths did not change [15].

Cancer: randomised trials do not show a clear overall increase, but there are signals and disagreements. A meta-analysis of 13 trials (49,406 people) found no significant effect of folic acid on total cancer incidence (RR 1.05, 95% CI 0.99 to 1.11), colorectal cancer (RR 1.01), prostate cancer (RR 1.17, 95% CI 0.84 to 1.62) or cancer deaths (RR 1.02) [18]. An earlier pooled analysis that weighted trials by how long supplements were taken found 21% more cancers in the supplemented groups (RR 1.21, 95% CI 1.05 to 1.39) [19]. A 2026 umbrella review of 67 syntheses concluded most evidence is weak: folic acid supplements showed no clear effect on total cancer incidence (RR 1.0, 95% CI 0.98 to 1.21), the few positive associations were all for prostate cancer, and observational data linked higher folic acid supplement intake with a lower risk of colorectal cancer (RR 0.88, 95% CI 0.81 to 0.95) [20]. Two possible harms come from specific trials. In 1,021 people with previous colorectal adenomas, 1 mg a day of folic acid did not reduce new adenomas (RR 1.13, 95% CI 0.93 to 1.37 in the second follow-up interval) and there were more advanced lesions (RR 1.67, 95% CI 1.00 to 2.80) and more people with three or more adenomas [21]; with longer treatment, sessile serrated polyps were more common (RR 1.94, 95% CI 1.02 to 3.68) [22]. In the same trial, prostate cancer was diagnosed in 9.7% of men on folic acid versus 3.3% on placebo over 10 years (hazard ratio 2.63, 95% CI 1.23 to 5.65) [23]. Against that, a trial in people with previous adenomas found no overall increase (RR 0.82, 95% CI 0.59 to 1.13) and a 39% lower recurrence in those who started with low folate (RR 0.61, 95% CI 0.42 to 0.90) [24], and a Chinese trial found that 1 mg a day halved first adenomas in people who had none before (14.9% versus 30.7%, RR 0.49, 95% CI 0.37 to 0.63) [25].

Depression: folate has been tested mostly as an add-on to antidepressants, with mixed results. A 2022 meta-analysis of L-methylfolate augmentation found better response than antidepressant alone (RR 1.25, 95% CI 1.08 to 1.46; three studies, 483 patients) and a small symptom reduction (standardised mean difference -0.38, 95% CI -0.59 to -0.17; four studies, 507 patients) [26]. Another meta-analysis of six trials found adjunctive methylfolate or folic acid lowered Hamilton depression scores (mean difference -2.16, 95% CI -3.62 to -0.69) and improved response (RR 1.36, 95% CI 1.16 to 1.59), with remission borderline significant (RR 1.39, 95% CI 1.00 to 1.92) [27]. A larger review of 11 randomised trials, however, found no short-term symptom benefit (SMD -0.12, 95% CI -0.45 to 0.22); it did find possible longer-term benefits - lower relapse (odds ratio 0.33, 95% CI 0.12 to 0.94 in one study) and fewer new cases in at-risk people (RR 0.65, 95% CI 0.43 to 0.98) [28]. A review built for guideline use found benefit only when trials were restricted to folic acid under 5 mg a day or 15 mg methylfolate once daily added to an SSRI, and rated all the evidence low or very low quality [29]. Several authors of the methylfolate meta-analysis were linked to companies connected to MTHFR genetic testing [26].

Doses used in studies

Neural tube defects: the trials used 400 micrograms a day or more, and the preventive effect did not clearly differ across that dose range; US guidance for women who could become pregnant is 0.4 to 0.8 mg a day [9][6]. Homocysteine lowering: in a dose-finding trial of 2,163 hypertensive adults, 8 weeks of folic acid lowered homocysteine up to about 1.2 mg a day, with little added benefit between 1.2 and 1.6 mg; carriers of the MTHFR TT genotype responded more [7]. Heart and stroke trials: CSPPT used 0.8 mg a day with enalapril for a median 4.5 years; HOPE-2 used 2.5 mg a day with 1 mg vitamin B12 and 50 mg vitamin B6 for about 5 years; NORVIT used 0.8 mg folic acid plus 0.4 mg vitamin B12, with or without 40 mg vitamin B6, for a median 40 months [15][14][13]. Cancer trials: 1 mg a day for 3 years or longer [21][25]. Depression trials: usually 15 mg methylfolate or folic acid below 5 mg a day added to an antidepressant [29]. Safety studies: unmetabolised folic acid has been measured after 400 micrograms a day for about 3 weeks and after 5 mg a day for 90 days [30][31].

Safety and interactions

Side effects: folic acid was generally well tolerated in the large trials. In CSPPT (20,702 people), adverse events did not differ between the folic acid and no-folic-acid groups [15]. Pooled across eight B-vitamin trials (35,788 people), serious adverse events were slightly more common with treatment than control (RR 1.07, 95% CI 1.00 to 1.14), with high-quality evidence [8]. Two older B-vitamin trials reported possible harm signals - a trend to more events with combined B vitamins after a heart attack (RR 1.22, 95% CI 1.00 to 1.50) and more admissions for unstable angina in HOPE-2 (RR 1.24, 95% CI 1.04 to 1.49) [13][14]. In depression add-on trials, no major side effects were reported apart from minor digestive upset [32].

Upper limit: the accepted upper level for synthetic folic acid is 1 mg (1,000 micrograms) a day; a 2020 review of the masking literature supports keeping that limit, and a 2026 review of folate forms advises monitoring for people taking more than 1,000 micrograms a day [33][2].

Vitamin B12 deficiency: high-dose folic acid can correct the anaemia of vitamin B12 deficiency while the nerve damage goes untreated, delaying diagnosis. A 2024 review of this "masking" question says the evidence is correlational and the proposed mechanism untested, but advises vigilance in older adults and anyone with low B12 intake or poor absorption [34]. A 2020 review argues the concern is overstated and that the 1 mg a day limit is safe [33]. For people at risk of B12 deficiency, assessing B12 status is advised before using high-dose folic acid [34].

Unmetabolised folic acid: at higher intakes, some folic acid circulates unmetabolised - 5 mg a day for 90 days raised it sharply in a study of 30 healthy adults (detectable in 96.6% after 45 days), and 400 micrograms a day for about three weeks produced detectable levels in most older participants [31][30]. What this means for health is not established; recent reviews call for research and for agreed upper limits for blood folate [20][35].

Medicines: folic acid interacts with methotrexate, a medicine that blocks folate metabolism. In six trials of people with rheumatoid arthritis taking low-dose methotrexate, adding folic or folinic acid reduced stomach and bowel side effects by 26% (RR 0.74, 95% CI 0.59 to 0.92), reduced abnormal liver enzyme rises by 77% (RR 0.23, 95% CI 0.15 to 0.34) and meant more people stayed on treatment (RR 0.39, 95% CI 0.28 to 0.53), without reducing the drug's effectiveness [36]. People taking methotrexate should ask their doctor before adding folic acid [36].

Pregnancy and breastfeeding: for women planning pregnancy or who could become pregnant, 0.4 to 0.8 mg of folic acid a day is recommended, because the benefit is substantial and harms at usual doses are small [6]. A US review of about 1.24 million people found no statistically significant harms from pregnancy folic acid supplementation for autism, twin pregnancy or maternal cancer [11]. A 2026 review of observational studies links very high folic acid intake or high blood folate in pregnancy with gestational diabetes, larger babies and neurodevelopmental concerns, but notes that "excess" is not clearly defined and calls for safe upper limits [35]. Folic acid is the only folate form that has been tested and shown to prevent neural tube defects in large trials [2].

Kidney problems: in people with impaired kidney function, B-vitamin treatment including high-dose cyanocobalamin did not prevent strokes (RR 1.04, 95% CI 0.84 to 1.27) in a pooled analysis of two large trials, and the authors note cyanocobalamin may accelerate kidney decline; benefit was seen only in people with normal kidney function (RR 0.78, 95% CI 0.67 to 0.90) [37].

Caveats and open questions

The pregnancy evidence rests on old trials: the Cochrane review found variable risk of bias, with only one of five trials at low risk, although it still rated the evidence high quality; the newest support is observational [9][11]. In the US review, the lower neural tube defect risk was seen in only one of two study periods, so it is not a clean story [11]. The stroke benefit is the biggest open question: it was concentrated in low-folate regions and absent in populations such as the Americas where diets already contain more folate, so whether extra folic acid helps people who already get enough is unresolved [17][8]. Two large recruiting trials in China, listed on this page, are testing folic acid together with blood-pressure medicines. Cancer remains unsettled: pooled analyses disagree, and the signals (prostate cancer, advanced adenomas and sessile serrated polyps in colon surveillance) come from single trials or subgroups while overall cancer incidence was not increased; the 2026 umbrella review rates most evidence weak and calls for monitoring [18][19][21][23][20]. Depression evidence is small and imperfect: trials are tiny, the methylfolate review had conflicts of interest, and evidence quality is low, so add-on folate is not established for treating depression [26][29][28]. The vitamin B12 masking question is still debated, and the health significance of unmetabolised folic acid is unknown [34][33][2].